book of the dev

iPad video footage capture and editing workflow: Part 3

Fri, 01 Jun 2012 14:20:45 +0100

After a slight wrong turn, I think I’ve found my capture solution for the forseeable future.

In the previous part of this series, I mentioned BlackMagic Design Intensity range as a possible solution for high quality, full resolution, full framerate capture from iPads and other HDMI capable devices. A week ago, I ordered the Intensity Shuttle for Thunderbolt. Even though my current Mac doesn’t have Thunderbolt (still waiting for those new Macs…), I knew I was going to spend a few days at my parents’ house, and my Dad has a Thunderbolt enabled Mac that I could use.

The BlackMagic Intensity Shuttle is a video capture card which, using either USB 3 or Thunderbolt, delivers full quality, non-compressed HD video from a HDMI or analog source. Which sounds exactly the ticket for my workflow. However, once I got hold of the unit (from the so-far excellent online retailer Jigsaw24), some problems emerged.

Firstly, the unit turned out to be faulty. After a minute or so turned on, the capture preview started losing frames, until eventually after a few minutes turning black completely. This was without even capturing to hard disk, so I knew a slow hard disk wasn’t to blame.

Secondly, even if I got this working, the data rate of a 720p 60fps capture is huge, meaning I would need to invest in a RAID 0 device. Whilst that’s not a huge stumbling block given the low price of such hardware these days, it’s definitely something to consider if you’re looking to get one of these zero compression capture solutions.

So after sending back the faulty Intensity Shuttle, I ordered the H.264 Pro Recorder, also by BlackMagic, and also sold by Jigsaw24. This is a bit more expensive, but compresses video on the fly to H.264, meaning the bandwidth used is much less. So much so that it can interface over a USB 2.0 cable, which means greater compatibility with hardware.

So far it has worked faultlessly, without frame skips, and at a high enough bitrate that video quality simply isn’t an issue. I’ve already pressed it into service creating video for an exciting promotion opportunity which I’ll be able to disclose in a couple of weeks time, and it got the job done with minimum fuss.

Whilst uncompressed video sounds great, it turns out that a high bitrate H.264 is more than ample for my needs, such that I doubt anyone will be able to tell the difference in the final videos. That the intermediate files generated are orders of magnitude smaller and produced flawlessly with modest hardware means the H.264 Pro Recorder is now my preferred solution for video capture.

I’ll post more video in a few days to show the quality that the hardware is capable of.

Food Run alpha aftermath: Part 1 - the survey

Fri, 25 May 2012 15:33:00 +0100

After three weeks of builds, testing, and feedback, the Food Run alpha is over. At the end of the period, I put out a survey to the participants to formalize the feedback and take some measurements. Super exciting pie charts inside.

Over the last three weeks, Food Run has been in the hands of thirty alpha testers. A big round of applause for them - their feedback has been invaluable. This is the first time I have run such a large testing and feedback session, so it’s been a good chance to learn some things about the process, as well as finding out what people think of Food Run.

Here’s the results of the final survey, deployed at the end of the alpha period.

Section 1

The first section of the survey was a grid of “how much do you agree / disagree with the following statements” questions. Possible answers were ranged from 1 to 5 in score:

Strongly disagree
Disagree
Neutral
Agree
Strongly agree

This section concentrated on seeing how Food Run as it currently stands was perceived by the group.

Good results so far. Still some work to do on sound effects, as these haven’t received as much attention as some of the rest of the game. Improvements are in plan. Navigation of the menus has definitely been improved during the alpha in response to feedback, and this decent score reflects that.

Section 2

The next section was in the form of yes / no questions.

That’s a good amount of people who found at least some secret areas. However, from feedback during the alpha, I know there are things I can do to improve the secret areas in general.

These two results can be taken together. It’s my aim that most people will, with perseverance, get through all of the levels in the final Food Run, but that only the more skilful and determined players will be able to get 100% in all the levels. This is looking about right.

As much as I like TestFlight, and it has certainly transformed the ability of us independent iOS developers to run feedback sessions like this, it still has some technical problems. Whether this can be attributed to TestFlight or Apple’s systems is unknown.

Section 3

This section concentrates on some changes I’m thinking of making to Food Run, in response to feedback gained during the alpha period.

This result surprised me. I had thought that most people would find the missing storyline strange. If I do incorporate some kind of storyline, it will be very light, nothing too prescriptive. A comic strip cel or two, nothing more. I also want to keep the undetermined nature of the food buddies - are they real food with legs, or people in suits? Not everything has to be nailed down, and I plan to play a little fast and loose with this.

Again, this surprised me. Most mobile games these days have aped the Angry Birds three star rating system, presumably because people like it. It’s good to remind yourself that you don’t have to blindly follow formula. I can still do more work to reward partial and full completion.

From the feedback and the response to this question, this is something I really need to look at. I forgot to ask the question, “would you like to see enemies in Food Run?”, but judging from informal feedback, that’s something I’ll be looking at along with this question, as my plan for the two are somewhat intertwined. I’m determined, however, that any solution will not increase the complexity of Food Run too much. Elegance and design simplicity has always been an aim for this project.

Section 4

The last section asked for any other comments, as well as providing an opportunity for the test group to suggest their favourite foods for inclusion in Food Run. Whilst I will refrain from printing any of the responses here, there were some great thoughts and feedback in this section, which I will take into account whilst building out the rest of the game.

What’s next?

Stay tuned for part 2, where I’ll disclose what I’m doing to respond to the feedback.

Food Run alpha now open to playtesters, journalists, publishers

Thu, 03 May 2012 16:30:00 +0100

As of today, I’m opening the first round of testing for Food Run. Spaces are limited, but open to the general public. Friends, fellow developers, journalists and bloggers most definitely included. I’d also like to invite potential publishing partners to take a look at the game. Details inside.

Food Run is now at the Alpha milestone! What this actually means varies according to where you work and who you ask, but for Food Run it means that all the core functionality is finished, and a quarter or so of the levels are done.

Which means it’s time to look outwards for some play testing, opinion, and feedback to make Food Run a better game. Anyone can take part, as long as they have access to an iPhone, iPod touch or iPad. First generation iPhones and iPod touch devices (launched in 2007) are incompatible, but everything else will work.

Testers

Play testers will receive a build of the game over the Internet via the excellent testflightapp.com. They will be able to play around 20 levels from Food Run.

All testers will get their name in the credits for Food Run, under the banner “Playtesters”, unless they wish to remain anonymous. In return, testers are asked only to play the game through at least once, or until they get stuck, or bored (hopefully not), and be prepared to give some opinions on what they liked and didn’t like, and why. If testers find any bugs, they should report these too.

Spaces are limited, so if you are interested, e-mail me at foodrunalpha (at) pixelsontoast.com as soon as possible.

Press

Journalists and bloggers who are interested in taking an early look at the game can get this build too - they won’t take up the limited slots allocated to testers.

E-mail me at kevin (at) pixelsontoast.com for more details.

Publishers

At this point I’m also looking for a publishing partner to help Food Run fulfil its potential. Specifically, I’m looking for a proactive partner who can add significant value, primarily in marketing (direct and indirect). If the correct partnership can’t be found, then Pixels on Toast will take the project forward by itself.

E-mail me at kevin (at) pixelsontoast.com for more details.

New website

The Food Run website has just been updated for Alpha. Check it out for the latest videos and screenshots.

http://foodrungame.com

Food Run video: first look at chapter 2

Wed, 25 Apr 2012 14:16:00 +0100

Here’s a first look at chapter 2 of Food Run, “Dessert Island”. Pun most definitely, although perhaps regrettably, intended.

I’d say the graphics are about 90% complete for chapter 2. I’m not going to lie, it’s been a graft getting it to this state. Keen eyes may notice that the backgrounds have changed from pre-rendered to flat vectors. After some to-ing and fro-ing, I found that whilst the radiosity renders were kind of nice, the vector look serves the game as a whole better, and makes the whole thing look a little more cohesive.

Even keener eyes may notice that I’ve dropped the “Pre-Alpha Version” sub moniker in favour of simply “Alpha Version”. Yes, that’s right, Food Run is (pretty much) at Alpha! That means I’ll be on the lookout for testers soon. Stay tuned for that announcement.

Food Run UI Progress: Part 1

Fri, 23 Mar 2012 09:46:00 +0000

Sometimes when you look at your own game during development you feel like something just isn’t cutting the mustard. This was definitely the case with the Food Run user interface. So it was time to redesign it. Progress and Photoshop mockups below.

Note: Click all three images for iPad Retina display resolution (2048 x 1536) images

When comparing to what I would consider the best of breed in super accessible mobile and tablet UI design; titles like Angry Birds, Spy Mouse, Bejeweled, and Draw Something, Food Run came up short. In a way, this is to be expected as 1.) I’m just one person competing against whole teams and 2.) my primary expertise is certainly not art, or graphic design. However, this is no excuse: the end user doesn’t know or care how a game is made, they just care about how good the game is.

I set about redesigning the user interface with the mantra, “bigger, bolder, breezier”, and outside of the alliteration, “simpler”. Although creating a design mantra is in danger of being pretentious, it can really help to guide your decision-making. Faced with a design choice, you can simply ask yourself if that choice fits the mantra. If it does, great. If not, you can safely ignore it. Design is all about choices, and so a system for navigating your way through the myriad becomes essential. The mantra is a razor that makes sure your design is focused on the right things.

I’ll present three screens in this post - the title screen, chapter select screen and level select screen. In this new design, the screens flow in a logical left to right order, and will slide over when transitioning between them.

Here’s the title screen. Main elements of focus are the logo and the play button. I always had the imagery of the 50s Americana diner in mind when thinking about Food Run, so the checkered background is a way of lightly evoking that without cluttering and becoming too twee. Fast food packaging often uses it in the same way. The gift box in the bottom left allows you to gift the game to someone else, taking the user directly into the gifting interface within the App Store.

The navigation bar at the bottom is an evolution of the design I used recently in WordCrasher Blitz. It seemed to work well there, as a constant of most of the UI screens, meaning the user quickly learns to depend on its position. The items in the nav bar are things you want to access from all places - the help, the OpenFeint system, and the settings.

The Food Run logo has been further refined, with the bevel becoming more cartoonish and less rendered looking. The sharper edges between colours rather than gradients have also come to the fore in this new UI design, as seen in the removal of soft shadows in favour of the sharply defined shadows. The remaining gradient, the red-through-beigey-yellow of the logo, is supposed to be reminiscent of food being baked, hence redder at the top. This was present in the previous design too.

Here’s the chapter select screen. Chapters will be portrayed as postcards, with the usual postcard tropes, fixtures and fittings, such as gaudy greeting notices and white borders, and probably in a later chapter, multiple photos collaged into one. Navigation is a vertical carousel, which makes sense as the screen flow runs from left to right. The back button will always be in the same position on all screens - top left, as pressing it slides the previous page in from the left.

Note the bevel style also carries through to the buttons.

The colour scheme is based on the classic theory of cooler colours receding, and warmer colours dominating. Therefore the user interaction elements (buttons and nav bar) are warmer than the rest of the design, whilst the background is blue.

Here’s the level select screen. Again, navigation is via a vertical carousel. The button at the top right shows your currently selected “leader” - the food which starts the level. Pressing it brings up a screen containing all the foods you’ve collected so far, so that you can select a new leader.

At this point in time, these screens exist only in Photoshop - however, they are constructed in such a way that I can quickly and easily rip them apart to produce assets for the final game. That will be the task for the next few days. And then, no doubt, further iteration.

WordCrasher Blitz will be iPad 3 Retina Display compatible

Thu, 08 Mar 2012 15:21:00 +0000

Just a quick note to say that WordCrasher Blitz will soon be iPad Retina Display compatible on the new iPad, soon after the iPad’s launch next week.

Mockups of Food Run running on the new iPad were shown a few days back, but it will be the recently launched WordCrasher Blitz which will see iPad Retina treatment first.

More thoughts on Retina Displays and donuts with legs

Mon, 05 Mar 2012 11:57:00 +0000

The previous blog post about the possibilities of Food Run on the rumoured high resolution iPad 3 has certainly caught the web’s attention. And amongst the punditry and comment threads, some valid and interesting points have arisen. Thoughts and responses.

First and most important: when do we get to play Food Run?

In general, there has been a great reaction to Food Run. It’s always a relief that when you let other people see your game, the response is positive. So it’s probably a good time to put the game in people’s hands and let them play the thing.

At the moment, I’d like to invite journalists to get in touch via e-mail and twitter, and I’ll send out early press preview copies soon. It’s also important to get feedback from the general public, so I’ll open up the beta testing programme shortly afterwards. I’ll be announcing details on this blog.

Other apps would benefit more from a Retina Display than games

I’ve seen this written in many places, and I would largely agree. Whilst the possibility for games is exciting, a Retina Display would mean more elsewhere. Web browsing, for example, would be transformed by such a display, with the ability to easily see the entire width of a page in portrait orientation.

But it’s the possibilities that it brings to magazines which interests me most. My experience of reading magazines on current iPads has been something that falls a little short of expectations. Most magazines are badly adapted for iPad, if at all. Some, like Wired, for example, have obviously gone to huge lengths to respect the format and realize it is somewhat different to the printed page. Other magazines seem to deal with iPad by simply “porting” their magazines over without any formatting or content changes. With these 1:1 translations, it is impossible to read the body text whilst viewing the whole page. Which means you have to zoom and pan to read the articles, something that doesn’t feel like reading a magazine at all.

A higher resolution display would allow you to read the fine print whilst still being able to appreciate the layout and graphic design of a magazine, which is one of the things that makes magazines stand out from web content. Although at this point, perhaps the 9.7” size of the screen will start to become the limiting factor. Magazines are, after all, typically larger than that.

Not all games would benefit from the new capabilities

Everyone loves a good retro pixelated game. The style and the artistry involved in a great retro game evoke a certain time, and a specific style of game. So no-one is suggesting that we should suddenly “upgrade” these titles and force them to run at high resolutions. But there are other types of games for which the upgrade makes sense. Part of what makes games visually interesting is the wide array of visual styles that can be explored. 2D or 3D, vector or bitmap, realism or abstraction, painterly, sketched, black-and-white - the possibilities are endless. And some of these visual styles will benefit more than others from a higher resolution.

My personal prediction for March 7th

The teaser invitations that Apple sends out always hold clues, and part of the fun (and indeed, marketing) of these invitations involves everyone and their dog trying to second guess what Apple are going to announce. So here’s my stab.

The invitation for March 7th reads, “We have something you really have to see. And touch.” My guess is that rather than solely describing the new iPad, this relates to two distinct devices: 1.) something to see, and 2.) something to touch. The “something to touch” being the iPad, and the “something to see” being something along the genetic lines of the Apple TV, whether that be an evolution of the current set top box, or a new unit in the TV form factor.

If true, both the iPad and the new Apple TV will surely feature Siri. That would just make sense.

Links to coverage

If you missed it, here’s the coverage of Food Run’s mockup iPad 3 screenshots (in no particular order):

The Unofficial Apple Weblog (TUAW)

What would Food Run look like on an iPad 3 Retina Display?

Thu, 01 Mar 2012 12:44:00 +0000

Speculation is rife that the upcoming iPad 3, heavily tipped to be making its debut at the Apple event on March 7th, will feature a Retina Display. What would Food Run look like on the rumoured device? Let’s find out. Full resolution shots inside.

JUST SHOW ME THE SCREENSHOTS ALREADY: Can’t be bothered with the texty bits huh? Here’s a Food Run screenshot from existing iPads (1024 x 768) and for the iPad 3 Retina Display (2048 x 1536)

After launching WordCrasher Blitz this week [Free | Trailer], I needed a change of pace and decided to figure out what an iPad 3 Retina Display would mean for my upcoming iOS game Food Run, and how it would look.

But first, some background. In Apple marketing speak, Retina Display means a display whose pixels are so small that at typical viewing distances, they cannot be individually distinguished. Apple coined the term for its launch of the iPhone 4, which featured a screen resolution of 960 x 640, compared to its predecessor’s resolution of 480 x 320. Twice as many dots along each direction, or four times as many pixels in total.

The first generation iPad and the iPad 2 both run at a resolution of 1024 x 768. If Apple were to go along the same route and double the resolution in each direction, then the iPad 3 will be running at 2048 x 1536. Again, four times as many pixels. That would be truly impressive for a 9.7-inch display.

So the question that us iOS developers are asking ourselves is, are we ready for this? If the display has four times as many pixels as the previous iPad, is our artwork ready? Luckily, Food Run was built with this possibility in mind. All of the graphics are created using vectors rather than bitmaps, so they can be scaled to any resolution without becoming pixelated.

I thought it would be fun to put this to the test and pump out the existing art at iPad 3 Retina Display resolution to see how it looks. Below, I’m going to show how the game will look on iPad 1 and 2 compared to the rumoured Retina Display of iPad 3. These shots are representative of actual gameplay, but rather than being screen-capped, they have been assembled from the vector source art in Photoshop.

Here’s a lo-res shot of the whole screen, so you can get an idea of the scale we’re looking at. It’s a typical scene from Food Run.

For clarity, let’s zoom in close and have a look at the food. Here’s how it looks on an iPad 1 or 2 - if you pressed your face against the screen.

And here’s how it would look on an iPad 3 Retina Display.

A huge difference, and considering how much we’ve zoomed in, a crazy amount of detail. Note the crisp black outlines, and the sharpness of the sesame seeds on the burger bun.

However, it’s worth bearing in mind that the advantage in real life at practical viewing distances would be somewhat less. You would have to peer closely at the screen to see such pixelation.

If you want to look at the whole screenshot, click here for the old school iPad (1024 x 768), and click here for the iPad 3 Retina Display (2048 x 1536). Compare and contrast. It will be interesting to see how that big screenshot looks when shrunk down to a 9.7” display.

How about the app icon, then? Regular iPad icons are 72 x 72 pixels in size, so assuming that the user interface on iPad 3 keeps the same icon size, that would mean a resolution of 144 x 144. Old-school iPad icon on the left, iPad 3 Retina icon on the right:

Again, a big difference. Much more detail is retained. The good news is that Apple requires app icons at 512 x 512 for the App Store anyway, so developers will be ready for this change straight away. Expect to see lots of updates on your iOS devices whose description simply reads, “Hi-res icon for iPad 3”. A clarion call for delta patch updates if ever there was one.

Will it happen?

So will we see an iPad 3 with Retina Display? Many portents are pointing to yes. On the other hand, there are some practical reasons that might make it difficult.

Four times as many pixels means four times as much video memory, and much larger assets. Bearing in mind that the 20MB mobile network download size for apps is already claustrophobia inducing, supporting Retina on iPad 3 whilst respecting the limit would be very hard. So if we do see a Retina iPad 3, expect to see that 20MB limit raised, even if only for iPad / universal apps.

With four times the pixels, the graphics card powering the iPad will need four times the fill-rate. That is, it will need to draw four times as many pixels per second. However, jumps of this magnitude between generations are not uncommon these days. With the graphics card, it is the expensive fast graphics RAM required which may prove to be the limiting factor.

I guess we’ll see if it happens on Wednesday.

What do developers have to do to be ready?

Not all iOS developers make their artwork using vectors. Many of the more hand-drawn titles will have pixel-based source art, meaning that unless they were working in high resolution to begin with, there is going to have to be a lot of reworking to fully support the new displays.

Titles using 3D graphics will be in a better state of readiness than bitmap games. It’s likely that just a few lines of code will enable these games to run on the new display. However, for best results, games developers are going to have to raise the resolutions of their textures, otherwise the 3D models will look smeary, albeit with sharply defined silhouettes.

As for Food Run, being 2D vector in nature, the source artwork is already Retina capable. In fact, go ahead and quadruple the resolution again, Apple - Food Run would eat it up. My only problem would be getting hold of an iPad 3 in a timely manner from my UK base. Likely not a small problem.

So what is Food Run anyway?

Some of you have probably come here for the iPad 3 screenshot mockups and don’t follow this blog or know what Food Run is about. Fair enough; let’s set that straight.

Food Run is a game coming out later this year for iPad and iPhone, featuring food… with legs. It’s a platform game with levels and obstacles much like a classic Mario game, but with very simple controls - simply touch to jump, hold the screen to jump higher. Collect other similarly bipedally endowed munchables along your way, and they’ll start following you. Lots of levels, lots of food to collect. Kind of Mario meets Canabalt, by way of Lemmings.

Here’s the latest video preview:

YouTube Link

Keep it tuned to this dev blog and the twitter account to follow the development of Food Run. And don’t forget to pick up WordCrasher Blitz (trailer), available now for the magical, mystical price of free.

WordCrasher Blitz is go

Mon, 27 Feb 2012 13:15:00 +0000

WordCrasher Blitz hit the App Store yesterday. Here’s the trailer and details.

Yep, it’s out and it’s FREE.

http://itunes.com/app/wordcrasherblitz

Go download it, and tell me what you think at @bookofthedev. I’ll be sharing some thoughts and experiences about free to play, banner ads, and in-app purchases as I gain more data.

Here’s the blurb:

WordCrasher Blitz is the all-new ultimate edition of WordCrasher, available to download for free on iPad, iPhone and iPod touch.

For those who haven’t played WordCrasher before, imagine Tetris meets Boggle meets physics, and you’re not far off.

New features

Full universal app - iPad, Retina, all the good stuff.
Letter multipliers - x2, x3 and x4 markers on some letters. A new layer of tactical wordplay.
Earn your WordMaster badges. Every game helps you to level up.
New “Quick Game” mode - 5 words, minimal time pressure. Great for the bus stop.
All new graphics
iOS 5 Twitter support
60fps gameplay (up from 40fps)
Free to play. There are ads at the top of the screen (AdMob), which you can optionally remove by paying $0.99 as an in-app purchase.

Classic features

Realistic tactile physics. Letters bounce, stack, roll and respond to tilt
Game Center and OpenFeint
Over 190,000 words recognized in US, UK and International English
Every *unique* word you enter is tracked, and the game tracks your vocabulary. The current WordCrasher champ has entered over 20,000 unique words!
Marathon and Flood Panic games modes (in addition to the new Quick Game mode)
Freeze time and Bomb powerups

Secret pet project reveal

Wed, 22 Feb 2012 11:42:00 +0000

Here’s the pet project I’ve been alluding to on Twitter for the last couple of weeks. WordCrasher Blitz - the ultimate version of WordCrasher, with the magical, mystical price of FREE, already submitted to the App Store and arriving in the next few days.

I knew in the back of my mind that by writing that I’ve finished working on WordCrasher back in October, I’d inevitably have to eat my own words. But there were some very compelling reasons to take a couple of weeks out of the Food Run schedule and work on this new version.

First Pixels on Toast product with ads. WordCrasher Blitz has banner ads, something I have no experience with. By using an existing project, I found I could quickly create a test bed to learn from. And at the same time, the advert management code went into my shared code repository for future projects.
First Pixels on Toast product with in-app purchases. At launch, there will only be one in-app purchase available - the ability to remove ads. Again, WordCrasher Blitz will be an important test bed for learnings about IAPs, which will benefit future products. And again, the IAP management code has been written in such a way that in can be re-used.
Bringing WordCrasher up to date. WordCrasher was originally launched in January 2010, and is showing its age a little. With iPad support, new graphics, new menus and features, WordCrasher Blitz now feels like a 2012 App Store game.

So really, it’s all about learning. I’m not expecting WordCrasher Blitz to be a huge earner, for the reasons I already went into in that October article, but it will help me to earn in the future by allowing me to dip my toe into the banner ad and in-app purchase pools. Which is just as important.

New features

Here are the features added to WordCrasher Blitz

Full universal app - iPad, Retina, all the good stuff.
Letter multipliers - x2, x3 and x4 markers on some letters.
Every game you play contributes to your WordMaster level. I’m not sure anyone will ever manage to max out at 50. We’ll see!
New “Quick Game” mode - 5 words, minimal time pressure. Great for the bus stop.
All new graphics
Twitter support under iOS 5
The magical price of FREE. This will be the first free-to-play Pixels on Toast game.
Re-balanced letter scores. The spread was probably too low in WordCrasher, so I’ve added more significance to rare letters over longer words. This ends up being a lot more satisfying.

The most important feature is probably the new “Quick Game” mode. Since there are ads at the top of the screen, it made sense to create a mode where there is very little time pressure on the player, to give them the chance to read and click on ads. Otherwise the ad monetisation would fall apart before it had started. The second, less cynical reason for the new mode is that I had always wanted to create a game mode in WordCrasher which would last a short amount of time, ideal for a minute of play at the bus stop, for example. The lack of time pressure also means you can comfortably play it whilst multitasking - e.g. whilst watching TV in the front room - something lacking in the previous version of WordCrasher.

The second feature which alludes to the banner ads is the WordMaster levelling up system. Along with the existing vocabulary tracking, this really has the potential to add to the stickiness of the title, important if you’re serving up banner ads.

I’m also excited by the letter multipliers. They really do change the game quite significantly. Lower scoring words can now be beefed up quite a lot with judicious use of multipliers. In Marathon and Flood Panic, you could adopt the risky strategy of keeping as many multiplied letters on screen at once to splurge into one über word, with the risk being that you’d be running with a board near to the brim. This feature adds a much needed new layer of strategy, as well as just the right sprinkling of lucky dip.

Classic features

For those unfamiliar with WordCrasher, here’s the base of existing features I’ve been able to build upon.

Realistic tactile physics. Letters bounce, stack, roll and respond to tilt
Game Center and OpenFeint
Over 190,000 words recognized in US, UK and International English
Every *unique* word you enter is tracked, and the game tracks your vocabulary. And yes there is a leader board for that - the current WordCrasher champ has entered over 20,000 unique words!
Marathon and Flood Panic games modes (in addition to the new Quick Game mode)
Freeze time and Bomb powerups

5 golden rules of app updates

Tue, 31 Jan 2012 15:16:00 +0000

The coming of the App Store (and other similar, non-capitalized app stores) means you can update your app indefinitely, adding features, responding to customer requests, changing the app until it eventually becomes a different beast altogether.

However, as with most things in life, just because you can, doesn’t mean you should. Here are my 5 golden rules for putting out app updates, designed to keep your customers happy, your sales high, and your sanity in check.

1. A buggy update is far, far worse than no update

If you’re only going to follow one rule, make it this one. Spend the extra time and make sure your update is fully tested on all platforms.

I was bitten by this with version 1.50 of WordCrasher on iOS, which showed graphical glitches for upgrades who were in the middle of a suspended game when updating (see above image). The result was several hectic days of scrambling to produce a patch, answering support requests and minimising damage to the WordCrasher brand.

2. Only add, never remove functionality

Removing existing functionality from one point release to another is a guaranteed way to receiving some damaging 1 star reviews on the App Store. The customer perception is, quite rightly, that they paid for functionality which you have subsequently removed.

I’m surprised we haven’t seen more litigation based on this.

3. With that in mind, be mindful of changing functionality

Changing functionality can be similar in damaging effect to removing functionality altogether. With games, changing a control scheme, even for the better, will alienate players who were used to the old way. And these players will not hesitate to give you a 1 star review.

The solution? Add an new alternative control scheme rather than changing the existing one. Make the controls default to the new scheme for new players, but keep the same scheme for old players, with the option of switching over.

The best solution? Get it right in the first place. The first release of an app is the most important release, as it sets out the expectations for future updates. It’s very hard to remove or change functionality after launch without invoking customer backlash. And since most reviews are based on the first version, there is extra incentive to get it right first time.

4. Consider your update granularity carefully

Unless you’re using regular updates as PR in a similar way to Doodle God, don’t update too often. Until the process is improved, updating apps is a chore for the end user. At the time of writing, I’m seeing a little red badge with the magical number “68” hovering above my App Store icon. I’m not sure I’ll ever catch up. And for the developer, more updates means more chance of buggy updates.

Of course, a good update now and then can reinvigorate interest in your app. If the updates are substantial enough, you may even be able to get some publicity out of the update. Releasing the same functionality in drips is much less likely to be newsworthy.

5. Think twice before removing support for a legacy OS version

As we move on through successive versions of both iOS and Android, it is inevitable that older devices get left behind. As you update your app, it may be tempting to remove support for older versions of the OS to clean up your code, or leverage new functionality exposed in later OS versions. However, bear in mind that doing so will cut off customers with older devices from subsequent updates. Worse still, if the customer is on (for example) iOS, and has multiple devices spanning multiple OS versions, iTunes may well update their app to the latest version, which is then incompatible with their older devices. Something that they will find, to their horror, the next time they sync with iTunes.

Food Run's level editor - a video walkthrough

Fri, 20 Jan 2012 13:51:00 +0000

The main tool for creating levels in Food Run is the in-house editor created specifically for this project, BreadBoxEd.

Here’s a two-part video walkthrough and tutorial. In these videos, a level is created from scratch and tested in the game.

Although created for Food Run, the editor could be used for any 2D grid-based game with level content, so hopefully it will see some re-use for future projects. It spits out XML files. It’s Mac only at the moment.

Part 2

Side note: like BreadBoxEd and the company name, Pixels on Toast, most of my tech is named after something toast-related.

Main tech namespace - toast
Rendering library - toastr (pronounced toaster)
Standard library - toastd (pronounced toasted)
STL library extensions - toastl
TestFlight wrapper library - toastflight
Etc.

iPad video footage capture and editing workflow: Part 2

Mon, 16 Jan 2012 14:03:00 +0000

At last, some movement on the iPad capture workflow, the results of which you can see in the new video of Food Run above. It’s not yet perfect, but it’s 90% of the way there. Details after the break.

This is the second part of this series, looking at iPad video capture options for developers. For part one, go here.

One of the options I used to employ for iOS video capture in the pre-iPad, pre-Retina days was running the game in the iPhone Simulator (part of the iOS SDK), and using desktop capture software to grab the frames. This worked very well, but with today’s high resolution devices (iPhone Retina at 960x640 and iPad at 1024x768), my ageing 2008 iMac cannot run both the game and capture at the same time without dropping frames.

Option 1 - Hardware Upgrade

The most obvious option here is a hardware upgrade. Whilst I’m sure I’ll have to do this at some point in the near future, I’m waiting out for the next generation of Mac refreshes, at the very earliest. Judging by previous performance and the number of months since the last hardware updates, we should be due in the first quarter of 2012, or thereabouts.

Option 2 - Native Capture

The next option would be video capture from the device itself. The solution of choice in this department would be the Blackmagic Design Intensity range, but a similar problem exists. In their current forms, the Intensity solution requires a PCI Express slot, USB 3.0 connection or Thunderbolt connection, none of which my iMac has.

Option 3 - Port to Mac

But then suddenly an obvious thought sprung to mind. I have already ported my iOS engine to Mac to produce WordCrasher for the Mac App Store, so why don’t I just port Food Run? In theory, it should run much faster than on the simulator, and perhaps fast enough to run desktop video capture in the usual manner.

The port took about half a day, it being a rather sloppy workable version rather than a polished end-user grade app. However, upon capture I found I was skipping frames, leaving a less than smooth result.

This won’t be an option for everyone. Some might be using a pre-rolled engine without Mac or PC support. But in my case, the option was entirely viable and low cost.

Smoothing the Framerate

The answer to the frame skipping problem was to change the Food Run code to render at 30 frames per second instead of 60, just for the purposes of capture. After all, the output video will be 30fps, so why not run and capture Food Run at that rate? This worked well, and the result is the video at the top of this article.

Further Refinements

That brings us up to date, to my current solution with which I’m 90% happy. The colours are accurate, the sharpness is great and the motion is fairly smooth. Why the qualification of that last statement? Well, by running at 30 frames per second, I’m still throwing half the frames away. Games which run at 60 frames per second have a lovely polished smooth look to them, and I’m not doing Food Run full justice by leaving half the frames behind.

There are two possibilities to bring the smoothness of 60fps to the video capture workflow.

Get the Mac hardware upgrade, capture at 60fps. Although the end result YouTube video will still be 30fps, I will be able to frame blend the 60fps capture in my video editing suite down to 30fps,which should in theory preserve some, but not all of the smoothness of a true 60fps video.
Do the frame blending in the Food Run Mac application itself. Render a single frame 30 times a second, but compose that frame from two frames taken 1/60th of a second apart, by rendering one of them to an intermediate buffer in OpenGL. This may initially seem just as costly performance wise as a 60fps capture, but the time savings are two-fold. Firstly, I will save time by running vertical syncs at 30fps, meaning less dead time waiting for the sync. Secondly, doing the actual capture at 30fps instead of 60fps should be a large saving.

Given my feelings about the timing of a Mac upgrade, I’ll probably attempt the second option first, but to achieve that I’ll have to build intermediate pixel buffer support into my engine first. Which shouldn’t take long at all.

To be continued…

Silent cinema

Thu, 12 Jan 2012 14:25:00 +0000

Hey, remember when video game trailers had sound effects?

A retrospective on the music-only game trailer.

Gears of War has a lot to answer for. Back in 2006, the original Gears trailer was a revelation, eschewing all foley for a low key emotional soundtrack, bleeding out the colour and playing up loss and the desperation of the situation. Innovative at the time, it still holds up today.

A genre was born: the music-only video game trailer. The traits of the genre are as follows:

No sounds effects
Emotional low tempo soundtrack
Scenes of destruction and violence, vividly contrasting with the lack of sound
An obvious emotional context
Cinematic direction

Whilst it is arguable whether the music-only treatment was invented in the original Gears of War trailer, this landmark video certainly brought it to mainstream gaming and made a huge impact. And inevitably, as we have seen all too much in the games industry of late, it was copied many times over in trailers for other games. Some of which were more successful than others.

Gears of War 2

Gears 2 and 3 revisited this treatment. This made sense, as the original trailer had a lot of impact, and the trailers for the subsequent games evoked the first successfully. They started to feel like a series touchstone.

The Gears 2 trailer probably tried a little too hard in the emotion stakes. The focus pull to the poppy and the weathered photo of family is a little trite.

Gears of War 3

The Gears 3 trailer is interesting because it adds some selective sound effects at key moments. Reverb is added to help the sound effects stand out as a directed moments rather than background.

Fable 3

Okay, let’s take the Gears formula and apply it to Fable 3. And……

This one doesn’t quite hit the mark. The tone is off, the treatment doesn’t match the source material. Fable 3 is a lighter, warmer, but less epic game than this trailer suggests. After all, this was a game that let you grab someone by the hair and fart in their face.

It’s such a shame; evoking the artistry of American Civil War paintings (and a touch of the Dutch masters) was a clever idea. The lighting and models are fantastic. But the music-only treatment seems forced, as if to fit a popular trend rather than being designed this way to achieve something in particular.

Assassin’s Creed: Revelations

The effect can, however, work well for games other than Gears of War. This clip is nicely directed, well put together. The emotional hook, so key to this genre, is seeing Ezio in an older, more vulnerable form, seemingly prone to attack. But then he turns on the ass kicking to sate our bloodlust.

A long running time of three minutes and twenty seconds means we’re in full-on music video territory here. As a result, it lacks the one minute punch of the Gears trailers, but mostly manages to keep the audience’s attention.

Like the Fable 3 trailer, the visuals are pre-rendered, where the Gears trailers were purposefully made to look very similar to the game engine itself.

Crysis 2

This trailer stands as testament to the importance of music choice is for this genre. Here, the vocal delivery is so cloying and saccharin as to prove a distraction. Even though production values are unquestionably high, the temptation to reach for the close window button is huge.

The treatment implies a little poetry, a little soul, where in fact there is very little to work with. The appearance of the guy in the muscle suit just underlines this.

Halo: Reach

Even Halo, the grandaddy of the console FPS, had to get in on the action. Production values are top notch as you would expect, with some epic shots of future war and destruction. It does suffer a little from the “guys in big suits” syndrome, until the emotional context is revealed with the smashing of the visor and the reveal of the (*gasp*) female soldier.

If there is a gripe, it is that it’s a little sad that the series that did so much to bring the FPS to the console mainstream falls in line behind Gears with this trailer. With this kind of budget and being such an industry heavyweight, there was the chance to innovate and create a bigger impact.

Infamous 2

This is what happens when you try to deliver this formula on a budget. Just a fly-around of the game world with the simulation paused. Whilst I have nothing against House of the Rising Sun in principal, it seems a poor fit for this trailer.

To be fair, many trailers are produced during the hype cycle for a game these days, some of which receive a lot of care and attention, whilst others are obviously a little more throw-away.

Killzone 3

And this is what happens when you leave the emotional hook behind for a couple of minutes of pure gunplay. As a result, it feels somewhat lacking. What is the story here? Lots of generic action, and nothing memorable. No beginning, middle, or end to speak of. Remember the first videos of Killzone 2, and how everyone was excited about it? Those early glimpses of Killzone 2 encapsulated the hopes for this generation of hardware. This trailer fell short of the mark of its predecessor.

So lots of trailers employing this effect, with varying amounts of success. When it works, it works well, but it seems that some of these videos use the treatment simply because everyone else is doing it, rather than part of some grand design.

It is easy to see why this formula is used so much. It’s a marketing shortcut which succinctly conveys to the audience that the game is an epic gaming blockbuster, worthy of your attention. It looks good in movie theatres, plays well on TV. The casual non-gaming viewer, who should not be ignored due to their present purchasing role, will get it. However, there are now so many of these trailers, the danger is that by employing this initially striking effect, you will be lost in the noise rather than make an impact. And at the end of the day, isn’t that what a game trailer is for? To stand out from the crowd, to attract attention?

Dead Island

You don’t have to break with the formula entirely to create a more interesting trailer. Take the Dead Island trailer for instance. Yes, it too owes a lot to the Gears formula, but it also adds a few neat twists. The footage is run in reverse, leaving you wondering not what happens next, but what happens before, a little like the film, Memento. This question is partly answered by the interleaved forward running action, with full-on visceral sound effects, which meets the timeline of the reverse footage in the middle. It hooks the viewer, it asks questions, and it answers them in a way which is memorable. And plus, they played the dead child card. Emotional checkbox well and truly ticked.

And what was the result? Dead Island went from completely off the radar to become one of the most anticipated game releases of 2011. In terms of marketing impact, that makes it one of the most successful video game trailers for some time.

Just goes to show, a little innovation can make all the difference.

Weekend web work

Mon, 09 Jan 2012 12:21:00 +0000

I spent some of the weekend adding to the Pixels on Toast website. I added Javascript animated slides for the main image, changing out every 10 seconds. There is a slide for each of the games, as well as a Pixels on Toast slide. You can also access each slide individually and statically by clicking on one of the game boxes beneath the main image.

Here are some things I noted whilst updating the site.

A change is (almost) as good as a rest

To prevent fatigue from working overtime, I’m purposefully switching tasks to something other than the regular work during extended hours. Tasks like updating the website. This means that evenings and weekends feel a little like a rest from the day job, even though they are spent achieving work that needs to be done just as much as the regular stuff. And when I return to the 9-5, the work still feels fresh.

Websites are iterative work

A website is never finished. There are several things I know I need to do to the site - the tweet button and Facebook “like” button are still awkward for one, but it’s better than it was before the weekend. And that’s OK, websites change and evolve over time to a greater extent than games do, which once they are finished are generally added to rather than modified.

BBEdit is probably great

I’ve started using BBEdit for my web site coding. It has a great range of HTML tools, including validation, link testing and code completion. It has Grep built right into its search and replace dialog.

I’m also a fan of its lightweight implementation of projects. Simply choose a folder on your hard drive: that’s your project. Changes on disk are automatically updated in the project tree. Compared to, say, Xcode projects which hold a separate representation of your project structure, this means that there is no data duplication. Maintaining both a project tree and a disk system tree can be painful. The simplicity of the singular representation reduces friction.

It’s early days with BBEdit, and I’ve only scratched the surface, but so far so good.

960.gs is definitely great

960.gs is a system for designing and coding websites through CSS that aims to speed up your productivity and help you implement a vertical grid system. This isn’t news for true web designers, but for casual tinkerers like myself, the discovery of the benefits of 960.gs this weekend was a nice surprise.

For those not familiar, a brief overview. The page you produce is always 960 pixels wide, which was chosen as it works well on most modern devices and is easily divisible into equal columns by many factors. 960.gs provides several CSS files for you to choose from, dividing your page vertically into 12, 16, or 24 columns. Design templates for pretty much any software package you can mention are included, from Photoshop to OmniGraffle.

Choose the number of columns which is appropriate for your site. It’s advisable to choose the lowest number that you need for your design, with 12 being sufficient for me, it being divisible by 3 (note the 3 game boxes below the main image on the website).

The CSS file lets you set HTML divs easily for size and position within the chosen grid system simply by setting the class. So class=”grid_4” would indicate a div 4 columns wide, and class=”grid_2 suffix_2” would indicate a div 2 columns wide, with a gap of 2 columns after it. Spaces between adjacent divs are automatically dealt with, and are 10 pixels wide.

It’s worth noting that the grid isn’t strictly enforced, it’s entirely optional and you can break out of it at any time by simply not using the CSS classes.

See the 960.gs website and Google for the many tutorials online for more information.

Epic waste of time vol. 2: Invisible Planet

Thu, 05 Jan 2012 13:32:00 +0000

More unwanted sprouts flicked onto the floor from the plate of indie development. This time, an early video treatment for a company name idea.

I quite liked this one. Invisible Planet Games, a name inspired by Planet X which supposedly remained somehow outside of our detection within our solar system. The scratchy mono footage and circular waves emanating from the centre were supposed to evoke early radio tech, and an era of space exploration. The video was early and unfinished. Eventually I might have added some kind of design for the invisible planet itself (contradictory as that seems), around which the waves would expand. This would be repeated in the static mono version of the logo.

Reason for rejection? Simple: the twitter name, @invisibleplanetgames, and the URL, www.invisibleplanetgames.com were just too long. Dropping the “games” suffix doesn’t help either, as there are already others using the name Invisible Planet in this way. “www.ipgames.com” might have worked, but I’m not a fan of abbreviation in URLs. Better to avoid that issue completely.

And a Happy New Year

Mon, 02 Jan 2012 14:44:00 +0000

Happy New Year! 2012 just sounds so much cooler and futuristic than 2011, doesn’t it? Therefore I’m boldly predicting good things for all of us. Magical ponies for everyone. Bring it on.

Here’s my development resolutions for the coming year.

1.) When it’s ready. I’m sure I’ve been frustrating some people with my answer to the eternal question “when will Food Run come out?”, but I’m determined that Food Run will be as good as it can be. I’m all too aware from experience how much good a couple of extra weeks of polish at the end of a game can do. Hence the clichéd answer: “When it’s ready.”

2.) More blog posts. Once a week isn’t really enough, but then these blog posts do take an awful long time to produce. So I’m endeavouring to produce more, but perhaps shorter, snappier blog posts in the New Year. We’ll see how this goes.

3.) No more analysis paralysis. I’m determined to spend less time pontificating, and more time iterating.

4.) Read more, learn more. In this business, if you’re not constantly learning, you’re stagnating.

5.) More communication with other devs. We, the indie dev community, should be helping each other out. Let’s do that more often. Need help? Advice? Cross promotion? Get in touch. I’ll be doing the same.

If you want your own magical pony, you can get it here.

Game of Marmite

Tue, 13 Dec 2011 12:08:00 +0000

I released Kick Flick HD, a game about football, in the summer of 2010 just as the World Cup was warming up. I had always wanted to make a football game, the sport being something I enjoy playing (badly) and watching (too much). Little did I know that what I was releasing was the Marmite of games: something, as the advertising for the yeast based extract is so ready to tell us, is loved and hated in equal measure, with little room for fence sitting. This post discusses why that happened, and what you can do as a developer to stop it happening.

Let’s get right to the App Store ratings. Reviews such as the following are not uncommon.

Conversely, very positive reviews are just as frequent.

As it stands, in the UK, here’s the overall breakdown of user ratings.

Not that great! An average of 3 stars is not what you want for an App Store release. A lot of 5 star ratings and lots of 1 star ratings - the Marmite effect. (Note: this doesn’t paint the whole picture. Later versions were rated higher than the overall average.)

The UK is well known for its love of football, generally speaking. So how do the user ratings for the US compare, a country in which whilst football (or its identical cousin, soccer) is growing in stature, when it comes to popularity, it pales in comparison to the nations top sports.

Turns out it fares better in the non-football nation. I shall leave to interpretation whether this is down to Kick Flick being unrepresentative of a good game of football, or an example of the less cynical nature of the US audience shining through. In any case, for the purposes of this article, the divisive nature of the ratings is still very much in effect.

The why

So why are reviews so polarised?

For good or bad, Kick Flick was designed with a singular vision - to emulate the feeling of scoring a curling free kick as viscerally as possible. The control system is unlike the other flick free kick type games available on the App Store. Instead of drawing a curve that you want the ball to travel along, or kicking and applying aftertouch once the ball is in flight, Kick Flick attempts to get as close to the kicking experience as possible on a touch screen.

Swiping through the ball kicks it in the direction of the swipe, with an amount of power related to the speed of the flick. However, just as in real life, you can slice through the side of the ball instead of striking straight through the middle to impart spin. Due to the Magnus Effect, the ball in flight will follow a curved trajectory, as the force imparted by the spin drags it from its original course. By judicious use of spin, you can curl the ball around the defensive wall of players and past the keeper.

Whilst I believe the goal was achieved of viscerally replicating the feeling of taking a free kick, it also meant Kick Flick has a steeper learning curve than the other football games. Unfortunately, in the world of 99¢ games, this is a no-no.

What this means on the App Store

The 99¢ App Store gamer has a very short attention span, with lots of games competing for their attention. If a game does not grab them in the first couple of minutes, they will be very quick to dismiss it. Games are much cheaper than their time. Kick Flick is more demanding than this, and it’s likely that for some gamers, after missing or hitting the defensive wall a few times, they will simply give up and move on. Perhaps they will even make a stop along the way to the exit to leave a 1 star review.

For those gamers who stick with it and start to score regularly, they will benefit from the intended effect of the design vision - the visceral reproduction of the art of free kicks. And these players, who feel like they have mastered something and thus entered a club of some relative exclusivity, are far more likely to leave a 5 star review.

Which leaves Kick Flick with an average rating somewhere between the two extremes. Unfortunately, the time-starved browser won’t see a game which takes time to master and becomes more satisfying as you learn the nuances. Rather, they will see a 3 star game.

The lesson

The learning from this experience was simple. The App Store is not the right place for games which are demanding of their players. Gratification must be delivered instantly, in that first play session, in that first two minutes. As developers started to learn this, they began tuning their games accordingly. This is responsible for a lot of shallow experiences on the App Store. However, when done correctly, a shallow experience can still be a fun one. In addition, the best of these games deliver instant gratification, but then follow that up by rewarding the player with further reward down the line.

In short, to create a game which has a better overall rating on the App Store, it is important that you cater for your player over the entire span of their play, from the first minute to the 1st hour and beyond. If you can keep the attention of some of those who might have bailed early, you might be able to convert them into long term, 5 star reviewers.

The right game for the right place

This is not to say that games which are demanding of their players are a bad thing. It’s more that they need to be deployed in the right place. Whilst the App Store skews towards more casual game experiences, there are other places where such games thrive. Steam, for example, is a far more friendly platform for the more involved game. Typical players are more likely to be core gamers than the grazers of the App Store. For more artful games, the PlayStation Network has long been friendly towards games which sit left of mainstream. And Xbox Live’s Indie Games channel is, despite recent misgivings, a place where such games can thrive.

Epilogue

I am perhaps painting too negative a picture for Kick Flick. It still remains my best seller, and is selling well some 18 months after release. Judging by recommendations on Twitter, it seems that those who enjoy the game seem to communicate this to others. As well as being my most returned game, it is also the game which is gifted to others most often.

However, when all is said and done, the fact that it has been a moderate success doesn’t mean that it couldn’t have fared much better if it had either a.) catered for casual players too, or b.) been distributed on a platform more conducive to longer learning curves and deeper gameplay.

Judge for yourself

Kick Flick HD is available on the App Store. It is a universal app for iPad and iPhone. For those readers who want to give it a try, tweet me at @bookofthedev or e-mail me at “kevin at pixelsontoast dot com”. I’ll send the first ten responses a promo code for a free copy of Kick Flick. Just don’t forget to leave a nice review :)

Epic waste of time vol. 1: One Button Defeat

Mon, 05 Dec 2011 14:39:00 +0000

There are roads you go down in the creative process which end up unravelling and leading you to the wrong side of nowhere. I thought it might be interesting to share some of these design ideas, prototypes, images and assets to show some of the hidden work that goes into independent games development. And so I present volume one, the lost company branding known as One Button Defeat.

Note that some of these images can be biggified via the usual method of clicking. Those that don’t respond to clicking don’t have larger versions.

I’ve already talked in great lengths about the horrifically elongated and extruded process I went through to come up with the new company name and branding, Pixels on Toast. But I had forgotten that I had spent quite a bit of time on some of the front runner candidate names in exploring logos and images I could use. One of these names was One Button Defeat.

The name itself comes from some late night gaming sessions with a good friend of mine. He was quite skilled and practiced with fighting games, the kind of person who learns all the special moves for all the characters and can pull them off with alarming regularity. In stark contrast, I am a complete idiot when it comes to fighting games. Whenever we played Street Fighter 2, my favourite luddite ploy was to choose E. Honda, the big boned sumo wrestler from Japan. And then when my finely nuanced friend came pirouetting near, I would activate the Hundred Hand Slap special move by mashing the punch button repeatedly. And I would use only this move, pressing only that button. Victories were rare, but when they came, I could claim a “One Button Defeat!” to add vinegar to fresh wounds.

On to the visual treatment. First I worked on a button design, created using layers and layer effects in Photoshop. The glyph on the button is a slightly grisly play on the 1UP button found on arcade machines, with the player’s head detached from the body and lying on the floor. I liked the way it seemed initially innocuous until you realise what it actually is. The button itself is based on classic arcade machine designs. The first attempt to produce the button image was a render, but it was hard to get the lighting exactly as I wanted it, and in fact it turned out to be much easier to control results in Photoshop.

Moving on from here, I started to look at splash screens, and experimented with taking the idea to its natural conclusion, ramping up the gory factor.

Too much? Maybe. I would definitely have problems using this logo with family friendly games such as Food Run. But maybe I could use it for a hardcore shooter game on XBLA or PSN.

Next I moved on to iPhone sized versions, and more age friendly treatments.

I also experimented with changing out the button decal for a bomb.

And another variation without decal. I was also thinking that the word “defeat” may be too negative, so I tried to change it to “victory”. Thinking back at this now, I don’t think this is necessary, but it was worth trying.

Finally, I had been inspired when watching the end credits of Blackadder, the third series set in the Regency period. I put this together in the style of theatre posters of the time. I still quite like this treatment.

So why did I end up rejecting One Button Defeat? Well it seemed like too much of an in-joke. And at the time I wasn’t sure about the word “defeat”, and didn’t like the chest beating qualities of the word “victory” either. Add to this that although I liked the player’s head being detached in the button decal, I was a little worried about how appropriate it was for a family audience. If I were in the business of making games solely for core gamers, this name probably would have been higher in my thoughts.

The search continued.

A bitmap font solution for games

Wed, 30 Nov 2011 12:10:00 +0000

I may have mentioned it before, but I’m not using any kind of engine or über framework such as Cocos2D or Unity. It’s not that I have anything against these libraries (in fact they probably represent the best way into iOS game development today), but when I started out on iOS, such solutions either didn’t exist, or were too early to depend on for production code.

This means that there are often holes, or missing functionality in my engine which I have to fill. One glaring shortfall has been decent bitmap font support. The old system lacked a font generation tool and support for Unicode and kerning pairs. Over the last two days I’ve found and implemented a solid solution that I’d like to share with those also on the search for a font system.

One of the drawbacks of “rolling your own” is that you have to write everything from scratch. Or at least, find an existing solution and adapt it to your needs. After doing some research, I ended up going for a hybrid approach - an existing file format and bitmap font generation tool, with my own back end integrated into my engine to support resourcing and rendering.

The format is known as BMFont, developed by AngelCode.

File format

The format exists in both text and binary format. Unfortunately, the text format is not XML so you’ll have to roll your own parser. I took this as an opportunity to write a more generic text parser to add to my engine - not strictly necessary, but it made sense and will hopefully bring benefits in the future when I have to deal with some other proprietary text format in the future. UPDATE: A reader points out that an XML format is indeed an option with BMFont. Can’t believe I missed that. In GlyphDesigner, you can find it in a combo box in the export dialog. A little more digging here would have saved me more time as I wouldn’t have to write a custom parser, I’m already using TinyXML elsewhere in code.

I’ve posted an example text BMFont here. The file format contains general font information, metrics for each glyph, and kerning data for specific glyph pairs. The format is fairly simple and self explanatory. There are several resources already available online, which AngelCode link to on their site: -

http://www.angelcode.com/products/bmfont/

Tools

There are now several tools available for the generation of BMFont files. These tools take a TrueType font and render them as bitmaps onto texture pages, and create the accompanying BMFont text file.

For Windows, AngelCode have made their tool available to the public.

http://www.angelcode.com/products/bmfont/

For Mac, 71squared have produced a slick and polished product, with the downside that it’s not free, selling at $29.99. You can download as a demo and unlock if it’s what you need. This is the application from which I took the screenshot for the main article image.

http://glyphdesigner.71squared.com/

For Java-enabled platforms, cokeandcode have produced Hiero.

slick.cokeandcode.com/demos/hiero.jnlp

Postscript: Coders are a strange breed

There is a tendency amongst us coders to write from scratch when an existing solution will suit our needs perfectly. This is known in layman’s terms as cutting off your nose to spite your face, and stems from some kind of innate distrust of code which hasn’t been written by our own fair hand. Finding an existing solution, and just coding the parts that I needed meant that I could integrate support for Unicode, nicely kerned fonts in just a few hours. Time regained which can be spent developing Food Run rather than creating a file format and tools from scratch.