Unity is an awesome tool I love using almost every day. It also feels like some things are painfully missing for too long. I realize that a lot of cool things are happening behind the scenes. And that there’s unseen complexity behind why some features take longer to bake than others. But…
Example 1: Nested Prefabs. They’re largely wonderful. But during the showcase video, even the team presenting them made comments along the lines of “when I first started using Unity it felt like this vital feature was missing”. They took too long. A lot of users would have loved to utilize even a basic working solution years ago.
Example 2: Timeline Signals. They’re ok? Underwhelming? Feedback has been mixed , and I don’t think it’s just because Unity users are unreasonable or too demanding (though that’s part of the story ;)). Timeline itself is a great tool in some ways but also feels completely lacking in others.
It’s really, really hard making an interactive product that will satisfy a large number of users and also fill a lot of smaller, niche use cases well. Unity has an incredible task doing what they do. I understand this (as should anyone who has spent time making games). Which is why I’d love for them to start making their own games.
Go beyond the cool tech demos and small-ish partner experiments and have teams in extremely close proximity under pressure that have to use these tools every day to deliver something on a tight budget. Take the FPS sample and make it a real thing that needs to be fun and largely free of bugs and doing cool new impossible sounding things requested last minute from a demanding producer. Take a few small demo-y projects and try releasing them on mobile…
I don’t doubt you already do a good amount of this and have close relationships with some number of publishers and studios. But it’s different when it’s part of your core. Eat your own dog food, please?
I’d much rather them work on an engine than either make a game studio from employees they have now or hire more staff to make a game. That is an exceptionally large investment with little chance for return in a hit driven industry that could better be put towards engine development.
Unity staff have and do make their own games, internal and otherwise (the fps sample project is already teaching unity many of the dogfood things you’re talking about).
Having Unity make their own games might improve the engine for game projects, but not for non-game projects Unity is increasingly being used for. It might satisfy some users, but not other users, and as such I am not sure what problems this would solve.
No, they just removed the old system before the new one was ready, and are currently relying on appropriate sections of the forums for feedback which is likely a much more effective approach as the actual developers for these systems are reading and posting regularly in them whereas I suspect most of them just ignored the feedback system.
It’s not inherently an either/or problem, they can feed into each other.
The FPS sample is great and I imagine helping a lot with figuring out what DOTs should and shouldn’t be. And I also imagine it’s not the only one they have, but simply one of many things they work on they’ve decided to publicly communicate. But there’s a big difference between a tech proving ground and running a project you expect to fully release one day.
I use Unity daily for both, you can substitute “experiences” for “games” here.
They really likely won’t because games are very expensive and often don’t make their money back though, especially larger ones. I do not want Unity to have to focus on making a marketable game while also trying to maintain the thing I use for money.
I have seen Unity being used for (and developed myself) software which would not be described as “experiences”. Data visualization is a big one, for example, and is developed very differently to a game.
It doesn’t need a AAA budget. You can imagine they already spend resources on feedback / standard user research, working closely with external teams (like the Adam demo), internal projects (like FPS sample or Megacity), etc. I wouldn’t advocate for getting rid of any of those things, but re-balancing.
The point being it’s not just maintaining the thing we “use for money”, but more like this: I see companies like Ubisoft show off a cool motion matching system years ago. It’s a relatively simple system to build, with a smallish team. But beyond our tight budget. As a small studio, Unity really is our “engine team”. I would have loved to have a tool like that for more than one of our projects at the time.
If they made their own games, would we have gotten that feature? Maybe, maybe not. (I am aware of kinematica btw, fingers crossed for something at GDC…). But when you’re directly under pressure, you get creative: necessity is the mother of invention. And it’s precisely the conditions of a small team with not a huge budget that can’t afford to throw man-hours at the problem that will discover and really advocate for that sort of thing.
[quote=“tbriley, post:9, topic: 735303, username:tbriley”]
I see companies like Ubisoft show off a cool motion matching system years ago. It’s a relatively simple system to build, with a smallish team. But beyond our tight budget. As a small studio, Unity really is our “engine team”. I would have loved to have a tool like that for more than one of our projects at the time.
[/quote]Well you should be happy since the person who did that at Ubisoft (Michael Buttner) is working on Unity’s kinematica.
i remember jumping out of my seat when i saw that announcement the approaches seem a bit different, but i’m hoping unity’s gives smaller studios greater ease of use + flexibility.
it’s also sort of my point. maybe a large amount of the work for kinematica started before michael buttner arrived. or similarly the DOTS work was well established before mike acton and andreas fredriksson joined. and i realize these accomplishments are made not just by these superhero programmers, it’s a confluence of a lot of hard work from a lot of domains / parts of the company. but i think we’d see more things like this if the external expertise feedback loop was more formally internalized.
@zenGarden iirc they showed a few things off towards the end of the year and then decided it needed a bit more time in the oven - you can find some cool demo videos on the interwebs fingers crossed for gdc.
one potential issue with a system like this is the space is developing pretty rapidly and it seems pretty regularly there is a cool new technique, for example:
and:
systems like these may be a bit more difficult to generalize, but we’ll see consumer grade tools for all kinds of approaches sooner rather than later