Official Introducing Gigaya: Unity's upcoming sample game

Subnautica was developed on a custom source version of Unity 5, so, yes, they have "rewritten the core of the engine" as well.

Not to say there are "absolute blockers" for making huge projects in Unity without source access, but Genshin and Subnautica are not good counterexamples—and I'm not sure I have seen many good counterexamples, particularly using editor versions later than 2019.

 

Oh, good to know—thanks for clarifying!

 

Not really.

 

Eh, the person who makes the claim that something exists, has to be the one to prove it, because proving the inexistence is near impossible.
https://yourlogicalfallacyis.com/burden-of-proof

 

interesting, so why don't you explain to me what he actually said there. because what I've got from that post is that there were issues with gigaya that were to costly to fix and they just preferred to lock the project down and fire everyone.

I'm really curios what you understood from that post.

 

One of the reasons why Gigaya was cancelled is because it doesn't make much sense to work on an game architecture that will start to become obsolete by the time DOTS start to get mass adopted. It doesn't serve any purpose to DOTS and is a huge project that requires maintenance and manpower after release. Also an incentive for people to stay with the old tech and keep them from upgrading to DOTS.
Unity should've worked on something like Gigaya many years ago, I also believe the chosen time was not great (financial instability, new incompatible tech coming, etc).
In my book, still a mistake made by bad management. Even if I pointed out some valid points, it is just plain wrong to start such a project, announce it to the whole world, then cut it short in the middle of production and fire the devs.
If my assumptions are right, Unity must now deliver on DOTS, and they should only focus on it. No more this package here or there, this new team on this weird feature that absolute no body cares to have for game development. Stop wasting money and manpower and put everyone on DOTS!
If Unity pause DOTS once again for whatever reasons, the ship will sink really fast.
[EDIT] This is of course my assumptions and probably pure speculation so be mindfull when quoting me!

 

There weren't 'issues' with Gigaya. It was a case where Unity management felt we had reached a flattened curve to value upon investment over time. Leonhard wrote exactly this. I personally disagree with that assessment (last 10% of shipping being hardest etc) but ultimately not my decision to make. We had areas to improve in performance and optimisations to apply but thats the case on every game and project ever. And those optimisations learnings is what every studio goes through with Unity so would have been valuable information.

 

Interesting post, but unfortunately not true at all. One of the processes we were starting to do was to test-upgrade parts of the project to newer features and tooling to evaluate the delta of performance, workflow, experience, etc. This way to highlight also missing functionality gaps etc.

We couldnt do that by just starting the project with future tooling that less than 0.1% of Unity's users are using currently.

 

Additionally much like most new things, broader adoption of new methods takes a long time. Even if dots officially came out this year unless it was a game changer and simple even for smaller use cases people would still use current methods for at least 3 years or so.

Way everyone talks about DOT's feels like it mostly applies to large scale and console quality mobile games. I don't see that causing broad adoption quickly like say PBR did.

 

I will give you the quotes as I understood them as I think you didn't read that correctly. Here are the quotes and what message I've got out of it

We already found out the editor core (not the engine) is broken. There is no point in continuing this uphill battle if the tools are broken.

The project will simply not work because the editor has issues. Cleaning up this project may be impossible or it would had taken months and we are not trusting the tools we have to deliver a quality example project.

We are going to get back to the drawing board and fix the tools we have instead of trying to continue struggling with this project. Actually a good idea IF they will invest in fixing the editor.

That is about it, but I have some final notes, instead of keeping the gigaya team on payroll to create other assets, learning stuff and on, they fired everyone. Because you can't create this stuff at the moment, there is no need to keep these people around. hard but true from the wallet point of view. What to do with all these people if we don't have a editor that works as it should? Once the editor is fixed (not the engine, which i think it works ok) we may get some of these people back.

 

@dorusoftware
You are literally arguing with the former main dev of Gigaya as to why their project was stopped...

 

You are reading between the lines alot and coming up with your own speculation as the truth. Im literally telling you exactly what happened, as one of the people who worked on the actual project since Day 1.

 

yeah - a lot of folks are fanning a ton of speculation on this with like ZERO evidence for it. the speculation about the in-between 'unsaid' on CodeMonkey is just out the f**king window - you could tell he did NOT want to make the video - at all - but was forced to address what had become an elephant in the room on his comment threads.

for me it's not about what management did, it's how they did it. if a product that Unity hyped on its own site needs to be canned, you wind it down, release a semi complete version with some documentation and then make it available to everyone. there would be grumbling about wasted potential from the community but generally it would at least look like Unity tried something, couldn't make it work and politely and gradually backed out. instead it gets instantly amputated with no official word until weeks after the incident. what that does is absolutely fan the flames of distrust and rampant speculation about Unity's interest and continued abandonment of projects. and as far as Unity hiring 3000 bodies, unless a lot of those are going to improve the engine itself, it changes nothing - AndyTouch says that right now they wouldn't be able to find anybody to release even a small part of Gigaya. that certainly indicates to me that most of those 3000 hires aren't going into game engine development.

ultimately most of this fallout is mindshare based with a section of dev community (i assume anyone happy with this decision and couldn't have cared less about Gigaya aren't on this topic thread). but eventually mindshare dissatisfaction can turn into market share and Unity management really doubled down on the worst way to handle it, which does not help it from a brand perspective among some game devs. right now to Unity management, the indie game dev community isn't important (aka profitable) enough to matter. they've still got some time to turn around or at least attempt some moves towards indie devs, but the longer they wait the more that lack of mindshare becomes a lack of market share.

 

My guy, you are literally arguing with someone who worked on Gigaya. He's saying the editor wasn't broken, but you'd rather pull conspiracy theories out of a super vague corporate PR post.

 

Hey thank you for your comments. You've been the best source of open information thus far regarding this whole debacle.

 

right, and i apologize if this is out of context. someone earlier asked about just releasing the Third Person Controller and a few other systems that were demoed earlier (rather than the whole project) and the answer was a pretty clear no.

i was referring to what they should have done, not what they did. i would have wound down the project and not amputated it.

 

No offense for anyone work quality, I'm not trying to insult anyone, myself know LESS than what he knows about programming and dev stuff, first he doesn't have the big picture the heavy devs at unity and top management have. And they will definitely not tell him anything about it.

Second, when he say, wait I can't speak like he is not in the room, sorry Let me refrain

When you Andy, report some bug, maybe the editor got laggy on your side, maybe you had an issue with a prefab, whatever, you may think is a matter of rewriting some system to make it work BUT in reality at the top they knew it was a deeper problem.

I don't pretend that I know what actual problem the editor or the engine has, but it does have an issue that forced them to stop development on all these promo materials, and as clearly stated, get back to the drawing board.

They can't came here and say, "ok the editor was broken for years, sorry guys, we will start fixing this right now." They can't legally do that because maybe someone will sue them or something. They can't admit the product is faulty. They can say from now on they will improve it, make it better. Corporate speak, if you read between the words the message is clear.

More clear than this? "We tried to do this and it didn't work as expected. There is an issue. We will get back to the lab and fix it." If the unity team member doesn't lie, then maybe in a year or two you can expect a better unity version.

What you have seen as stuff that needs optimization to make it work, the big guys knew it was not feasible.

I'm not sure why all of you people get triggered in defending unity when some of you got fired, the CEO insulted every game dev out there. Only positive thing is that they finally half admit they have an issue.

 

You are free to speculate, say 'maybe...' alot and come up with your own version of events; but none of what you said is accurate at all.

 

I imagine that if the issues were as serious as described by dorusoftware, Unity would have never been considered a worthy engine by big companies and ditched a long time ago.

 

I figured everything, all the mail to people that don't exist

 

Just wanted to note: There are many, many various reasons companies join these foundations. Some legal (avoiding anti-competition claims), some tax related, some even as part of another deal with a different company that has a vested interest in the foundation, etc, etc.

The LAST reason would be that they are so generous and caring that they want to fund their direct competition, like some modern-day corporate hero. LMAO! The idea is ridiculous!

Companies simply want to make money, avoid lawsuits, and pay minimum tax. The end.
There is no moral lesson to be learned here.

 

A little off-topic, but I just wanted to correct the statements I made in this same thread about Unity 2021 LTS: (so we are all up to date on this)

I just upgraded (again) my project to the latest Unity 2021 LTS to check if all the bugs were fixed.
(I already did that when Unity 2021 LTS first came out, but it was a complete disaster of crashing bugs.)

I'm glad to inform everyone that Unity 2021 LTS seems to be running great now! (after no less than SIX updates, mind you)
No more editor crashes or anything.
HOWEVER: shader compilation has been modified and as a result I have spent more than 6 hours debugging my custom uber-shaders to make them compile as they were on Unity 2020 LTS. (specifically, using #pragma shader_feature_local with more than one keyword per line is behaving very differently than in previous versions, which systematically disabled some keywords and messed up my shaders... Also, some global defines have changed their behavior. And all of that is undocumented, of course )

I haven't seen any performance increase in builds, but neither a decrease, so that's fine.

The editor is much slower to present the properties of an object in the inspector when you click on it than Unity 2020 LTS though... (you click on an object then it takes like one or two seconds to update the inspector... wtf?)
It doesn't do that anymore. It's fine. (??)

UPDATE: it seems like HDRP is working very badly on the editor in this version of Unity.
If you are using built-in as I do, it's alright. But if you use the scriptable render pipelines, they are very buggy here.

You can check if it's worth your time upgrading here:
https://docs.unity3d.com/Manual/WhatsNew2021LTS.html

@The_Island it seems like Android Vulkan video is working great now on Unity 2021 LTS! no more frame stuttering and crashing, so good work and thank you! Be assured your work is appreciated!

 

I was really looking forward to Gigaya. Very disappointed. At least put it on the Asset Store as is.

 

Not going to happen.

"Introducing Gigaya: Unity's not coming sample game"

 

Spoiler

Spoiler

 

I also have similar issues with shader comp where for some reason multiple shaders are reporting using more than 16 texture samples when they are deffinitely not.

Amoung other weird shader compilation issues where stuff that worked fine before upgrading now doesn't work.

More things that might be helped if projects like Gigaya existed long term to be part of testing before release

 

Well, yes, I wasn't exaggerating when I said I spent 6 hours fixing my shaders LOL

Oh, I'm 100% confident Gigaya would had ONLY worked in the VERY SPECIFIC version of Unity it would had been built.
Just look at the Official Unity HDRP Assets™ in the store LOL

 

At the very least upgrading between unity versions might be a little less painful.

Like they'd have something to try

 

Check out the shader variants your shader has, some use extra textures like lightmaps and light cookies. If that's indeed the problem and it's related to variants you aren't using, try to disable them by removing the relevant #pragma_multicomple or #shader_feature. I think a of the keywords used when compiling the shader that has an error is shown in the debug log window, you can use that to figure out the variants.

 

As I said before, most of Unity's open source projects do not work out of the box. I think we would still benefit I think from seeing some of the code and use it as a reference possibly...or perhaps they could allow the community to work on it like they did with Chop Chop!

They put work into it, might as well not let it completely go to waste.

 

Go back to built-in, you won't have that combinatory explosion of shader variants with the Standard shader! Plus both Shader Graph and ASE editor are compatible with built-in!
Ok that was a bad taste joke! :V

 

Thing is this wasn't a problem before, it actually started happening after upgrading from 2021.3.0 to 2021.3.3 and also still in 2021.3.6

And i've pretty much been using these same textures and shaders (with improvements) since like Unity 2017, 2019 for URP tho..

also not the only one...
maximum ps_5_0 sampler register index - Unity Forum

 

This issue is probably due to an unwanted combination of new variants that uses more interpolators. Also I don't get why there's a limit of 16, SM5 interpolators limit is 32, not 16. Prob another lowest common denominator crappy hard limit Unity put for mobile platforms.
Tbh, you should probably stop upgrading Unity if you don't need any of the new features/fixes because it's getting worse and worse overtime. Either they improve thigns and break other stuff, or they add new things and introduce new bugs.
I also have lot of problems and performance issues with newer versions of Unity. And this is not a joke, this is reality. Been locked with 2019 since forever... Time to stop my licenses...

 

I have been upgrading my project since Unity 2017 up to 2021 LTS now and I haven't noticed performance issues other than physics stuttering on 2021 at low fps (which I fixed by updating the physics manually).
I'm on Built-in RP though. Testing on PC and Android Nvidia Shield.
Where are you noticing performance downgrades?

 

Like they say, there is the first 90% and then the second 90%.

 

I always thought the limit was shader model, but this happens regardless.

I don't usually have issues like this and I upgrade fairly regularly. But it seems 2021.3.6 is funky for several reasons.

Problem with IPointerEnterHandler IPointerExitHandler and SetActive - Unity Forum


Apparently, this fix is breaking everyone's UI (specifically Doozy UI). I don't fully get it but I am experiencing the problem one way or another.

2021.3.5 is apparently fine and has the fix I was looking to try anyway which might fix the windows touch issue i've been nagging about for months.

Imagine if Unity had a reasonably scaled project to test these things on...

 

They have... our projects. Well, at least when we do bother diverting our resources into creating and uploading repro projects and convincing the bug reporting team that the bug is indeed real.

 

Im fully aware.

 

That's what has made me stop sending bug reports. It's like you're presenting a case for trial or something.
I have my own stuff to do, I can't spend my time debugging others software.
Literally: I can't afford it.

 

Hence my idea of being able to just pay for it. It would probably cost less money to just pay unity to fix the damn problem and figure it out than to try and diagnose it yourself.

 

You're actually a Legend.

 

Oh, but they already offer that service!
It's called PREMIUM SUPPORT and they send someone from Unity to actually fix the engine.
Prepare the bank account though.

 

I think this thread is beginning to come to a close. Ample time has passed for all voices to be heard, and it seems that things are going around in circles now. Just general angst. So make any final comments you need to make, everyone.

 

All the best for @Andy-Touch and the rest of the Gigaya team. You can be pround of your achievements. Good luck guys .

 

^ Nice closing words! 100%