Gigaya was probably too ambitious and maybe not the best way for Unity to get experience for what it’s like to actually publish and maintain a game. Maybe making the art, developing systems, etc. was asking too much. That said, Unity really needs to be publishing games (especially for mobile) to do quality control on new features and to demonstrate best practices to its user base.
I recently submitted a confirmed bug in 2022.1 using AR Foundation (release) and URP (release) that resulted in a black screen with no background rendering. Not a big deal, but also the same bug that’s cropped up repeatedly over the years. This is not a bug we would see if Unity was actively maintaining games using their core features and latest release versions of packages and sharing that project with users.
I spend more and more development time finding the magic combination of features, renderers, UI, input systems, etc. that work and provide the best results. On any given day, I can’t honestly say what is the recommended setup for new projects. Bare bones AR examples use a different set of packages and versions compared to bare bones feature examples. Often, the examples use packages that are unavailable or incompatible with each other, making it even harder to pinpoint what is broken. In the old days, I could just download a completed game like Bootcamp and use that as a guidepost to how Unity thought a game should be set up.
Development has become so fragmented that I find it’s hard to collaborate on projects. If I ask for a particle effect, my motivated assistant looks for ‘the best’ effect tutorial out there and gives me a VFX effect. I then have to explain that won’t work on Android, so use the other particle system. I get the new shuriken effect, but apparently they did not hear me when I asked for URP.
How to solve this and at the same time decrease the amount of tech support answering the same questions over and over?
Unity should simply buy several existing games built in Unity that tackle the most common development tasks we all have.
Tweak these games, publish them and make them working examples for how things should be done. Maybe hire the original developer every 6 months or so for a week to tweak them and redeploy to the stores with the latest recommended versions of everything. In the process, they’ll fix problems with renderers, privacy updates and all the other stuff that happens when you actually publish a game.
This could even support the new monetization priorities by making sure that everything just works and developers know the correct versions of everything to use.
-
An AR mobile game
-
With a good tracked image mechanic and place on tabletop interface ( the AR Foundation examples work, but are so barebones and not at all professional level)
-
A screenshot gallery and add to native gallery option
-
Mobile Ads
-
Recommended renderer, UI and Input system that is most compatible to modern devices
It doesn’t have to be a great AR game, just one that is always working and on Github. Simply having this game would answer 50% of the questions I see repeated on the forums constantly.
-
A 2D Platformer
-
With a silky smooth character controller
-
A high score system tied into a sharing system
-
A save preference system
-
Character customization
-
A FPS multiplayer game
-
With a silky smooth character controller
-
A tuned camera system using the recommended setups for 1st and 3rd person
-
Multiplayer functions that are expected in this sort of game
-
A RTS game
I don’t really have the answer for exactly what games they should publish, but there are many instances of common problems that have already been solved. The games should all be as small and simple as possible. Heck, just use the 2D flying saucer assets and publish it as a game! But, given how many games are out there, surely there are some great, simple games whose owners would gladly sell to Unity for this purpose.
I’m hoping with all the Gigaya backlash, somebody at Unity might actually see this and realize that it is actually quite important that they test their engine out by publishing games and not just give up on that idea. In fact publish MORE games that are less ambitious than Gigaya but that actually demonstrate how to use the engine and serve as tech support reference points!