It would be cool if UT make "games"!

It doesn't need to be big games with >30 hours of gameplay, but games that lasts around 1-2 hours, it would be cool already.
So, they would know better what developers need, there are dozens of stuff right now that is extremely annoying to work with in the editor, and I'm not talking about the big stuff like terrain or nested prefabs, I'm talking about simple things like those in this recent official thread. Also, when they find a bug they would be able to debug it right away and fix it without the help of customers, the perfect repro case would be right there, in their hands, and there are things that you only get to know when you really use the thing in a real life case, one could say, "well, the features are there, they were tested, so what's the matter?" - use it in a real (and big) project and you will see, I think that only testing it is not enough... there are lots of things in the editor that I look at (lots of them were posted in the artist workflow feedback thread so I hope I get to see a lot of them fixed) and can't help but think: whoever implemented this feature the way it is right now never really used it - yes, it works - but it's not suitable for repetitive tasks and other situations.

I know this is a rant, sorry! But I think it would be awesome if they took this path!

Oh, and while they develop their games they could also make those "Happy Hour Gamedev" streams where they add a feature in the game or fix a bug and we get to watch and interact with the developers, I wouldn't miss one of those streams!

and sorry for my English.

 

My thoughts on this are very represented by others the previous times this has been discussed, but covered best by @Aurore's post here.

Making games and making game engines aren't necessarily the same skill set, and honing one doesn't give you an advantage in the other. Their time is best spent processing and integrating feedback provided to them by their very involved community.

 

I read that post from Aurore right before I post and I was quite reluctant about creating the thread, but like I said, it doesn't need to be a fully fledged and long game, just need to have all the features that a game normally would have, in another words, make use of all the unity features, that's it, also, she says this: "just because we've achieved in creating a successful and popular game engine doesn't mean we have the perfect formula to create an amazing and successful game."...
It doesn't need to be an amazing and successful game, what I was thinking would be just another way of teaching different approaches of how to implement a feature and making this feature come along with all the other features.

Sure, that's why they would need to start another team with 3 to 6 developers to make this, developers who make games of course, I think it's enough to create a quality game, small, but quality - like Angry Bots were, this team would have close bonds with the rest of the UT.

 

All that thread really says to me is I haven't done enough in the past three years..

@Rodolfo-Rubens

Personally I would of liked to see something come of the "official" threads that were posted around here (with a continued effort). Like the terrain system thread looked really promising and then kinda devloved into obscurity.

There's of course nothing like first hand experience (which I don't even need to guess Unity has staff with plenty of) but I can't see at this point how much they'd gain? Unity has references, feedback, examples, other projects they "consult" on (some of them very expensive projects).. If that appears not to cause much in the way of influence, why would their own internal development team fair better?

 

Keep at it! People like us are just a slower burn.

 

I agree, I think it would be useful. Especially in the case of demos, they seem to focus on cinematics at all costs, which are always very finely tuned and constructed in a way that a game cannot be.

Here's an example: Mecanim gets a fair bit of criticism, maybe if the Adam demo had been an interactive game it would have forced them to find ways to make a generic rig work better in cinematic sequences, and improve the system that we all use. But when giving out the characters on the store (which of course is very nice of them!) it seems they had to be re-rigged to be used outside of the demo context, and quote "[we] made a small rigging tool which tries to replicate some of Adam’s shoulders functionality." Also they used a third-party plugin for cloth simulation, so that had to be removed and the clothes re-rigged for Unity's physics. One wonders why, if Unity's native cloth physics is good, a third party plugin was needed at all?

I'm not trying to make a big deal of this particular example, but it makes one wonder if by creating game demos instead of highly customised cinematic sequences which have to be gutted before becoming suitable for general distribution, Unity would not end up providing a much better set of tools.

 

Yes, agreed with everything you said, I think the same goes for Blacksmith, making a cinematic demo is one thing, making a game is a complete different thing, and that's not what is Unity primarily made for...

Well, newcomers would gain, also, even though I like Unity a lot, as a company, the community and the software and I'm pretty sure I will use it forever but I can't say I trust their software 100%, yes, we have lots of games that prove what Unity is capable of so I will not mention that every other major engines has their games, even though I'm not talking about these types of games, is just that... sometimes it feels uncomfortable using the editor, there are some things that they never address and we need to code our own workarounds or buy it on the asset store, and I'm not talking about big features like nested prefabs or terrain, etc, I'm talking about very, very simple things like, the project settings, graphics settings, input, etc, they open in the shared inspector window instead of having their own tabbed window.
Of course, this is just one of the items in the list, and again: I hope many of these workflow flaws gets fixed when they start addressing the feedback in the feedback thread.

About the games, very short, you can even call demo, demonstrating how to approach different types of elements in different genres, all packed in the same project, not like the ones in the learn section, more features, more complex things going on, more like the Angry Bots. Right now I can say that I can do almost everything I can think of in my games, thanks to the easiness of the Unity way and the internet, well, almost. I know I still have A LOT to learn, but even if I knew everything that there is to know I would want this because I like to see how other people would make something I did
But I remember being lost in many aspects of the process of the development, mainly about UI, I think everyone gets a little lost when start, it's true that we have many examples, tutorials and even some projects but it's not like having complete game projects with everything working together, all the objects in a scene wired up, multiple scenes in the project, etc (like the tanks, which was made to demonstrate the unet features) maybe releasing these small game projects would shed a small light in some subjects, and like I mentioned before, make some streams fixing bugs, implementing a feature, etc, would be very nice too.

What I'm suggesting here are not games to be commercially successful, I'm suggesting this for two things:

1. UT acknowledge by themselves, as users, what is the best for the engine, what can be improved in the engine to speed up different workflows, make the use of engine more comfortable, etc.
2. One more type of material in the Learn section (we have the tutorials section with some projects but they are way too simple and they come in a very slow pace, it's because of the fact that those projects are developed by the same team behind the engine development, I know, I know!)
   • If the game is still under development, show just a banner with the game working title or the type of game that is being developed, in the project page we would have a description, maybe an introductory video, all the streams that was already made, a button to be remembered when there will be more streams, a comment section or a link to the project's thread in the forums so people can suggest features, etc.
   • If the project was already finished, the same things as above, but also a short video showing how is the game like, their features and etc, a link to download the playable demo and a link to download the project itself, maybe only make these available for plus subscribers, so there would be another incentive to get a subscription. But leave the playable compiled game and the dev streams for free users too.

I don't know guys, I'm very hungry right now, been working non-stop, so maybe I'm just delirious with these ideas, sorry. Sorry about my English too.

 

Unity Technologies makes technology. We use it, give them feedback, they iterate and learn what works best from what we say and they see.

If you say they should use it, then give us what they think is best then we really get a delayed product, less communication and have a significantly smaller impact on the product as a whole.

So no. I prefer it this way.

 

That's the idea, but I don't think it always works like that. When you have a tool that you don't use yourself, you tend to focus too much on the things that make it easy to use, or build games at beginner level, in order to attract more new customers, rather than stuff that builds the 'guts' of a more complex game.

The terrain system is a good example of something that's been clearly wrong for a long time, and for much of that nobody really knew what was going on with it.

 

They reached out years ago polling the community on Terrain and have already said they're working on it.

Again, an internal team that builds *entire* games is going to be a big investment and it sounds like the goals will generally reduce community interaction.

 

It's not that simple. And the terrain system is not an isolated case, there are many other things that is just like it. Take this thread as an example, is not just the terrain system... but I'm not rushing them, I can wait.

You say that like this was already done... and failed.

If they make the games appealing enough and only release the source to plus subscribers, that could mean an increase in their income, I honestly don't see anything in the plus subscription package right now that make me want to get it, not even the dark skin, but demo games provided by UT... even if it was in the pro package.

How come? You mean interaction between the UT and the users?

 

Yes. From a business standpoint they seem to rely on the community quite a bit for feedback and direction. If they made the business choice to migrate that to in-house then it stands to reason that they wouldn't be communicating with us for feedback as much since they would have committed to the in-house group for user feedback.

 

But it's not like they would just abandon us, I said before, just one more type of material in the learn section, all the rest should stay the same, at least that's what I had in mind, how I see is, they would have one more source of feedback, mostly for bugs then features/editor usability improvements, a source that is very close to them that would provide better bug reports - because they would have the project in hands, the perfect repro case would be right there.

 

That one more thing represents significant work, time and investment on their part.

 

I know I'm not suggesting something small, it's big, but it would also be a big help for the community.
Also, investment is a good thing, it means you will profit somehow (plus and pro only ), so both parts win.

 

The funny thing is that, if you go by what some people like @ShadowK are saying, engines like UE are rolling out the equivalent of all of the major features of the asset store quite fast, as well as creating their own games.

I think that part of Unity's quality is due to how carefully they update the engine, but I'm not at all certain that things are anywhere near as efficient right now as they could be in terms of getting us the features we need.

And as much as I like the fact that Unity seems to be doing a better job of seeking feedback, they have a startling way of opening the discussion sometimes like they haven't any idea what people have been clamoring about on the forums for years.

Anyway, in terms of time and investment - I think that when there's a will, there's a way.

 

Actually I think that would be quite cool and they'd probably have a pretty good chance of selling a bunch to all the Unity developers who are fans already.

 

They do use it as previously stated.
https://forum.unity3d.com/threads/t...e-without-in-house-games.279014/#post-1842948
"We in fact have 3 teams that create content for internal and external use with Unity, all with different goals. We also have a close working relationships with our premium and enterprise support customers, we get a lot of feedback from them, it also stemmed our sustained engineering stuff. Additionally there's the closed Alpha and Beta groups and the community site of course."

Unless that info is outdated - which it might be.

 

Well they have at least one game released, and nothing prevented it from becoming massive hit except the lack of players, although they got a lot of downloads.

I think they should spend more time making the Unity Editor be able to create games and simulations that a teacher or student could use the sort language and symbolism they'd actually use in class and not these programming languages and Unity specific technical specifications. e.g., Having a comprehensive library of human animations and telling you want char a to walk from address 1 to address 5 but run while threatened from char c. On the way char a should look for a set containing objects [0 - A] sort of a lisp/prologue way of simplifying game / simulation development.

So basically Unity would be a complete simulation environment that users directed their story or game with. LOL, that's a lot of work and data collection to do that so I wouldn't hold me breath waiting for it.

 

This is literally the opposite of what would be considered a good idea. This idea is, if I may be so bold, very bad.

 

Games though? Because imo these cinematic demos are probably not the most relevant thing for most customers. You can customize everything to a tee for quality when you're making a quick demo, but making a game forces you to make the sorts of things that people actually need - efficient, flexible, scalable things that also look good.

Anyway, this is the way I see it. If they had made a game demo out of the Adam demo, that people could download and fiddle with in the same way that UE does with unreal tournament, I imagine that not only would people be better able to appreciate what Unity can do, but the feedback regarding what it should be able to do would be a lot more relevant and precise. As it is, all we really get from that demo are a bunch of game models on the store that, while quite pretty, don't really teach very much about game development in Unity.

 

Opening up Unreal Tournament to get everyone involved was a cool idea, not only was it a great learning experience for the community (working on an actual large scale project w/ a crack team of pro's).. But Epic got plenty of free help..

Something like that could work?

 

They been doing that from the very start, I even remember working with ue2 and ue3 always loading up that generations UT game and see how they did different things.

 

The terrain system thread was started almost 3 years ago. I don't really consider that something to be proud of anymore :|

 

If they started a game three years ago, would they be done today?

 

Possibly, depends on the game. I personally don't have a strong opinion on that.

I was remarking more on the comment made about the terrain system official thread, asking what everyone wanted. I guess the #1 response should have been "consistent communication."

 

Although I can see the hypothetical benefits of your and others suggestions, I really think Unity, using user feedback, internal teams, and premium customers (who are working on high quality games) provide enough input to keep the engine moving forward, and leading the pack as it is.

Unreal tourney (pick one) took 3+ years to create, that isn't time well spent creating a game instead of developing the engine. And although I participated in several mod teams back in the day using UT, I got really tired of fps mods with altered guns and stats. Nearly nobody was able enough to convert from the rank and file, twitch based fps shooter.

If Unity did create a commercially released game - it would HAVE to be at least as good as the best indie project on the market still yet to come - to PROVE and SHOW what is capable with the engine. That would take tremendous effort much better served developing advances in the engine for us to use.

 

Well, obviously there are users wanting 4 different things here:

1. Complete and fully fledged game projects like UT (Even if they come in >3 years)
2. Small game projects
3. New important features to be delivered faster + more communication
4. Everything is fine as it is

 

I'm not so sure, I mean UE make games - mobile games, PC games - and on top of that free demos and unreal tournament, and last but not least seem to roll out stuff regularly that can only be found on the asset store for Unity. I think the biggest problem for UE users is how much stuff is changed and added every time they download a new version, which seems to come quite often.

Like I said I think Unity's great approachability and ease of use is partly due to them taking more time, but it seems to me as if they perhaps have too much of a focus on marketing Unity to new devs, and flattening out the learning curve, and not enough catering to those who need robust advanced features. Most of the complaints about lack of features I hear are things that a newbie probably won't encounter for months or years.

On top of that what annoys me is that every time they make a demo, it is completely stripped before arriving on the asset store as what basically amounts to a pack of game models. Whereas most of what Epic does arrives as-is to the user, and there's no doubt a lot to learn from them.

Maybe coming up with a game is not the most efficient way to improve things, but at minimum it would force them to come up with something substantial and representative enough of the core engine that they can live with the prospect of everybody getting their hands on it, and which could teach us a lot as well.

 

This is a good point, but I don't think creating a game would resolve this. Possibly, but the game "tweakable/modability" would also have to cater to all skill levels, unless it was explicitly explained, this game is for advanced users, which would probably not 'sell' well with the overall user base. idk

This is also a good point. I agree I'd like Unity to use the tool and its limitations to develop the content that is presented to show what is doable in the engine. I've not messed with the Adam content, but the Blacksmith content (characters specifically) were 'customized' with setup/tools not available at that time in the engine.

 

You could be right, but there's already projects that people can download for beginners. I think a more advanced game (or even just a short project) would keep everybody interested through the beginner stage in hopes that they will one day be able to make something like it (and in the meantime poke around in it).

Not just the blacksmith demo, but it seems the Adam demo characters too - they have been "re-rigged" for general use and a tool has been created which "tries to replicate some of Adam's shoulder functionality".

But the biggest disappointment is that we simply won't be able to download, play and modify/poke around in the original demo which I think could have taught a lot about how to create cinematics. Epic have no problem handing over the infiltrator demo. I can only imagine that the Adam demo is full of hacks and third-party integrations that make it impossible to distribute.

 

It is the opposite, actually.

Making a game forces you to make inflexible non-scalable choices that do the job for the feature that needs to be implemented right now. You'll get bunch of non-reusable hacks instead of better tools.

Basically, when you need to implement something specific, the obvious choice is to go after inflexible solution, because they're easiest/fastest to implement. Designing an API (which is what unity is doing) is an entirely different beast.

 

Even if they worked on a full game with 3d and 2d UI I think they will find their API is very clunky, or maybe my use of it is clunky. Either way a UI heavy game would be great learning material for all.

 

Well the thing is that hacks and inflexible solutions are for end-users to implement if the choose. Unity should, imo, demonstrate the core functionality that Unity provides in the best way possible. Sure that might not be precisely what everyone uses for their game, or quite as good and fitting as a custom solution made by an experienced developer, but that's what the product comes with so that's what should be shown.

I mean, if you make an asset store product, the demo scene should represent functionality that a user can reasonably expect to find straight away - and if something is hacked it should be clearly marked as such (and even then it's not really a good idea imo, it should probably have been implemented as a feature if it's so important).

I mean imagine if Unity went on a shopping spree on the store, downloaded SEGI, CaronteFX, megasplat and a host of other things that don't come with the engine, and made a demo with it, I think it's fair to say that it makes it difficult to accurately assess what the software you're downloading from the Unity site is capable of out of the box.

I'm not trying to make a huge deal out of this, and I don't think the Asset store is a bad thing at all - it's a great thing that Unity have enabled people to provide solutions to uncommon problems. But I do think that Unity should definitely make their own demonstrations of the engine without using any third party assets at a minimum, and make the original demos available.

 

I don't think your getting it, only when Unity make a lot of games will they start to play with and cover all aspects of their own API and then they can make huge improvements to their API and how it works.

If I had my way I would sign every Unity employee including the CEO/HR/SALES up to the One Game A Month, they only get paid if they release a peer reviewed game that month. With the side note that they can work in teams on bigger projects but all must contribute equally.

Additional Note: Thousands of people make a game for ludum dare in 48 hours so this should be coffee break < 1 hour work for game engine developers.

 

I think the problem is with more advanced features though, I think Unity does a pretty perfect job of making small games as easy as possible.

 

Hacks and inflexible solutions are a BAD thing when they end up as a part of API you're working with.

It doesn't need to demonstrate anything - there were games built in unity, and the engine was around for a while.

If the argument was "But unreal/crytek makes games...", then at least in Unreal engine quality of programming api is significantly worse - precisely because they were making specific games. Haven't worked with Cryengine, heard horror stories about documentation quality, though.

Unity is too big for an asset store product.

 

That's exactly what I said! The only thing I'm saying is that if a hack is essential to the demo (i.e., the demo would look much worse without it) then it should be incorporated as a feature and not a hack.

It could help people learn about how to do things in Unity, by example.

I can't really say, but they came up with a whole Blueprint system so it's not like they do not have time to make their API easier to deal with.

Principle is the same though. When you see a demo of a game engine, you sort of expect that if you strip out the art, sound effects, level design etc, then what remains is going to be quite possible in the software you download. I don't think anybody is easily going to be able to cook up CaronteFX, neither is it part of Unity's core product, and so it should not have been included.

Anyway, for me this is not a big deal - what is a big deal is the fact that these integrations make distribution of the demo impossible. So essentially we get nothing from it, except a pack of models that cannot be used in our games (which is fine and expected, but probably not very valuable to most game developers). And not only that but it transfers reliance of the ability to use these features off Unity's shoulders and onto third parties, often individuals.

Anyway, this is the way I see things, I think there is way too much reliance on the asset store and third party assets for things, and too much obscurity surrounding the development of Unity's demos which many people would probably like to learn how to make. Maybe this is changing because they seem to be integrating things much more frequently, so that's good.

And I also like the fact that there is more communication between Unity and people on the forum, although when I see threads pop up asking about what we'd like to see, I think it would be great if the initial post reflected at least an acknowledgement of what had already been brought up in threads many times, often with some participation by Unity team members themselves.

 

No. Just no. API should remain separate from hacks, because hacks in API have development/maintenance cost. A very high one.

Uh, there's a sea of information about doing things in unity by now. Manual, learnign section, and an ocean of tutorials. If someone STILL needs something else, then perhaps their gamedev career wasn't meant to happen.

Blueprints use the same api, and it is messy.

Not the same, because scale is way too different.
I do not think there are any assets on asset store that took 10 years to develop.
On this kind of scale, development of your demo could easily take a few years, and by the time it is done, it'll be no longer relevant, because it'll be essentially using a fork of the old version of an engine.

 

I think you don't understand what I mean by 'hack'. I'm talking about a feature that may be suitable for integration, but which is implemented in an inefficient way because the system has not been designed to include it. If it is integrated, then it is not a hack, in terms of what I'm talking about at least.

If it's just a quick solution for an uncommon problem, then it's a different kind of hack - and what I'm saying is that a demo does not need to include this latter kind of hack to demonstrate the product, and so it should simply not include it.

So if Unity's standard shader needs to be hacked to include subsurface scattering for skin or something, and the demo cannot do without it, then this should be integrated as a separate shader that's included in the download.

 

Nope.

Feature included into standard download should be supported and tested on multiple platforms, plus well thought out. Something that has been hot wired and quickly thrown together to work right in one demo and only under right conditions should not be included as a feature.

You need to realize that unity is not a toy project, and that different kind of effort goes into making a part of API used for large number of purposes compared quick solution that is supposed to work in a demo. A quick hack coudl be documented somewhere as a possible way to do something, but it definitely should never be made a part of standard distribution.

 

We're talking different things.

If a hack is absolutely necessary to make a demo of a product, then the hack is by definition not something that is just useful for one scenario.

If it is not necessary to make the demo, it should probably not be included at all.

 

one shader is never going to cover everyone uses cases, so it makes sense for the standard not to have subsurface scattering, and for people to just build there own character shaders.

 

We're not talking about different things.

If a hack is absolutely necessary to make a certain demo, then by definition it is still something that most likely is useful only for that particular scenario and that particular demo, and not something useful for everybody.

Basically, if you need X to make Y, it does not mean that people who don't make Y need X.

Honestly, just take a look at Qt 5. It is the most amazing GUI toolkit out there. They don't exactly make any flagship gui applications by themselves.

 

Not part of the standard shader, but a different shader.

A demo's purpose is to demonstrate the product itself, not what it could be hacked into.

But anyway, I'm not particularly interested in the topic of hacks - if Unity want to implement hacks in a demo that's perfectly fine with me - as long as I get the demo and learn all about it. That's all I care about.

 

With this kind of logic, you'll end up with a demo that demonstrates some moving cubes and pretty much nothing else. Or maybe one of those untextured prototyping assets. Because art for the game is something you're suppsoed to do yourself.

 

Do you think the standard shader is the equivalent of custom art, and should not be included in Unity?

 

This doesn't make sense at all.
And it's not only the shaders we are talking about here, the atmospheric scattering, area lights, the cloth physics, etc. In the blacksmith demo we have a pseudo-pcss, also sub-surface scattering, the timeline was used to make it while we still didn't have access to it (which is cool considering it's a cinematic demo), many other things that we don't have access to are used in these demos, and if we are, it's not the same that was used in the demo, like the adam model in the asset store which was mentioned many times in the thread.

 

Sorry for double-posting but is this what I think it is:
https://forum.unity3d.com/threads/unity-starter-kits-what-do-you-want.471647/
?

 

It's possible for both to be true I'd say.

With the resources and skill set they have, I wouldn't be advocating Unity investing lots of resources into a game team.

However at the same time, the fact that Epic makes games means that the engine gets lots of those little improvements and features that you only encounter when you are making games day to day, with game teams creating fixes and them coming back into the main branch.

Look at the marketplace for example. The Unity VR example package is a great starting point. The best UE4 VR example package on the other hand is the full project files for Robo Recall, which anyone can download and pull apart.

Again, with the type of company they are I don't think Unity should expend resources making big budget games, however the UE4 approach means that the examples are much more usable. Take ReCore as an example - Unity put up a great blog post talking about the lighting challenges that were encountered, and discussed how they were got around by using a custom branch of the engine. The feature listed (baking lightprobes without needing to bake the whole environment) is a fantastic idea but since it was done for that and then forgotten it seems to have dropped off being usable in the public version of the engine.

edit: Just saw this thread: https://forum.unity3d.com/threads/unity-starter-kits-what-do-you-want.471647/ and it sounds fantastic. If Unity puts out some projects like this and has teams actually trying to make stuff with the tools (in a way where fixes can be put into the main branch) it will lead to great improvements.