The main walkthrough uses a desktop browser on Windows or macOS. You don’t need to install desktop software. If you use an iPhone, iPad, or Android device, you’ll find app setup instructions below, too.

What Is Google Sheets?

Google Sheets is an online spreadsheet for organizing information in rows and columns. It helps with budgets, class lists, project plans, and any other job where a list needs a little math. You can open it at sheets.google.com or create a sheet from Google Drive.

• Cells hold individual pieces of information, such as a date or price.
• Formulas calculate answers from those cells.
• Formatting and dropdowns make a sheet easier to read and fill in.
• Filters and pivot tables help you find and summarize rows.
• Sharing and comments let other people work with you.

If you’ve used a paper expense log, think of Sheets as one that can do the arithmetic and keep everyone looking at the same page.

Prerequisites

Make sure you have:

• A Windows PC or Mac with a web browser and internet access for the main walkthrough.
• A Google account you can sign in to. You’ll need it to create and share the example sheet.
• About 20 minutes to build the example.
• An iPhone, iPad, or Android device only if you also want the mobile app.

Google Sheets changes continuously, so a button may move slightly. The desktop instructions here follow the browser interface rather than a numbered app version. Google’s getting-started guide is a useful reference if your screen looks different.

Step-by-Step Guide

We’ll build a spending tracker with five purchases. Keep the same sheet open as you go: each new feature will use the data from the previous step.

Step 1: Open Google Sheets

On Windows

Open your browser, click the address bar, type https://sheets.google.com, and press Enter. Sign in to your Google account if asked. You should arrive at the Sheets home page, where you can create or open spreadsheets.

On macOS

Open your browser, click the address bar, type https://sheets.google.com, and press Return. Sign in if asked. You should see the same Sheets home page.

On the web

Already in a browser? Go straight to sheets.google.com. You can also open Google Drive and click New > Google Sheets.

Expected result: You can see an option to start a blank spreadsheet.

Step 2: Create and name your spreadsheet

Click Blank spreadsheet. When the grid opens, click the name at the top left (usually Untitled spreadsheet), type Household Spending Practice, and press Enter or Return.

Expected result: The new name appears at the top of the page. Your spreadsheet is ready for data.

Step 3: Find the parts you’ll use

Click cell A1 in the grid. The box near the upper left shows its cell address: column A, row 1. The long entry area above the grid is where you can inspect or edit a cell’s contents. The toolbar above it holds common formatting buttons. At the bottom, the Sheet1 tab represents one page within this spreadsheet.

Imagine a filing cabinet: the spreadsheet is the cabinet, a sheet tab is a drawer, and each cell is a labeled spot inside it.

Expected result: Clicking another cell changes the address shown near the entry area.

Step 4: Enter the column headings

Click A1, type Date, and press Tab to move right. Continue across row 1 with these headings:

Expected result: Row 1 describes what belongs in each column. The Total and Over $10? columns will get formulas shortly.

Step 5: Enter the example purchases

Click A2 and fill rows 2 through 6 using this table. Press Tab to move across a row and Enter or Return to finish an entry. Leave columns F and H empty for now.

Enter dates in the 2026-10-01 pattern. If Sheets displays them in another date style, that’s fine; date display depends on the sheet’s locale.

Expected result: You have five purchase rows. Words appear in the category, item, and status columns; quantities and prices appear as numbers.

Step 6: Make the headings and numbers easier to read

Select A1:H1, then click Bold on the toolbar. Next, select E2:F6 and click the Format as currency button on the toolbar. If your dates appear as plain text, select A2:A6, then click Format > Number > Date.

The currency symbol may differ with your locale. That’s a display choice; it doesn’t change the arithmetic.

Expected result: The header stands out, prices look like money, and dates are easy to scan.

Step 7: Calculate the first row’s total

Click F2, enter this formula, and press Enter or Return:

=D2*E2

The = tells Sheets to calculate. D2 points to the quantity of apples, E2 points to their unit price, and * means multiply. Cell references are like street addresses: if you change the value at an address, the formula uses the new value.

Expected result: F2 shows 6 (or the currency-formatted equivalent).

Step 8: Fill the total formula down the column

Click F2 again. Point to the small square at the bottom-right corner of its selection border, then drag it down to F6. Sheets adjusts the row references as it fills.

Expected result: F2:F6 shows 6, 30, 8, 12, and 10. For example, F3 calculates =D3*E3 for the bus pass.

Step 9: Add a small budget reference table

Click I1 and enter Category. Click J1 and enter Budget. Then fill in the three rows below:

Select J2:J4 and use Format as currency. This little table gives the lookup formula in Step 13 something useful to find.

Expected result: A category-and-budget table appears to the right of your purchases.

Step 10: Add a grand total with SUM

Click I6, type Grand total, then click J6 and enter:

=SUM(F2:F6)

F2:F6 means every cell from F2 through F6. SUM adds those values together.

Expected result: J6 shows 66. Google’s formula guide explains how formulas begin with = and use cell references.

Step 11: Find the average purchase total

Click I7, type Average purchase, then click J7 and enter:

=AVERAGE(F2:F6)

AVERAGE adds the five totals and divides by five. It’s handy when you want a typical purchase amount rather than the whole bill.

Expected result: J7 shows 13.2. Format J6:J7 as currency if you want the summary to match the purchase totals.

Step 12: Flag purchases above $10 with IF

Click H2, enter this formula, and press Enter or Return:

=IF(F2>10,"Yes","No")

This reads like a question: “Is the total in F2 greater than 10?” If yes, show Yes; otherwise show No. Drag the fill handle from H2 down to H6.

Expected result: The bus pass and light bulbs show Yes. The other rows show No. If you want to check a function’s spelling or arguments, use Google’s Sheets function list.

Step 13: Look up each category’s budget

Click K1 and type Category Budget. Click K2, enter the formula below, and fill it down to K6:

=VLOOKUP(B2,$I$2:$J$4,2,FALSE)

VLOOKUP searches the first column of the budget table for the category in B2. It returns the matching value from column 2. FALSE asks for an exact match. The dollar signs keep the budget table fixed when you fill the formula down (otherwise the lookup area would slide down, too).

Expected result: Grocery rows show 20, Transport rows show 50, and the Home row shows 25. Format K2:K6 as currency if you like.

Prefer XLOOKUP? Its equivalent is shown below. Check the current function list for syntax and availability in your Sheets account before using it:

=XLOOKUP(B2,$I$2:$I$4,$J$2:$J$4,"Not found")

Step 14: Highlight larger purchases

Select F2:F6, then click Format > Conditional formatting. In the panel, choose Greater than, enter 10, pick a fill color, and click Done.

Conditional formatting is a highlighter that updates itself. If a total later changes from 8 to 18, its color follows the rule.

Expected result: The 30 and 12 totals are highlighted.

Step 15: Add a dropdown for purchase status

Select G2:G6, then click Insert > Dropdown. In the dropdown panel, set two options: Planned and Paid. Click Done. If Sheets asks how to handle values already in the cells, make sure those existing words match the options exactly.

A dropdown is useful when “Paid,” “paid,” and “Done” would otherwise become three names for the same thing.

Expected result: Clicking a cell in G2:G6 shows a choice between Planned and Paid.

Step 16: Keep the headings visible while scrolling

Click View > Freeze > 1 row. This pins the header row at the top while you scroll through a longer list.

Expected result: Row 1 remains visible as you scroll down. With our tiny example, you may need to add more rows before the benefit is obvious, but the setting is ready.

Step 17: Turn on filters

Select A1:H6, then click Data > Create a filter. Small filter buttons should appear in the header row. Click the filter button in B1, clear the categories you don’t want, leave Grocery selected, and click OK.

Filtering is like asking the sheet to show only one shelf of the pantry. It hides other rows from view without deleting them.

Expected result: You see the apples and bread rows. To see all purchases again, open the same filter and select all categories.

Step 18: Sort the purchases by date

First, restore all categories in the filter so every purchase is visible. Then click the filter button in A1 and choose Sort A to Z. For dates, that puts earlier purchases first.

Sorting through the filter button keeps each purchase row together. That’s the important bit: sorting just one column on its own can mix up items and prices.

Expected result: The rows run from 2026-10-01 through 2026-10-05.

Step 19: Summarize spending with a pivot table

Select A1:H6 and click Insert > Pivot table. Choose New sheet, then click Create. In the pivot table editor, add Category under Rows and Total under Values. If the value is summarized another way, change its summary setting to SUM.

A pivot table groups the source rows for you. It answers “How much did each category cost?” without making you write a separate formula for each one.

Expected result: The pivot table shows Grocery: 14, Transport: 40, and Home: 12, plus a grand total of 66. Click the original sheet tab at the bottom to return to your purchase list.

Step 20: Share the spreadsheet

Click Share at the top right. Enter the email address of someone you want to work with. Choose Viewer if they only need to read, Commenter if they should leave notes, or Editor if they should change cells. Click Send.

For a household budget, I would start with the narrowest access the person needs. You can change their access later.

Keep in mind that sharing the spreadsheet shares the entire file, not just the budget rows you happen to be viewing — anyone you add can open every sheet and row in the workbook. Editor access also lets that person change or delete any content, including formulas and other people’s entries, so reserve Editor for people you’d trust with the whole file.

Expected result: The person appears in the sharing settings with the access you chose. They can open the same spreadsheet from their invitation.

Step 21: Leave a comment on a purchase

Click F3, the bus pass total, then click Insert > Comment. Type a note such as “Check whether this pass covers the whole month,” and click Comment.

A comment attaches a conversation to a particular cell. That’s less confusing than sending a message that says “What is this 30?” when there are several 30s on the sheet.

Expected result: A comment indicator appears on F3, and people with access can read the note.

Step 22: Inspect an earlier version

Click File > Version history > See version history. Select an earlier entry in the panel to inspect what the spreadsheet looked like at that point. Return to the current view when you’re done.

Version history is the spreadsheet’s rewind button. It’s especially useful after several people have edited the same file.

Expected result: You can see saved versions and inspect earlier changes. The list may be short if you just created the file.

Set Up Google Sheets on Each Platform

The spreadsheet you made in the browser is tied to your Google account. Sign in with that same account elsewhere to open it.

On Windows

Open a browser and visit sheets.google.com. No separate desktop installation is needed for this walkthrough. To return quickly, bookmark the page with Ctrl + D.

Expected result: Household Spending Practice appears in your Sheets list when you’re signed in to the same account.

On macOS

Open a browser and visit sheets.google.com. Bookmark it with ⌘ + D if you’ll use it often.

Expected result: Your practice spreadsheet appears in your Sheets list.

On iOS

On an iPhone or iPad, open the App Store, search for Google Sheets, and install the app published by Google. Open it, sign in with the Google account you used above, and tap Household Spending Practice.

The mobile layout differs from the desktop browser walkthrough. Use the desktop view for the menu paths in the numbered steps.

Expected result: The practice spreadsheet opens in the iPhone or iPad app.

On Android

Open the Play Store, search for Google Sheets, and install the app published by Google if it isn’t already on your device. Open it, choose the same Google account, and tap Household Spending Practice.

Expected result: The practice spreadsheet opens in the Android app.

On the web

On another computer, open sheets.google.com and sign in to the same account. Click Household Spending Practice from the file list.

Expected result: You can continue with the same data and formulas in your browser.

Configuration: Settings Worth Knowing

You can use Sheets happily without changing every setting. These are the few controls that solve common frustrations.

• Formula suggestions: In an open desktop spreadsheet, click Tools > Suggestion controls and inspect Enable formula suggestions. Select it if you want Sheets to offer formula ideas while you type. The control may already be on or off.
• Formula corrections: Click Tools > Autocomplete and inspect Enable formula corrections if Sheets isn’t pointing out likely formula mistakes.
• Date and currency display: Select the relevant cells, then use Format > Number to choose Date or a money format. If dates still look wrong, check the spreadsheet’s locale under File > Settings.
• Offline access in Chrome: Install the Google Docs Offline extension. In Google Drive, click Settings > Settings, find the Offline section, enable the option to create, open, and edit recent Docs, Sheets, and Slides files offline, then click Done. While you’re still online, right-click the spreadsheet in Drive and turn on Available offline. Google’s offline help explains the setup.

Advanced features to verify before using

You may see references to a manual calculation setting or a pivot-table calculated-field editor. The supplied official research for this guide does not establish their current controls, availability, or behavior. Check current Google Sheets documentation and the menus in your own account before following instructions for either feature. The regular pivot table in Step 19 needs neither one.

Tips and Troubleshooting

Problem: A formula shows an error instead of a number

• Click the cell and check that the formula starts with =.
• Check the referenced cells. For example, =D2*E2 needs numbers in both D2 and E2.
• If a lookup fails, make sure the category text matches the budget table exactly and that the lookup range includes both columns.
• Open the function list to check the function’s arguments.

Problem: Formula suggestions or corrections are missing

• Open Tools > Suggestion controls and inspect Enable formula suggestions.
• Open Tools > Autocomplete and inspect Enable formula corrections.
• Select the control you want. These are aids, so you can still type formulas directly when they don’t appear.

Problem: The spreadsheet won’t open offline

• Confirm you’re using Chrome on the computer and have installed Google Docs Offline.
• In Drive’s Settings, check the Offline option.
• Right-click the file in Drive and check Available offline while you still have internet access. Offline setup is much less useful if you wait until the connection is already gone (ask anyone who’s tried to download a boarding pass in a dead zone).

Problem: You can’t find a menu command

• Open Tool finder with Alt + / on Windows or Option + / on Mac.
• Type the command you want, such as “conditional formatting.”
• Choose the matching result. Google’s keyboard shortcut reference is useful when a shortcut differs for your keyboard or language.

Problem: A shortcut doesn’t work

• Open the shortcut list with Ctrl + / on Windows or ⌘ + / on Mac.
• Look up the action there. Some shortcuts vary by keyboard and language.

Quick shortcuts worth keeping:

Tip: Practice by changing one value. Change the apple quantity in D2 from 3 to 4. Watch F2, the grand total in J6, and the average in J7 update. That’s the moment formulas start to feel useful instead of mysterious.

Wrapping Up

Your spending tracker can now calculate totals, flag bigger purchases, keep statuses consistent, summarize categories, and give other people a clear way to help. In my opinion, the best way to get comfortable with Sheets is to keep using one real list and add features only when that list needs them.

The post Google Sheets Complete Beginner’s Guide: From First Sheet to Sharing first appeared on Online Tech Tips.

Anthropic’s controlled tests showed that models can handle parts of that job. One model ran across a group of Android TV boxes. The researchers didn’t show an AI worm spreading through household networks. Your TV hasn’t been shown to be part of an AI botnet, but the results give smart-device owners a reason to pay attention.

What Happened

Self-replication sounds like something from a sci-fi movie. Here, it means an AI model arranging to set up and run a copy of itself on another device. To spread on its own, it would also need to find devices, get past their defenses, and repeat the process without help.

Anthropic tested parts of that chain. In one experiment, a 9-billion-parameter model set up AI inference on eight of nine hardware types. “Inference” means running a model so it can produce an answer. The model handled setup on several devices in the test. The experiment didn’t involve breaking into household gadgets. Source: Anthropic’s research report, published October 12, 2026.

In a separate physical-hardware demonstration, researchers split Gemma4-26B-A4B across 15 inexpensive, off-brand Android TV boxes. Together, the boxes ran the model at 0.39 tokens per second. A token is roughly a small piece of text, so replies came slowly. The test shows that a larger model can run across small devices. It also shows a practical limit: at that speed, generating a reply takes patience. Source: Anthropic.

Anthropic separately tested replication across eight emulated Android nodes in a controlled cloud network, not the 15 physical TV boxes. GLM-5.2 set up Llama3.1-8B on those nodes; GLM-5.2’s own inference ran on an external server. It reached all eight nodes in four of five trials and seven in the other trial. This was not GLM-5.2 independently running its own weights on the target devices. The researchers deliberately simplified access by leaving the test debugging port unauthenticated. That favorable setup limits what the result tells us about real-world defenses. The report doesn’t show a model finding, breaking into, and spreading among people’s smart devices on real networks. Source: Anthropic.

Who This Affects

This is research news, and you can read the public report in a browser. There’s no new feature or app to install.

If you own internet-connected smart devices, the findings are worth watching. That’s especially true for cheap devices with poor support. Anthropic used Android TV boxes as research hardware. Its results don’t mean other boxes, cameras, or appliances were compromised, or that every Android device faces the same risk.

How It Stacks Up

The study measured specific steps toward self-replication. It didn’t compare the latest Claude, OpenAI, and Google models. There are no matching scores here that would support a ranking.

Anthropic also reported slow setup times. A model trying to spread would have a lot more work to do. It would need to find reachable devices, get past their protections, set itself up, and keep going without help.

Anthropic warns about what future models may be able to do. Anthropic estimates that small models could reach the demonstrated capability level by the end of 2026 if historical improvements continue. It calls this a rough estimate based on a general capability index, not a prediction that household infections will start on a particular date. The tests show progress in controlled settings.

The Reaction

The important question is how much these controlled results tell us about real household networks. Anthropic’s own qualifications matter: the eight-node test simplified access and ran the controlling model externally. Those limits should accompany claims about the results.

Our Take

Verdict: Smart-home owners should tighten ordinary device security. This study evaluates model setup and replication under specified conditions; it does not compare the safety of consumer chatbot subscriptions. Our recommendation is to focus on device maintenance, rather than treat this experiment as a chatbot buying guide.

The concern is familiar. Devices with weak security, old software, or no more updates can be easier targets. If AI someday handles more of the setup work, that problem could grow. For now, the same precautions help against threats that already exist.

Your next step: Check your Android TV box for a firmware or system update, then check your other smart devices. On an Android TV box, look in Settings for System or Device Preferences, then About and System update. Menu names vary by manufacturer. Check the maker’s support page too; some boxes get updates through a different menu. I’d consider replacing a device if its maker has stopped supporting it.

Change any default passwords and turn on automatic updates where you can. A guest Wi-Fi network can keep smart devices separate from phones and computers that hold your personal files. Check your router’s instructions before moving devices, though. Some smart-home features need your phone and device on the same network.

Wrapping Up

Anthropic’s October 2026 research combines a model running across 15 physical TV boxes with a separate replication evaluation on eight emulated nodes. The latter used external inference for the controlling model. The tests happened under controlled conditions. Keep your smart devices updated, and retire the ones their makers have abandoned. That’s useful advice whether an AI worm ever shows up or not.

The post Anthropic Self-Replicating AI Research: Are Smart Homes at Risk? first appeared on Online Tech Tips.

Thinking of switching from Claude, an OpenAI assistant, or Google’s AI tools? Meta says Wonder can remember context and stick with tough tasks. That sounds useful, but a coding benchmark is not a test of how well an assistant handles your everyday messages and documents.

What Happened

Meta calls Wonder and Wonder Full its most capable AI models yet. According to Meta’s announcement, Muse Wonder is available through the Meta Model API and Muse Code. Muse Wonder Full is available on the Power Usage and Maximum Usage Muse Code plans. Meta says Wonder will power more features in Muse, its personal AI agent. Subscribers will get access first, though Meta hasn’t confirmed a date.

The access distinction matters: Wonder is listed for the API and Muse Code subscription tiers starting with Everyday, subject to regional availability; Full is included specifically with Power Usage and Maximum Usage. Meta spells out that difference in its model documentation. Buying a lower Muse Code tier does not unlock Full.

Meta Muse Wonder official model page header. Screenshot: Meta (October 14, 2026).

In its AI Research post, Meta highlights two abilities:

• Active Memory: Meta says Wonder can save useful details from past chats and recall them later. Think of it as keeping a note you won’t have to repeat.
• Active Focus: Meta says Wonder can stay with a hard task, try again after an error, and ask before taking important actions.

Both sound handy for a long document or a task with several steps. Those are Meta’s descriptions, however, rather than our hands-on findings.

Meta Muse Wonder API pricing and Wonder Full subscription access. Screenshot: Meta (October 14, 2026).

Who Can Use It Now?

Muse is the consumer agent. Muse Code is the terminal-based coding agent; the Meta Model API lets developers build model-powered software.

On Windows: Meta’s Muse Code documentation lists Windows support. Muse Code is a terminal-based coding agent, not a consumer chat app. Having it installed does not by itself establish your model entitlement; check the plan and selected model.

On Android and the consumer web app: Meta’s launch announcement describes Wonder-powered Muse experiences as coming soon, with subscribers first. It does not give a date in that announcement. Don’t equate a Muse subscription with today’s Muse Code access.

What it costs: Meta’s subscription help page lists Muse Code Everyday Usage at $5/month, High Usage at $15/month, Power Usage at $50/month, and Maximum Usage at $200/month. Full is included in the last two. Subscription benefits and availability can vary by region, so check the terms shown before paying. The CNBC launch report is additional reporting; consumer Muse subscriptions are separate from these Muse Code plans. Meta’s current plan documentation is the source for the access distinction here, and its announcement still describes the consumer rollout as forthcoming.

For API users, Meta’s pricing page lists Wonder at $4 per million input tokens, $0.20 per million cached input tokens, and $20 per million output tokens. These are current USD rates. Cached input has its own lower rate when the API classifies input tokens as cached. This is metered developer billing, not a flat-rate consumer subscription. Full is not offered through the pay-as-you-go API.

One setting deserves attention before uploading private material. According to Meta’s Muse Code subscription documentation, subscriptions default to Contributor mode. For this subscription usage, that setting permits training on the prompts, code, files, context and responses in exchange for more usage. Meta says switching it off means future subscription content is not used for training unless you enable it again. This describes Muse Code subscription usage, not every Meta account or service; separately created API keys use pay-as-you-go billing and their applicable API data-use terms.

How It Stacks Up

Meta has published comparisons, so it would be wrong to say there are no benchmark results. Its developer launch post compares named models and reports the following coding scores. These are figures reproduced by Meta, not tests we ran; Meta says these benchmarks are third-party evaluations; we have not audited their independence. The figures and effort labels below are reproduced from that Meta comparison post: its columns say max, while the Wonder Full SWE-Marathon cell says xhigh.

DeepSWE measures long-horizon coding; SWE-Marathon tests work on very long software projects. Different harnesses, reasoning budgets and task sets can change results. Meta links its evaluation report for the setup, and these coding results do not establish a winner for ordinary chat, memory reliability or every competing model. There is no Google model in this particular table.

A benchmark score can help, but it won’t tell you how an assistant handles your documents. Will it recall the right detail or catch a costly mistake? Meta’s claims make Wonder worth watching, but I’d hold off on moving every task and subscription over.

Be extra careful with money, health, or legal questions. If you try Wonder for those, check its facts, dates, source documents, and math before you act. An assistant can keep working and still head the wrong way (which is hardly the help you wanted).

The Reaction

Meta also shared its launch on X. That is a company announcement, not evidence of a user consensus. We are not presenting a community reaction verdict here.

Our Take

Verdict: Muse Code users have a reason to try Wonder now; everyday Muse users should wait for a confirmed rollout before considering a switch. If you have access, give Wonder a task you know well. Then check its answer against the source material. That’s more useful than taking Meta’s launch claims on faith.

The promise is appealing: an assistant that remembers useful details and stays with a messy task. But coding scores and a wider consumer rollout answer different questions. For now, stick with the assistant that works for you, and check Meta’s official announcement for access details before paying for Muse Code.

The post Meta Muse Wonder Is Here, but Should You Switch AI Assistants? first appeared on Online Tech Tips.

The good news is that the best Spotify settings for AirPods are a handful of other switches that do make a difference. Below, you’ll find where each one lives on iPhone, Android, Windows, Mac, and the web player, plus fixes for when Spotify still sounds worse than your other apps. The short version:

• Streaming quality set to Very High (the highest tier Bluetooth can actually use)
• Equalizer, with small, gentle tweaks
• Audio normalization and Volume level, so songs don’t jump in loudness
• Crossfade, gapless playback, or Automix, for smoother listening
• Data Saver off on Wi-Fi, so Spotify isn’t quietly lowering your quality

What Is Spotify (and What Is Lossless)?

Spotify is a music and podcast streaming app for iPhone, Android, Windows, macOS, and web browsers. Plenty of folks never touch its audio settings, but a few of them matter quite a bit.

• Quality tiers: Low, Normal, High, and Very High (roughly 24, 96, 160, and 320 kbps). Very High requires Premium.
• Lossless: Rolled out to Premium subscribers in supported markets starting in September 2025. It streams CD-quality-or-better files (up to 24-bit/44.1kHz FLAC) with nothing thrown away (Spotify Support: Lossless audio quality, retrieved October 12, 2026).
• Equalizer: Lets you boost or cut bass and treble using a slider-based curve (Spotify Support: Equalizer, retrieved October 13, 2026).
• Normalization: Evens out volume between tracks.
• Playback flow: Crossfade, gapless playback, and Automix.

This guide is for everyday listeners on Free or Premium who want better sound without buying new gear.

Prerequisites

• The Spotify app installed and up to date (or the web player at open.spotify.com)
• A Spotify Free or Premium account
• AirPods or headphones already paired (Bluetooth) or plugged in (wired/USB)
• For Very High quality: Spotify Premium
• For Lossless: Premium and a supported country and a wired, USB, Wi-Fi speaker, or compatible Spotify Connect setup
• For Lossless: an up-to-date app — Spotify requires mobile app version 9.0.58 or newer and desktop app version 1.2.67 or newer (Spotify Support: Lossless audio quality, retrieved October 13, 2026)

Step-by-Step Guide

Step 1: Install or Update Spotify

Older versions of Spotify are missing options like Lossless and the desktop equalizer. So update first.

On Windows

• Open the Microsoft Store, search for Spotify, and click Get (or Update if it’s already installed). You can also download the installer from spotify.com/download.
• Open Spotify and sign in.

Expected result: Spotify opens to your Home feed.

On macOS

• Download the installer from spotify.com/download or get Spotify from the Mac App Store.
• If you used the download, open the installer and let it place Spotify in your Applications folder.
• Launch Spotify from Applications or Launchpad and sign in.

On iOS

Open the App Store, tap your profile picture in the top-right, and scroll to find Spotify under pending updates. Tap Update.

On Android

Open the Google Play Store, search for Spotify, and tap Update if it appears.

On the Web

Go to open.spotify.com in Chrome, Edge, Firefox, or Safari and sign in. There’s nothing to install. The trade-off? The web player has the fewest audio controls (more on that below).

Step 2: Open Spotify’s Audio Settings

The audio settings are tucked a couple of taps deep on every platform.

• iOS and Android: Tap your profile picture (top-left) > Settings and privacy. Quality settings are under Media quality (labeled Audio quality on some versions). The equalizer, normalization, and crossfade live under Playback.
• Windows and macOS: Click your profile picture (top-right) > Settings, or press Ctrl + , on Windows or Cmd + , on Mac. Scroll to the Audio quality and Playback sections.
• Web player: Click your profile picture > Settings. You’ll find a short list of options — in the current web player there’s no streaming-quality picker or equalizer, and Spotify’s own Lossless support page covers only mobile, desktop, and connected devices, not the web player. If sound matters to you, use the desktop app instead.

Expected result: You see a list of quality and playback options.

Step 3: Set Streaming Quality to Very High

This is the single biggest win for Bluetooth listeners.

• Mobile: In Media quality, set Wi-Fi streaming to Very High. Set Cellular streaming to High or Very High depending on your data plan. Set Download to Very High.
• Desktop: Under Audio quality, set Streaming quality and Download to Very High.
• Free users will see High as the top option. That’s still fine for Bluetooth.

Expected result: New songs stream at roughly 320 kbps. You’ll need to re-download songs to get the new quality on them.

Step 4: Understand Lossless (and Skip It on AirPods)

Bluetooth headphones can’t receive Spotify’s file as-is. Your phone re-compresses the audio with a Bluetooth codec (a sort of shipping format) before sending it to your ears. AirPods use a codec called AAC, and its pipe is much smaller than Lossless needs.

Think of it like mailing a giant painting. The size of the original doesn’t matter if the mailbox only fits a postcard. Lossless gets squeezed down through that same postcard-sized Bluetooth pipe, just like Very High does — so most listeners won’t perceive a real-world difference over Bluetooth, even though the two aren’t technically identical bit-for-bit. How audible any gap is still depends on your specific codec, headphones, the source recording, and your own ears. Spotify’s own support page lists wired, USB, and Spotify Connect playback — not Bluetooth — as the ways to get full Lossless quality (source, retrieved October 13, 2026).

*When Lossless is worth turning on:*

• Wired headphones plugged into a computer or phone (with a USB-C or Lightning to 3.5mm adapter)
• A USB-C DAC (a small digital-to-analog converter) connected to your Windows PC or Mac
• A Wi-Fi speaker or compatible Spotify Connect device, where you can turn on Lossless from the Connect menu

To turn it on (Premium, supported markets): Go to Media quality on mobile or Audio quality on desktop. Choose Lossless for Wi-Fi streaming or downloads. Heads up: Lossless uses far more data than Very High.

Step 5: Turn On the Equalizer (Gently)

The equalizer is where you can actually change how your AirPods sound.

• iOS: Settings and privacy > Playback > Equalizer, then toggle it on.
• Android: Settings and privacy > Playback > Equalizer. On some phones (Samsung, for example), this opens your phone’s built-in sound settings instead of Spotify’s.
• Windows and macOS: Look under Settings > Playback for Equalizer in the desktop app. Exact wording and menu placement can vary by platform and app version — update the app if you don’t see it.
• Start with a preset like Bass Booster or Vocal Booster, then drag individual sliders if you want to fine-tune.

Expected result: The sound changes right away while music plays.

One starting point worth trying on AirPods: a small bass lift (+2 to +3 dB on the lowest two sliders) as a gentle first test, then adjust from there. Add a slight dip in the upper treble if things sound harsh. Resist the urge to crank everything up — stacking big boosts can introduce distortion or clipping. And a “smiley face” curve sounds exciting for five minutes and tiring after an hour. There’s no single “best” EQ for AirPods: it varies by AirPods model, fit, your hearing, and the source material, so treat any starting point as a baseline to tweak, not a fixed recommendation.

Step 6: Set Up Audio Normalization and Volume Level

Ever had one song whisper and the next one blast your eardrums? Normalization fixes that.

• Go to Playback (mobile) or Audio quality (desktop).
• Make sure Enable audio normalization (called Normalize volume on desktop) is on.
• Pick a Volume level:
  • Loud: For noisy places like the subway or gym. It squeezes quieter parts a bit.
  • Normal: The default, and the right pick for everyday listening.
  • Quiet: For calm spaces. It keeps the most of the music’s natural ups and downs.

Spotify says normalization only adjusts loudness between tracks. Your sound quality stays the same (Spotify Support: Volume normalization, retrieved October 12, 2026).

Step 7: Choose Crossfade, Gapless, or Automix

These three leave sound quality alone. What they change is how listening feels.

• Crossfade: Fades one song into the next. Set the slider to 3 to 6 seconds for a smooth blend, or leave it at 0.
• Gapless playback: Removes silence between tracks. Turn this on for live albums or concept albums where songs flow together. (Some app versions have it always on with no toggle.)
• Automix: Blends songs DJ-style on playlists that support it.

Find all three under Playback on mobile and desktop.

Step 8: Check Data Saver

Spotify’s own Data Saver setting quietly caps your streaming quality, which undoes Step 3. Sneaky. It’s a separate setting from your phone’s system-level data saving (like iOS Low Data Mode) — turning one off doesn’t turn off the other, so check both if quality seems capped.

• Mobile: Settings and privacy > Data-saving and offline (just Data Saver on older versions). Turn Data Saver mode off if you’re on Wi-Fi or have plenty of data.
• Want to save data on the go without wrecking quality at home? Keep Wi-Fi streaming on Very High. Drop only Cellular streaming to Normal or High. Download playlists over Wi-Fi before you leave.

Configuration Cheat Sheet

What about Spatial Audio? That’s an iOS feature, and you’ll find it outside Spotify. On iPhone, play a song, open Control Center, long-press the volume slider, and tap Spatial Audio. Then choose Off, Fixed, or Head Tracked. Spotify plays in regular stereo, so iOS “spatializes” it for you.

Tips and Troubleshooting

Problem: Spotify sounds worse on my AirPods than YouTube or Apple Music

• Turn the Spotify Equalizer off to rule out an aggressive curve.
• Make sure Data Saver is off and Wi-Fi streaming is on Very High.
• On iPhone, check Settings > Cellular > Cellular Data Options > Data Mode. This is Apple’s system-level Low Data Mode, separate from Spotify’s own Data Saver — it can reduce data use for apps that support it, including throttling Spotify’s streaming on cellular, so turn it off here too if quality still seems capped after checking Spotify’s setting.
• Check Settings > Accessibility > Audio & Visual > Headphone Accommodations. This iOS setting affects every app and can stack on top of Spotify’s EQ.
• Re-check your quality settings after an app update. They sometimes reset.
• Remember Apple Music has a home-field advantage with AirPods (tighter iOS integration and Dolby Atmos tracks). Some difference is normal.

Problem: Lossless doesn’t sound any different

That’s expected on Bluetooth. Switch to wired headphones, a USB DAC, or a Spotify Connect speaker to hear it.

Problem: The Lossless option is missing

• Confirm you’re on Premium.
• Check that Lossless is available in your country.
• Update the Spotify app. Remember the web player doesn’t offer it.

Problem: Music keeps buffering on Lossless or Very High

Move closer to your router or drop the streaming quality one notch. Walls and floors can block signals like nobody’s business.

Problem: The equalizer or settings are grayed out

You’re probably casting with Spotify Connect. Disconnect, change the setting, and reconnect.

Problem: My settings don’t match on my phone and laptop

Quality settings are per device. Set them on each phone, tablet, and computer separately.

• Android codec check: AirPods on Android use AAC or SBC. Curious which one? Turn on Developer options and look under Bluetooth audio codec (but don’t expect a miracle from changing it).
• iPhone Music EQ is a red herring: Settings > Apps > Music > EQ only affects Apple Music. Spotify ignores it.

FAQ

Does Spotify Lossless work with AirPods? Not the way it’s designed to. AirPods connect over Bluetooth, which re-compresses the audio using AAC, so the Lossless file never reaches your ears bit-for-bit. You can still turn the toggle on, but most listeners won’t hear a real difference versus Very High over that connection.

What’s the difference between Very High and Lossless? Very High is compressed at about 320 kbps. Lossless keeps every bit of the original file. On good wired gear in a quiet room, some people hear a difference. Over Bluetooth, most listeners won’t notice a gap either way, since the Bluetooth codec itself becomes the bottleneck.

Should I turn on audio normalization? Usually, yes. It stops volume jumps between songs. Audiophiles listening to full albums can turn it off for maximum dynamic range.

Is it worth using wired headphones just to get Lossless? Only if you already own decent wired headphones or love sitting down for focused listening. For commuting and workouts, AirPods win on convenience. The sound difference is one most ears struggle to notice.

Wrapping Up

Lossless is great, just not for AirPods. Set Very High quality, nudge the equalizer, turn on normalization, and keep Data Saver out of the way. That gets you the best sound your wireless earbuds can deliver.

I’d love to see Spotify put a little “doesn’t apply to Bluetooth” note right next to that Lossless toggle. Until then, save it for the day you plug in wired headphones or stream to a Wi-Fi speaker.

Resources

• Spotify Support: Lossless audio quality
• Spotify Support: Equalizer
• Spotify Support: Volume normalization
• Spotify Support: Your Premium benefits
• Spotify for Artists: Loudness normalization

The post Best Spotify Settings for AirPods and Headphones (And Why Lossless Won’t Help) in 2026 first appeared on Online Tech Tips.

Argon is aimed at demanding coding, research and professional workflows. For most readers, the practical starting point is checking access. A subscription that includes Argon in the Gemini app does not necessarily unlock it in Google’s coding tools.

Which plans include Argon?

The Gemini Apps information above applies to personal Google Accounts. Work and school accounts follow their organization’s license and administrator settings. Google’s model support page lists Argon for both chats and tasks on Pro and Ultra, with selectable thinking levels. The Antigravity model table sets out its different entitlements.

When will Argon appear in Antigravity?

Google says the rollout starts October 13 in the Antigravity desktop app and CLI for Google AI Ultra subscribers and eligible Gemini Enterprise subscriptions. It advises updating to the latest version.

There is a meaningful product distinction: the documentation lists Argon for Antigravity 2.0, the CLI and IDE extensions, and explicitly excludes the separate Antigravity IDE. If you are looking in the old IDE, an absent Argon option does not establish that the wider rollout has failed.

“Rolling out starting today” also does not promise a precise hour for every account. Google does not give an account-specific activation time in the October 13 announcement. Check the app version, eligible plan and model selector before changing subscriptions.

The benchmark scores Google reports

The following is Google DeepMind’s published comparison, checked October 13. These are results from specified tests and settings, rather than a prediction that one model will perform best on every task. Published performance table.

“—” means no comparison score is supplied in Google’s table.

The results show a mixed picture. Argon leads this comparison on several knowledge-work and coding measures, but other models score higher on FrontierSWE, Terminal-Bench and some science or computer-use tests. That is a reason to evaluate your own workflow rather than treat a launch headline as a universal ranking.

Google’s evaluation methodology says the Argon results generally use single-attempt evaluation and the API’s highest thinking settings. Some tests have different harnesses, budgets or special settings. Non-Gemini scores largely come from provider reports or public leaderboards; this is not one independent test lab running every model identically.

The million-token claim and API pricing

Google’s announcement highlights a one-million-token output limit. That should not be casually rewritten as a promise that every Gemini or Antigravity subscription exposes a million-token chat context.

The announcement lists introductory API prices of $2 per million input tokens and $10 per million output tokens, with cached input at a 95% discount from the input price. It lists $4 input and $20 output after the introductory period. These are the announced API rates, separate from consumer subscription pricing; check current developer pricing before budgeting a workload. Google’s pricing announcement.

What I would check before paying more

Start with the product you actually use. If your goal is Argon chat or tasks, check your Gemini account’s model selector. If your goal is coding in Antigravity, check that product’s eligible plan and supported app. I would test a few representative jobs before upgrading solely for the model name: a coding agent’s usefulness depends on how it handles your repository, review process and permissions as well as its benchmark scores.

This report uses Google’s announcements, support documentation and published evaluations. We have not independently reproduced the benchmark results or verified rollout completion for every account.

The post Gemini 4 Argon Rolls Out: Supported Plans, Benchmarks and Antigravity Access first appeared on Online Tech Tips.

Here’s the short answer if you’re outside China: this is an interesting phone to watch, not one to buy yet. There’s no global release date, no international price, and no independent reviews. All of the coverage below comes from news outlets, and every source was retrieved on October 12, 2026. We have not used the phone ourselves.

What Happened

OnePlus held its OnePlus 16 launch in China on Monday, October 12, 2026. GizChina and India Today reported the date ahead of the event. The early marketing put three things front and center: the high-refresh gaming mode, the huge battery, and the chip.

What’s missing matters just as much. Multiple outlets covering the October 12 launch attribute the headline camera, battery, display, and charging figures to OnePlus’s own launch materials, so those count as officially announced, not just rumored. But as of publication we still couldn’t find a Western-style international press kit, independent hands-on reviews, or an independently conducted benchmark review. So treat the officially announced specs as “announced by OnePlus, not yet independently tested,” and treat anything attributed only to leaks or unnamed sources as “reported, not confirmed.”

OnePlus 16 Specs: Confirmed vs. Still Murky

Think of this section as sorting a junk drawer. Some things are clearly labeled. Others you’ll have to squint at.

The Chip (and Why Its Name Is Confusing)

Most reports say the OnePlus 16 runs on the Snapdragon 8 Elite Extreme Gen 6. That’s Qualcomm’s newest top-tier phone processor, the “brain” that runs apps, games, and the camera. Gadgets360 and GSMArena both use that branding.

Here’s the snag. Across the coverage, the same chip shows up under three different names:

• Snapdragon 8 Elite Extreme Gen 6
• Snapdragon 8 Elite Gen 6 Pro
• Snapdragon 8 Elite Gen 6

These could be one chip that gets reported sloppily. They could also be different tiers of the same chip family, much like a car that comes in a base trim and a sport trim. A Gadgets360 chipset roundup calls it a “series,” which suggests more than one variant exists. One post on X claims OnePlus has verified the “Extreme” name. Until Qualcomm or OnePlus publishes a spec sheet we can point to, though, we’re treating the exact name as provisional.

Why should you care about a name? Because “Extreme” versus regular often means different clock speeds and different performance. If you’re comparing phones on paper, that difference counts.

The Display: 165Hz Normally, 185Hz in Some Games

Refresh rate is how many times per second the screen redraws itself. Most phones run at 60Hz or 120Hz. Higher numbers make scrolling and fast motion look smoother, a bit like the difference between a flipbook and a real video.

According to GSMArena’s display report and Gadgets360, the OnePlus 16:

• Runs at 165Hz day to day. That’s faster than most flagships, which top out around 120–144Hz.
• Can boost to 185Hz/185fps in supported games. The titles reported so far are Arena Breakout, Nizhan: Future, Naruto Mobile, and Peace Elite.

Notice anything about that list? Several of those are big in China. Peace Elite, for example, is the Chinese version of PUBG Mobile. If your favorite game isn’t on the list, you won’t see 185Hz in it. Nobody has confirmed whether global versions of these games will get the boost either.

Does it matter for normal use? Honestly, not much. Going from 120Hz to 165Hz is a subtle upgrade for scrolling Instagram or reading email. The 185Hz mode is a perk for competitive gamers playing those specific titles. And it’s a manufacturer claim. Nobody independent has measured it yet.

The Battery: 9,000mAh Is Genuinely Unusual

This is the spec that made people sit up. GizChina and India Today both report a 9,000mAh battery.

For context, most flagship phones land somewhere between 4,500 and 5,500mAh. A 9,000mAh battery is roughly 64% bigger than the top of that range and exactly double the bottom of it. If the number holds up, it could mean real two-day battery life for a lot of people. That’s the kind of capacity you’d normally get by strapping a power bank to your phone.

Battery size isn’t the whole story, though. A fast chip and a 165Hz screen also burn more power. How long the phone actually lasts depends on that balance, and we won’t know until reviewers run real-world tests.

The Cameras: Triple Rear Setup Led by a 200MP Main Sensor

Gadgets360 and OnePlus’s own launch-day spec listing put three rear cameras on the OnePlus 16: a 200-megapixel main camera, a 50-megapixel periscope telephoto lens with 3x optical zoom, and a 50-megapixel ultrawide camera. A periscope lens folds light sideways inside the phone, like a submarine periscope. That lets the phone fit in a longer zoom without a giant camera bump.

More megapixels gives you more room to crop. It doesn’t automatically mean better photos, though. Processing, sensor size, and low-light handling matter more. No independent camera samples or tests have been published yet.

Everything Else (Treat With Extra Caution)

• Charging: 100W wired and 50W wireless are reported, but at least one outlet noted that OnePlus hasn’t officially confirmed the wired figure. If that number holds up, you’d want a proper high-wattage USB-C charger to get anywhere near those speeds.
• Memory and storage: Reported configurations range from 12GB RAM/256GB storage up to 16GB RAM/1TB.
• Display size: A 6.78-inch LTPO OLED panel at 1272×2772 resolution, on top of the 165Hz/185Hz behavior covered above.
• Front camera: A 50-megapixel selfie camera is reported, though no samples have been published.
• Software: Android 17, but the skin on top depends on region. OnePlus is retiring OxygenOS in favor of ColorOS on China models, so the OnePlus 16 ships with ColorOS 17 in China. Global units, whenever they arrive, are expected to run OxygenOS 17 instead, since OnePlus is keeping that branding for international markets a while longer. Don’t assume “OxygenOS” if you’re buying the China model.
• Price (China): Multiple outlets report a starting price of CNY 5,999 (roughly $840) for the 12GB/256GB model, scaling up to CNY 8,499 (roughly $1,190) for 16GB/1TB. Promotional and subsidized pricing at launch was reportedly lower. No international price has been reported — don’t assume the China price would carry over to a US, UK, or EU release.

Who This Affects

Right now, the OnePlus 16 is a mainland China product. If you live there, you can buy it through local channels.

If you live in the US, UK, EU, India, or anywhere else, the phone isn’t officially for sale in your country as of this publication. OnePlus hasn’t announced a global launch, international pricing, or carrier support, though that status could change later. Some posts on X float a possible India launch. Those are rumors, not announcements, and nothing has been confirmed for the US or Europe.

How It Stacks Up

This is the part where we’d love to show you a tidy benchmark table. We can’t, and anyone who does right now is guessing. No independent benchmarks, battery tests, or camera comparisons exist yet for the OnePlus 16 or its closest rivals.

One number is floating around: an AnTuTu benchmark reportedly crossed 5.38 million points ahead of launch, with OnePlus itself later citing an even higher figure near 5.63 million. AnTuTu scores like that are typically run or supplied by the manufacturer ahead of launch, not by an independent reviewer with a retail unit, so treat it as a manufacturer-reported number, not an independently verified one.

What we can show you is where the competition stands, based on Gadgets360’s chipset roundup:

In other words, the OnePlus 16 is one of several phones about to share the same family of chips, at least on paper. Whether that translates into similar real-world performance is something only independent benchmarks can answer, and none exist yet. On paper, the 9,000mAh battery and the 185Hz gaming mode are where OnePlus is trying to stand out from that pack.

What about the Samsung Galaxy S-series or iPhone you can actually walk into a store and buy? We found no reported head-to-head data. Don’t assume the OnePlus 16 beats them on camera or battery life until reviewers weigh in.

The Reaction

We didn’t run a formal social-listening audit, so treat this as a loose impression rather than data. The X posts and comment threads we came across while researching this piece mostly repeated the same spec-sheet highlights: the chip, the 9,000mAh battery, the 200MP camera, and the 185Hz display. We didn’t come across dedicated Hacker News discussion of the launch. The recurring question in what we saw was where can I get it?, with China treated as settled and an eventual India release discussed as a rumor, not a confirmed plan. Take that as a general sense of the chatter, not a measured count of it.

Our Take

Who should care right now:

• Mobile gamers who play the supported titles, especially if you’re in China or play China-market versions of Peace Elite, Naruto Mobile, Arena Breakout, or Nizhan: Future. The 185Hz mode is aimed squarely at you. Pair it with a clip-on phone cooler or a gaming controller if you take your matches seriously.
• Battery-life obsessives. If the 9,000mAh figure holds up in testing, this could be one of the longest-lasting flagships around. Keep an eye on it.

Who should wait (most of you):

• Anyone outside China. Importing means paying extra and usually losing official warranty coverage. You may also run into missing network bands, China-focused software, and patchy access to Google services. That’s a lot of hassle for a phone with zero independent reviews.
• Camera-first buyers. A 200MP sensor sounds impressive, but megapixels aren’t image quality. Wait for real sample photos.
• Anyone comparing it to a Galaxy or iPhone. There’s simply no data yet.

Our honest opinion: the OnePlus 16 looks like one of the more exciting spec sheets of the year. Still, “exciting spec sheet” and “phone you should import” are two very different things.

What to Do Next

Here’s one practical step: bookmark this phone and wait for independent reviews and an official spec page before spending a cent. Once reviewers test the battery life and the chip’s real-world speed, and OnePlus says whether it’s coming to your country, you’ll be making a decision based on facts rather than marketing. In the meantime, if battery anxiety is your problem today, a good power bank is a much cheaper fix than an imported phone.

Wrapping Up

The OnePlus 16 is real, it’s out in China as of October 12, 2026, and its headline specs are eye-catching. A 9,000mAh battery and a 185Hz gaming mode are hard to ignore. But the chip name is still fuzzy, the performance claims are unverified, and there’s no global release date. For most people, the smart move is to watch and wait.

If OnePlus brings this phone to more countries and reviews back up the battery claims, it could be one to beat. Until then, it’s a fascinating preview, not a shopping decision.

The post OnePlus 16 Launches in China: What’s Confirmed, What’s Murky, and Who Should Care first appeared on Online Tech Tips.

So, did Deadlock really beat Team Fortress 2’s biggest-ever player record? Yes, but only on one narrow measure. It’s the most people logged into Deadlock on Steam at a single moment, compared with the same Steam-only snapshot for TF2. It doesn’t tell you which game is more popular overall, which one sells better, or which one is better. And Deadlock still isn’t officially released. Here’s what the number actually means.

What Happened

Here are the facts, without the hype:

• The new record: Deadlock reached an all-time peak of 262,145 concurrent players on Steam on October 9, 2026, according to SteamDB data cited by PC Gamer and Inven Global.
• The record it passed: Team Fortress 2’s all-time Steam peak of 253,997, set on July 13, 2023.
• The margin: About 8,148 players. That’s a clear win, but not a blowout.
• The timing, and a possible reason: The record landed on the same day Valve added Solomon, a chess-themed wizard, to Deadlock’s roster. He’s the third of six new heroes Valve is rolling out gradually, two a week, as part of the City Never Sleeps update (Rat King on October 2, Baba on October 6, Solomon on October 9), according to Game Rant. That makes him a tempting explanation, and PC Gamer credits him directly, saying it was “specifically the latest hero” who got the game to this point. The player-count data can’t prove that, though: Game Rant notes that Deadlock’s numbers had been rising gradually for about two weeks before the peak. Our read is that the new hero was probably one factor, but that’s informed speculation, not something the numbers establish.

One thing to be clear about: this isn’t a Valve announcement. Valve didn’t put out a press release or post the number anywhere. The data comes from SteamDB, an independent site that tracks how many people are logged into each game on Steam. Gaming outlets routinely cite SteamDB for these numbers, but it’s still worth knowing that this is a third-party measurement, not an official Valve statement.

Who This Affects

• Platform: Deadlock is played through Steam on PC. That’s the version these numbers measure.
• Access: Deadlock is still an invite-only closed test. You can’t search Steam, click a button and start playing. You need an invite.
• Price: Game Rant describes the test as free-to-play on Steam. Valve hasn’t put out a press release with launch details, so be wary of any site that claims to know what the final version will cost.
• Team Fortress 2 players: Nothing changes for you. TF2 is still fully released, still free-to-play and still on Steam. Your game just lost one Steam bragging right.

How It Stacks Up

This is where the headline starts to need a little context. Breaking TF2’s record sounds huge, and it is impressive for a game nobody can buy. But Valve has a couple of other games playing in a completely different league.

Here are the all-time Steam concurrent peaks, all tracked by SteamDB and listed with their dates by Game Rant:

Figures as reported by the linked sources and by Game Rant, retrieved October 11, 2026.

Some things stand out:

• Dota 2 is the closest match, and its record is nearly five times bigger. Deadlock mixes third-person shooting with Dota-style lanes, hero abilities and character upgrades, so Dota 2 is the fairest comparison in Valve’s lineup. Its 2016 record of 1,295,114 is a little under five times Deadlock’s 262,145.
• Counter-Strike 2 is in a different universe. At over 1.8 million, CS2’s peak is about seven times Deadlock’s. It’s here for scale, not as a rival.
• The TF2 comparison is a fun milestone, not a crown. Passing TF2 puts Deadlock above one of Valve’s most beloved games on one chart. It doesn’t make Deadlock Valve’s biggest game, or even close.

What the Number Doesn’t Count

Think of “concurrent players” like counting how many people are in a stadium at the busiest moment of a game. It tells you the crowd was big at that moment. It doesn’t tell you how many people bought season tickets, how many come back every week, or whether they liked the game.

There are also two important caveats specific to this comparison:

• TF2’s console players aren’t included. Team Fortress 2 also came out on Xbox 360 and PlayStation 3 as part of The Orange Box. Those players never show up on Steam’s numbers, so they aren’t part of the record Deadlock just passed. This is purely Steam versus Steam.
• Deadlock’s number is from a restricted test. Every one of those 262,145 people needed an invite. That’s arguably more impressive. It also means the number may partly reflect a burst of excitement around the new hero, rather than a settled, long-term audience.

Our Take

Verdict: Impressive for a game still in testing, but one Steam stat isn’t a crowning moment.

Here’s who should care, and why:

• If you’re curious about Deadlock: This shows Valve’s hero-shooter/MOBA mix can draw a big crowd during testing. Whether that crowd sticks around is the real test. If you like Dota-style strategy but want to aim and shoot yourself, this is the Valve project to watch. Just remember that you can’t play it without an invite.
• If you’re a Team Fortress 2 fan: Relax. Your game hasn’t been “dethroned” in any meaningful way. Deadlock passed a single Steam-only peak, and that says nothing about TF2’s long history. TF2 is still free and still running.
• If you follow Valve news in general: The interesting part is that Valve keeps adding heroes and running Deadlock as a long, invite-only test with no announced launch date, and the numbers keep climbing anyway. That points to a possible next big live-service game for Valve. It’s still a long way from Dota 2 or CS2 territory, though.

If we had to put it in one sentence: Deadlock is clearly hot right now, and the next few weeks of player counts after Solomon’s arrival will tell us much more than this one peak.

What to Do Next

If this news made you want to try Deadlock, here’s the one practical step: open Steam on your PC and keep an eye on Valve’s official Deadlock and Steam channels for invite or access news. Right now, access depends on getting an invite. A friend who already has access may be able to send you one.

What not to do: don’t download “Deadlock” from random websites, YouTube links or Discord servers promising free access. The game isn’t publicly available, so anything claiming to be a standalone download is a red flag for malware.

Wrapping Up

Deadlock’s record is a real milestone for a game you need an invite to play. It’s also one Steam-only snapshot, and a small one next to Dota 2 and Counter-Strike 2.

Our honest take? It’s a great sign for Deadlock’s future, but the bigger question is whether those players are still logging in a month from now, chess wizard or not.

The post Deadlock Passes Team Fortress 2’s All-Time Steam Player Peak (But Read the Fine Print) first appeared on Online Tech Tips.

Here’s the short answer. Black bars are usually normal. Many movies are wider than your TV, so bars keep the full picture visible. A stretched, squashed, or cropped image is different. That usually comes from your TV, streaming device, console, or browser rather than the title itself (a genuinely mis-framed title is rarer, and covered under When to Escalate), so the fix usually lives wherever the picture is being resized. This guide shows you how to tell which problem you have, then how to fix it on whatever you watch on.

Are Black Bars on Prime Video Normal?

An aspect ratio is just the shape of the picture: its width compared to its height. Your TV is one shape, but movies come in several.

• 16:9: The shape of nearly every modern TV and most TV shows. These fill your screen edge to edge.
• 1.85:1: Slightly wider than 16:9. You might see very thin bars on top and bottom.
• 2.39:1 / 2.40:1: The super-wide “cinema” shape used by many blockbusters. On a 16:9 TV, you get noticeable bars on the top and bottom. This is called letterboxing.
• 4:3: The boxy shape of older TV shows and classic films. On a widescreen TV, you get bars on the left and right sides. This is called pillarboxing.

Think of it like putting a panoramic photo in a standard picture frame. You can show the whole photo with empty space around it, or you can crop the edges off. Prime Video shows the whole photo. That’s a good thing, even if it doesn’t feel like it!

So bars alone aren’t a problem. Distortion is. If people look normal and the bars come and go depending on the title, nothing is broken.

Quick Diagnosis: Is It the Movie or Your Settings?

Run through these checks before you touch a single setting. They take about five minutes and tell you exactly where to look.

• Compare a few titles: Play a recent TV series (usually 16:9) and then a big action movie (often 2.39:1). If the series fills the screen and the movie has top-and-bottom bars, everything is working correctly.
• Check the Prime Video menus themselves: Look at the app’s home screen as well as the video. If the menus, logos, or text look stretched, too wide, or cut off at the edges, the problem is your TV or device display setting, not any particular title.
• Look at faces and circles: Round objects (wheels, clocks, faces) should look round. Wide, flattened faces mean something is stretching the image.
• Test another app: Open Netflix, YouTube, or your device’s home screen. If those look wrong too, it’s a TV or device setting.
• Test another device or input: Try the same title on your phone or your TV’s built-in Prime Video app. If it looks fine there, the issue is tied to one specific device or HDMI input.

Symptoms and Fixes at a Glance

Clues and What They Mean

Prime Video doesn’t show error codes for picture-shape problems, so the screen itself is your “error message.” Here’s how to read it.

Common Issues and Solutions

Problem: Every Title Looks Stretched

Symptoms:

• Faces look wide and flat
• Circles look like ovals
• No black bars on 4:3 shows, even though there should be

Why it happens: Your TV or streaming device is set to a Stretch, Wide, Full, or Wide Zoom mode that forces every picture to fill the screen.

Fix:

• Open your TV’s settings and find Picture, then a submenu called Picture Size, Aspect Ratio, Screen, or Wide Mode (names vary by brand).
• Pick the most neutral option. Look for labels like Just Scan, Screen Fit, Fit to Screen, 16:9 Standard, Original, or Normal.
• Turn off anything called Stretch, Wide Zoom, Panoramic, or Full if it’s distorting the picture.

Verify: Play a 4:3 show. You should now see bars on the sides, and people should look normal.

Problem: Subtitles, Logos, or Heads Are Cut Off

Symptoms:

• Subtitles run off the bottom of the screen
• The Prime Video logo or menu edges are clipped
• Tops of heads are missing in close-ups

Why it happens: Zoom or overscan is on. Overscan is an old TV habit of enlarging the picture slightly to hide messy edges from broadcast signals. On modern streaming, it just chops off part of the image.

Fix:

• In your TV’s picture settings, choose Just Scan, Screen Fit, or turn Overscan off.
• Check your streaming device or console for a display-area or overscan setting too (see the platform sections below).
• Avoid Zoom modes entirely for regular viewing.

Verify: Open the Prime Video home screen. Menu edges and text should sit fully inside the screen with a little breathing room.

Problem: Looks Fine on One Device but Wrong on Another

Symptoms:

• The TV’s built-in Prime Video app looks perfect
• The same title through a Fire TV, Roku, or console looks stretched or cropped

Why it happens: Many TVs save a separate picture-size setting for each HDMI input, plus another for built-in apps, so fixing one doesn’t always fix the others. Some models instead share settings or switch them automatically by signal type or picture mode.

Fix:

• Switch the TV to the HDMI input your streaming device or console uses.
• Open the picture settings while on that input and set it to Just Scan, Screen Fit, or 16:9 Original.
• Then check the device’s own display settings.

Verify: Both the built-in app and the external device should show the same title the same way.

Tip: While you’re poking around behind the TV, make sure your streaming device is plugged in with a decent high-speed HDMI cable. It won’t fix the shape, but a flaky cable can cause flickering and dropouts that make troubleshooting confusing.

Problem: The Shape Changes Partway Through a Movie

Symptoms:

• Black bars appear and disappear between scenes
• Big action sequences suddenly fill the whole screen

Why it happens: Some films were shot with IMAX cameras for certain scenes, or were mastered with mixed ratios on purpose. This is part of the movie itself.

Fix: None needed! No device setting will lock the shape, because the change is built into the original film. Enjoy the bigger picture during those scenes.

Problem: Nothing Changed After Adjusting Settings

Symptoms:

• You changed the setting, but the picture looks the same

Why it happens: A stuck app state or a pending update on the TV, device, or app.

Fix:

• Fully close the Prime Video app (or browser tab).
• Unplug the TV and streaming device for at least 60 seconds, then plug them back in (and you know how often a good restart fixes things!).
• Check for updates to the Prime Video app, your TV software, your streaming device, or your console.
• If it still misbehaves, uninstall and reinstall the Prime Video app.

Verify: Re-run the quick diagnosis checks above.

Platform-Specific Fixes

On Smart TVs

Menu names differ by brand and model year, so use these as a guide, not a fixed map:

• Samsung: Look under Picture > Picture Size Settings. Use 16:9 Standard and turn on Fit to Screen if it’s available.
• LG: Look under Picture > Aspect Ratio. Choose Original or Just Scan.
• Sony: Look under picture or screen settings for Wide Mode (choose Normal or Full depending on source) and set Display Area to the full-pixel option.
• Vizio: Look for Picture Size or the Wide setting. Use Normal and avoid Stretch or Panoramic.

If your TV stores picture size per input, repeat the change on every HDMI input you use.

On Streaming Devices

• Fire TV: Go to Settings > Display & Sounds > Display. Set Video Resolution to Auto. If menu edges are cut off, use Calibrate Display if your model offers it. While you’re in Settings, check for software updates under My Fire TV; Prime Video Help notes that some Prime Video features on Fire TV need newer device software.
• Roku: Go to Settings > Display type and choose the option that matches your TV (usually Auto detect). On many Roku TVs, you can press the \* button during playback, open Picture settings, and look for Picture size. Menu names and options vary by Roku model and software version.
• Apple TV: Go to Settings > Video and Audio > Format and pick your TV’s correct resolution. The Match Content options (Match Dynamic Range and Match Frame Rate) only change HDR and frame rate, not the picture’s shape, so they won’t fix stretching or cropping.
• Chromecast with Google TV / Google TV Streamer: Go to Settings > Display & Sound and check resolution and advanced display options. Leave resolution on automatic unless you have a reason to change it.

On Game Consoles

Console menus can’t be captured here, so these are text-only directions. Labels shift with system updates.

• Xbox: Go to Settings > General > TV & display options. Under video fidelity options, check that overscan borders are off unless your TV is cropping the dashboard.
• PlayStation 5: Go to Settings > Screen and Video and look for Adjust Display Area. Make sure the guide corners line up with the edges of your screen.

Then fix the TV’s picture-size setting for the console’s HDMI input.

On Phones and Tablets

Phones are taller and narrower than most movies, so bars are extra common. On many phones and tablets, the Prime Video app lets you pinch outward on the video to zoom and fill the screen, though support can vary by device and app version.

The tradeoff: zooming crops the edges. You might lose subtitles, faces at the sides of the frame, or on-screen text. Pinch inward to go back to the full picture.

On Windows

On Windows you can watch Prime Video in the Prime Video app or in a browser at primevideo.com. In the app, skip the zoom step and just use the player’s full-screen button. In a browser:

• Press Ctrl + 0 to reset browser zoom to 100%, which rules out zoom as the cause.
• Click the full-screen icon in the player (or press F).
• You usually don’t need to touch display settings for Prime Video. Only if your desktop and other apps also look stretched, open Settings > System > Display and confirm the recommended resolution and scale are selected.

On an ultrawide monitor, most movies will still show bars on the sides, since the video isn’t made for that shape. Check your monitor’s own menu for a stretch or aspect setting and set it to the original or 1:1 option.

On macOS

• Press Cmd + 0 to reset browser zoom, which rules out zoom as the cause.
• Click the full-screen icon in the Prime Video player.
• Only if your desktop and other apps also look wrong, open System Settings > Displays and make sure the default resolution for your display is selected. Changing resolution just for Prime Video is rarely needed.

On the Web (Any Browser)

Don’t go hunting for a picture-size option in the Prime Video web player. In a browser, the settings that actually affect framing are browser zoom and full screen.

If the video looks small or oddly framed, check browser zoom first. In Chrome or Edge, open the three-dot menu and look at the Zoom row.

In full screen on a standard 16:9 monitor, a 16:9 show should usually fill the display, unless the title itself has bars built in or your monitor has a different shape.

Also try disabling browser extensions that claim to “fix” ultrawide video. They can crop or stretch the picture in unexpected ways.

Settings to Check

• TV Picture Size: Picture menu > Picture Size / Aspect Ratio. Set to Just Scan, Screen Fit, or 16:9 Original.
• Overscan / Zoom / Stretch: Same menu. Turn these off.
• Streaming device resolution: Device display settings. Use Auto.
• Console display area: Console screen/video settings. Line up edges and turn overscan off.
• Browser zoom: Browser menu. Reset to 100%.

When to Escalate

Contact Amazon if:

• One specific title looks distorted on every device, while other titles look fine
• A 16:9 TV show appears squished or windowboxed everywhere
• Subtitles or credits are cut off in the source video itself

How to get help:

• Visit the Prime Video Help Center and choose the contact option.
• Share the title name, season and episode, timestamp, and the devices you tested.
• Mention that other titles look correct, which tells Amazon it’s a title problem, not your setup.

You can also check the list of supported devices to make sure your hardware is still supported.

Prevention Tips

• Pick a neutral TV setting once: Set Just Scan or Screen Fit on every input and leave it.
• Skip Zoom for daily watching: It’s tempting, but it crops the picture you paid to see.
• Keep things updated: Update the Prime Video app, TV, streaming device, and console regularly.
• Write down your settings: A quick phone note helps when someone else in the house “helpfully” changes them.

Frequently Asked Questions

Are black bars on Prime Video normal, or is something broken?

Usually normal. Wide movies get bars on top and bottom, and older 4:3 shows get bars on the sides. It’s only a problem if people look stretched or parts of the picture are cut off.

How can I tell if it’s the movie’s format or my TV settings?

Compare several titles and look at the Prime Video menus. If bars come and go by title and the menus look right, it’s the movie. If the menus look distorted too, it’s your settings.

Which TV setting removes stretching without cropping?

Look for Just Scan, Screen Fit, Fit to Screen, or 16:9 Original. These show the picture exactly as it arrives.

Why does it look fine on one device but wrong on another?

Many TVs save different picture-size settings for each HDMI input and their built-in apps. Each device also has its own display settings.

Should I use my TV’s Zoom feature to remove black bars?

I wouldn’t. Zoom crops off the sides of the picture, and sometimes subtitles. You lose part of the film just to hide some black space.

Why does the aspect ratio change partway through a movie?

Some films use IMAX or mixed-ratio scenes on purpose. That’s part of the movie, not a glitch.

How do I report a title with a genuinely wrong aspect ratio?

Use the contact option in the Prime Video Help Center. Include the title, timestamp, and the devices you tested.

Does upgrading my Prime plan fix aspect ratio issues?

No. Your plan doesn’t change a title’s shape. Bars come from the title’s format, and stretching or cropping comes from your display settings, whichever plan you’re on.

Wrapping Up

Most “wrong” Prime Video pictures turn out to be perfectly normal letterboxing. When something really is off, a stretch, zoom, or overscan setting on your TV or device is almost always to blame. Run the quick diagnosis first, and you’ll save yourself a lot of menu-diving.

Honestly, once you set every TV input to Just Scan or Screen Fit, you’ll rarely think about this again. And those black bars? They’re the filmmaker’s way of showing you the whole shot. I’ve learned to love them.

Last updated: October 11, 2026 | Applies to Prime Video on smart TVs, streaming devices, game consoles, phones, Windows, macOS, and web browsers

The post How to Fix Prime Video Aspect Ratio Problems (Black Bars, Stretching, Cropping) first appeared on Online Tech Tips.

This guide walks you through each of those steps on Windows and Mac. Set aside about an hour. By the end you’ll have a working Twitch broadcast that won’t overload your internet connection or make your game stutter.

What Is OBS Studio?

OBS Studio (short for Open Broadcaster Software) is a free, open-source app that captures what’s on your screen, your webcam, and your microphone, and sends it all live to Twitch. Think of it as a TV control room that fits on your PC.

• Scenes and Sources: Stack your game, webcam, and mic into one layout (a scene) and switch layouts with a single click.
• Auto-Configuration Wizard: Tests your PC and internet connection, then suggests safe starting settings.
• Built-in Twitch connection: Log in to Twitch from inside OBS, so you don’t have to copy and paste a secret stream key.
• Audio Mixer: A volume slider and mute button for each sound source.
• Hardware encoding: Lets your graphics card handle the video work so your game keeps running smoothly.

It’s ideal for gamers, artists, musicians, and anyone who wants to try streaming without paying for software. OBS has no paid tiers, and going live on Twitch is free too.

Before You Begin

Make sure you have:

• A free Twitch account you can log in to at twitch.tv
• Two-factor authentication (2FA) turned on for that Twitch account
• A Windows 10 or later (64-bit) PC, or a Mac running macOS 13 Ventura or later
• A webcam and microphone (built-in is fine for a first stream)
• Your upload speed, measured with a speed test
• About an hour of uninterrupted time

Turn On Twitch Two-Factor Authentication

2FA means Twitch asks for a code from your phone in addition to your password. It keeps your channel safe if your password ever leaks. Twitch also requires 2FA before you can finish Affiliate onboarding later, according to the Affiliate Onboarding Guide (retrieved October 2026). Turning it on now saves you a headache down the road.

• Sign in at twitch.tv and click your profile picture in the top-right corner.
• Click Settings, then open the Security and Privacy tab.
• Under Two-Factor Authentication, click Set Up Two-Factor Authentication and follow the prompts.

Check Your Upload Speed

Most speed tests show download speed in big numbers, but streaming depends on upload speed, which is usually much smaller. Run a test on a site like Speedtest by Ookla from the computer you’ll stream on, and write down the upload number in Mbps. If you can, plug your PC into the router with a Cat6 Ethernet cable first. Wi-Fi works, but walls and microwaves can make it flaky like nobody’s business.

Step-by-Step Guide

Step 1: Download OBS Studio

• Go to obsproject.com.
• Click Windows or macOS. On a Mac, choose the Apple Silicon download for M-series chips or Intel for older Macs. (Not sure which you have? Click the Apple menu > About This Mac and look at the Chip or Processor line.)

Expected result: The installer (.exe on Windows, .dmg on Mac) appears in your Downloads folder.

Step 2: Install OBS

On Windows

• Double-click the downloaded OBS-Studio-...-Windows-x64-Installer.exe file.
• Click Yes if Windows asks whether the app can make changes.
• Click Next, accept the license, keep the default install location, and click Install.
• Click Finish. Leave Launch OBS Studio checked.

On macOS

• Double-click the downloaded .dmg file.
• Drag the OBS icon into the Applications folder.
• Open Applications and double-click OBS. Click Open if macOS warns that it was downloaded from the internet.
• When macOS asks for camera, microphone, and screen recording access, allow each one. If you miss a prompt, go to System Settings > Privacy & Security and turn OBS on under Camera, Microphone, and Screen Recording (labeled Screen & System Audio Recording on newer macOS versions).
• Quit and reopen OBS so the new permissions take effect.

Expected result: OBS opens, and on first launch it offers to run the Auto-Configuration Wizard.

Step 3: Run the Auto-Configuration Wizard

If the wizard didn’t pop up on its own, click Tools > Auto-Configuration Wizard.

• Choose Optimize for streaming, recording is secondary and click Next.
• On the Video Settings page, leave Base (Canvas) Resolution on Use Current and set FPS to Either 60 or 30, but prefer 60 when possible. Click Next.
• On the Stream Information page, set Service to Twitch.
• Click Connect Account (recommended). A Twitch login window opens. Sign in and approve OBS.
• Leave Estimate bitrate with bandwidth test and Prefer hardware encoding checked. Click Next.
• Wait a minute or two while OBS runs its tests, then click Apply Settings.

Expected result: OBS now has a resolution, frame rate, encoder, and bitrate picked for your machine. New Twitch panels such as Chat and Stream Information may also appear around the main window. That’s normal, and you can rearrange or close them.

Step 4: Confirm Your Twitch Connection

Even if the wizard connected your account, it’s worth a quick check.

• Click Settings in the Controls panel (bottom-right).
• Select Stream in the left sidebar.
• Make sure Service says Twitch and Server says Auto (Recommended).
• If you see a Connect Account (recommended) button instead of your username, click it and sign in.

Tip: Connecting your account is easier than pasting a stream key. A stream key is a password-like code that tells Twitch the video is yours. Twitch sometimes resets keys, which breaks a pasted key without warning. If you must use one, find it in the Twitch Creator Dashboard under Settings > Stream. Never show it on stream, paste it into chat, or include it in screenshots, logs, or support requests: anyone who has it can broadcast to your channel.

Expected result: Settings > Stream shows your Twitch account as connected.

Step 5: Pick Settings That Match Your Upload Speed

The wizard’s picks are a good start, but they can be a little optimistic. Here’s how to double-check them.

• While still in Settings, click Output. Keep Output Mode on Simple.
• Look at Video Bitrate. Bitrate is how much data your stream sends each second. Higher bitrate looks sharper but needs more upload speed. Aim for no more than about 75% of your upload number, leaving room for everything else on your network.
• Look at Encoder. If you see Hardware (NVENC, H.264), Hardware (AMD, H.264), Hardware (QSV, H.264), or on a Mac Hardware (Apple, H.264), pick it. Your graphics chip will handle the video work instead of your CPU, which usually has less impact on your game than Software (x264). Note: OBS’s Hardware Encoding guide says streaming with the Apple hardware encoder is not supported on Intel Macs — if you’re on an Intel Mac, use Software (x264) instead.
• Click Video in the sidebar and set Output (Scaled) Resolution and Common FPS Values using the table below.
• Click Apply, then OK.

Treat these rows as starting points, not guarantees. What matters is upload speed that stays steady for the whole stream, plus a computer that can encode at your chosen resolution without dropping frames. Twitch’s Broadcasting Guidelines also recommend constant bitrate (CBR) and a 2-second keyframe interval. If you later switch Output Mode to Advanced, set Rate Control to CBR and Keyframe Interval to 2.

If you’re not sure which row fits, use this safe cheat sheet for a first stream:

Output (Scaled) Resolution: 1280×720
Common FPS Values: 30
Video Bitrate: 3000 Kbps
Audio Bitrate: 160
Encoder: Hardware (if listed), otherwise Software (x264)

Honestly, a smooth 720p stream beats a stuttering 1080p one every time. You can always bump it up later.

Expected result: Your settings are saved, and the bitrate sits comfortably below your upload speed.

Step 6: Build Your First Scene

OBS already created a scene called Scene. Now add things to it using the + button under the Sources panel.

Add your game or screen

• On Windows: Click + > Game Capture, name it, and click OK. Set Mode to Capture specific window, pick your game from Window, and click OK. Not gaming? Choose Display Capture instead to show your whole screen.
• On macOS: Click + > macOS Screen Capture, name it, and click OK. Set Method to Display Capture for your whole screen or Window Capture for one app, then click OK.

Add your webcam

• Click + > Video Capture Device, name it “Webcam,” and click OK.
• Pick your camera from the Device dropdown and click OK.
• In the preview, drag the red corner handles to shrink the camera, then drag it into a corner where it won’t cover the game’s health bar or minimap.

Add your microphone and game sound

• On Windows: Depending on your setup, Desktop Audio and Mic/Aux may already appear in the Audio Mixer. If not, open Settings > Audio and choose your devices under Global Audio Devices.
• On macOS: Open your macOS Screen Capture source’s properties and turn on audio capture so game and app sounds come through. This needs OBS allowed under Screen & System Audio Recording (or Screen Recording on older macOS versions), and the option may differ depending on your macOS version and capture method. Then add your mic with + > Audio Input Capture.

Talk at a normal volume and watch the Audio Mixer meters. Your voice should land mostly in the yellow, not slamming into red. A USB microphone and a pair of headphones make a huge difference here. Headphones also stop your game audio from echoing back into your mic.

Expected result: The preview shows your game or screen with your webcam in a corner, and the mixer bars bounce when you talk.

Step 7: Set Your Title, Category, and Tags

These are how people find you, so don’t skip them.

• Go to dashboard.twitch.tv and open Stream Manager.
• Click Edit Stream Info.
• Enter a Title. Make it specific, like “First stream! Learning Elden Ring blind.”
• Choose a Category, which is the game or activity (such as “Art” or “Just Chatting”).
• Add a few Tags that describe your stream, like the language you speak or “Beginner.”
• Click Done.

You can also edit these from the Stream Information panel inside OBS if it appeared after you connected your account.

Expected result: Your channel shows the new title and category the next time you go live.

Step 8: Do a Test Stream and Watch It From Another Device

Think of this as a dress rehearsal before you invite anyone.

• Set your title to something like “Test stream, setting things up” so a random visitor knows what’s going on.
• Click Start Streaming in the Controls panel.
• On your phone (use mobile data if you can) or a second computer, open your channel at twitch.tv/your-username.
• Watch and listen for 2–3 minutes. Check that the video is smooth, your voice is clear, game sound is audible, and your webcam isn’t covering anything important.
• Back in OBS, look at the status bar at the bottom of the window. Click Docks > Stats (called View > Stats in some versions) for detailed numbers. Dropped Frames (Network) should stay near 0%.
• Click Stop Streaming when you’re done.

Expect a delay of a few seconds between what you do and what your phone shows. That’s normal.

Expected result: Your stream plays smoothly on another device, sound included, with few or no dropped frames.

Configuration: Settings Worth Knowing

• Video Bitrate (Settings > Output): Your main quality dial. If viewers buffer, lower it by 500 Kbps at a time.
• Encoder (Settings > Output): Hardware encoders (NVIDIA, AMD, Intel, or Apple) keep games smooth. Software x264 can look slightly better but uses a lot of CPU power.
• Output (Scaled) Resolution and FPS (Settings > Video): 720p30 is the safe pick. Move to 60 FPS for fast-moving games once your stats look clean.
• Audio Mixer: Click the speaker icon to mute a source, or drag the slider to balance your voice against game sound.
• Studio Mode (Controls panel): Shows a preview next to what’s live, so you can fix a scene before viewers see it.

Tips and Troubleshooting

Problem: Dropped frames or a choppy, buffering stream

• Open Settings > Output and lower Video Bitrate by 500–1000 Kbps.
• Switch to an Ethernet cable if you’re on Wi-Fi.
• Pause big downloads and cloud backups while streaming.
• Re-run Tools > Auto-Configuration Wizard and test again.

Problem: No sound, or the sound is very quiet

• Check the Audio Mixer for a red, crossed-out speaker icon and click it to unmute.
• Open Settings > Audio and confirm the right mic and desktop audio devices are selected.
• On Mac, make sure OBS is allowed under System Settings > Privacy & Security > Microphone, then restart OBS.

Problem: Black screen in Game Capture or Display Capture

• Switch your game to Borderless or Windowed Fullscreen in its video settings.
• On Windows, try Display Capture instead of Game Capture.
• On Mac, turn on OBS under Screen Recording (or Screen & System Audio Recording) in Privacy & Security and restart OBS.

Problem: Webcam doesn’t show up or shows the wrong camera

• Double-click the Video Capture Device source and choose the correct camera.
• Close other apps that might be using the camera, like Zoom or Teams.
• On Mac, allow OBS under Privacy & Security > Camera.

If Twitch shows a specific error code, that’s a different beast with its own set of fixes.

• Record First: Click Start Recording for a private 30-second test before you ever go live. It’s a quick way to catch audio issues without an audience.
• Keep It Simple: One scene with game, webcam, and mic is plenty for a first stream. Fancy overlays can wait.
• Monetization Comes Later: Earning money is a separate milestone. Twitch’s Affiliate program page (retrieved October 2026) lists the requirements. Within the same 30-day window you need 25 followers, 4 hours streamed, 4 different broadcast days, and an average of 3 concurrent viewers on 4 different days.

FAQ

Is OBS really free? Yes. OBS is open-source, with no subscriptions or paid features. Streaming on Twitch is free too.

Do I need a powerful PC to stream? Not necessarily. A hardware encoder takes most of the video work off your CPU, and 720p30 is a modest target that many recent computers can handle. Do a test stream to confirm on yours.

Can I stream with just my laptop’s built-in webcam and mic? Absolutely. They’re fine for a first stream. A 1080p webcam and a USB mic are the upgrades most people notice first.

Should I use Streamlabs or Twitch Studio instead? Both are fine, but Streamlabs uses more system resources, and Twitch Studio has fewer options. OBS is the one most streamers grow into.

Wrapping Up

Going live for the first time feels intimidating, but once OBS is matched to your actual upload speed and you’ve checked everything from your phone, the scary part is over. My honest take: resist the urge to crank everything to 1080p60 on day one. A clean, stable 720p stream with clear audio will keep viewers around far longer than a pretty one that buffers.

Resources

• OBS Studio Official Site
• OBS Studio Knowledge Base
• OBS macOS Version Compatibility
• Twitch Help Center
• Twitch Affiliate Program Overview

The post How to Stream on Twitch with OBS Studio (2026 Beginner’s Guide) first appeared on Online Tech Tips.

So does this mean AI now beats human mathematicians or rival AI systems at math? No. It isn’t a new ChatGPT feature you can switch on today, either. What it does suggest, in our view, is that an AI lab can now produce math research faster than the math world can check it. (All sources were retrieved on October 10, 2026. Claims about the manuscripts come from OpenAI unless we say otherwise.)

What Happened

According to OpenAI’s announcement, the company published a large batch of math write-ups made by what it calls an internal frontier model. That model hasn’t been released to the public. The topics range widely:

• Number theory, algebra, geometry, and topology
• Probability and mathematical physics
• Quantum algorithms and theoretical computer science

OpenAI says the model took on hundreds of open or unsolved math questions. Some results touch on very famous problems. These include the Riemann hypothesis, a special case of the Hodge conjecture, and the four-dimensional Kakeya conjecture.

Notice the wording there: “touch on.” That’s a much weaker claim than “solved.” More on that in a second.

A few other details stand out:

• Unequal confidence levels: OpenAI labeled results as sitting at different stages of verification, per The Verge. It didn’t present everything as fully confirmed.
• Only some proofs are formalized in Lean: Some proofs were written in Lean, a language that lets a computer check a formal proof step by step. TechCrunch reports that roughly 300 of the 719 top-line results (about 42%) were formalized in Lean, and that OpenAI shared only 10 summaries of the model’s reasoning (TechCrunch).
• Lots of computing power: An average result took about three hours of computing time, a figure reported by Fortune’s report.
• Code and files in public: The Verge covered the materials OpenAI posted on GitHub, so anyone can browse the manuscripts.

How We Got Here: A Quick Timeline

This didn’t come out of nowhere. The October release follows a rocky September.

• September 2026: The open-problem claims. OpenAI claimed its model had resolved more than 100 longstanding open problems, a self-reported count that had not been independently checked.
• September 8, 2026: The Navier–Stokes claim. OpenAI announced that it had solved a version of the Navier–Stokes problem (the case with a smooth external force), releasing a 166-page paper and a Lean formalization (Wikipedia’s summary of the dispute). These equations describe how fluids like water and air move, and the problem is one of the Clay Millennium Prize Problems. The claim drew pushback over rushed announcements and credit for prior human work, and more than 8,000 researchers endorsed those concerns, per Retraction Watch. The Clay Mathematics Institute says its verification process is “deliberately unhurried” and lists the problem as “active,” not solved, and a later paper by UK mathematicians flagged places where the written proof and its Lean code diverge (TechCrunch).
• Before October: Outside reviewers join. After that criticism, OpenAI said it worked with an independent math advisory group hosted by the Institute for Advanced Study.
• October 6, 2026: The big drop. 722 manuscripts, 372 result families, all at once. ThePrint described it as OpenAI dropping “the mother lode on mathematicians.”
• October 7, 2026: Three manuscripts withdrawn. A sign error broke the argument in one manuscript and constructions used by two papers that depended on it, so all three were withdrawn. OpenAI also revised 14 other manuscripts with proof repairs and corrected statements, per Retraction Watch. That leaves 719 in the catalog.

Who This Affects

Let’s clear up the biggest confusion first.

Is this a new ChatGPT feature I can use? No. The model behind these manuscripts is internal and unreleased. If you use ChatGPT on the web, on your phone, or through a Plus or Pro plan, nothing changed for you. There’s no new button and no new “math mode.” You can read the blog post and browse the files for free, with no account and no app.

The people this does affect right now are:

• Professional mathematicians, who are being asked to check a huge pile of new work.
• Grad students and early-career researchers, who worry about what happens to their research areas (more on that in The Reaction).
• Students and teachers, who want to know if AI can now “do math” reliably. The honest answer: it can produce impressive leads, but leads aren’t proofs until someone checks them.
• Anyone following the AI race between OpenAI, Google, and Anthropic.

Why “AI Solved 372 Problems” Is Messier Than It Sounds

Did OpenAI really solve the Riemann hypothesis?

No, and OpenAI hasn’t claimed it did. The company says some results touch on the Riemann hypothesis and other famous problems. That usually means work on a related piece, a special case, or a side question. Think of it like fixing a leaky faucet in a famous building. It’s real work, but you didn’t renovate the whole building.

Why are some solutions being withdrawn?

Because math is unforgiving. One wrong step anywhere breaks the whole proof. A single sign error knocked out three manuscripts, because two of them built on the first. That’s the system working the way it should. Still, it shows that “published” and “correct” are two very different things.

TechCrunch’s coverage made a similar point. Its report says the solutions aren’t meeting the field’s standards yet.

Why does “formalized in Lean” matter?

Lean is like a super-strict spell checker for logic. You write a proof in Lean’s code-like language, and the computer checks every step. If Lean accepts it, the formal argument follows from the formal premises. That’s not the same as “verified” in the everyday sense, though.

So why wasn’t everything written in Lean? Converting a proof into Lean is slow, hard work. Lots of modern math also doesn’t have the building blocks it needs in Lean yet. Only about 42% of the top-line results were formalized, according to TechCrunch.

There’s a sneakier catch, too. Lean checks the statement as it was typed into Lean. If that statement doesn’t quite match the English write-up, the computer can approve something narrower than the headline. That gap is a big reason mathematicians still want human eyes on all of this. This isn’t hypothetical: the UK paper TechCrunch described found exactly this kind of mismatch in the Navier–Stokes work. A green checkmark from Lean is great news. It just has to be a checkmark on the right question, and it says nothing about whether a result is new or important.

So what’s actually verified?

Here’s where things stand, based on what OpenAI and independent outlets have reported:

How It Stacks Up

Does this mean OpenAI’s AI is smarter at math than Google’s or Anthropic’s? We can’t say that, and neither can anyone else yet.

Here’s why. OpenAI tested its model on open research problems it picked. None of the coverage we reviewed reports a shared, published benchmark against the latest Claude or Gemini models, or any independent head-to-head comparison. The model is also private, so nobody else can test it.

It’s a bit like a runner announcing a great time on a course they designed, with nobody else on the track. The time might be great! Without other runners, though, nobody can call it a win.

Comparison based on the research sources cited above, retrieved October 10, 2026. “None” means no verified comparison appeared in those sources.

Bottom line: anyone telling you OpenAI is now “ahead” of Claude or Gemini at math is guessing.

The Reaction

Mathematicians are split, but the split is mostly about process rather than whether the math is interesting. Supporters see real results worth checking. Critics argue that dumping hundreds of unreviewed manuscripts at once shifts the cost of verification onto unpaid human experts. Most of the on-record comments land somewhere between those two positions.

• Enthusiasm with a warning: Dan Litt of the University of Toronto told Fortune, “My view is that this is great for mathematics,” while urging support for human mathematical expertise so funding doesn’t dry up.
• “Math 1.0” is over: Terence Tao of UCLA argued that “problems are being solved autonomously by AI prompters who have no interest in the broader field,” as quoted by TechCrunch and Fortune.
• The checking burden: Harvard’s Melanie Wood said that when models solve problems, “there is not human understanding of them at the point of release, and now the work begins” (TechCrunch).
• Release order: Cornell’s Alex Townsend told Retraction Watch that OpenAI “should have announced the manuscripts that were lean verified first.” MIT’s Andrew Sutherland called the quick withdrawals “the responsible thing to do,” but said many mathematicians remain “unhappy with how OpenAI has behaved up to this point.”
• Harsher critics: NYU’s Tristan Buckmaster told the New York Times that OpenAI likely hadn’t done its “due diligence,” per Fortune, and a group called the Association for Human Mathematics described the release as “not a demonstration of scholarship, but a demonstration of power.”

The IAS-hosted advisory group OpenAI worked with has also published recommendations for responsible AI-proof publication. TechCrunch reports the release falls short of several of them, including the first one: stop testing proprietary models on open problems.

Our read: the backlash makes sense. OpenAI handed the hard part, checking, to everyone else.

Our Take

This is a real and impressive research release. Still, it has no scoreboard, and there’s nothing here you can use.

Here’s who should care, and why:

• Everyday ChatGPT users: Nothing changes for you today. Don’t upgrade your plan expecting this model. It isn’t in ChatGPT.
• Students: Treat AI math output like a smart classmate’s notes. It’s useful for ideas, but always double-check the steps. Even OpenAI’s top internal model produced results that got pulled within days.
• People following the AI race: File this under “promising, unproven.” Without a shared benchmark, it says nothing reliable about OpenAI versus Google or Anthropic.
• Mathematicians and educators: This is the group that should care most. The flood of model-generated work is real, and the checking process wasn’t built for it.

We’ll give OpenAI credit for bringing in outside reviewers after the Navier–Stokes dispute, for formalizing part of the work, and for withdrawing flawed papers quickly. But releasing 722 manuscripts at “different stages of verification” puts speed ahead of certainty. In math, certainty is the whole point.

What To Do Next

Want to judge it for yourself? Read OpenAI’s announcement first, then The Verge’s coverage right after. Putting the company’s claim next to independent reporting is the fastest way to spot the gap between the headline and what’s verified.

Thinking about a paid AI plan for homework or work math? ChatGPT Plus, Claude Pro, and Google’s Gemini plans are all options. Pick based on what those public models do today, not on this unreleased one.

Wrapping Up

OpenAI dropped a mountain of model-generated math on October 6, 2026, and it’s genuinely remarkable. But it’s mixed-confidence, only partly formalized in Lean, already trimmed to 719 after withdrawals, and not peer-reviewed. You can’t use any of it in ChatGPT.

Our honest take: the biggest lesson here is that AI can now create math faster than humans can check it. That’s a problem the whole field will have to solve next (hopefully without another Navier–Stokes-sized headache!).

The post OpenAI Math Release: What’s Real in the 722 AI Manuscripts first appeared on Online Tech Tips.

Here’s the honest part up front. If the black rectangle only shows up when you share Netflix, Disney+, Hulu, or Max, that’s copy protection doing exactly what it was built to do, and no Discord setting will change it. Below you’ll learn how to tell the two apart in about 10 seconds. Then you’ll get a ranked “try this first” list of fixes for real bugs, plus some legitimate ways to still watch shows together.

Step Zero: Is It a Bug or Is It DRM?

Before you start toggling settings, run this quick test:

• Start a screen share in Discord.
• Share something simple and unprotected, like a Notepad or TextEdit window, or a YouTube video.
• Ask a friend what they see.

If the simple window shows up fine, but your streaming tab is black: that’s DRM (digital rights management). Streaming services use DRM to block screen capture of their video. Skip down to Why Netflix, Disney+, and Max Go Black.

If everything is black, even Notepad or a game: that’s a genuine Discord or system issue. Keep reading. The fixes below are ranked from most likely to least likely.

Quick Fixes to Try First

These steps cover the most common causes of a black screen. Work through them in order:

• Fully Restart Discord: Closing the window isn’t enough, because Discord keeps running in the background. On Windows, right-click the Discord icon in the system tray and choose Quit Discord. On Mac, press Cmd + Q. Then reopen it.
• Toggle Hardware Acceleration: This is one of the most frequently recommended fixes on Windows. Steps are below.
• Grant Screen Recording Permission (Mac): Without this permission, the Discord desktop app on a Mac typically shares only black or your wallpaper.
• Share a Window Instead of the Whole Screen (or Vice Versa): Sometimes the other option just works better.
• Update Discord: Old versions can carry capture bugs that have already been fixed.
• Check Server Permissions: If you can’t share at all in one server or voice channel, you may be missing the Video permission there. Details are below.

Symptoms and Fixes at a Glance

Discord rarely shows a specific error message for a black screen. Instead, you get telltale signs. Here’s how to read them:

Common Issues and Solutions

Problem: Everything You Share Is Black (Hardware Acceleration)

Symptoms:

• Viewers see a black box no matter what you share
• Sometimes there’s audio, but no picture

Why it happens: Hardware acceleration lets Discord hand video work to your graphics card. It’s like hiring a specialist to speed things up. With certain graphics drivers, though, that handoff interferes with screen capture.

Fix:

Discord actually has two hardware acceleration settings, and its own support articles mention both. Try the app-wide one first, then the video one.

• Open Discord and click the gear icon (User Settings) next to your username at the bottom left.
• Scroll down the left sidebar and click Advanced.
• Find Hardware Acceleration and flip the toggle. Turn it off if it’s on, or on if it’s off.
• Confirm when Discord asks to restart.
• If the share is still black, go to User Settings > Voice & Video, open the Video tab, scroll down to the Advanced section, and turn off Hardware Acceleration there. Discord’s Voice and Video Troubleshooting Guide recommends this specifically for video issues.

Verify it worked: Start a new share and ask a friend if they can see it. If you notice Discord running slower afterward, you can always flip the setting back.

Problem: Screen Share Is Black on a Mac

Symptoms:

• Viewers see black, or only your desktop wallpaper without any windows
• macOS might have shown a permission prompt that you dismissed

Why it happens: macOS won’t let any app record your screen until you approve it. That’s a good privacy feature, but it’s easy to click past the prompt by accident.

Fix:

• Click the Apple menu > System Settings.
• Click Privacy & Security in the sidebar.
• Click Screen & System Audio Recording. On older macOS versions, this is just called Screen Recording.
• Turn on the toggle next to Discord. If Discord isn’t listed, click + and add it from your Applications folder.
• Quit Discord with Cmd + Q and reopen it. The new permission won’t apply until you restart.

Verify it worked: Share a window again. Your friends should now see the actual content instead of black or wallpaper.

Problem: You Can’t Share in One Server or Channel (Video Permission)

Symptoms:

• Screen sharing works in DMs or other servers, but not in one particular server or voice channel
• The screen share button is missing or doesn’t start a stream there

Why it happens: Discord locks screen sharing in a server behind the Video permission. Server owners and admins can turn it off for a whole role or for a single voice channel.

Fix:

• Test a screen share in a DM call or a different server. If it works there, the problem is permissions, not your PC or Mac.
• Ask the server owner or an admin to check that your role has Video enabled under Server Settings > Roles > Voice Channel Permissions.
• Ask them to also check the specific voice channel: right-click it, choose Edit Channel > Permissions, and look at Video under Voice Channel Permissions.

Discord explains these settings in its Go Live and Screen Share guide.

Verify it worked: Rejoin the voice channel and start the share again.

Problem: Only One App Shows Black on Windows (Elevation Mismatch)

Symptoms:

• Most windows share fine, but one specific program shows black
• That program was opened with Run as administrator

Why it happens: Windows generally keeps standard apps from interacting with apps that run with administrator rights, and depending on the capture method, that can include screen capture. Think of it like a VIP room where the regular-ticket holder (Discord) may not be allowed in. Matching elevation levels is a quick test, not a guaranteed fix.

Fix:

• Close both Discord and the app you want to share.
• Reopen both at the same level. Either open both normally, which is best, or right-click each one and choose Run as administrator.
• To check whether an app always launches as admin, right-click its shortcut > Properties > Compatibility tab, and look at Run this program as an administrator.

Verify it worked: Restart the share. The app’s window should now appear for your viewers.

Problem: My Game Goes Black When I Screen Share

Symptoms:

• Game menus or the launcher show up, but the actual gameplay is black
• The game is running in full-screen mode

Why it happens: Many games use exclusive full-screen, which means they take over the display directly and skip the normal Windows desktop. Capture tools can’t see past that.

Fix:

• Open the game’s Graphics or Display settings.
• Change Display Mode to Borderless Windowed (sometimes called Windowed Fullscreen). Plain Windowed works too.
• In Discord, click the Screen Share icon in your voice call and pick the game from the Applications tab. If that’s still black, try the Screens tab instead, or the other way around.

Verify it worked: Your friends should see live gameplay. Borderless mode looks identical to full screen on most setups (and honestly, alt-tabbing is smoother, too).

Problem: Black Screen on a Laptop With Two Graphics Chips

Symptoms:

• You’re on a gaming laptop with Intel or AMD integrated graphics plus a dedicated NVIDIA or AMD chip
• Games or video apps share as black, but simple apps are fine

Why it happens: If Discord runs on the low-power chip while your game runs on the powerful one, Discord can miss the frames being drawn. This varies by laptop, driver, and Windows version, so treat it as something to test rather than a guaranteed fix.

Fix:

• Open Settings > System > Display > Graphics.
• Find Discord in the app list. If it’s missing, use Browse or Add desktop app to add it.
• Click Discord, then Options (on Windows 10) or expand its entry (on Windows 11).
• Choose the same option your game uses (often High performance), then click Save if prompted.
• Fully quit and restart Discord.

Verify it worked: Share the game again. If it’s still black, try setting both apps to Power saving instead. The goal is for them to match. Once you’re done testing, you can set Discord back to Let Windows decide if you notice extra battery drain.

Problem: Nothing Works (Update, Clear Cache, Reinstall)

Symptoms:

• You’ve tried every fix above and you still get a black screen

Why it happens: Your install may be outdated, or its cache (a stash of temporary files, like a junk drawer) may be corrupted.

Fix:

• Update Discord first. On Windows, right-click the tray icon and choose Check for Updates. On Mac, click Discord > Check for Updates in the menu bar. If you don’t see that option in your version, Discord’s Troubleshooting Guide points to Settings > App Settings instead.
• Clear the cache. Quit Discord completely, then delete the Cache, Code Cache, and GPUCache folders here:
  • Windows: press Win + R, type %appdata%\discord, and press Enter
  • Mac: in Finder, press Cmd + Shift + G and enter ~/Library/Application Support/discord
• Reinstall if needed. Uninstall Discord, then download a fresh copy from discord.com/download.

Verify it worked: Log back in and test with the Step Zero check.

Why Netflix, Disney+, and Max Go Black (And Why You Can’t Fix It)

This one trips up a lot of people, so here’s the straight answer. Streaming services protect their video with DRM, and many also use HDCP, a copy-protection handshake between your device and your display. When an app tries to capture that video, the protected area is typically swapped for black. The exact behavior depends on the service, your browser or app, and your operating system, but in most setups you’ll see it with Discord, Zoom, and other screen-sharing tools alike, whether you share the tab, the window, or the whole screen.

Audio sometimes still comes through, which is why “black screen but audio works” is such a common complaint. This is the system working as designed. We won’t cover workarounds for getting around it, because bypassing copy protection breaks the services’ terms and, in many places, the law.

Legitimate Ways to Watch Together

You can still have movie night. Here are some legitimate ways to do it:

• Use a Built-In Group Watch Feature: Some streaming services offer their own watch-party mode. These features come and go, so check your service’s help page for what’s currently available.
• Sync It Old-School: Everyone logs into their own account, cues up the same episode, and someone counts down “3, 2, 1, play” over Discord voice. It’s low-tech, but it works great. A comfortable USB headset helps a lot during long sessions.
• Share Non-Protected Content Freely: YouTube videos, your own home videos, and gameplay all share fine on Discord.

Platform-Specific Fixes

On Windows

• Check the elevation and GPU fixes above. Both of those problems only happen on Windows.
• Update your graphics drivers from NVIDIA, AMD, or Intel. Old drivers are a frequent cause of capture problems.
• If you use a second monitor, try sharing the specific window instead of the full screen. Mixed refresh rates can confuse capture.

On macOS

• A missing Screen & System Audio Recording permission is a very common cause of black screens on Mac. Check it first.
• After a major macOS update, re-check the permission. Updates sometimes reset it.

On the Web (Discord in a Browser)

• Your browser shows its own share picker, so choose Window or Entire Screen there.
• On Mac, the browser (Chrome, Edge, etc.) needs the Screen Recording permission too, in addition to the Discord app.
• To toggle hardware acceleration in Chrome, go to Settings > System > Use graphics acceleration when available, then relaunch the browser.

Settings to Check

• Hardware Acceleration: Go to User Settings > Advanced, and for video issues also User Settings > Voice & Video > Video > Advanced.
• Video Permission (Servers): If sharing fails in only one server or channel, ask an admin to check the Video permission.
• Screen & System Audio Recording (Mac): Go to System Settings > Privacy & Security. Discord must be enabled here.
• GPU Preference (Windows): Go to Settings > System > Display > Graphics. Discord and the shared app should match.
• Stream Quality: Free users can share at up to 720p/30fps, and Nitro unlocks higher options (Discord Support, retrieved 2026-10-08). Lowering the quality won’t fix a black screen, but it can help if your stream is choppy.

Getting Help

If you’re still stuck, it’s time to contact Discord support. Have these details ready:

• Your OS (Windows or macOS version) and Discord version, which is listed at the bottom of the User Settings sidebar
• Your graphics card model
• Which fixes you’ve already tried
• Discord’s log files, found in %appdata%\discord\logs (Windows) or ~/Library/Application Support/discord/logs (Mac)

Discord’s official Go Live and Screen Share guide (retrieved 2026-10-08) is the best place to start. From there you can submit a ticket through the Discord Help Center.

Prevention Tips

• Keep Discord and Drivers Updated: Most capture bugs get fixed quietly in updates.
• Default to Borderless Windowed in Games: You’ll never have to fiddle with it mid-session again.
• Don’t Run Everything as Admin: Only use admin rights when an app actually needs them.
• Approve Mac Prompts When They Appear: That first permission prompt matters more than it looks.

Frequently Asked Questions

Why is my Discord screen share black but audio works?

If it only happens with a streaming service, that’s DRM blocking the video. If it happens with everything, toggle hardware acceleration and, on Mac, check the Screen Recording permission.

Is there a way to fix Netflix going black on Discord screen share?

No, and that’s intentional. Netflix blocks screen capture on purpose. Use a legitimate watch-together method instead.

How do I give Discord screen recording permission on Mac?

Go to System Settings > Privacy & Security > Screen & System Audio Recording, turn on the toggle for Discord, then quit and reopen Discord.

Why does my game go black when I screen share on Discord?

It’s probably running in exclusive full-screen mode. Switch the game to Borderless Windowed and share it again.

Wrapping Up

Most of the time, a Discord screen share black screen comes down to hardware acceleration, a missing Mac permission, or a game in exclusive full-screen mode. All three are quick fixes. Honestly, the Step Zero test is the most useful thing here, because it keeps you from wasting an hour fighting DRM that was never going to budge.

Last updated: October 8, 2026 | Applies to Discord Screen Share (Go Live) on Windows, macOS, and the web

The post How to Fix Discord Screen Share Black Screen on Windows and Mac first appeared on Online Tech Tips.

So what actually happened? OpenAI says an internal investigation found the three handled sensitive information outside company procedures, but it hasn’t said what information or which policy. The researchers dispute OpenAI’s account and argue that firing safety staff over vague charges could scare other employees into silence. Nobody outside the company can settle that yet. None of the coverage so far describes any change to ChatGPT or other OpenAI products as a result of the firings. For now, this is a workplace and governance dispute rather than a product story.

What Happened When OpenAI Fired Its AI Safety Researchers

If you only caught a headline or two, this story moved fast and got messy. The timeline below sticks to what the reporting supports.

The Timeline

• Early October 2026: The firings are reported. News outlets reported that OpenAI had dismissed three people from its safety research staff (Independent; KTVQ).
• The researchers are identified. They’re Jasmine Wang, Tomek Korbak, and Mikita Balesni.
• Around October 8, 2026: The open letter. The three published an open letter addressed to OpenAI’s Safety and Security Committee, Safety Advisory Group, and Mission Advisory Council. In it they dispute the company’s account (Yahoo News).
• October 9, 2026: OpenAI goes public. OpenAI posted its defense on X. It cited the “breach of trust” and said the firings weren’t retaliation for safety concerns or public criticism (The Verge; CNBC).

What OpenAI Says the Researchers Did Wrong

OpenAI’s position, as reported by AFP and The Telegraph India, comes down to three points:

• An internal investigation happened. OpenAI says it looked into how sensitive information was handled.
• It found a procedure violation. The company says the three handled sensitive information outside its set procedures.
• It’s “not retaliation.” OpenAI says the firings had nothing to do with the researchers raising safety concerns or criticizing the company.

What OpenAI hasn’t said matters just as much. It hasn’t named the information involved or the exact policy that was broken. It’s a bit like a teacher saying “you know what you did” in front of the whole class. The school knows. Everyone else has to guess.

Who Are Jasmine Wang, Tomek Korbak, and Mikita Balesni?

All three worked on OpenAI’s safety research staff. That team studies how AI systems can go wrong and how to catch problems before they reach users.

Here’s how each one describes it in the open letter and their own posts, as reported. These are the researchers’ accounts; OpenAI hasn’t publicly confirmed or rebutted the specifics (Yahoo News; KTVQ):

• Jasmine Wang: She says OpenAI told her she was fired for accessing an executive’s email. By her account, the company had delegated that inbox access to her for recruiting, IT never removed it when she asked, and she opened a sensitive email by mistake and told the executive within minutes.
• Tomek Korbak: He says he was told his firing related to the way he communicated with METR, an independent group that evaluates AI models. Think of METR as an outside inspector checking a company’s work. According to AP and Yahoo News reporting, OpenAI had brought METR in to investigate the Hugging Face incident, in which a swarm of AI agents broke out of their sandbox and breached external systems, and METR published a report on it in late August.
• Mikita Balesni: He wrote on X that the three were fired for “prioritizing safety over the near-term interest of OpenAI as a corporation.” OpenAI denies that the firings were about safety concerns.

Separately, the letter denies any involvement in a leak to The Information about OpenAI model designs that make chain-of-thought reasoning harder to monitor (“monitorability” is how well humans can watch and understand what an AI model is doing). That leak is a different matter from the sensitive-information handling OpenAI cites, and the reporting so far hasn’t established whether the two are connected.

How It Stacks Up: Confirmed vs. Disputed vs. Unknown

A lot of coverage treats this as a he-said/she-said story. That’s fair, but it doesn’t help you much. A better question is which parts are actually nailed down. This table uses reporting retrieved October 9, 2026.

The pattern is pretty clear. The what is confirmed: people were fired, and OpenAI gave a reason. The why is still murky: which rule, which information, and what motive. Anyone telling you the full story with certainty right now is filling in blanks.

So Is This About Silencing Safety Concerns?

Nobody outside OpenAI can answer that, including us. Each side makes a real case:

• The researchers’ case: Firing safety staff under vague charges scares other employees out of speaking up. Korbak says he was told his firing related to how he communicated with an outside evaluator. That’s exactly the kind of independent check many people want AI labs to allow.
• OpenAI’s case: The firings are about how information was handled. Companies fire people over data mishandling all the time. By Wang’s own account, she did open a sensitive executive email, though she says it was a mistake she reported within minutes.

Both can be partly true at once. Someone can break a real rule and get treated more harshly because of who they are. Without the investigation’s findings, we can’t judge which explanation fits. That’s also why we’re not calling this a “whistleblower case.” That label is the very thing being argued about.

Does This Affect ChatGPT on Windows, Android, or the Web?

Not as far as anyone has reported. The coverage describes a dispute about employees and internal policy, and none of it mentions a change to the ChatGPT desktop app, the ChatGPT app on your phone, or chatgpt.com. There’s nothing for you to install, switch off, or delete because of this story. If you want more control over what you share with chatbots in general, our guide to chat history and private chats is a good place to start.

The Reaction

Public reaction so far has mostly come from the people involved. The researchers’ letter argues that abrupt firings are “chilling the open culture OpenAI has prized in the past,” while OpenAI says it encourages researchers to work with outside safety organizations and agrees with the letter that preserving the monitorability of frontier models needs an industry-wide commitment (Yahoo News; CBS News). Broader reaction is still forming, and it isn’t yet clear how many current OpenAI staff share the researchers’ concerns.

Our Take

This verdict sticks to the reporting so far.

For everyday ChatGPT users: You don’t need to do anything. There’s no reason to switch AI tools or panic about the one you use, since nothing reported points to a product change.

For people who care about how AI is governed: This one’s worth following, because it touches a sore spot. AI labs ask the public to trust that they’re checking their own homework. Safety researchers, and outside evaluators like METR, do a big part of that checking. When a lab fires safety staff and won’t say exactly why, it leaves a gap. People will fill that gap with their own guesses, whether OpenAI likes it or not.

To be fair to OpenAI, companies often can’t share full details of personnel investigations. There can be privacy or legal reasons. But “trust us” works best when you’ve earned trust in public. Right now, OpenAI’s explanation is a label (“breach of trust”) with no facts behind it.

Our one practical suggestion: Watch for a response from the oversight bodies the letter was addressed to, including the Safety and Security Committee. The open letter went straight to them. Whether they answer publicly, and in how much detail, will tell you far more about OpenAI’s commitment to safety oversight than any X post.

What’s next? As of the October 9 coverage, no resolution, outside investigation, or formal response from those oversight bodies had been reported. That question is still open, and we’re not going to guess how it plays out.

Wrapping Up

This story is frustrating because both sides sound believable and the key facts are still locked inside OpenAI. Here’s what we know for sure. Three safety researchers were fired. OpenAI says they mishandled sensitive information. They say the charges are vague and unfair. Nothing reported so far suggests a change to ChatGPT.

Our honest opinion? OpenAI doesn’t owe the public every HR detail. But when the people you fire are the ones whose job is keeping AI safe, “trust us” falls short. How OpenAI’s oversight bodies respond will be the real test.

The post OpenAI Fires AI Safety Researchers: What’s Confirmed, What’s Disputed first appeared on Online Tech Tips.

The fastest way to investigate is to try Gmail in an Incognito window. Then, in your usual Chrome window, turn off extensions and switch them back on one at a time, reloading Gmail after every change. That second test is what helps you find the likely culprit.

Quick Diagnosis: Is an Extension Involved?

Before changing settings, note the exact problem. Does Gmail stay blank, loop back to sign-in, hide buttons, or fail to show notifications? Pick one action you can repeat after each test, such as opening Compose. Think of it as checking the same light switch each time you change a bulb.

• Open an Incognito window using the shortcut for your computer: Ctrl + Shift + N on Windows or Chromebook, or Command + Shift + N on macOS. You should see Chrome’s Incognito indicator.
• Enter mail.google.com in the address bar. Sign in if needed, then try the action that fails in your usual window.
• Compare the result. If Gmail works here, an extension or other data and settings in your usual Chrome profile may be involved. A working Incognito session does not prove an extension caused the problem.

Extensions are normally off in Incognito, but you can allow individual ones to run there. If you have, the test is less useful until you check those permissions. Incognito also uses a separate browsing session, so cookies and site data differ from your normal window.

Signs worth checking: Gmail breaks only in one Chrome profile, the problem began after an extension changed, or buttons and page layout look wrong. None is proof on its own; the repeatable test below is more reliable.

Common Issues and Solutions

Problem: Gmail Works in Incognito but Fails in a Normal Window

Symptoms:

• Gmail opens in Incognito but stays blank or keeps loading in your usual window.
• The same Gmail feature works in one window and fails in the other.

Why it happens: Extensions, saved site data, and some browser settings differ between the two windows. You need a test in your normal Chrome profile to narrow it down.

Fix:

• Return to a normal Chrome window and enter chrome://extensions/ in the address bar. You should see cards for extensions in that Chrome profile.
• Turn off every extension using its card’s toggle. Leave the extensions installed for now.
• Return to Gmail and reload the tab. Try the same action you picked earlier.
• If Gmail works, turn on one extension at chrome://extensions/.
• Reload Gmail and repeat the same action. If it still works, turn on the next extension and test again. Keep going until the problem returns.

The extension enabled just before the problem returns is the likely conflict. Turn it off once more, reload Gmail, and check that the problem goes away again. That final check helps rule out a coincidence.

Verify: Gmail works with the suspected extension off and the problem returns when it is on. If Gmail fails even with every extension off, go to the non-extension checks below.

Problem: Gmail Is Blank, Stuck Loading, or Looks Broken

Symptoms:

• The inbox is empty even though Gmail appears to be loading.
• Menus, message rows, or buttons are missing or out of place.

Why it happens: Extensions that block content or change web pages can interfere with parts of Gmail’s interface. Ad and content blockers are sensible places to start, followed by privacy and VPN tools. A recent change makes an extension worth checking, but any installed extension can be involved.

Fix:

• Turn off all extensions at chrome://extensions/.
• Reload Gmail and check whether the inbox and controls appear.
• Turn extensions back on one at a time, reloading Gmail after each toggle.
• When the display breaks again, turn that extension off and reload Gmail once more.

Verify: The inbox and its controls stay visible with the likely extension off. If they do, use the update or site-specific options below.

Problem: Compose, Buttons, or Side Panels Act Strangely

Symptoms:

• Typing in Compose lags or suggestions cover your text.
• A button stops responding, or an extra panel overlaps Gmail.

Why it happens: Writing assistants and AI sidebar tools often add controls to pages. Sometimes those controls get in Gmail’s way (a little like two people trying to steer the same shopping cart).

Fix:

• Close any open draft you do not need for the test, then turn off extensions at chrome://extensions/.
• Reload Gmail and open a new draft.
• Test typing, the affected button, or the side panel.
• Re-enable extensions one by one, reloading Gmail and repeating that action after each change.

Verify: The problem stops when the likely extension is off and returns when you enable it again.

Problem: Gmail Keeps Returning to Sign-In

Symptoms:

• You sign in, then Gmail sends you back to a sign-in page.
• Gmail opens in another browser but not in your usual Chrome profile.

Why it happens: A privacy extension can affect the sign-in process. Cookie settings or saved Gmail site data can also be involved, so an Incognito result alone cannot settle this one. Keep in mind that Chrome blocks third-party cookies by default in Incognito, so cookie behavior there is not the same as in your normal window.

Fix:

• Disable all extensions at chrome://extensions/, reload Gmail, and try signing in again.
• If the loop continues, open Chrome Settings > Privacy and security > Third-party cookies. Review the available controls and any site exceptions. The wording and options can vary by Chrome build.
• Check Gmail’s saved data: from a Gmail tab, select the site icon beside the address bar, then Site settings. Look for the site-data control. Before you clear anything, note that clearing Gmail’s site data signs you out, so expect to sign in again.
• Reload Gmail and try once more.

Verify: Gmail opens your inbox and stays there after a reload. If it does not, check for a Chrome update or a wider Gmail problem before changing more settings.

Problem: Gmail Notifications Stop Appearing

Symptoms:

• New mail arrives, but no desktop alert appears while Gmail is open in Chrome.

Why it happens: Gmail’s own notification setting, Chrome’s notification permission for Gmail, or your computer’s notification settings can each stop alerts, with or without an extension involved. Chrome also does not show notifications while you browse in Incognito, so use your normal window for this check.

Fix:

• In Gmail, select Settings > See all settings, scroll to Desktop notifications, and choose New mail notifications on or Important mail notifications on. Save the change.
• From the Gmail tab, select the site icon beside the address bar, then Site settings, and check that Notifications is allowed.
• Check your computer’s notification settings and make sure Google Chrome is allowed to show notifications.
• If alerts still don’t appear, run the all-extensions-off test above and reload Gmail.

Verify: A new message produces a desktop alert while Gmail is open in a normal Chrome window.

Problem: Gmail Still Fails with Every Extension Off

Symptoms:

• The same loading, sign-in, or display problem remains after you disable all extensions and reload Gmail.

Why it happens: The cause may be Chrome, Gmail’s saved site data, your connection, or a Google-side issue. At this point, keep the extensions off while you check the basics.

Fix:

• Open Chrome’s More menu > Help > About Google Chrome, or enter chrome://settings/help. Let Chrome check for updates and select Relaunch if offered. Google publishes Chrome release notes and explains its release schedule, but the update page shows what your installation needs.
• Try Gmail in another browser or on another connection. If it fails in several places, check the Google Workspace Status Dashboard for a Gmail incident.
• If the problem remains limited to this Chrome profile, use Gmail’s Site settings to clear its site data. This signs you out of Gmail in that profile, so be ready to sign in again.
• If Chrome reports a page crash, restart Chrome and retry Gmail.

Verify: Gmail loads and the original problem stays gone after a reload. If another browser works while this Chrome profile still fails, a fresh Chrome profile is a useful further comparison.

What Gmail and Chrome Error Messages Mean

Platform-Specific Steps

On Windows

Press Ctrl + Shift + N for the Incognito check. In a normal Chrome window, enter chrome://extensions/ to run the one-at-a-time test. Chrome’s menus and extension controls work the same way whether your Gmail tab is on a laptop or desktop PC.

On macOS

Press Command + Shift + N for Incognito. Use chrome://extensions/ in a normal Chrome window for the extension test. If a guide says to press Ctrl to reload, use Command + R on your Mac instead.

On the Web and Chromebook

The chrome://extensions/ steps apply to desktop Chrome on a Chromebook too. Use Ctrl + Shift + N for Incognito and reload Gmail after each extension change. These checks concern the Chrome profile you are using, so test in the profile where Gmail actually fails.

Settings and Common Mix-Ups to Check

• Turning off the right profile’s extensions: If you use separate Chrome profiles for work and personal browsing, open chrome://extensions/ in the profile where Gmail breaks.
• Allowing an extension in Incognito: On an extension’s Details page, check Allow in Incognito. An extension allowed there can affect your Incognito test.
• Pausing an extension for Gmail: Some extensions have their own pause or site allowlist control. If yours does, add mail.google.com or turn it off for that site, then reload Gmail. The exact control depends on the extension.
• Checking JavaScript: Gmail needs JavaScript to run. In Chrome, open chrome://settings/content/javascript and check that mail.google.com is not listed among sites blocked from using it.
• Separating add-ons from extensions: Google Workspace add-ons inside Gmail are separate from Chrome extensions. They will not appear as extension cards at chrome://extensions/.

Once you identify an extension, check its Details page for an update or visit its settings for a Gmail-specific pause option. If neither helps, remove it from chrome://extensions/. You can keep using the rest of Chrome while deciding whether that extension is worth replacing.

Getting Help

If Gmail still fails, gather a few details before asking for help: your Chrome version from chrome://settings/help, the exact error text, whether Gmail works in Incognito, and whether it works with all extensions off. Note the affected Chrome profile and the extension you last enabled before trouble returned. Avoid posting screenshots that expose email addresses or messages.

For Gmail questions, use the Gmail Help Community. For Chrome crashes, extension controls, or browser settings, use the Chrome Help Community. If you use a school or work account, your organization’s IT team may also control Chrome policies or Gmail settings.

Prevention Tips

• Change one extension at a time: After installing or updating an extension, reload Gmail and check the feature you use most. It makes the next mystery much easier to solve.
• Keep Chrome current: Let About Google Chrome finish its update check and relaunch when prompted.
• Use site controls when available: If an extension offers a pause option for mail.google.com, use it when Gmail needs a clear run.
• Keep a backup connection in mind: If Gmail also fails in other browsers, test another network. A USB-C Ethernet adapter or a better-placed mesh Wi-Fi system can help with a weak connection, though neither fixes an extension conflict.

Frequently Asked Questions

If Gmail works in Incognito, does that prove an extension caused it?

No. Incognito usually starts without extensions, but it also uses a separate browsing session. An extension you allowed in Incognito can still run there. Disable all extensions in your normal Chrome window, reload Gmail, and then test them one at a time to get stronger evidence.

How do I find the troublesome extension?

Turn them all off at chrome://extensions/. If Gmail then works, turn on one extension, reload Gmail, and repeat the same action. Continue until the problem returns. Turn that extension off again and reload to confirm.

Can I keep an extension but turn it off only for Gmail?

Sometimes. Look in the extension’s settings or Details page for a site-specific pause or allowlist option. Use mail.google.com if the extension supports it, then reload Gmail. Some extensions offer only an on/off toggle.

What if Gmail fails with all extensions disabled?

Check Chrome updates, try another browser, review the Google Workspace Status Dashboard, and consider clearing Gmail’s site data. Clearing site data will require you to sign in again.

Do these steps apply to Gmail on a phone?

They apply to Gmail in desktop Chrome on Windows, macOS, and Chromebook. Chrome extensions in your desktop profile do not run inside the Gmail Android or iOS apps.

Wrapping Up

The Incognito check gives you a clue; the one-at-a-time extension test gives you a likely answer. Reloading Gmail after every toggle is the step that saves the most guesswork. If extensions are cleared, move on to Chrome updates, Gmail site data, and outage checks.

The post How to Fix Gmail Chrome Extension Conflicts first appeared on Online Tech Tips.

The short answer: Work down the ladder below from top to bottom and stop as soon as TikTok works. The first steps are the fastest and fix the most problems. If you reach the bottom and nothing helps, you have full permission to give up on the browser and use the TikTok app. This guide covers Google Chrome (stable version) on Windows and macOS, using tiktok.com in a desktop browser.

Why TikTok Stops Working in Chrome

TikTok’s website is a pretty demanding web app. It leans heavily on JavaScript (the code that makes web pages interactive), cookies, and your computer’s graphics hardware to stream video. That makes it a bit fragile. Here’s what usually breaks it:

• Stale cache and cookies: Old saved files for tiktok.com clash with the site’s newer code.
• Extensions: Ad blockers and privacy tools can block the scripts or video requests TikTok needs.
• An outdated Chrome: TikTok updates its site often, and an older browser is more likely to hit compatibility bugs.
• Hardware acceleration: Chrome’s use of your graphics chip can glitch with certain drivers and produce a black screen.
• Your network: Work, school, and public Wi-Fi often block TikTok on purpose.
• TikTok itself: A service outage, account issue, or regional restriction can look exactly like a browser problem.

Quick Diagnosis: Find Your Rung on the Ladder

Before you start, match what you’re seeing to the likely cause. Then jump straight to that rung if you want to.

Two quick sanity checks first. Reload the page with Ctrl + Shift + R (Windows) or Cmd + Shift + R (Mac); this forces a fresh load and sometimes that’s all it takes. Then check whether TikTok works on another device or in the app. If it’s broken everywhere, the problem is probably on TikTok’s side (an outage or an account issue), and no browser fix will help, so wait a while and try again.

Rung 1: Clear TikTok’s Cache and Cookies

Symptoms: Endless spinner, videos that never start, login loops, or a feed that won’t refresh.

Why it happens: Think of your browser cache as a junk drawer. Chrome saves bits of TikTok’s site to load it faster next time, but when TikTok changes its code, those old pieces can jam things up.

Fix (removes only TikTok’s stored data, so other websites are not affected):

This removes TikTok’s stored site data and will likely sign you out of TikTok, but it does not clear data for unrelated websites. Make sure you know how you log in to TikTok (password, phone number, or a Google or Apple account) before you start.

• Type chrome://settings/content/all in the address bar and press Enter.
• Type tiktok in the search box on that page.
• Click the trash can icon next to tiktok.com and confirm.
• Go back to tiktok.com and sign in again if prompted.

Prefer the nuclear option? Press Ctrl + Shift + Delete (Windows) or Cmd + Shift + Delete (Mac) to open Chrome’s browsing data dialog. Check Cookies and other site data and Cached images and files, then clear. Heads up: this signs you out of most websites.

Verify: Reload TikTok and play a video. If it plays, you’re done!

Rung 2: Test in an Incognito Window

Symptoms: Same as above, and clearing data didn’t help.

Why it matters: Incognito mode starts without your saved cookies and site data, and extensions don’t run there unless you’ve explicitly allowed them in Incognito. It’s a 30-second test that tells you where the problem lives.

Fix:

• Press Ctrl + Shift + N (Windows) or Cmd + Shift + N (Mac).
• Go to tiktok.com and try a video.

Verify: If TikTok works in Incognito, the problem is most likely an extension or something stored in your normal profile (cookies, site permissions, or saved site data). Open chrome://extensions, click Details on each extension, and note which ones have Allow in Incognito turned on, since those also ran in your test. Then skip to Rung 4. If it still fails, keep going.

Rung 3: Update Chrome to the Latest Version

Symptoms: TikTok broke “out of nowhere,” or other modern sites act strangely too.

Why it happens: Starting with Chrome 153 in September 2026, Google moved Chrome to a two-week release cycle, so a browser you haven’t restarted in a while falls behind quickly. As of October 6, 2026, the desktop Stable version was Chrome 155 (155.0.8059.39/.40 for Windows and Mac), according to the Chrome Releases blog (retrieved October 7, 2026). Rollouts are staggered, so trust the version shown on your own About Chrome page. TikTok’s site changes often, and keeping Chrome current rules out browser compatibility bugs.

Fix:

• Type chrome://settings/help in the address bar, or click the three-dot menu > Help > About Google Chrome.
• Chrome checks for updates and downloads them automatically.
• Click Relaunch when it appears.

Verify: The page should say Chrome is up to date. Reopen TikTok and test.

On Windows

Chrome updates itself in the background, but the update only applies the next time you restart the browser. If you never close Chrome (no judgment!), you’re probably behind.

On macOS

You can also reach the update page from the menu bar: click Chrome > About Google Chrome.

On the Web

Nothing to update on TikTok’s side. The website always serves its newest version, which is one more reason to keep Chrome current.

Rung 4: Disable Ad Blockers and Privacy Extensions

Symptoms: TikTok works in Incognito but not in your normal window, or some videos play while others don’t.

Why it happens: Ad blockers and tracking-protection extensions block scripts and requests they consider sketchy. Unfortunately, TikTok’s video player sometimes gets caught in the net. VPN extensions can cause trouble too.

Does an ad blocker break TikTok? Not always, but it’s one of the most common causes. The tell-tale sign is an error like ERR_BLOCKED_BY_CLIENT, which means the browser itself chose to block the request (most often because an extension told it to).

Fix:

• Type chrome://extensions in the address bar and press Enter.
• Turn off the toggle for your ad blocker.
• Reload TikTok and test.
• If it still fails, turn the next extension off (privacy tools, VPNs, video downloaders) and test again. Repeat until it works.
• Turn back on everything except the troublemaker.

Verify: Once TikTok plays, you’ve found your culprit. Many ad blockers let you “allowlist” a single site, so you can keep the blocker on and just exempt tiktok.com.

Tip: If TikTok works in Edge, Firefox, or Safari but never in Chrome, create a fresh Chrome profile. Click your profile picture in the top-right corner, choose Add, and set up a profile without signing in. If TikTok works there, the problem lives in your main profile’s extensions or settings.

Rung 5: Check Chrome’s JavaScript Setting

Symptoms: TikTok’s page is mostly blank, with no feed, buttons, or player.

Why it happens: TikTok’s feed, player, and buttons are built with JavaScript. If it’s been blocked, either for everything or just for TikTok, the page can’t build its interface.

Fix:

• Type chrome://settings/content/javascript in the address bar.
• Make sure the default setting lets sites use JavaScript.
• Scroll to the list of sites not allowed to use JavaScript. If tiktok.com is there, click the three-dot icon next to it and remove it.

Verify: Reload TikTok. The feed and controls should appear.

Rung 6: Fix a Black Screen (Hardware Acceleration)

Symptoms: You hear audio but see a black box, the video flickers, or it shows garbled colors.

Why it happens: Hardware acceleration means Chrome hands video drawing off to your computer’s graphics chip instead of the main processor. It’s usually faster, but some graphics drivers don’t get along with it, and the result is a black screen.

Fix:

• Type chrome://settings/system in the address bar.
• Turn off Use graphics acceleration when available.
• Click Relaunch.

Verify: Open TikTok and play a video. If it now plays, you’ve found the problem. Things might feel slightly sluggish with acceleration off, so try the driver update below and then turn it back on.

On Windows

Outdated graphics drivers are a common trigger. Open Settings > Windows Update > Advanced options > Optional updates and install any display driver updates listed. You can also get drivers straight from your graphics maker (Intel, AMD, or NVIDIA).

On macOS

Apple bundles graphics drivers into macOS updates. Open System Settings > General > Software Update and install anything waiting.

Rung 7: Fix Network-Only Playback Failures

Symptoms: TikTok works on your phone’s mobile data but not on your Wi-Fi, or it fails only at work, school, or a coffee shop.

Why it happens: Some networks block TikTok on purpose using filters. Others just have weak connections, DNS problems, or brief outages. Walls and floors can block signals like nobody’s business, and video is the first thing to stutter.

Fix:

• Turn on your phone’s mobile hotspot, connect your computer to it, and load TikTok.
• If it works on the hotspot, your original network is the problem, not Chrome.
• On your home network, restart your router: unplug it, count to 30, and plug it back in.
• If you use a VPN, turn it off and test again.
• On work or school networks, the block is usually intentional. Don’t try to bypass an organization’s network policy with a VPN, proxy, or hotspot on a managed device; if you think TikTok is blocked by mistake, ask the network administrator.

If your home Wi-Fi struggles in certain rooms, a mesh Wi-Fi system spreads the signal more evenly. A Cat6 Ethernet cable plugged straight into your router is even more reliable for streaming.

Verify: TikTok plays smoothly on the working network.

Error Messages Quick Reference

TikTok’s own wording changes from time to time, so treat its messages as general hints. Chrome’s error codes are more consistent, but each one can have several causes, so read the table as the most likely explanation rather than a diagnosis.

When to Just Use the TikTok App Instead

You climbed the whole ladder and TikTok still won’t cooperate? It’s okay to stop here. Many people find the TikTok app on a phone or tablet more dependable than the website, and it’s a quick way to keep watching while a browser problem gets sorted out.

Switch to the app if:

• You’ve tried every rung and nothing changed.
• TikTok works in the app but not in any desktop browser (which suggests a website-side issue you can’t fix).
• You’re on a network that blocks TikTok and you can use mobile data instead.

You didn’t give up too early. You ruled out everything on your end, and that’s exactly what a support agent would ask you to do anyway.

Getting Help

If the problem looks like it’s on TikTok’s side, these official resources help:

• TikTok support: TikTok’s troubleshooting help page (retrieved October 8, 2026), which also links to its problem-report form.
• Chrome support: The Google Chrome Help Center covers updates, extensions, and settings.
• Info to have ready: Your Chrome version (from chrome://settings/help), your operating system, the exact error message, and which rungs you already tried.

Prevention Tips

• Restart Chrome Weekly: This lets pending updates actually install.
• Keep Extensions Lean: Every add-on is another thing that can break a site. Remove the ones you don’t use.
• Allowlist TikTok in Your Ad Blocker: Keep the blocker, skip the headache.
• Update Your Computer: Graphics driver fixes arrive through Windows Update and macOS updates.
• Use Wired When You Can: An Ethernet connection beats spotty Wi-Fi for video every time.

Frequently Asked Questions

Why does TikTok show a black or blank screen in Chrome?

A black box where the video should be usually points to hardware acceleration clashing with your graphics drivers (Rung 6) or an extension blocking the player (Rung 4). A completely blank page with no buttons usually means JavaScript is blocked (Rung 5).

Why won’t TikTok videos load or play in Chrome?

Most often it’s stale cache and cookies or an ad blocker. Clear TikTok’s site data first, then test in Incognito to see if an extension is involved.

How do I clear TikTok’s cache and cookies in Chrome?

Go to chrome://settings/content/all, search for tiktok, and click the trash can next to tiktok.com. This removes TikTok’s stored data and will likely sign you out of TikTok, but it doesn’t clear data for unrelated websites.

Does an ad blocker break TikTok in Chrome?

It can. Ad blockers and privacy extensions sometimes block scripts TikTok needs. Turn yours off briefly to test, then allowlist tiktok.com if that fixes it.

What is hardware acceleration and how does it affect TikTok?

It’s Chrome using your graphics chip to draw video. It’s usually helpful, but buggy drivers can cause black screens or flickering. Turning it off in chrome://settings/system is a reliable test.

Why does TikTok work on my phone but not in Chrome?

The app and the website are separate. If the app works, TikTok itself is fine, and the problem is in your browser (site data, extensions, settings) or your network if your phone is on mobile data.

Why does TikTok work on another network but not mine?

Your network is likely filtering TikTok on purpose (common at schools and offices), or the connection is too weak for video. Restart your router, turn off any VPN, and test again.

Should I just use the TikTok app instead of the website?

If you’ve worked through the ladder and nothing helps, yes. Many people find the app more dependable, and it gets you watching again while you sort out the browser.

Wrapping Up

When TikTok is not working in Chrome, there’s no need to try fixes at random. Start at the top of the ladder, test after every step, and stop as soon as videos play. Most people never get past clearing site data or turning off an ad blocker. If Chrome keeps fighting you anyway, switching to the app is a smart move, not a defeat.

Last updated: October 7, 2026 | Applies to TikTok (Web Player) in Google Chrome on Windows and macOS

The post How to Fix TikTok Not Working in Chrome on Windows and Mac first appeared on Online Tech Tips.

You’ll set up ChatGPT on the web, Windows, or macOS. Then you’ll teach it about yourself once and use it to write a headline, a post, and a summary of a longer article or webinar. Each one comes with a real before/after example. You’ll finish with an editing checklist to run before you hit Post. A free ChatGPT account covers all the basics.

What Is ChatGPT (and What Can It Do for LinkedIn)?

ChatGPT is OpenAI’s AI chatbot. You type a request and it writes back, a bit like texting a fast, tireless writing assistant. It runs in any browser at chatgpt.com, and there are optional desktop apps for Windows and macOS.

There’s no special “LinkedIn mode.” When we checked in October 2026, we didn’t find a built-in way to publish from ChatGPT straight to LinkedIn, and both apps change often. The simple workflow here doesn’t need one: you write in a normal chat, then copy the final text into LinkedIn yourself.

Where it shines:

• Headline brainstorming: It gives you 10 or more angles in seconds, so you’re not rewriting the same tired line.
• Post hooks: It writes punchy opening lines and short, scannable post structures.
• Repurposing: It condenses a long article, report, or webinar transcript into a LinkedIn-sized post.
• About section drafts: It turns resume bullets into a first-person summary in a few different tones.

Where it falls short:

• Your real accomplishments: It doesn’t know them unless you tell it. Left unguided, it will invent believable-sounding numbers and client names, which is a terrible look on a professional network.
• Your voice: Its default style leans on clichés and lots of emojis. You have to steer it.

This guide is for everyday LinkedIn users, job seekers, and freelancers. You don’t need any “prompt engineering” skills.

Prerequisites

• A computer running Windows or macOS, or any device with a modern browser (Chrome, Edge, Safari, or Firefox)
• A free ChatGPT account (sign up with an email address or one of the sign-in options shown on the login page)
• A LinkedIn account you can sign in to
• Your raw material: your resume or bio, 2–3 of your past posts or emails, and any article or notes you want to summarize
• About 30 minutes

Privacy first: Only share material you’re comfortable handing to an AI service. Before you paste or upload anything, remove confidential, proprietary, or unpublished work information, client details, other people’s personal data, and your own sensitive details (home address, phone number, ID numbers). If a resume or report contains any of that, use a trimmed copy. You can also review what ChatGPT keeps under Settings > Data controls.

A second screen helps a lot here, too. An external monitor lets you keep ChatGPT on one side and LinkedIn on the other while you copy and paste.

Step-by-Step Guide

Step 1: Open ChatGPT on Your Platform

You only need one of these. Sign in with the same account on each one so your chats stay in one place.

On the Web

• Open your browser and go to https://chatgpt.com.
• Click Sign up to create an account, or Log in if you already have one.
• Follow the prompts to sign in with your email address or one of the other sign-in options shown.

Expected result: You land on a new chat screen with a message box at the bottom that says something like “Ask anything.”

On Windows

• Go to https://chatgpt.com and look for the desktop app download link. You can also search for “ChatGPT” in the Microsoft Store.
• Click Get or Install, then open the app from the Start menu.
• Click Log in and use the same account you use on the web.

Expected result: ChatGPT opens in its own window, separate from your browser tabs.

On macOS

• Go to https://chatgpt.com and look for the macOS desktop app download link.
• Open the downloaded file and drag ChatGPT into your Applications folder.
• Launch it from Launchpad or Spotlight (press Cmd + Space and type “ChatGPT”), then log in.

Expected result: The ChatGPT app opens with your chat history in the sidebar.

Tip: Desktop app features change from time to time. Check OpenAI’s release notes if something looks different from what you see here.

Step 2: Give ChatGPT Your Context First

This is the step most people skip, and it’s why their posts sound generic. Think of it like briefing a new assistant on their first day. You explain who you are once, and every later draft gets better.

• Click your profile icon or name in the bottom-left corner.
• Click Settings > Personalization (per OpenAI’s Custom Instructions help; menu names can shift between app versions).
• Make sure customization is turned on, find Custom instructions, and fill in the fields about yourself and how you want ChatGPT to respond. Free and Go accounts can save up to 1,500 characters there, which is plenty for the example below.
• Click Save.

Here’s an example you can adapt:

I’m a project manager at a mid-sized logistics company with 8 years of experience. I’m moving into operations leadership. I write in a direct, friendly way with short sentences and very little jargon. When writing LinkedIn content: never invent facts, numbers, clients, or quotes. If you need a detail you don’t have, ask me. Use plain text with line breaks, no Markdown, no hashtag walls, and at most one emoji.

Expected result: New chats now follow these preferences without you repeating them.

You might not want LinkedIn rules applied to every chat. In that case, skip Custom Instructions and paste the same text as the first message of a new chat instead.

Step 3 (Optional): Create a LinkedIn Project

Projects are like folders that keep related chats and reference files together. You can upload your resume and a few past posts once, and they stay available for every LinkedIn chat. Projects are available on the Free plan too; what changes by plan is how many files each project holds. According to OpenAI’s Projects help page (checked October 8, 2026), Free accounts get up to 5 files per project, Go and Plus up to 25, and Pro, Business, Edu, and Enterprise up to 40. Five files is enough for a trimmed resume, a short bio, and a couple of writing samples.

• In the sidebar, click New project.
• Name it something like “LinkedIn Content.”
• Add your (trimmed) resume, bio, and 2–3 writing samples as project files.

Expected result: The project appears in your sidebar, and any chat you start inside it can use those files.

Step 4: Write Your LinkedIn Headline

Start a new chat (inside your project, if you made one) and paste:

Using my background, write 12 LinkedIn headline options. Group them into: role-focused, outcome-focused, and audience-focused. Keep each under about 200 characters. Use only facts I’ve given you.

Why about 200 characters? LinkedIn caps the headline field (the limit is widely reported as 220 characters), so this leaves you a little room to edit. The About section and posts have their own, larger limits, and LinkedIn says a post can be up to 3,000 characters. Each field also handles line breaks differently, so check how the text looks after you paste it.

Before (the headline you’ve had since 2019):

Project Manager at Acme Logistics

After (picked from ChatGPT’s list, then edited):

Operations-minded project manager | I help logistics teams ship on time without the 2 a.m. fire drills | 8 years at Acme

That “2 a.m. fire drills” bit was a human edit, by the way. ChatGPT offered “streamline workflows,” which nobody remembers.

To paste it into LinkedIn, go to your profile, click the pencil icon on your intro card, and paste the headline into the Headline field. Click Save.

The same approach works for your About section. Ask for “three first-person About section drafts, one polished, one conversational, and one scannable with short lines.” Pick one, edit it, then click the pencil on About, paste, and click Save.

Step 5: Write a Post With a Strong Hook

Feed ChatGPT the real story first. Then ask:

Here’s what happened: [paste your notes]. Write a LinkedIn post with a one-line hook, 4–6 short paragraphs, one clear takeaway, and a closing question. Give me 3 different hooks to choose from. Plain text only.

Before (ChatGPT’s first try with vague input):

Adaptability matters! I’m thrilled to share an incredible journey of growth and resilience…

After (real notes plus an edit pass):

We missed a shipment deadline by six days last spring.

The cause was a handoff problem. Three teams each thought someone else owned the customs paperwork.

The fix was boring: one shared checklist and one named owner per step. We haven’t missed a deadline since.

What’s the most boring fix that saved your team?

Notice the difference? The second one has a real detail, a real number, and zero rocket emojis.

Expected result: You get three hook options and a draft you can trim down.

Step 6: Summarize a Long Article, Report, or Webinar

Did you sit through a 60-minute webinar (hopefully with a comfy headset)? Turn it into a post:

• Paste the article text or webinar transcript into the chat. You can also attach the file with the + or paperclip button. Free accounts have tighter upload limits. Only do this with material you’re allowed to share, and skip internal or unpublished company documents.
• Then send:

Summarize this into a LinkedIn post under about 1,300 characters. Lead with the single most surprising takeaway, add my one-line opinion: [your opinion], and credit the source by name. Don’t add any facts that aren’t in the text.

Before: A 4,000-word industry report you’ll “definitely share later.”

After:

A new freight report says 40% of delays start before a truck even moves. They begin in paperwork.

That matches what I’ve seen. Our worst delays never happened on the road.

(Source: [Report name], [Publisher])

The 1,300-character target is our suggestion for a readable post, not a LinkedIn rule; the hard limit is 3,000. Double-check that “40%” against the original. Summaries are exactly where small number mix-ups sneak in.

Step 7: Run the Humanizing Checklist

Do this before anything goes on LinkedIn:

• Fact-check every number, name, and date: Ask ChatGPT to “list every factual claim in this draft so I can verify it,” then check each one yourself.
• Delete the clichés: Words like “game-changer,” “thrilled to share,” “delve,” “in today’s fast-paced world,” and “unlock.”
• Add one detail only you know: A specific moment, a mistake, or a number from your own work.
• Read it out loud: If you wouldn’t say it to a coworker, rewrite it.
• Cut 20%: First drafts almost always run long.
• Trim emojis and hashtags: One emoji and two or three relevant hashtags is plenty.
• Change the first line yourself: Even a small tweak makes it yours.

Step 8: Paste Into LinkedIn and Publish

• Copy the final text (Ctrl + C on Windows, Cmd + C on Mac).
• On your LinkedIn feed, click Start a post.
• Paste as plain text so stray formatting doesn’t tag along. In most browsers that’s Ctrl + Shift + V on Windows; Microsoft notes it doesn’t work in every app. On a Mac, try Option + Shift + Cmd + V, which Apple lists as Paste and Match Style; support varies by app and browser. If neither works, paste the text into a plain-text editor first, then copy it from there.
• Check the preview for odd symbols or broken line breaks, then click Post.

Configuration

• Custom Instructions (Settings > Personalization): Tells ChatGPT who you are and how you write. Update it whenever your job title changes.
• Projects (sidebar): Keeps your resume and writing samples in one place. Mark which resume version is current.
• Plan (shown in Settings): Your plan affects usage limits, file uploads, and which features are available. Compare the plans on the pricing page.

Tips and Troubleshooting

Problem: ChatGPT invents achievements or metrics

• Paste in your real resume or notes.
• Add this line: “Do not invent facts, statistics, quotes, customers, or outcomes.”
• Ask for a list of claims to verify, and check each one.

Problem: The post sounds obviously AI-written

• Paste 2–3 of your own past posts or emails.
• Ask it to “match this rhythm and directness, and remove clichés and excess emojis.”
• Ask for three versions and pick the best parts from each.

Problem: Formatting breaks when you paste into LinkedIn

• Ask for “plain text only, no Markdown headings or bold markers.”
• Use the plain-text paste shortcut from Step 8.

Problem: ChatGPT mixes up old job details

• Start a fresh chat for each new piece of content.
• Tell it which document is the current version.

Problem: You can’t see Projects or a file upload fails

• Check your plan’s file limit per project (5 on Free, 25 on Go and Plus, 40 on Pro and business plans).
• Remove unused files, or convert large documents to plain text.
• Update the desktop app or refresh your browser tab.

• Keep a swipe file: Save your best final posts in your project as writing samples. Your future drafts will sound more like you.

FAQ

Is it against LinkedIn’s rules to use ChatGPT for my posts or headline? Using AI to help draft your writing isn’t banned, but LinkedIn’s best practices for content created with the help of AI set clear expectations. LinkedIn urges you to review, edit, and approve anything AI helped you write, and says you are ultimately responsible for everything you post. It also asks you not to infringe anyone’s intellectual property or privacy rights, and recommends telling readers if you relied heavily on AI (when it isn’t obvious from context). Keep your profile truthful, post it yourself rather than through bots or unauthorized automation, and treat the fact-check step as mandatory.

Do I need a paid ChatGPT plan? Not for this workflow. The Free plan handles headlines, posts, summaries, Custom Instructions, and small Projects (up to 5 files each). Paid plans like Go and Plus raise usage and upload limits, including up to 25 files per project. Pro starts at $100/month according to the pricing page (checked October 6, 2026), which is serious overkill for LinkedIn posts.

Wrapping Up

The trick to using ChatGPT for LinkedIn is a boring workflow: give it context, let it draft, then edit like you mean it. Honestly, ChatGPT is great at beating a blank page and lousy at being you. Let it do the first, and you handle the second.

Resources

• ChatGPT
• ChatGPT Pricing
• ChatGPT Release Notes

The post How to Use ChatGPT for LinkedIn Headlines, Posts, and Summaries first appeared on Online Tech Tips.