Cubetrix - Changelog

This is the list of interesting changes and updates for Cubetrix. To read about Cubetrix itself, go to the Cubetrix blog post.

• 2024-08-30 3(1.2) - On more devices
• 2024-08-30 4(1.4) - High score, and drag-to-look
• High score
• Drag to look
• 2024-09-01 5(1.5) - high maxValue, and better dragging
• high maxValue
• Dimmed high values
• Drag gesture start
• Recentering gravity
• 2024-09-01 1.6 - Reverted the drag gesture
• 2024-09-03 2.0 - Hamburger menu, and more
• Created hamburger menu
• Accidental showing/hiding of the buttons
• Added settings
• Fixed arrow buttons
• 2024-09-04 2.1 - Settings tweaks
• Foreground colouring in the settings
• Sections
• Optional toggle on tap
• 2024-09-09 2.2 - Instructions and Translations
• Instructions
• Translations
• 2024-09-12 - Ratio
• What
• Aesthetic
• How the number behaves

I tested Cubetrix across several devices of different form factors, and ages. So many, in fact, that when my wife walked in and saw my desk covered in devices, she made the comment that it looks like I’m running an intricate drug ring.

Then when I released the app, and sent the link to several friends. The first person I chatted with wasn’t able to install it. Specifically:

The game uses the phone’s gyros for navigating around the 3D space. I was expecting some very old phones to not have these, but apparently even some modern budget phones don’t have them yet. I already had the fall-back functionality for this use-case, but had accidentally set gyros as required in Google Play. So I’ve now made them optional, but you will definitely have a better experience if your phone has gyros.

I’ve skipped 1.3 so that versionCode and versionName loosely match. I was bending my head on the previous version when trying to debug an issue with a friend.

All future games you play will now keep the high score on your device. Like all other saved data, nothing is transmitted off your device.

The high score is displayed in a fainter yellow, just above the current score, which is a stronger yellow.

Probably the more exciting update is drag-to-look. This is a way of replicating the gyro functionality on a device that doesn’t have a gyro, simply by dragging your finger around.

Normally, to do a swipe, you brush your finger across the screen in a particular direction. To do a drag, you simply hold your finger longer than about half a second. The number displays will all go grey to give you feedback that you have successfully entered this mode. At this point, you can move your finger around to look around.

The normal re-centering gravity is turned off while you are dragging, because I found the gravity annoying while testing it on a phone that doesn’t have a gyro.

In this screenshot:

I’ve added the all-time high max value (dim green). It’s like the dim yellow all-time high score, but for the green highest value block.

I know that some people use the high score to mark their progress, which is totally fine. But I find the max value to be a more interesting measurement of progress.

A common bit of feedback was that the high score was a bit distracting. Now that I’ve added the high max value, this would be even more the case. So I’ve dimmed both sets of numbers some more.

I had actually played with this when I first introduced it, but with different phone screens having different gamma values, one of the phones that I tested on felt too dark. But the majority of phones were way too light. So it seems right to bring it down a bit.

I’ve made the drag gesture start dragging from when it becomes active instead of when the finger first made contact. This has the affect of not having a sudden jump when the drag becomes active. I’ve reverted this. See the 1.6 update.

I used to apply the recentering gravity (the force that slowly moves the cube back into a neutral position) when ever gyro values were processed. This seemed logical at the time. But it has the effect that gravity is only calculated and applied at those times, which can be significantly less frequent that the frame rate. I’ve therefore moved those calculations to the draw function so that it’s consistently applied when a new frame is needed. It’s a light calculation that has a trivial impact on the frame generation time.

Reverted the drag gesture behaviour. Users didn’t like it, because it made the action less predictable, which inhibited muscle memory. Now it will “catch-up” to your finger again rather than starting from when it becomes active.

There used to be a reset button (for starting a new game) in the top left of the screen. Instead there is now a button to show the hamburger menu, which contains a “New game” button.

Below that button, there are some settings, which you can change to change how the game and User Interface behaves.

It used to be that you’d tap anywhere on the screen to show/hide the arrow buttons. I found this really annoying, so I’ve moved it to the hamburger menu.

• Show buttons - Replaces the old functionality where you’d tap on the screen to show/hide the buttons.
• Hide buttons on drag - The buttons used to automatically disappear every time you’d drag to look around the cube. This is now optional.
• Invert buttons - Some people like to rotate the cube. Others like to rotate the view of the cube. This option addresses that.
• Auto recenter - This slowly brings the cube back to the center. It’s to cater to the slow drift that gyros have. If your device doesn’t have gyros, or you want to get the perfect angle for a screenshot, this is now optional.
• Rotate on flick - This option is mostly for devices that don’t have gyros, but might be useful for people who would prefer to use the buttons. In either case, you can now disable the flick gesture.

Something I forgot to put in the release notes was that the arrow buttons didn’t show correctly on some devices. This seemed to be rare, but did happen. So I’ve improved that, and haven’t been able to find any devices that still show the problem. If you still see it, I want to know about it.

• Created a hamburger menu and placed configuration there.
• Got rid of annoying behaviour that led to accidentally showing or hiding the buttons. Instead it is now the “Show buttons” toggle in the hamburger menu.
• Created settings for commonly requested & disagreed-on choices.

This is an “own goal” moment. While implementing the settings interface, I anticipated the problem of the text being the same colour as the background on some devices. But when I tested it on a phone that I expected to have this problem, it was fine. So I chose not to do anything about it. It would have only taken me a minute, but I didn’t do it. This was the wrong decision, but it’s fixed now.

Above: The text was previously unreadable on some devices.

Above: Now it’s visible.

There were only 5 settings before, but they were already feeling a bit cluttered. I’ve now created sections to make them easier to skim.

Above: There are now titled sections to make quickly skimming the settings easier.

Added a setting restoring the ability to toggle the buttons by tapping anywhere on the screen. I’m not sure if anyone was relying on this, but I referred to it in the original video describing how to use Cubetrix. So in the interests of making things easy, it makes sense to keep that functionality for now.

I forgot to put this in the release notes on the Google play listing. It is:

A common bit of feedback I’ve received from people who don’t speak English, is that they aren’t sure how to play the game. This is because the video that I encourage people to watch is only in English.

To solve that, I’ve added text instructions within the app, and then…

I’ve just abstracted out all of the user-facing strings so that I can add support for different languages. In each case, I’ve used Google translate to make the translation, and when I’ve had the knowledge to do so, I’ve corrected context mistakes. There will be mistakes that I’ve missed. But it’s a start.

I’ve chosen the languages that I have chosen to start with, because I know someone who speaks that language, so it might be nice for them. And hopefully if I’ve got any serious typos they might let me know.

The currently supported languages are:

• English.
• Arabic.
• German.
• Spanish.
• French.
• Luxembourgish.
• Norwegian.
• Portuguese.
• Russian.

And you can use them by choosing the desired language in Android’s settings.

This has been a really interesting exercise, and one that I specifically waited until now because of the inertia it will add to changes involving text.

Of particular interest to me was how Arabic would behave (being a right-to-left language), and to the best of my understanding, the libraries that I’ve used have nailed it. It was also interesting to see how differently Android behaved as well (gesture directions).

This update is entirely about the ratio. Let’s dive into that.

I’ve always been interested to see what the ratio is between the maxValue and the number of moves I’ve made. It’s a measure of how efficiently I’m attaining each maxValue.

I’m conscious that this may not be interesting to everyone. So I’ve made it faint so that it doesn’t distract too much. Yet I love how it contributes to the hackers-style aesthetic that I’m going for. So I didn’t want it too dim.

I’m also using the extended display segments to display the decimal point. Yet, in the current opacity, they aren’t particularly visible. I played with the opacity of the illuminated segments being as they are now, while the non-illuminated segments were the same as the non-dimmed displays. The problem with that was that it wasn’t as readable as I’d like. - I’m likely to play with this further to optimise readability vs looking cool.

The number is calculated by dividing the maxValue by the number of moves. So every time you make a move without increasing the maxValue, the ratio gets a little bit smaller. But then when you increase the maxValue the number jumps up.

So effectively, the ratio is a saw tooth wave. It slowly decays, and then makes a big jump from time to time.

A high number (eg 10) might be a good indication that you are playing well. While a low number (eg 2) might be an indicator that you’d not. Or maybe not… Maybe you’re biding your time and will make a big jump up soon.

In any case, I found it interesting.