GIGAYA

I think it's pretty clear that benefit of the doubt is something Unity needs to win back at this point.

 

Here's the thing: if they only listen to internal feedback that's still a massive problem. They've had tens of thousands of people using the engine over the years and enough feedback that you could print it all off and fill the Library of Congress. They have been so bad at listening to feedback that they closed off their own dedicated feedback site.

I don't care what they see as limitations of the engine from an internal standpoint because we have been dealing with the results of that viewpoint for ages now. They need to listen to people who actually use the engine in a production capacity and have for a long period of time, something even a modestly large demo, which this is, will never compare to.

 

I agree, but it's at least a start. With the degree of tunnel vision Unity has been afflicted with so far, even a match-3 game would do to begin with, as long as they have to build it and make it actually work on a wide variety of machines.

 

One project won't solve all Unity problems under the sun like some people here expect it to. It's a good start imo. Way better than any previous demos, for sure.

 

The content in the video already looks quite polished and they will do a full on release on Steam.

 

How about we wait for the project to become available before deciding Unity's efforts are a good start? Lots of unproductive speculation in this thread.

 

You don't know that yet. To me it looks like a lovely looking demo that has a character controller - so far.

The problem with demos are that Unity's very good at them. It's the perfect engine for them, but as the project grows, most people on these forums get bogged down by them, where the real problems start.

 

What does "full on release" mean exactly?

 

SO FAR I take it to mean releasing a free demo scene with about 30-45 mins playtime.

 

I'm trying to discern the difference between "full on release on Steam" vs "dumping it on Steam" and so far I can't come up with anything meaningful.

 

But promise us that you show all the dramas.

 

There is a lot more systems in Gigaya than just URP Assets and a CC.

Just some of them:
- Save and Load
- Async Scene Loading
- Quests
- Enemy AI (that has pathfinding, states, you can combat, etc)
- Pause
- Checkpoints
- Enemy AI (two types)
- 'Machines' -> Elevators, Interactive Levers, Activation Plates, etc
- Collectable System
- SFX Tools
- NPCs with contextual quest dialogue (also localized)
- Surface tagging and detection (for playing contextual sfx and VFX like footsteps and foot imprints)
- Cross-Scene Object Pooling (Used by many things like SFX, VFX, Projectiles, etc)
- Cross-Scene Referencing
- Cutscene and camera 'volume' tools (IE: walk into an area and it switches to a different camera or plays a cutscene)
- And more!


And plans for more common things (like quality settings, I really want to tackle a decent controller remapping setup, etc). We are still working on it.

Granted that this is not every scenario for every type of game; I think its certainly the right direction compared to previous efforts. And atleast starts with a little bit of everything.

 

I am the first to blame Unity for lack of taking feedback seriously and lacking leadership who is able to accept external critique and incorporate that in meaningful decisions. The whole process of how DOTS came to be up to where it is right now, speaks for itself.

BUT I definitely give credit to what they are doing at the moment. The team has a reasonable size, the project has a clearly described lifecycle and @Andy-Touch gets special credit for coming down to the shark tank here (pun intended) and trying to discuss issues in a very calm and constructive way. I really appreciate the last part, because constructive communication is not necessarily a core trait of all other dev teams.

@Murgilod raises some very good points, albeit maybe in a bit salty manner in general. I do get the frustration, believe me. Ambitionless roadmaps that are not even stuck to can leave us behind very, very frustrated. The whole progress of what the Animation team seems to have been doing for 2 years and the next two years - or more precisely the total lack of visible results and feedback opportunities shall stand as an example for where that frustration comes from.
Further added by very questionable acquisitions as of late instead of focusing money on things that would actually matter (e.g. why build shadergraph instead of just buying the team at Amplify? Their product is still light years ahead in UX, just switch your custom shaders between renderpipelines to find out why ) or pushing extra paid service upon service upon the user base.

The main point, that is still standing and that needs to be addressed is: take user feedback seriously and act meaningfully upon it. Add more staff to QA and bugfix-teams. It's painful to see the current response times to bug reports and sometimes the pure lack of effort in handling them.
That being said - having an actual production team, that is working on shipping a real, living product does not cancel out the other. And honestly? If you are able to buy Weta, the money for a small 15 ppl dev team and their product should not really be an issue. Some people here forget, that Unity has about 3.400 employees as of now. We are not talking about a small indie studio here - it's a tech corporation.

I would really ask for this project to be taken seriously though - e.g.: Ship to different platforms (switch, PC and maybe consoles) - put effort in some sort of meaningful gameplay to incentivize people to actually download and play it - an maintain it after release with more content. Especially the part of "provide more content after release" should be a priority, because shipping patches/content and keeping a game alive over a certain duration is a challenge, where unity clearly is lacking proper workflows/tools.

Oh - and if I can make a wish? If you consider a new project after this: Make it multiplayer! No one tells you to build fortnite - just look at what Dani did with Crab Game or something along those lines - a casual, arcade game. It could serve another important part of feedback on topics Unity lacks and does not require the effort and money in terms of content/world-building/etc.

 

With out naming and getting into specifics (for legal reasons) in majority of scenarios its not as simple as 'just buy company X or product Y and job is done'. Could be many reasons why its not that simple; legal, money, ownership, independence, direction, technology choices, patent, etc. In some cases the stars do align and it works out (IE: Cinemachine! Although im waiting for AcidArrow to jump in and say 'it sucks'. )

(Disclaimer; im not on the team who make acquisition decisions)

 

Some more fun info; we use Plastic for version control (no one on the team had used it before so that was a fun first week! ) and automate nightly/weekly builds through Unity Cloud Build for various branches (Main, QA, etc). And then have a pipeline for those builds to be distributed internally through Steam for playtesting etc.

 

On paper this is better already because it actually has long-term support effort behind. It's way larger in scope than any other demo. And they're using default engine tooling without rewriting half the engine with source access.

 

@Andy-Touch I know and agree. I always assumed it was probably not necessarily the lack of Unity's wish to buy them and more the actual availability.
It just served as an example - of which there are a few more - of acquisitions which would make sense in comparisons to some, who seem to not really add up in the long run.

And for things that got bought, there seems to be a certain issue with "following through". For example: Why does Odin Inspector cost this much money now and is not getting integrated to become "Standard Unity featureset"? Why is ProBuilder/ProGrids getting so slow in development and seemingly is stuck in limbo since a while? Why is breaking legacy compatibility absolutely ok for many core features, but integrating text mesh pro as a standard Text-component in Unity is just not happening? Why are some of the great tool ideas behind MARS all put into a separate product instead of making it available to the masses (aka "democratized")?

 

We set ourselves a very strict 'only public unity builds, public APIs and public packages' rule. We started on 2021.1 and currently on 2021.2 (but will move to 2021 LTS when is ready; we already have it running on it but wanted to do the upgrade after GDC ).

 

Worth a quote so people don't all get too harsh! <3

 

Some links for the chronic contrarians and naysayers who watched the video but didn't look into it any further before sharing their opinions with the world:

https://blog.unity.com/games/introducing-unitys-latest-sample-game-gigaya

https://unity.com/demos/gigaya

 

I never take online discussion personally.

 

I appreciate your view but the FPS project promised the same thing then stopped. I mean at that stage, the editor would not move more than 1 frame a second on my fairly good rig at the time. I think it taught Unity a lot but they did actually push that project too in the same manner until they didn't (basically when a showstopper came up, since to work around would defeat the point).

 

Thank goodness since if (and frankly it ought to be when) we have a cuppa I don't want my lights punched out

 

I wasn't interested in the FPS sample so I can't comment. I do hope the history doesn't repeat itself in that case.

 

Never! Always would be happy to have a brew and chin-wag!

 

You should have used Collaborate, you cowards.

 

Given Collaborate isn't long for this world... https://blog.unity.com/games/upgrading-from-collaborate-to-unity-plastic-scm

 

I mean, so is half of Unity's current feature set.

I'm thinking about this in terms of making a quality reality show (I volunteer to direct it, pro-bono), I'm not saying it in order to help improve the engine / editor, I'm convinced that can't happen.

 

I dunno why everyone is so salty. Unity has 5,000 employees and simply spent some time and resources to dogfood their latest systems in a small playable demo that they're giving out source for free.

Yeah SRP is terrible, there's still plenty of holes in Unity, it could have been a different type of game, etc, but they're still incrementally building projects that will actively feed back into the editor and other packages.

I think it looks great, I'd love to check out Andy's CC and see how they're doing things. Good job guys.

 

Glad they're doing it.

Movement looks kind of weird. When Wondu on top of the gear you can see his feet sliding with movement (and there, as well as on the roof, you can see his feet aren't grounded).

The cadence of "where the Gigaya" is wrong - the "where" happens too fast, and Minda's voice shouldn't rise to "the Gigaya". The vocal effects on that voice and the robot voice make it a bit hard to understand them. The robot's "I have no purpose if there are no settlers" sounds like a non-sequitor--it's not relevant to the previous sentence and the general thrust of the trailer's narrative.

And "Time to try something new" as well: you're creating an expectation (like my colon there a few words ago) for some type of introduction, but the scene kind of ends. That might be a meta reference to this whole project, but within the context of the trailer (where we just get a title drop) it doesn't work.

This is all nitpicking. I'm glad this exists! I look forward to the benefits that arise from it.

 

@Andy-Touch I'm glad that this is finally happening!

Not too long ago we had a thread about how Unity should make some games of their own. This is a much needed step in the right direction!

I just hope the game isn't abandoned and that your team gets to keep making games every now and then to validate the current state of the engine.

Kudos!

 

Ill look into that. Probably a collider sizing issue.

Minda is a spiritual ethereal entity and the robot is old and scrappy; its intentional for them to speak in different annunciations than humans!

 

For someone very dismissive of every single thing that Unity does, I do sometimes wonder why you frequent the Unity forums every single day.

(Please don't take this the wrong way; im just confused because if I disliked every single thing about a product/company I just simply would not spend all day every day for years posting on their official forums. Then again, everyone is free to how they use their time!)

 

tbh unity did that to themselves. In a way, he's the product of unity's long streak of interesting decisions.
I don't think there's anything like unity and it's workflow, so personally speaking for me, even If I switched to a different product and wasn't a fan of recent decisions I'd still frequent this forum.

Plus, it's very active here. Forum is a win for unity

 

To compensate for past failures in that respect I think they probably need to pledge to keep this particular dogfooding project fully alive across a good number of LTS releases to come. And the pledge should state a specific minimum number of LTS releases that they will keep this going for. And any future successor projects that may be more ambitious and feature the next generation of technologies should only be started with a fair period of overlap with the original project before the original project is allowed to rest in peace.

A lot of the conversations featuring criticisms over the years have related to specific tech areas, specific examples of over promising and under-delivering, having to abandon new systems and go back to the drawing board half way through, etc. These discussions only capture part of the problem because they are driven by outsiders, and as outsiders we cannot fully get to the bottom of any structural problems that may be more obvious internally. eg if there are problems with specific people, management politics, silo mentality issues etc. But a proper, sustained dogfooding exercise can at least bring such issues into focus internally, or at least attempt to bypass the counterproductive crap, get priorities all pointing in the right direction in ways that are harder to politick around, etc. And should gradually be able to demonstrate to the broader audience that progress is being made and that the right priorities are winning out.

Hopefully this exercise began in time and is allowed the necessary room to tread on some toes internally, otherwise there will be louder calls for sharkfooding in future

 

While their previous dogfood project was the DOTs based FPS sample, this project is basically a tall vertical slice (for features). If, internally they can make the assets duplicated 4 or more times, and the scope extended with those duplicate assets, of course mixing asset usage - they will have a sufficiently wide test project. This is where stuff starts to break down, and force the game to rely on streaming, stress physics, audio, rendering and so on.

So long as Unity doesn't try to work around problems, it will qualify as dog fooding. When you work around, you demonstrate no engine problems.

So it's great they went for tall first.

 

I would have started with the sort of project they've gone for this time because I dont think they need a wake up call about how far short Unity has fallen in some of those other areas and games beyond a certain scale - something already happened to their understanding of where they were really at in the last few years that caused their roadmap ambitions to fall off a cliff. I found that hard to come to terms with in some ways due to the level they had previously set expectations at, but I will concede that its a good thing they learnt that lesson and found some more reasonable limits to their ambitions for now.

So I think stuff mentioned in the last few posts is important when considering what happens next, when DOTS is more comprehensively in place for example, and how they should dogfood the scale of things such tech should enable. But I dont think that moment arrived yet, they dont need to even sniff the dogfood to establish that as of the current state of Unity, a world of workarounds and 3rd party dev/systems are required to hope to function at such a scale.

 

Do you plan to share your optimizations in detail to get it running? Wanted to port our new game to Switch because our previous one sold very well on there, but even on lowest quality settings (URP) a terrain with a few trees drops fps below 30 and looks super bad there, not to mention any gameplay stuff.

 

Your engagement and answering of questions is greatly appreciated.

 

Really sucks how the editor/desktop sharing is low res. They're showing editor footage and I literally can't read any of the menus, and the game looks like a PS1 title because of it.

They say it's because they're using zoom, this begs the question.. why? why not discord? This seriously hurts the stream. No testing beforehand? How does something like this happen?
If it was due to the developer's slow internet I can understand, but this I really can't fathom.




At max quality.

Good luck reading anything.

 

I am sure there will be a write up and any questions drop them in here so that they can post the high res bits you couldn't see!

 

Thanks for the demonstration on Twitch @Andy-Touch , it's a nice sample project and I look forward to learning from it. It would be cool if appropriate elements of it were made available as separate packages that are great for bringing into other projects easily and regularly. things like the responsive vegetation as a package, the water as one, things like that. character controller maybe too, and so on.

beyond learning from the whole project, there is definitely a lot that people can use here, i think it will help in it being used more to package-ize it's useful content

 

fun fact for those who might not know, gigaya uses Volumetric fog & mist asset from krconnect.
It's a good asset! Hopefully URP gets volumetric fog one day

Thanks for the demonstration! Was fun to watch and learned a lot, especially about character controllers and animations. Wish I was better with physics
I'm now watching the saved broadcast since I was a bit late.

 

pretty nice mechanics, that revolving ground.

I guess it's no rootmotion physics capsule, built on the new third-person controller?
with mecanim and custom IK implementation.

and nice ocean. is there, maybe buoyancy? the new physics in 2022 will make interesting possibilities

 

I believe the answer to that is yes. They're using the water from the boat attack sample, which has buoyancy.

 

@Andy-Touch Hey, if the team ends up using Addressables for GIGAYA, please consider adding modding because it'll be released on Steam, and also to see the difficulty of making Addressables mod-friendly, especially when combining multiple different mods from different sources using catalogs. Showing how you're able to solve that would be very valuable for those of us trying to get it done using Addressables. Thanks!

 

Thank goodness nobody really cares about performance.

 

Actually, I'd be interested in an example project that uses addressables. I haven't wrapped my mind around this feature yet.

In any case, a detail postmortem for this project would be cool. One where all of the technical decisions are explained in detail. As with the 3D Game Kit, I imagine this project has a lot of potential as a teaching example.

 

I actually want to see that it also passes Nintendo's, Microsoft's and Sony's technical requirements checklist and be able to download it as a free demo from those stores. So we have to fight less on the TRC end and Unity Technologies identify issues related to that more quickly.