Permalink

]]>Random Acts of Photography, Autumn EditionVloggingVideoStreetAutumnThomas FitzgeraldThu, 05 Nov 2026 12:21:48 +0000https://blog.thomasfitzgeraldphotography.com/blog/2026/11/i-got-back-to-photo-vlogging-a-video-about-autumn-in-dublin-city53c631e4e4b0c4d68989cbef:540c893ee4b04215b3085eaf:6aec74dce9a48b2f485e0f69

This past weekend, I did something I haven’t really done in four years: I made a photography vlog. A proper, talking-to-camera, while out and about, taking photos, and showing you around type of old-school vlog. And I actually think that it’s pretty good. I’ve had this idea floating around my head for a few years, and it’s been on my to-do list for about as long. The idea was to do an “Autumn in the city” video. When you think of Autumn, you think of landscapes, forests, rivers and mountains and so on, but I wanted to show just how many Autumn photography opportunities there are in the city too.

I had been putting this off for so long, for no real reason, really. I really struggled with talking to the camera outside over the past few years. Much of this is my own issues, but some of it stems from having issues with my knee and ankle problems. (I had a torn meniscus, and then a ruptured tendon in my ankle, all within months of each other - not fun) I actually found that I got quite nervous on the streets when I had reduced mobility. It was all in my head, but it took me a long time to get over it. was also concerned that I was getting too old for this, as I’m in my 50s now (noooooo), but I’ve been watching a lot of older YouTubers lately too, and so I shouldn’t really let that stop me. I missed making these types of YouTube videos, so I’ve been slowly trying to get myself used to it again.

So at the weekend, I just made myself do it. It turns out, all you need to do is get started. Once you get going, it just sort of flows. I had originally said to myself, “Just try it; if it doesn’t work, you don’t have to use it”. I was going to take baby steps and figure out what I was doing again.

I had already shot some stills around the place over the previous few days, so I already had a good few stills. But the light was quite different when I was shooting video, so I shot quite a few more while I was at it.

The way I actually shot the video has changed quite a bit since I used to shoot my “street photo diary” series. I used to use a GoPro attached to my camera to shoot video while I was shooting stills, but this time I mostly just used my Canon R6II for both stills and video. If I saw something I wanted as a still, I just shot a few seconds of video too. I think this works better than the GoPro way, but there were some downsides to this too.

I also had my Osmo Pocket 3, which I got to do these kinds of videos, but I had yet to use it as I had intended, so there was some of this in there too. I was using a Rode shotgun mic on the Camera, which I thought would be an issue, as Dublin is surprisingly loud, but Final Cut Pro’s “Voice Isolation” works really well. This is something that didn’t exist when I used to shoot before. When using the Osmo Pocket, I was using the lav mic (actually clipped onto my camera bag strap as opposed to holding it!) and the sound is much better. I actually do have a Rode wireless lav mic too for my main camera, but it’s a bit of a pain, and switching was going to be an issue, although I might look at an alternative in the future, as the wireless mic was much better.

The Osmo Pocket is a bit of a pain, actually. The auto exposure isn’t great on it. I could use manual, but as I’m constantly moving in and out of shade, it’s not ideal. I shot in Log, but still had trouble grading it to match the rest. I probably just need to get better at using it and figure out the settings. The Canon video is great - most of the time. It does suffer from the Canon IBIS wobble, though, especially at 24mm, which is a real pain in the you-know-what. I actually nearly prefer the video I get from the Canon to the stills, but that’s another story.

Of course, the immutable rule of shooting a video is that no matter how much footage you think you have, you probably don’t have enough. When I got back and did a rough cut, I liked how it was working overall, but realised I was lacking something. I was at two minds whether or not to go back into the city to get more footage the following day, but when I woke up, it was dull and wouldn’t match the light. However, when I was out getting coffee, the sky cleared up, and so I figured I might as well take the opportunity. Unfortunately, I didn’t have my camera with me, but I did have my phone, and I also had my little pocket Canon G7X in my bag. So I grabbed the tram into the city and set about grabbing some B-Roll.

I was just planning on getting a few extra shots, but I ended up shooting 20 minutes of footage and a whole bunch of stills too. I shot in log using the Blackmagic app. It’s good, but it’s a pain to grade, and it kind of falls apart. I think the bitrate is quite low in the Blackmagic app, or I’m doing something stupid. My on-device storage is too low to shoot ProRes, and I didn’t have an external card or drive with me. It was fine, but I wish I had a proper camera with me. It’s funny, because I got into a discussion with someone on social media about how no one needs a vlogging camera now that iPhones are just as good, but that’s not my experience. Again, maybe if I were shooting in ProRes. If you watch the video, can you tell which shots were from the iPhone?

The one thing that really surprised me was how much the leaves had changed in the space of a few days. It had only been a week since I had shot the first set of stills, and the leaves had turned significantly browner and far more had fallen off in the short space of time. It made continuity a little difficult, but in the end I just went with it.

Anyway, the whole thing came together pretty well, I think. I remembered how much I liked making these kinds of videos, and I’m glad I gave it another go. I started the process as a practice attempt, but I ended up with what I think is a good result, so you tell me. Of course, the biggest issue is the YouTube algorithm and whether it even shows it to anyone. I’ve done my best with the thumbnail and the title, and will promote it as much as possible, but I guess time will tell. I really want to do these kinds of videos more, as my channel has gotten a bit stale with all the Lightroom tutorials that no one watches. It will take some time to build up momentum, though. I really don’t understand the YouTube algorithm. Anyway, I hope you enjoy it.

I will have a corresponding Photo Journal post out shortly too, with some of the stills I took, so stay tuned for that. I’m also bringing back the newsletter so that you can get notified of new posts (well, a weekly digest) in case you miss anything, so you can sign up for that here.

Capture One has released a new update with a focus on portrait and colour tools, while Luminar’s autumn release combines revised automatic adjustments with a new printing workflow. Apple has also announced changes to Pixelmator Pro, and ON1 has launched its 2027 edition of its signature software.

Capture One 16.9 brings more retouching into the RAW workflow

Capture One 16.9 was released on 29 October, alongside version 4.2.0 of its mobile application. The headline additions are Stray Hair Retouching, Skin Tone Retouching and Body Skin Retouching, aimed at reducing repetitive portrait clean-up. Capture One says the new retouching tools process photographs locally, without uploading them to the cloud. Its official feature overview explains the additions.

Stray Hair Retouching provides separate controls for hair inside and outside the main hair area, with protection brushes for details you want to retain. The skin tools address differences such as red hands, colour casts and uneven texture across the body. For portrait and wedding photographers, these could reduce the number of images that need a separate Photoshop pass.

The Advanced Colour Editor also gains editable RGB and HSL targets, reference-colour matching and expanded uniformity controls. Capture One also reports approximately 30 per cent faster Enhanced Denoise processing for Bayer files.

One quick note before you upgrade: Sessions and Catalogues must be upgraded to version 16.9, after which older versions cannot reopen them without restoring a backup. Multi-User Sessions has left beta, while Assisted Review remains in beta. Full details are available in the release notes.

Luminar updates its editing tools and adds printing

Skylum’s Luminar Fall Update became available on 28 October. The new release revises Enhance, Atmosphere, crop suggestions and the AI Assistant, while adding signature and print ordering tools.

According to Skylum, Enhance now adapts its adjustments more closely to the photograph’s lighting, colour and subject. Atmosphere uses depth information to place volumetric fog within a scene, while Smart Compositions proposes crops based on the subject and compositional principles. The assistant can suggest adjustments in response to a description of the desired result.

You can now order prints directly from within Luminar too, with the software letting you order prints directly to your doorstep without having to use a separate app or website, although it’s not clear what lab or service they are using.

Pixelmator Pro will gain Generate Fill and closer Final Cut integration

Apple announced its next Creator Studio update on 27 October, with availability planned for November. For photographers, the main addition is Generate Fill in Pixelmator Pro. Users will be able to select an area and describe a change, with generated content placed into existing layers or added as new layers with masks.

Apple also announced a direct editing workflow between Final Cut Pro and Pixelmator Pro. An image in a video timeline can be opened in Pixelmator, edited and saved back to the project automatically. These remain forthcoming features, rather than a confirmed release today. Apple’s announcement provides the details.

These changes appear to be only for the Creator Studio version of the software, so owners of the standalone Pixelmator Pro should check feature eligibility rather than assume every subscription feature is included in their purchase. Unfortunately, Apple still insists on the god-awful Liquid glass redesign of Pixelmator Pro, which, in my opinion, makes the application borderline unusable, as the interface now blends into your content and makes it very difficult to see controls, not to mention being very distracting, but you may be ok with it.

ON1 launches Photo RAW 2027

ON1’s Photo RAW 2027 is now available. The new update includes a new RAW processing engine, revised masking, updated NoNoise AI and an AI Culling Assistant in the MAX edition. Most were announced previously; the concrete launch date is the new development.

Once the release is available, the RAW engine deserves particular attention. Comparing the same files in the previous version, Photo RAW 2027, Lightroom and PhotoLab should reveal more than promotional examples about changes to detail, colour and highlight handling. I will try and do a review in the next few weeks, or at least an overview

Help Support the Blog

Check out my eBooks and Presets

Check out my photography eBooks , Capture One Styles, and Lightroom Presets available on GumRoad.

Subscribe to my YouTube Channel

Check out my YouTube channel for tutorials, Vlogs, and more!

Buy me a coffee!

If you want to say thanks or help, then you can feed my caffeine habit and buy me a coffee via PayPal with a one off donation to my PayPal tip jar. (Please note that PayPal doesn’t make it easy to respond to these so just know you are thanked in advance) Note that this post contains paid affiliate links. We get a small commission for purchases made through these links, which helps run this site.

Permalink

]]>Thoughts on the new Touchscreen MacBook Pro from a Creative ProfessionalmetadataEditorialLightroomThomas FitzgeraldWed, 28 Oct 2026 16:29:13 +0000https://blog.thomasfitzgeraldphotography.com/blog/2026/10/why-captions-keywords-and-other-metadata-still-matters-in-202653c631e4e4b0c4d68989cbef:540c893ee4b04215b3085eaf:6ae222da2f77ec187b8231a3

After my recent video on using Siri to add captions and keywords to photos, someone on social media asked a question that kind of took me aback: “What's the point?” I had assumed everyone who was in any way into photography understood the benefit of adding keywords and captions, but I guess not. The more I thought about it, the more it made sense in a way. Phones recognise faces. Most software (except Lightroom Classic, because…Adobe) can find "dog on a beach" without a single piece of user-generated metadata, and AI seems to know what's in every picture. So why bother writing anything by hand?

The short answer is that metadata is still very important, even in 2026. It’s how stock agencies and picture editors find your images, and how search engines and AI assistants understand them. Othe metadata is how your name and copyright travel with the file, and how you state whether a picture can be used for AI training. It’s how you embed your usage rights and a link to your website. Titles and Captions in particular can be a bit of a pain, but tools like Siri now make writing them almost effortless. In the current era, they matter for more reasons than ever, not fewer.

(Just to be clear, I understand that many of my readers probably understand this already and may find some of this patronising, but I’m really not trying to be. It’s important to realise that maybe not everyone knows this stuff, so that’s why I’m writing this)

Captions and Keywords Oh My!

So what exactly is a caption and what is a keyword?

• Caption: a sentence or two describing the photo: who, what, where, when, and sometimes why. For example: Commuters crossing O'Connell Bridge in the rain, Dublin, on the morning of the first storm of the season. This is often a more detailed description than a straigth up “title” although the title is important too.

• Keywords: short, searchable tags. For example: Dublin, O'Connell Bridge, rain, umbrellas, commuters, street photography, Ireland.

Other similar metadata, which may be required depending on where you’re sending the file, is a separate “title”: usually a shortened version of the Caption and maybe location data. If you’re a press photographer, there is a bunch of other stuff you need to submit too.

It’s also important to understand that this metadata is attached to the file when you export it (unless you explicitly tell it to strip that data out) and isn’t application-specific. In apps like Lightroom, Capture One, or Apple Photos, captions and keywords can be written into the image file itself using a standard format (such as IPTC metadata). One caveat here is that while all applications should treat this equally, some can be problematic, especially when it comes to keywords. There’s also a whole thing about hierarchical keywords, but I’m not going to get into that here.

There are several cases where you actually need to have metadata in your image files.

1. If you sell or license photos.

For anyone who earns money from photography, adding Captions and Keywords is often not optional.

Stock agencies require keywords. Upload to iStock, Adobe Stock or any of the others, and their search engines rely heavily on the title, caption and keywords you supply. A great photo with no keywords is, in practical terms, a photo nobody can find.

It's the same with direct licensing. When a picture editor or a company's marketing team needs, for example, "Dublin street scene, rain, people with umbrellas", they search either your online archive, or ask you to search your own database. If your archive is captioned and keyworded, you can answer that request in minutes instead of digging through folders. When I've licensed images to clients in the past, being able to quickly find and quickly prove the details of the right shot has mattered as much as the shot itself.

Captions and metadata can also carry the facts a client actually needs: where it was taken, when, and who's in it. That's the difference between "nice photo" and "usable photo" for editorial work.

2. Search engines, even with things like Google’s AI answers, still read your captions.

If you post photos on your own website, words are still how they get discovered. Search engines look at the alt text, the caption, the filename and the text around an image to work out what it shows. Captions are also one of the most-read parts of any page: people who skim an article still tend to stop and read them.

A lot of people now ask an AI assistant rather than typing into a search box, and those tools lean on the same text signals to understand what an image on your page is. A photo with a clear, specific caption gives them something concrete to work with. A photo with nothing gives them nothing.

There's a nice practical benefit too. Some website platforms and plugins can pull the caption or title from your image's metadata when you upload it, so writing it once in your photo app can save you from writing it again on the web. WordPress does this, for example. Squarespace used to do this, but it’s been broken for quite some time now (but that’s another rant). Some will set this as alt text, although technically Alt text is a different field.

3. Accessibility and the importance of good alt text

People who use screen readers can't see your photos. They rely on alt text, which is read aloud in place of the image. If that text says "IMG_4821" or nothing at all, they get nothing.

Alt text and captions aren't quite the same thing. Alt text describes what's visible; a caption adds context. But if you've already written a careful caption, writing alt text can be a lot quicker, and in many cases, you can use the same thing. Some people think alt text is just for search engines; it’s not. And you may be thinking, why does a blind person want to look at a photograph? But alt text can help people with all kinds of visual impairments and not just complete blindness. The IPTC standard even has dedicated fields for alt text and extended descriptions, so it can live in the file alongside everything else, although most online sites may not read this data.

4. Finding your own images in the future

Even if you don’t submit to stock photo agencies or post online, it’s still useful to have Captions and Keywords on your images when it comes to finding them in the future. While some tools, such as Apple’s Photos and Lightroom Desktop (cloud version of Lightroom), support AI-based image recognition search, others, such as Lightroom Classic and Capture One, do not. Search in these applications still relies on metadata, so if you want to be able to find files in the future, you need to have metadata added to your images.

The importance of other metadata when it comes to your rights, and AI training

The same metadata panel where you type a caption also holds your name, a copyright notice and licensing links. Those fields are not just for show. Google Images reads the Creator, Credit Line and Copyright Notice fields embedded in a photo and displays them next to it in search results. Fill in the Web Statement of Rights and Licensor URL fields, and Google can show a "Licensable" badge with a link to where someone can license the image (IPTC guide).

Then there's AI. The IPTC standard now includes a Data Mining field, developed with the PLUS Coalition, that lets you state in the file itself whether the image may be used for AI or machine-learning training (IPTC). It's not a magic shield: it relies on AI companies choosing to honour it, and plenty of websites and social platforms still strip metadata on upload. But in places like the EU, where the law allows rights holders to reserve their work from text and data mining, a clear machine-readable statement attached to the image is exactly the kind of signal that matters.

At a time when it's getting harder to tell real photographs from generated ones, an image that carries its photographer's name, a caption with a real place and date, and a clear rights statement is simply more credible. Metadata is part of how a photograph says "a person made this".

Unfortunately, at the moment, this doesn’t appear to be available in Lightroom, but hopefully in a future version. You can still put this in the copyright field

AI doesn't make this obsolete. It makes it easier

That's really why I was so excited when I figured out how to do this with Siri. The reason most people never caption or keyword their photos isn't that it's useless. It's that it's tedious. Typing a description for every image after a long shoot is nobody's idea of fun.

Dictating a caption to Siri while you look at the photo takes a few seconds. Letting a tool suggest keywords and then deleting the wrong ones is far quicker than typing them from scratch. The machine handles the grunt work; you supply the facts only you know.

Of course, if you use Lightroom Classic, the interface doesn’t help either, with the tiny text field down in one corner of the display. I still wish Adobe would add a whole metadata module to Lightroom Classic to make it easier to manage and enter metadata. There are some plug-ins that can help. I use one called Search + Replace, which helps batch-process metadata, with tools to find and replace metadata text across images, append metadata, and so on. For standalone applications, the industry standard is Photo Mechanic from Camera Bits.

So, what was the point again?

The point is that if you send your work to clients, submit to stock sites, post online or even if you just want to make it easier to group and find your work in the future, you really should be adding captions, keywords and other metadata. Yes, some of this can be a pain, but it will save you much more work in the future. Using AI tools to help with this is one way to get past the initial “do I really have to do this” hurdle and can speed up the process, but it also shouldn’t be the end of the process either. It’s important to add context to the photo too, and that’s still something only a human can do.

In this week’s photo software news….Lightroom’s latest mobile tools offer new ways to rescue photographs, while Topaz has released fixes that should interest Mac users. The controversial PhotoCraft also continues its development, although there are still good reasons to approach it cautiously.

Lightroom can now open closed eyes using generative AI

Adobe introduced Open Eyes for Lightroom Mobile on 19 October, offering a quick correction for portraits in which somebody has blinked. When Lightroom detects closed eyes, it presents Open Eyes as a Quick Action and generates a replacement. Adobe says the feature is available free of charge. Its official announcement also explains that photographs of the same person from your Lightroom library can help match their eye colour, shape and size.

It addresses a common problem that many photographers have, especially with group shots where everybody looks good except for one person. Photographers have long repaired these images in Photoshop by copying open eyes from another frame, but that requires a suitable alternative photograph and some careful compositing.

The distinction is that Open Eyes can generate detail without a reference, although it does look for reference photos before it does its thing. I just did a quick test on a selfie, and it does actually work, and it works reasonably well too, although I’ve only tried one photo yet. But my eyes are kind of weird, and it still managed to fix them, although it’s not 100%. The problem is, especially with close-ups, that closing your eyes affects the surrounding muscles too, so there’s more that changes than just your eyes. Still, it’s a useful tool, although I don’t understand why this is limited to Lightroom Mobile.

Speaking of Lightroom Mobile, Generative Expand is now available on Lightroom Mobile, allowing photographs to be extended beyond their original boundaries. This could be useful when adapting a landscape photograph to a vertical format or adding space around a tightly framed portrait. Both tools are presented as available mobile features, rather than previews.

Topaz Photo fixes processing problems, with an important Mac caveat

Topaz Photo 1.7.1, released on 22 October, increases the input limit for local Denoise Max processing and fixes several problems involving installation, upscaling and GPU compatibility. These include failures when installing Wonder, issues with High Fidelity v3 and Standard MAX on Intel GPUs, and output dimensions being approximately one pixel off when specific enlargement dimensions were requested. The official release notes also list a fix to Topaz’s processing backend on macOS 27.

The most significant qualification concerns Apple M6 Macs running macOS 27. Topaz says initial image analysis can take 15 to 30 minutes while the AI models initialise and build their cache. Subsequent processing should return to normal speeds, but the company acknowledges that the initial delay is unintended and says its investigation is continuing.

That is worth knowing before assuming the application has frozen. It also means this update should not be read as confirmation that all compatibility problems on those machines have been resolved. For affected photographers, the distinction between a long first load and consistently slow processing will matter considerably.

PhotoCraft continues development, and strangely, hasn’t been sued yet.

PhotoCraft 0.16.0 arrived on 25 October, continuing the fast development of ArtCraft’s open-source Vibe-coded Photoshop clone. According to its release notes, the update improves aspects of PSD compatibility, keeps several smart filters editable, corrects behaviour in 32-bit adjustments and addresses problems across colour handling, cropping and layers.

I remain sceptical of treating PhotoCraft as a dependable Photoshop replacement at this stage. In my opinion, it is, to all intents and purposes, an attempt to pirate Photoshop, albeit through an unorthodox and roundabout way. I’m just surprised that Adobe hasn’t issued a cease and desist yet, although I suspect that lawyers for Adobe realise that this could end up being a game of whack-a-mole if they don’t address the root cause, which is vibe-coded reverse engineering, and I suspect that will take longer to deal with. They may also be hoping the buzz around this dies down or that it ends up being so bad that no one uses it, because the company is in a lose-lose situation here. There will be substantial bad press for them if they go after the developer, but at the same time, they have to protect their patents. Anyway, we will see what happens.

If you want a Photoshop alternative, I strongly suggest that you consider something like Affinity instead, as it’s a proper application created by a legitimate developer who isn’t attempting to steal another piece of software through the back door. Even if you hate Adobe with all your passion, this is just wrong. In my opinion, anyway. And if you do hate Adobe so much, why do you want an exact copy of an application that you claim to hate anyway?

Bonus Photo Hardware News: Sigma announces the FP2

Sigma’s 23 October announcement confirms that the FP2 will become available on 11 November 2026. It continues the modular approach of the original fp series, with a 24.5-megapixel partially stacked full-frame sensor, a tilting screen, improved autofocus and shooting at up to approximately 20 frames per second.

The camera also includes 70GB of internal storage alongside its SD card slot. Video features include 6K Open Gate recording, ProRes RAW and Leica L-Log, while 32-bit float audio requires the appropriate optional hot-shoe unit and compatible Panasonic accessories.

For anyone who also shoots video, the attraction is a small camera that can be adapted to different jobs. Whether that compactness survives the addition of a viewfinder, audio accessories and a practical shooting rig remains to be seen, although it is ideal for things like gimbals and drones where you need a bare-bones setup. PetaPixel did a good review of this on YouTube if you’re interested.

For many photographers and creatives, “AI” has become something of a dirty word. I’ll be honest, I’m a bit of an AI sceptic myself, especially when it comes to generative AI. Concerns around “slop”, intellectual property theft during training, and environmental issues are perfectly valid concerns. However, AI is a broad term and not all technology that falls under the “artificial intelligence” umbrella is the same. This has led some people to avoid or criticise tools that have little in common with image generators. AI denoising, like the Denoise feature in Lightroom, is one of the most common examples. It carries the same label, but it works differently and does a very different job.

I’ve actually covered Denoise quite a bit on this blog and on my YouTube channel. Recently, someone left an irate comment on one of my videos complaining about how Denoise was just slop and how it wasn’t a real image afterwards. They complained that if you use Denoise, why even bother taking a photo in the first place? I was actually taken aback by this because I had assumed people understood the difference. It’s easy to dismiss such comments as someone trolling, but with such anti-AI sentiment out there these days, it occurs to me that maybe some people think that AI Denoise tools are using generative AI and are replacing your image with an entirely synthesised one.

But this isn’t what happens at all.

Traditional noise reduction mostly worked by blurring neighbouring pixels together. That removes the grain but also takes fine detail with it, leaving skin waxy and textures mushy. I mean, the algorithms are a bit more complicated than that, but that’s the gist of it. AI denoising uses a neural network trained on large numbers of image pairs, each showing the same scene with and without noise. When you run it on your photo, it looks at each area and estimates what the clean version of those exact pixels most likely was. It removes the speckle and keeps the edges and textures that were really there. Lightroom's version even works on the raw file before it becomes a normal image, which is part of why the results look so natural.

In other words, it uses machine learning and statistical analysis based on thousands of images to figure out how to adjust your image on a pixel-by-pixel basis to achieve the best noise reduction. This is a very simplified version of what is happening. Adobe’s own blog has more details on how their Denoise works, but most other tools that have similar technology will work in the same way.

But the important thing to note here is that it is NOT using generative AI.

So what’s the difference?

Generative AI creates content that was never captured. Ask a tool like Generative Fill to add a sky or remove a person, and it invents new pixels based on what it has learned images usually look like. A denoise tool’s job is to improve your image: it modifies your image rather than creating a new one. A generative tool's job is creation: it is free to make up something plausible where there was nothing. You’ll often see this in more recent tools like Lightroom’s (cloud) prompt-to-edit feature, where it tries to edit your image, but because it uses generative AI, it’s effectively creating a new version. This often leads to different details in your image, so people might appear where none were there before, and some other fine details might completely change. It’s perfectly ok, in my opinion, to be mad at this kind of thing. But that’s not how AI Denoise works at all.

There is some crossover, of course, and they are built on similar technology such as machine learning and neural networks, but that’s where the comparison really ends. Denoise isn’t perfect, and can occasionally result in artefacts and unnatural-looking images. Still, a denoised photo is your photo, just cleaner. Or to put it another way, denoising edits your existing image data, whereas generative AI creates entirely new image data. Even if you’re using it to edit your photo, it’s still a new image.

AI Denoise isn’t new either. It’s been around for a while now. Adobe introduced it in 2023, and many other software companies now have also added similar features. In particular, there is DXO’s Pure RAW, which uses DXO’s own flavour of denoising technology in combination with its popular lens correction profiles, and Capture One also recently added AI denoising to its toolset. Other notable mentions are Luminar and ON1 No Noise AI.

I hope, like Denoise itself, this has made things a bit clearer. (Nerdy and terrible Denoise joke!)

Support the Blog

Check out my eBooks and Presets

Check out my photography eBooks , Capture One Styles, and Lightroom Presets available on GumRoad.

Subscribe to my YouTube Channel

Check out my YouTube channel for tutorials, Vlogs, and more!

Buy me a coffee!

If you want to say thanks or help, then you can feed my caffeine habit and buy me a coffee via PayPal with a one-off donation to my PayPal tip jar.

Note that this post contains paid affiliate links. We get a small commission for purchases made through these links, which helps run this site.

Next Wednesday, Apple is holding its "Pro and Tell" event. If the rumours are right, we're about to get something many of us never thought we'd see: a MacBook Pro with a touchscreen. It's reportedly an OLED model with a Dynamic Island where the notch used to be, and it would be the first Mac you can actually touch.

For a long time, that seemed impossible. Back in 2010, at the "Back to the Mac" event, Steve Jobs dismissed the idea in a way that has been quoted ever since:

"We've done tons of user tests on this, and it turns out it doesn't work. Touch surfaces don't want to be vertical. It gives great demo, but after a short period of time you start to fatigue, and after an extended period of time, your arm wants to fall off. It doesn't work. It's ergonomically terrible."

For years, that was more or less Apple's position. The iPad was for touch and the Mac was for the keyboard and trackpad, and the two weren't supposed to meet. So what changed?

The evidence is already in macOS 27

This is probably the worst-kept secret Apple has had in a long time (up there with the Home View). Apart from the fact that it has been rumoured for months, the changes in recent versions of macOS make it pretty clear that Apple has a touchscreen in mind.

• Sidecar got full touch support. Until now, using an iPad as a second display for your Mac meant the Apple Pencil and a few basic gestures. In macOS 27, you can use your fingers.

• Pull to refresh arrived on the Mac. It's a small thing, but it's a touch gesture, and seems a bit redundant trackpad-only computer, although I use it all the time now, so maybe not!

• The 27.2 betas have even more. Code in Photos, Notes, Safari and Quick Look references touch gesture recognisers, and the Dock now has a new long-press animation for icons.

• The interface has been moving this way for a while. Liquid Glass, with its bigger, rounder, more spaced-out controls, makes a lot more sense once you picture a finger instead of a cursor. I wrote about Liquid Glass from a designer's perspective last year, and in hindsight, a lot of it looks like groundwork.

Apple doesn't usually rebuild parts of its operating system for no reason (I’ll resist the obvious Liquid Glass joke here). In a way, the evidence was hiding in plain sight. That, and the rumours, were pretty explicit.

What about the Apple Pencil?

This is the part I'm most curious about as a photographer and a designer. If Apple is putting a touchscreen on a MacBook Pro, Apple Pencil support doesn't seem like a huge leap. Sidecar already supports the Pencil, so the groundwork is there. There’s been no mention of this in the rumours, and it probably wouldn’t work too well, but it is possibly something that could happen.

For photo editing, it could be useful. Precise masking, retouching, dodging and burning: there are plenty of tasks where a pen would be better than a mouse or trackpad. Being able to reach up and paint in a mask in Lightroom or Photoshop without reaching for a separate tablet has some appeal.

That said, a laptop is definitely not the best form factor for this kind of thing. Drawing on a vertical screen has the same problem Jobs described. You can't rest your hand properly, and the screen wobbles every time you touch it, even with the best Apple-designed hinge. The iPad works for drawing because you can lay it flat or hold it like a piece of paper. Unless Apple has done something radical in mind (or a kickstand), I suspect Pencil support on a MacBook would be highly unlikely. But you never know. I’m just thinking out loud here! Of course, there’s always the possibility they allow it to flip over into a tablet mode, but that seems even more unlikely.

Was Jobs wrong, or has the world changed?

Steve Jobs wasn't wrong about the ergonomics. Holding your arm up to a laptop screen for long periods is still tiring, and I don't think anyone wants to do all their work by poking the display. Not to mention the inevitable fingerprint problem. Of course, that’s always a way to sell more of the infamous Apple polishing cloths. However, things have changed since Steve made those comments, and most people now spend so much time on phones and tablets that touching a screen is instinctive. I've caught myself doing it a few times after switching from my iPad on a project.

Having said that, I’ve been watching the commentary around this on social media, and according to many users who use touchscreen PCs, they hardly ever use the touch function. But then social media isn’t a reliable barometer of user experience, so who knows.

We'll find out next week how Apple has actually approached it. I'll have more thoughts once it's been announced.

The Tandem OLED upgrade may be far more interesting than a touch screen

The thing I’d be far more interested in is the potential of an OLED screen. This in itself could be a big upgrade over the already excellent screens on the MacBook Pro. Given the iPad Pro uses a Tandem OLED, I’d expect Apple to use the same tandem OLED technology. That stacks two OLED layers on top of each other, which roughly doubles the brightness. On the iPad Pro, Apple quotes 1,000 nits full-screen and 1,600 nits peak for HDR. For photographers, the big win isn’t really brightness, since the current mini-LED MacBook Pro already hits similar figures. It’s that every pixel produces its own light. That means true blacks and no “blooming” halos around bright objects on dark backgrounds, which is exactly the kind of thing that makes judging HDR edits difficult right now.

It’s worth remembering that OLED has a long history in the professional video world. Sony’s BVM-X300 OLED reference monitor was for years the standard for grading and mastering HDR content. I have seen one of these in action, and they’re amazing. Incredibly expensive, but amazing. I also have a friend who works as a professional colourist on feature films, and they recommend the iPad Pro with its OLED screen to clients for watching back edits, as it’s the most accurate screen compared to their studio monitors for the price. Having one of these in a MacBook would be amazing. This is probably more of a thing for video and film work, mind you, than photography, but still! It will be expensive, mind you.

Over to you

I've never used a touchscreen laptop for any length of time, so I'm curious: if you've used a touchscreen Windows PC or a 2-in-1, how did you find it? Did you actually use the touchscreen, or did you forget it was there after a while? What do you want to see in a high end MacBook Pro? Let me know in the comments.

Basically, the idea is to use visual Intelligence to create captions for any image. Visual Intelligence uses the screenshot tool to select part of the screen, and then you just ask Siri about what you have selected. In this case, you can ask it to create captions for you.

To activate Visual Intelligence, you use the same command key for taking a screenshot selection, so it’s Command + Shift + 4 and then just press Command again to go into Visual Intelligence mode. You will notice that the cursor changes, and now the marquee has a glass look when you drag it. Drag it around the image on the screen, and then you just ask Siri to create a caption for you. You can also get it to generate keywords. Another thing I cover is how to be more specific, so if you know something about the location of the photo, include that in your prompt.

In the video, I also explain how to use Apple’s new “describe a shortcut” feature to automate this a bit. You can basically use the shortcuts app to create a shortcut that will do all of this in one step for you and copy the results to the clipboard. the prompt I used to create my shortcut is below:

Describe a Shortcut Prompt: “Take a screenshot and create a caption for the image in the screenshot, ignoring the user interface. Copy that caption to the clipboard. Wait five seconds, then create keywords for the same image, also ignoring the user interface. Then copy those to the clipboard”

Once you run this, it will create the caption, copy it to the clipboard, and wait 5 seconds before creating keywords and copying those to the clipboard. The wait is so you have time to paste the caption before the keywords are copied to the clipboard.

DXO has released ViewPoint 6, the latest version of its software for correcting geometry, perspective and distortion. The big change in this version is the addition of AI masks to ReShape Fusion, the tool for adjusting individual parts of a scene within the frame. You can now apply a geometric correction to one element of an image using AI selection to mask the area.

The software also works with RAW files and has basic RAW editing tools, including exposure compensation and white balance. I've written about how useful ViewPoint can be before, when I pointed out how to use it to fix the "stretchy people" effect you get at the edges of wide-angle shots.

I should say up front that I haven't had much time to try ViewPoint 6 out myself yet, so this is more of an announcement than a review. I may follow up with my own thoughts once I've spent more time with it. For now, here's what's new in version 6.

AI Masks in ReShape Fusion

If you have elements within a scene that still look distorted, even with conventional distortion correction, View Point lets you correct that element independently. This tool is in ViewPoint 5, but you had to manually isolate and correct the element you wanted to correct. With the new AI masks, ReShape Fusion can isolate that element for you, and you can control how far the change spreads into the rest of the image.

One example of where this can be useful is when cropping. Conventional perspective tools work on the whole image. When you straighten converging verticals, the entire picture plane tilts, the rectangle becomes a trapezoid, and you have to crop to get a rectangle back. The stronger the correction, the more of the image you lose. By working on a single element instead, DXO says that for many images the correction no longer costs you the edges of the frame.

DXO is also keen to point out that nothing is generated here. ViewPoint doesn't create new pixels or fill anything in at the edges. It rearranges what's already in the photo, and the dimensions of the frame stay the same. That's a different approach from the generative fill tools some other applications now use for this kind of problem.

AI Selection, Not Generative AI

It’s important to note that the AI part of this tool is limited to making the selection. The corrections themselves are still done by hand.

You start by drawing a rectangle around the subject. ViewPoint then detects the element inside it and builds a mask that follows its outline. The mesh used for the correction is built to match that outline, so the control points sit on the subject itself rather than on a uniform grid across the whole image. The density of the mesh also varies, with a coarse grid in areas that don't need changing and a finer one where you're working.

From there, you adjust the mesh directly. You can:

• Subdivide regions for finer control, or merge them where you don't need the detail

• Direct the subdivision inside or outside the selection

• Switch individual control points on or off

• Control how far a change reaches into the rest of the image

That last one is important, because reshaping one part of an image always pulls on whatever is around it. DXO gives the example of a face near the edge of a group portrait, which can be brought back towards normal proportions while limiting how much the adjustment affects the people beside it. The same mesh can also be used to reposition or resize an element that looks out of proportion with the rest of the scene.

The Other Geometry Tools

The rest of ViewPoint's toolset carries over from the previous version. The software is built around three common geometry problems, all of which become more noticeable with wide-angle lenses:

• Barrel distortion, where straight lines bow outwards

• Converging verticals, caused by tilting the camera

• Volume deformation, where subjects near the edge of the frame look stretched

Lens distortion is corrected automatically using DXO's lens modules, which are based on lab measurements of thousands of camera and lens combinations, and the maximum field of view is preserved. Volume Deformation corrects stretching at the edges horizontally, vertically or both. There are also automatic and manual perspective tools for full-frame corrections, with ReShape Fusion handling the localised adjustments, along with horizon correction and the Miniature Effect.

ViewPoint tends to be thought of as a tool for architecture, and that's where it's most obviously useful. But the same problems come up in interiors, real estate, events, travel, street, portrait and landscape photography, so it's applicable to a lot more than buildings.

RAW Editing

ViewPoint 6 can open RAW files directly in the standalone application, along with TIFF and JPEG, without converting them first or round-tripping through another application. RAW files are rendered using DXO's own camera profile in the DXO Wide Gamut colour space. ViewPoint also reads the image metadata to download and apply the matching DXO Camera-Matching Profile, and corrects distortion automatically.

There are two new controls for preparing the image before you start on the geometry:

• Exposure Compensation, with a range of -4.00 to +4.00 EV, available for all formats

• White Balance, with presets, Temperature and Tint sliders and an eyedropper for RAW files. JPEG and TIFF files get a single slider and the same eyedropper.

Editing is non-destructive. Corrections are saved to a sidecar file next to the image, so the original file is never changed, and your edits are still there when you reopen the image later. ViewPoint's sidecar files are kept separate from the ones PhotoLab writes, so the two applications won't interfere with each other. However, to bake in the distortion correction for use in another application, you will still need to export the corrected image. Finished images can be exported as 8-bit or 16-bit TIFF, or as JPEG.

Standalone, Plug-in or Inside PhotoLab

ViewPoint 6 works as a standalone application and as a plug-in for Lightroom Classic, Photoshop and Capture One. If you own PhotoLab 10, the ViewPoint tools are integrated directly into PhotoLab's Geometry palette.

Availability and Pricing

ViewPoint 6 is available now from the DXO website for macOS and Windows. Pricing is as follows:

New Licence: $129.99 | €119.99 | £109.99

Upgrade (from ViewPoint 5): $84.99 | €74.99 | £69.99

There's a 30-day free trial available from the DXO website.

You can use my code: ThomasFitzgerald15 to get 15% off (new customers only, I'm afraid).

This isn't a sponsored post; however, I'm on DXO's affiliate programme, so if you use my code, I get a small commission, but DXO has no input into what I write.

A new browser-based version of Luminar has been released, bringing RAW processing and several of Skylum’s familiar editing tools to the web. There are also a couple of smaller releases worth knowing about: fixes to Exposure’s rendering on Apple Silicon Macs and a revised preview system in Topaz Gigapixel.

Luminar brings RAW editing to the browser

Skylum launched Luminar on Web on 15 October, offering browser-based RAW editing alongside tools including GenErase, photo restoration, sky replacement and Light Depth lighting adjustments. Its launch announcement also states that there will be cross-device synchronisation, with basic functionality available free and full access advertised at US$69.

For photographers who move between different computers or devices, it’s easy to understand why this might be useful: being able to access your editing software without having to install an application. Whether it becomes useful for more demanding work will depend on how well it works, and things like camera support, export quality, responsiveness and how reliably edits move between devices.

I had a quick try of it, and it does work, but the Free trial requires you to sign up with a credit card, so - not doing that - and the basic edits are a bit slow (but still usable), to be honest. Still, in a pinch, if you’re a Luminar user, I can see this being quite useful. I haven’t really been keeping up with Luminar recently, but I must have another look at it. If you use it regularly, please let me know your thoughts in the comments below.

Exposure X7 addresses incorrect rendering on Apple Silicon

Exposure X7 version 7.3.1, released on 14 October, fixes a problem with Shadow and Highlight adjustments not rendering correctly when using GPU acceleration on Apple Silicon Macs. The update also addresses scattered white pixels in the shadows of monochrome RAW files, including those from the Leica Q2 Monochrom. Both changes appear in the official release notes. Exposure Software. The release also fixes preset-related error messages when Exposure is used as a Photoshop or Lightroom plug-in, addresses several crashes and adds Canon EOS R3 support.

I actually thought that they stopped making this application. I used to use it ages ago, and it showed promise, but then for a while they hadn’t updated it. I miss the old alien skin plug-ins that Exposure evolved from, but I must give the application itself another try. They seem to be a bit slow at adding camera support, mind you. My Canon R6II was only added this year, and that’s a three-year-old camera that has already been superseded. And incidentally, I tried the trial version, and despite supposedly being supported, it still doesn’t see RAW files from my Canon.

Gigapixel moves previews to the GPU

Topaz released Gigapixel 1.4.0 on 12 October, with changes aimed at making large-image previews faster. The updated preview system uses the GPU, while a preference allows users to return to the previous behaviour. The release also fixes Autopilot stalling when reloading an image and updates Topaz’s processing backend for Mac support. Topaz’s announcement describes this as a performance and reliability release. Topaz Community

For photographers preparing large prints, preview speed can have a considerable effect on the editing process. Comparing settings becomes tedious when every adjustment requires a lengthy wait, so a faster preview could be useful even without a new enhancement model.

There are still limitations to bear in mind. Topaz lists known colour issues involving images with Apple’s wide-colour sharing profile and CMYK export-queue thumbnails. Those make colour consistency worth checking when evaluating the update, alongside the speed improvements.

Bonus Gear News!! Fujifilm updates its square-format Instax camera

Fujifilm announced the instax SQUARE 2 on 14 October, with availability scheduled from 28 October. It is joined by SOFT HAZE square-format film, which Fujifilm says uses a newly developed emulsion to produce softer light effects and warmer colour tones. The official announcement covers both products. Funnily enough, I never actually dipped my toe in the Instax pool, being more of a Polaroid man myself!

Help Support the Blog

Check out my eBooks and Presets

Check out my photography eBooks , Capture One Styles, and Lightroom Presets available on GumRoad.

Subscribe to my YouTube Channel

Check out my YouTube channel for tutorials, Vlogs, and more!

Buy me a coffee!

If you want to say thanks or help, then you can feed my caffeine habit and buy me a coffee via PayPal with a one off donation to my PayPal tip jar. (Please note that PayPal doesn’t make it easy to respond to these, so just know you are thanked in advance)
Note that this post contains paid affiliate links. We get a small commission for purchases made through these links, which helps run this site.

There have been a few interesting developments in photography software over the past week, with Adobe introducing another way to edit photographs using generative AI, ON1 releasing a new version of its noise reduction software, and some practical improvements arriving for Lightroom and Photoshop users.

Some of these features are available now, while others are still in early access or beta. Here is a closer look at what has changed and what is new in the world of photo software.

Lightroom introduces Prompt to Edit

The September update to Lightroom desktop, version 9.6, introduces an early-access feature called Prompt to Edit. This lets you describe changes you want to make to a photograph, or choose from example prompts for tasks such as restoring colours, fixing dust and scratches, and reducing blur.

This brings generative editing directly into Lightroom, without having to send it to Firefly. It should also be noted that this doesn’t work like an AI agent controlling Lightroom. This uses generative AI to edit your image, and so it can change things you may not want changed.

However, there are some details worth understanding before treating it as another everyday adjustment tool. Adobe currently lists Nano Banana 2 output options at 2K and 4K, and the generated result is saved as a DNG alongside the original photograph. The interface also lets you know how many AI credits this will use.

While this saves as a DNG, it’s not a raw-type DNG, but rather flat pixel information, so any raw data from your camera will be lost using this.

To be honest, this isn’t something I would ever use, especially for client or public work. Because it uses generative AI models, it completely regenerates your image, and while some people may be ok with that, others definitely won’t. I would use this with extreme caution, especially if you’re using this on client work. In the example below, I asked it to brighten the image and add sunset colours to the sky, and while it did this, it also changed things liek the railings, added people where there were none and so on. It’s now a different image totally, and not the one I shot. This is the major issue with this and all generative AI editing. I get that some people won’t care, but I have major issues with this kind of thing, and I wish they would leave it out of Lightroom.

It should be noted that these tools were recently added to Photoshop too, and the same caveats apply. This shouldn’t be confused with the “chat to edit” feature in the Photoshop Beta, which works by letting AI control the Photoshop interface, but more on that in a future post (I tried it and it’s not very good)

Anyway, “Prompt to Edit” is currently an early-access feature in Lightroom desktop.

ON1 releases NoNoise AI 2027

ON1 released NoNoise AI 2027 on 1 October, bringing updated noise reduction and sharpening tools to its standalone application and plugins for Lightroom Classic and Photoshop.

According to ON1, the release includes rebuilt noise reduction and TackSharp AI models, along with a selector offering six models matched to different types of subjects. The idea is that different photographs need different treatment when removing noise and preserving detail. Fine feathers, skin texture and stars can all suffer when noise reduction is applied too aggressively.

I haven’t actually tried it yet, but it would be interesting to compare it with Lightroom’s denoising and DxO’s RAW processing tools, such as PureRAW (my current favourite denoising tool)

NoNoise AI 2027 is available for Mac and Windows, and ON1 continues to offer an outright purchase option and a trial version. It should not be confused with Photo RAW 2027, which remains forthcoming, with a release targeted for late October or early November.

DxO PhotoLab 10.1 adds native Affinity integration

Another recent update worth mentioning is DxO PhotoLab 10.1, released on 22 September as a free update for PhotoLab 10 owners. This introduces direct export to Affinity’s native .af format, making it easier to move from RAW development into more detailed editing. Once you have finished your adjustments in PhotoLab, including optical corrections and DeepPRIME noise reduction, choosing Export to Affinity creates a document with those corrections applied and opens it directly in Affinity. There is also an option to include the uncorrected image as a second layer, allowing you to compare the two versions or selectively blend them using a mask.

By the way - if you’re buying anything from DXO, you can use my affiliate code: ThomasFitzgerald15 to get 15% off (new customers only, I’m afraid. This is a paid affiliate code and I get a small commission if you use this.)

Better Reflection Removal in Lightroom

Lightroom Classic 15.6 and Lightroom desktop 9.6 also bring improvements to Reflection Removal. Adobe says the updated processing produces fewer smudges and ghosting artefacts when dealing with reflections on windows, glass and water.

This is particularly relevant to travel and city photography, and I recently used this extensively (the old version, not the new version) when I was shooting from the observation deck of the Shard in London. It will be interesting to see if the changes actually make any difference.

These improvements are included in the September updates. (Yes, I know, I’m late getting to it!)

Camera Raw improvements for Photoshop users

There are also some interesting developments in Camera Raw.

Camera Raw 18.7, announced on 29 September, adds a thumbnail-based Grid view for browsing, comparing and rating groups of photographs. This is interesting, as traditionally, Camera RAW was paired with Bridge for this kind of thing, so make of that what you will. It also introduces automatic Match Exposure, including an advanced option designed to compensate for changing light conditions across a sequence.

Separately, the Photoshop beta build released on 1 October introduces multithreaded saving for PSD and PSB files. This allows qualifying documents to be written using parallel processing, potentially reducing the time spent waiting for large layered files to save.

The benefit depends on the document size and your hardware, so there is no single speed improvement that will apply to everyone. Nevertheless, for anyone working with complex retouching files or large composites, saving performance can make a noticeable difference over the course of a working day.

This feature remains in beta. Adobe recommends using Save As or Save a Copy when testing it, so any experiments should be carried out on duplicate files.

Help Support the Blog

Check out my eBooks and Presets

Check out my photography eBooks , Capture One Styles, and Lightroom Presets available on GumRoad.

Subscribe to my YouTube Channel

Check out my YouTube channel for tutorials, Vlogs, and more!

Buy me a coffee!

If you want to say thanks or help, then you can feed my caffeine habit and buy me a coffee via PayPal with a one off donation to my PayPal tip jar. (Please note that PayPal doesn’t make it easy to respond to these so just know you are thanked in advance)
Note that this post contains paid affiliate links. We get a small commission for purchases made through these links, which helps run this site.

Permalink

]]>Random Acts of Photography - Wales EditionSite NewsThomas FitzgeraldTue, 08 Sep 2026 15:56:27 +0000https://blog.thomasfitzgeraldphotography.com/blog/2026/9/the-website-situation-for-photographers-is-getting-ridiculous53c631e4e4b0c4d68989cbef:540c893ee4b04215b3085eaf:6aa0302c5ee77f5a64d1a863

I thought I was done with website woes when I recently upgraded to the latest version of Squarespace, but alas, the AI-dominated world we live in now has other plans. I could go into the ramifications of Google switching to its AI summaries as its primary search result and the result that is having, but I’m saving that for a bigger post. The TL;DR version is I’m becoming increasingly reluctant to write articles here, when they’re only going to be stolen and repurposed by Google, while it simultaneously kills any search traffic to the site. But there’s another problem that I’m sure most other photographers and other website owners have had to contend with, and that’s the problem of bots.

You know this is affecting many websites due to the huge rise in Cloudflare “are you a human” authentication boxes. I even saw some twit on X the other day complaining because his personal Claude bot couldn’t get past one of these dialogs. It’s actually becoming a serious problem, and unfortunately, at the moment, there’s no easy way to prevent them on Squarespace.

Just this past week, I had a traffic spike of 25,000 page views that were entirely bots. This totally messes with analytics, and more importantly, they’re almost certainly up to no good, either scanning the site for malicious reasons, attempting to hack it, or just stealing my content. And the most frustrating thing is that there is a whole cohort of people out there who don’t see a problem with this, and that we should just roll over and accept our fate to the AI overlords. It’s incredibly depressing.

I actually have no idea any more if the content I write is successful or not. The analytics in Squarespace are basically useless now because they don’t have any bot prevention, so half the traffic is basically meaningless. Google Analytics is also not much use because, in its case, the bot filtering is perhaps too aggressive. On top of that, the cookie law means that at least 20% of people turn cookies off, blocking Google Analytics. (the Squarespace one captures more because some of it is cookie-agnostic.) On top of that, traffic continues to decline as Google is slowly choking off search traffic.

Anyway, this is the reason I’ve been posting less on here. I’m basically just writing content for Google and other AI companies to steal now, and until I come up with a solution, I’ll be dialling back my writing on this website. If anyone has any ideas on how to address this situation, I would be grateful. There is an option on Squarespace to ask AI not to scrape, but it basically does nothing because most of these companies or third parties aren’t exactly ethical.

I will still write here, just a little less often until I can find a way to stop this. There’s also my other photography blog, which doesn’t seem to suffer from the same problem yet, and you can also find my writing over on my Medium blog, designtography.

DxO has released PhotoLab 10, the latest version of its RAW photo editing software for macOS and Windows. The update adds new ways to make selective adjustments, expands the software’s AI-assisted masking capabilities and introduces a new colour grading tool.

The release also adds a direct connection to Affinity for pixel-level editing, along with several changes aimed at improving photo library management and everyday editing workflows.

Depth Masks for distance-based adjustments

One of the main additions in PhotoLab 10 is Depth Masks. The feature uses AI to analyse an image and create a depth map, allowing adjustments to be targeted according to an element’s distance from the camera.

This means photographers can make adjustments to foreground, midground or background areas without needing embedded depth information in the image. Once an adjustment has been applied, sliders and feathering controls can be used to refine where the effect appears.

The approach provides another option for selective editing alongside PhotoLab’s existing masking tools.

A new Colour Grading tool

PhotoLab 10 also introduces a new Colour Grading tool built around a single colour wheel. It allows shadows, midtones, highlights and overall colour balance to be adjusted independently on a single colour wheel.

More AI-assisted masking options

AI Masks have been expanded in several areas.

The software can now identify individual people in a scene. Photographers can create one mask covering everyone, combine selected people into a single mask, or create separate masks for individual people. Individual facial features can also be selected when more targeted adjustments are needed.

PhotoLab 10 adds further face segmentation categories, including eyebrows, eyes, irises, pupils, sclera, teeth, facial hair, face skin and lips. These sit alongside existing AI selection categories such as sky, subject, background, people, animals, flowers and vehicles.

AI Masks can also now be refined with U Point™ technology. After creating an AI-based selection, the U Point eyedropper can be used to expand or contract the selected area based on colour and luminosity.

Another AI-assisted addition is automatic sensor dust removal. PhotoLab 10 can detect and remove sensor dust and small imperfections at the RAW level before they become part of the editing workflow.

More flexibility with U Point adjustments

PhotoLab’s local adjustment tools have also been expanded. A new Control Brush combines conventional brush-based painting with U Point’s colour- and luminosity-based selection.

Control Points can now be switched from circular to elliptical shapes, which can be useful when an elliptical selection better matches the subject being adjusted.

Together with AI Masks and Depth Masks, these options give photographers several different ways to create and refine local adjustments depending on the image and the area they want to edit.

Direct editing in Affinity

PhotoLab 10 adds a new button alongside the existing export options that sends a photograph directly to Affinity for pixel-level editing.

This provides a route from PhotoLab’s RAW processing environment into Affinity for tasks such as retouching, compositing and detailed pixel manipulation. DxO describes this as the first stage of an ongoing integration between the two applications, with further developments planned for future releases.

The one thing I will say about this that might be a little confusing is that PhotoLab 10 does not support the Affinity file format, despite Canva making it open. So you’ll have to use TIFF. I would have suspected that they would have added this given how this is a big part of the announcement, but anyway. Given the note about further collaboration, perhaps this will be added in the future.

Other workflow changes

There are also several smaller workflow improvements in PhotoLab 10. These include support for as-shot crop ratios, histogram adjustments directly within PhotoLibrary, improved database backup and restoration, and simplified folder relocation management.

These changes are aimed at making it easier to organise and work with larger photo libraries alongside the new editing features.

Pricing and availability

DxO PhotoLab 10 is available now for macOS and Windows,

New license $249.99 / 249,99 € / £229.99

Upgrade from DxO PhotoLab 8 or 9: $129.99 / 129,99 € / £114.99

The release is available from DxO’s Online Store, and a 30-day trial is also available from the DXO website.

You can also use my code: ThomasFitzgerald15 to get 15% off (new customers only, I’m afraid).

Note that this post contains a paid affiliate code. We get a small commission for purchases made using this code, which helps run this site.

I missed this when the recent round of Adobe Photography updates dropped for August 2026, but they added a much-requested and long-missing feature for Lightroom on the iPad. There’s not much point in me teasing it, as it’s in the title, but yes, Lightroom on iPad finally gets Denoise. And it actually works pretty well.

How to use Denoise in Lightroom on iPad

To use Denoise on the iPad, you’ll need to be using Lightroom version 11.5 or later, and I suspect it’s only available on certain iPads. I’m using an older iPad Air with an M1 processor, and it works fine for me. I can’t find any proper support notes which list system requirements, so let me know if you have an iPad that’s not supported. If not, I’ll assume anyone capable of running this version will work.

To use the new Denoise feature, just go to the detail panel in the edit section and tap on the Denoise button to turn it on. Lightroom will first download the full-resolution image if it’s only using a smart preview or proxy. Then it needs to download the AI model for it to work. So the first time you use Denoise on the iPad, it will take a little longer. Once it’s done this, it will then apply Denoise itself. On my older M1 iPad Air, it took about 30 seconds, although I didn’t time it exactly. I suspect if you have an iPad Pro or an iPad Air with a newer processor, this would be much faster.

Conclusion

That’s pretty much all there is to it. Other than having to download the model the first time, it pretty much works the same as the desktop version. There’s no sign of super resolution yet, and the iPhone version of Lightroom gets generative expand, but the iPad version does not for whatever reason.

I had taken some photos over the weekend, and I was processing them in Lightroom when I noticed that the software was acting unusually. When I applied a preset, the preview worked perfectly, but it took a few seconds to actually apply. At first I thought that maybe this was an issue with the new version of Lightroom, so I quit and restarted. It was then that I noticed another weird issue: the software was still trying to write the metadata to disk. So I thought that maybe my disk was running out of space. I checked, and there was still lots of room. Curiouser and curiouser I thought. While it turned out not to be a disk space issue, I was actually close to the source of the problem, thinking it was disk-related. But let me back up a minute.

Were SSD issues slowing down Lightroom?

My main workstation is a Mac Studio with an M1 Max processor and a 1 TB internal drive. As this isn’t much space, I keep my Lightroom catalogue on an external drive, a Samsung T7 SSD. I basically use this as an overflow for stuff that I would normally keep on my system drive. My actual photos are stored on a different SSD. I’m always having an issue with space, especially on my system disk. It’s recommended that you keep at least 10-15% of an SSD free for optimal performance and to avoid issues, and I normally do that, so that’s why I checked to see if I had been running out of space.

I don’t know what made me think of it, but just to be doubly sure something wasn’t gumming up the works somewhere, I opened Blackmagic Speed Disk to test the speed of the SSD. I was thinking it was a possible problem with too many things connected to my Mac Studio, but anyway, the result quickly found the cause of the problem. My Samsung T7 was running slowly. Ridiculously slowly.

The drive normally has a transfer speed of around 800 MB/s, but the write speed had slowed to just 60 MB/s.

“Well, there’s your problem”!

Now, there are lots of things that can slow down an SSD to this extent. I also had my Parallels disk images on this for when I need to run Windows, and I had been running it, so I thought maybe this had used all the disk space temporarily and the drive was still clearing its cache. But it wasn’t that. I tried clearing some more space and leaving it for an hour to see if it would improve, and it did, but only by a few MB/s. At this point I was beginning to think the drive might be dying, which was the last thing I needed, but a quick Google search revealed that this is a common problem with the Samsung T7. The solutions weren’t exactly forthcoming, though. Eventually, I just copied everything off, reformatted it and put everything back. That solved the issue. For now, anyway. If it happens again, I will definitely replace the drive. One thing I did do was change the format. It had been “Mac OS Extended Journaled”, but I took the opportunity to change the partition to the newer APFS. Samsung does have software to manage their drives, but its Mac version is terrible and requires you to mess around with the boot settings and security settings on your Mac, and I wasn’t in the mood for that.

So what does this mean for Lightroom speed?

Once I had everything back on the newly formatted drive, I checked the speed again, and everything was back to normal. The drive was performing at its standard speed, and so I hoped this had solved the issue. I waited another while just to be sure and checked again, and it was still fine, so I fired up Lightroom, and sure enough, it was back to its old self.

So what’s the moral of the story?

Almost every day I see someone on social media complaining about Lightroom that it suddenly slowed down and assuming it was all Adobe’s fault. This is especially the case after an update. The thing is, though, it’s rarely the software itself causing an issue directly. Now, you could argue that it should be more robust, and it should be better able to handle things, and that could be a fair point. But I have been testing Lightroom thoroughly for years, including running beta versions, and it’s almost always something wrong on your system or configuration if the software suddenly slows down. Well, in Lightroom Classic anyway. Unfortunately, you have less control over things in Lightroom Desktop. Occasionally there are issues with new versions, but these are usually well reported and are usually fixed pretty quickly, at least in recent years.

Note that this post contains paid affiliate links. We get a small commission for purchases made through these links, which helps run this site.

Adobe has released the latest round of Lightroom and Camera RAW updates, and in this article we’re going to concentrate on Lightroom Classic. Aside from the usual under-the-hood tweaks and bug fixes, Lightroom Classic 15.5 adds three new key features, one of which I’ve been asking for for a long time!

Speaking of Lightroom, don’t forget to check out my Lightroom presets over on my Gumroad store! Shameless plug!!!

Mask Feathering and Adjustments in Lightroom Classic 15.5

Ever since Adobe introduced AI-generated masks, I’ve been asking them to add the ability to modify those masks. In fact, I’ve been wanting these features since before AI masks. Capture One has always had the ability to blur a mask, and so it’s good to see Lightroom finally gain this ability. Well, sort of.

If you use any of the AI masks, you will now get two new sliders. One to feather the mask and one to control the edges. The feather slider basically blurs the edge of the mask, and the Edge slider expands or contracts the mask. If you’re familiar with green screen keying in video, these are the equivalent of a matte choker tool (yes, that is the actual term!)

The only thing about this that I find a little disappointing is that you can only apply it to AI-generated masks. You can apply it to brushed-on masks, for example, which is something that you can do in Capture One. Also, if you combine multiple masks, the mask adjustment sliders are on the individual masks, and not the overall combined mask.

Adjust the opacity of Lightroom’s Crop Tool in Lightroom Classic 15.5

The second new feature allows you to adjust the opacity of the outside of the crop tool when cropping in Lightroom Classic. What this means, basically, is that when you crop, outside the crop tool is normally pretty dark to indicate which areas are being cropped out. Well, in the new version, you can now adjust the opacity of this area, so if you are trying to get a really precise crop and the cropped area was too dark, you can adjust the opacity of this now.

To adjust this, when the crop tool is activated, right-click on the image and from the contextual menu that pops up, you should get a list of opacity options, ranging from 100% to Off. Simply select whichever one works for you!

Render to DNG

Adobe has added the ability to render an image to DNG. This will flatten your image and bake in all the edits, giving you a DNG of the image, and it will reimport it into Lightroom. This is not a RAW file but a flattened file containing all your edits baked into the raw file. It’s as if you had exported it as a TIFF and reimported it. Interestingly though, while it doesn’t create a RAW file, the resulting DNG does retain some exposure latitude. If you look at the DNG using the HDR option in the Develop module, you will see that it does retain some information over 100% brightness. This is an interesting one, because, on the one hand, people have been asking me about a similar feature for years, but on the other hand, I suspect some people won’t be happy that this doesn’t do something like the old Enhance tool did, and create a linearised RAW file. But there are some good uses of this.

One of the big advantages of this Render to DNG feature is that this will allow you to do denoising and super-resolution on the same image. Now, you could do this in the past, but it involved considerable hoop jumping, as just exporting to DNG and re-importing didn’t work. Now, all you have to do is apply Denoise, render to DNG and then do Super Resolution on the final image.

It’s also useful if you do a lot of edits to an image, and then want to do an overall look on top of changes you might already have made. In this case, you can render to DNG and then apply a film preset to the DNG, adjusting the opacity of the preset to get the right look.

The biggest downside of this Render to DNG feature will be the file size. On an image taken with my Canon R6II, this was 128 MB compared to roughly 50 MB for a RAW file. Now, if you’re not doing a lot of these, it’s not that big a deal, but if you’re very tight on space, it may be an issue.

Conclusion

I wrote a while ago that Camera RAW seemed to be getting all the new features, and that Lightroom Classic was starting to get left behind. This update takes a small step in the right direction by adding at least one of those features to Lightroom Classic. However, the disparity continues with new features added to this round of Camera RAW too. What’s more, there is starting to be a divergence between Lightroom Desktop (Lightroom Mac & PC to use the official name) and Lightroom Classic, as the version of Lightroom Desktop just released adds generative fill to the crop tool, which isn’t in this version of Classic.

On a recent trip to London, I took quite a few photos and some video at the top of the Shard, one of London’s most famous buildings. I have already posted a few of these, and there are a lot more to come. However, in this post, I wanted to talk about the challenges I faced editing a subset of the images I took and how I addressed them. I ended up using both Lightroom and DaVinci Resolve’s new photo editing mode for these. The latter is something I particularly want to talk about, as it has some interesting abilities, as well as some quirks that those not familiar with the software should be aware of.

Pre-Processing in Lightroom

I started with these images in Lightroom. The reason for this is twofold. Firstly, Resolve’s actual RAW conversion is still not great, to be honest. It lacks lens corrections and proper colour profiles. The second reason is that I wanted to remove the reflections. I was shooting through glass, and so it was difficult to avoid reflections because I didn’t have a polariser when I was shooting these. I also didn’t want to get right up to the edge to try and block the reflections (it’s very high!)

Luckily, Lightroom’s “Remove Reflections” works a treat on photos like this and totally removed the reflections. I actually batch processed the entire set of photos I took from the shard with the reflection removal tool. It took some time to process, but luckily it happens in the background now in newer versions of Lightroom.

Once I had done this, I made sure that there was no image information clipped and then exported them as TIFF files for import into Resolve.

Editing in Resolve

So you may be wondering, why bother editing in DaVinci Resolve at all? Why not just do it all in Lightroom? Well, the issue I was having was one of consistency. Because the light was constantly changing and I was shooting through glass, at different angles, and in different surroundings, I was having trouble getting consistent results. This is where Resolve’s much greater control over colour comes in. In Resolve, it’s much easier to do targeted and finer adjustments than you can in Lightroom.

The colour wheels offer much finer control in Resolve, not to mention that there are various modes for the colour correctors. They just work much better than the ones in Lightroom, which are sort of a blunt instrument. You can also specifically target colours and areas of the image more easily than you can in Lightroom, plus you get a visual reference for your corrections in the form of the node tree. Now, you absolutely could make these adjustments in Lightroom; it’s just much easier and quicker to get the results you want in Resolve. That is, of course, if you know how this kind of colour correction works. And this kind of node-based editing is something I’ve been doing for years, so it’s second nature to me.

Breaking down the node tree

One of the advantages of working with a node-based editor, is that it’s easy to break down the colour corrections, making it easier to see what you’re editing and easier to go back to it later. This also makes it easier when you apply those edits to another image.

If we look at the node tree I used for the image above, I’ll talk you through what I did. There are no hard-and-fast rules for this kind of thing, and you can work however you want, but this is how I approached it.

Node1: The first node is for overall brightness and tonal fixes. I haven’t done much here, except reduce the highlights slightly. Think of it as more of a technical fix before passing it to the second node.

Node2: This is where I did the bulk of the colour correction, using the standard and the HDR colour correction tools. The HDR three-way colour corrector lets you select finer areas of brightness to work on. It’s hard to explain in a short post. To be honest, I’m not even sure I fully get it myself; I just know how to use it!!!

Nodes 3 and 4: I used these two parallel nodes with the gradient mask to tweak the sky and ground values separately. There’s no direct “clarity” tool that I’m aware of in the colour editor, although there may be in the effects, but that’s one of the limitations that I’ll get to later. I did manage to add some details to the foreground, though, using a reverse blur modifier. (I’m not entirely sure how this works either)

Node 6: is an overall contrast addition using the curves tool. This was just because I thought it was a little flat and needed some more oomph

Node 7: This node adds some off-screen lens flare. I used the Red Giant “Real Lens Flares” plug-in for this. The reason I added this was that the reflections were gone, it lacked some interest in the sky, and there were still a few mismatched colours and artifacts where the reflection removal didn’t do the job properly.

Node 8: This node adds some subtle grain to the image

Node 9: I didn’t use this node in the end, so it’s turned off. It was a colour space transform to convert the colour space to RGB, but I ended up doing this a different way. More on this below.

Limitations and Issues

There are two things you need to be aware of if you plan on editing photos in Resolve. Actually, there are three. The first I’ve already covered: the RAW conversion for stills is still lacking, so you’re best off using something else for that.

Secondly, if you’re using the free version of Resolve, you are limited in the output resolution. You can also only use certain effects and are limited in the number of effects. For example, I was using the free version here, and I could only output at about half the resolution that the image actually was.

Thirdly, you need to be aware of the colour space you are using. This isn’t something that you have to really deal with too much when working with Lightroom, but it’s an issue with Resolve. Because it’s set up for working with video, you’ll have issues if you try to open the resulting images in Photoshop. So the best thing to do is set the output colour space in the project preferences. You can also just put a colour space transform node at the end of the node tree, but this will alter how it looks in Resolve. I would do this if you were working on a mixed project, but otherwise, it’s best to set the output to sRGB in the project settings.

Permalink

]]>London in Miniature