What would it take for Unity to become the go to engine of AAA developers?

Many AAA game developers have their own bespoke game engine technology e.g. Battfield's Frostbite engine.

As Unity improves the gap between what Unity can do and these high end AAA engines will diminish. So at some point in the future it could be more cost effective for those companies to use Unity.

The question is what is in the gap between a Frostbite level AAA engine and Unity?

 

The most important gap is that Unity has to be many things to many people... Frostbite just has to excel in few things for a small list of studios.

Thus you are comparing a general purpose engine with a long list of build targets to a very FPS oriented PC/Console engine.


Don't think anyone who had access to Frostbite and the resources to work with it (don't know the toolset, but unless somebody REALLY poured some money into usability I am expecting Unity to trump it when it comes to the usability of the interface), and needed to create a PC or Console FPS would pick Unity over Frostbite... like... ever.

The same could be said for Unreal, really. Maybe even CryEngine, don't know how that thing compares to Frostbite (I guess not even an obscure inhouse engines toolset can be so crappy in the usability like CryEngine was when I tried).


At the same time Unity and Unreal allow the smaller studios and Indies without access to Publisher inhouse engines, or without the resources to develop their own engine to get within 5% distance of what a similar sized studio can do with an engine like Frostbite, if they know how to optimize their games in Unity or Unreal. Maybe missing on some of the glitz... but then, can a smaller dev really use all that glitz?
Horses for courses really.

 

Split it in two, do tools in unity, export to native engine, ..., PROFIT?

 

Do you really want that?

It might mean Unity catering really heavily to AAA needs, which might be different than indie needs. It may become a worse engine for you and I then.

 

I mean unity is too generic as it much work on every platform, so it's optimized for none (jack of all trades and all of that). AAA is all about that performance bit, they make lemonade by using every possible bit of the lemon as they can. But a second big part of AAA is having pipeline that works flawlessy to have efficient production, it mean creating tools and adapting them for each game every time. Tool make a difference between the failing project and the smooth sailing. The hick is that generally tools are integrated in teh engine to have wysiwyg result, so they would need to integrate unity to their engine as a platform to unify all tools and allow anyone to extend them easily.

 

Enterprise is not cheap. And as mentioned above, too generic. AAAs require more flexibility, and already are invested in their own tools. In house engines mean they can change core tools on the fly and can also future tech. Given the long time lines for those games, they are often building for platforms a year or two out. It's just not practical. Or necessary. Unity is perfect for it's market, it doesn't need to be a solution for every market.

 

literally not possible, large studios are very invested in there own tools, and the publishers dont like relying on others. What if a publisher had all of its studios building on Unity for titles that are a few years away from release. Than say Apple or a similar company bought unity, that could destroy a publisher.

Also why would you want this, it could not cater to the need of AAA and indies at the same time, and someone will be left behind if it does.

 

Let's not forget that's exactly what unreal do though, but they are much more focus in AAA games (as they know how to make AAA games and what it implies, unlike some people) and basically sell a starting point to customize. And what's left is more or less adapter for indies but don't cover it as much like unity does, it's more like an opportunity.

 

That's a really good point. Aiming at big studios can easily make it unusable to everybody else.

Unreal model has its own issues. IIRC changes to API in unreal engine are more rapid. So something you rely on may end up being refactored or disappear altogether between minor engine releases.

 

This. Unity's model/market is clearly successful, not much logic in upending the model to follow a less successful competitor.

 

They corner different market really, I mean unreal just got big into the Japanese market recently, a market that used to have not invented here syndrome big time, Given that square enix, ie one of the biggest tech leader of this market, start using it (ff7, the new dragon quest). Since teh question is about AAA, I would say unreal got that market better than unity (I mean how did ReCore fare?).

So basically what @zombiegorilla said, each have their core market and expending beyond opportunity might not be worth the cost. AAA is only relevant to unity to sell the dream to indies (see that other thread).

 

I'm no AAA developer, but I can speculate on what I would want in an engine as one. I might even get close. Here are my top picks:

Purpose built for the game genre. If I'm going to spend five years on a FPS game, I want an FPS engine. More so if I intend to push out sequels every year after that. Same if I was doing open world, or RTS, or whatever. The engine should be built with the specific game type in mind.

Amenable to large teams. The engine shouldn't require my artists and my designers and my programmers to talk to each other. Heck, my rock texturing team shouldn't even need to talk to the rock modelling team. Large corporations work best when the work can be effectively segmented between teams.

Single platform. Or at least specialised to the two or three platforms I'm using. If I'm doing a PC game, I don't want mobile builds, and vice versa. On a console I might even go to something exclusive to the console.

All of these changes would likely make Unity a worse choice for the average indie or solo developer. So I'm hoping Unity doesn't go that way.

 

So most of you argue that Unity needs to provide a wide set of platforms.

What if you look at it another way, when in time would Unity have been a go to AAA platform, the 80's, 90's, 00's.

80's definitely, pixel art arcade games with 3d wireframe standouts like STAR WARS.





90's things start going 3d




00's





I think Unity is good up to about here, going into the 2010's how does it compare, what games and engines outshine what we can do with Unity?

Or where is the gap, is it as some have said engines dedicated to specific genres, although most good game engines cross genres e.g. Frostbite is used in Battlefield FPS games and Need For Speed. Also the Battlefield games include flying, floating and driving vehicles.

So I don't think the genre specific argument holds up as well as it used to. e.g. modern FPS games also have boats, vehicles, planes and mechs.

Then again what about sports games, just checked FIFA 2017 uses Frostbite 3.

Is Frostbite the game engine Unity should be competing with?

 

I definitely did not see this kind of universal agreement in previous posts.

In 80/90/00 unity would've been unusable due to insanely (for the epoch) high hardware requirements. Well, maybe it would've been usable in 00s - in their later half. Maybe.

One of the Forstbite titles I played was dragon age inquisition. Wasn't very impressed.
When I was thinking about buying mass effect andromeda, I looked into screens, and my impression is that Frostbite 3 was inferior to previously used unreal engine. Might be wrong about it, plus modeling and animation job in that game didn't help the impression.

In the end I think that competing with this in-house framework would be a bad idea.

--edit--
In the end it is not about engine, but about available people and their skillsets.
So unity should remain unity, without trying to chase every engine on the block.

 

Anyway, why care about "AAA"? If you don't have 80 million budget, you are not AAA, so features an AAA would benefit from won't be very usable for you.

What's the point?

 

No. Most of us argued that the ideal AAA engine would support less platforms. Unity is a wide based generic engine. The in house AAA engine is a narrow specific engine.

Flying a plane and controlling a character aren't hugely different. But you don't see Assasians Creed style parkour in a FPS game. Nor do you see Kerball style docking. Or Pond Wars style 2D wave mechanics. The variety of games that can be produced in Unity is immense, and far more then is required by any AAA studio.

Or do you really mean to tell me you think an AAA engine for COD games should be capable of producing a match 3 game for mobile? That's a stretch, even for you.

 

Frostbit is a graphical engine I think, not a complete engine?

 

Definitely not an engine in the context as unity.

 

The part is so pointlessly hypothetical, it doesn't deserve a response.

The second part is pretty much the same realm but more ingnorant than pointless. No it shouldn't be 'competing' with frostbite, the are different tools for different purposes. It's like asking if Photoshop should be competing with Maya.

 

So is Unity only good for 2D then?

 

Oh my god, stop. That isn't what's being said at all.

 

How did you get that from what he said? He wasn't drawing a direct comparison. It's the techy equivalent of saying you are comparing apples with oranges. You wouldn't then ask if Unity was only good for fruit.

 

Better lights, shaders and rendering. Check the SEGI asset on the store - this is within reach for Unity and should be a priority.

 

Meh, you would be pretty hard pressed to find any engine that's purpose built for anything these days. Think about how EA's Frostbite engine is used in everything from Madden to Mass Effect (just don't sort things alphabetically). At this point, the only studios with a specialized engine have staked their life in that genre for at least the last two console cycles and carried their engine through it all.

It's like two-thirds graphics engine and a third engine framework (plus a bunch of tools), from what I understand. You wouldn't be able to find the one true frostbite engine since every game tailors it to their needs. I don't think it's far off to think of it as a lego engine, in theory at least.

 

No, but it exemplifies the level ignorance I was referring to. Making comparisons and suggestions based on no understanding of the topics at hand is pointless. It would like me suggesting that those little peddles on a piano should also be used on the keyboard so pianos could compete with accordions. I know virtually nothing about pianos or accordions, and never bothered to even try to actually understand how they work, so it is basically nonsense.

 

This, and it seems that most people tend to conflate the engine with the editor.

 

Are you even reading responses you get?

 

From a technical perspective, I guess sure anything is possible with enough time and money.

But from a business standpoint this is, frankly, a ridiculous question. Unity *created* a market segment, and then went on to absolutely dominate it with a install base of, what, 2 million users or something like that. That market is vast and has very specific needs, needs they are 100% focused on serving, and doing it quite successfully. I've been here since 2009 and seen it first hand.

The AAA market is very different. First, you are mostly competing against internal tool sets that are highly customized to studios' game genres and workflows. Not to mention a not invented here mindset. Source: my son works for Blizzard.

And even if Unity could compete there, any resources dedicated to that segment run the risk of making it difficult to continue addressing their core market, which then leaves them open to attack.

I would also ask how many AAA engines have gone to mothballs because they just couldn't continue competing successfully with those internal tool sets and the few premier AAA solutions that do exist.

So, like I said, makes no business sense at all in my opinion.

 

Man, what's so damn great about AAA games anyway?

It's a market cursed with gambler's addiction and hit chasing.

All they do is snowball their wealth into larger and larger projects until one day they make a flop and the company instantly goes up in smoke.

It's down right foolish if you ask me.

 

It'll be interesting how the new rendering pipeline coming (hopefully this year) will change Unity's perception regarding AAA suitability.

 

Your problem wasn't looking at Frostbite, it was looking at Bioware games

They've never been very good at art. Their original ME trilogy on UE3 was nothing special to look at (it got almost no graphical updates between 2 and 3, 2010 and 2012), and even their own in-house engine they used for DA Origins and DA ][ were not visually significant.

If you want a showcase of what Frostbite can do, your best point is games DICE makes, like the new Battlefront 2 reveal trailer.

Consider too that this is the "PS4 experience," not the full PC experience. Frostbite is a fantastic engine, Bioware just isn't good at the art side of things (especially not when they place their newbies in charge of their flagship franchise, a la Mass Effect Andromeda).

Gambler's addiction? You sure you aren't talking about mobile games, or specific types of multiplayer AAA games?

What's so "good" about them (some of them, at least) is that they can offer a grand experience that indies simply can't. Name me an indie that has half the world-building and detail of Skyrim. Name me an indie that has anywhere near the cinematic design and 30-hour story full of voiced dialog and animation and setpieces that Mass Effect 3 does. Name me an indie that has the depth and sheer level of content and detail and writing quality--all at once!--that The Witcher 3 does.

AAA is great because it does it all. Indies are all about satisfying one little niche, while AAAs (and AAs to an extent - see The Technomancer, Nier Automata, etc.) provide a wealth of complementary mechanics and content that all enhance the experience.

It's a mistake to conflate all AAA as one specific exploitative type, or dismiss it entirely.

 

1) It's gambler's addiction because after every iteration of a AAA cycle the studio takes the gained capital and doubles down. They do this again and again and again, but no matter how many times they "win" eventually they WILL lose and the house is going to clean them out.

2) I really was talking about it from a business perspective. But if you insist...

Minecraft.

Everything Tell Tale has ever done.

Okay, that's super easy.
The Witcher 3.

3) Deliberately missing someone's point is not the same thing as winning an argument.

 

Mass Effect 3 and titles like Star Wars: Knight of the Old Republic looked good when they were released.

The issue is that on screenshots and video fragments I saw Andromeda looked like a downgrade compared to Mass Effect 3. Which does not work well to convince me that Frostbite is any better - because with a better engine same artists should be provide better quality visual output.

Cool demo reel somewhere only proves that it is possible to make cool visuals if you throw a lot of money at the project.

Honestly I do not understand interest in Frostbite engine. Most likely the only reason why it is even a thing is EA not wanting to pay royalties/licensing fees, not because their tech is cooler.

I think the primary problem (in threads like this one) is that at least some people assume that engine is the main factor that determines visual quality, and not high budget and effort of many skilled artists.

The way I see it from AAA perspective choice of the engine is largely irrelevant, because even if some feature is missing, it will be possible to license it from someone or even implement it from scratch.Regardless of the engine 6 months in the project will have its custom tools that will be unlike pretty much anything used by other people. Witcher 3 had custom engine, remember? And its first installation used heavily modified engine from Neverwinter 2 of all things.

 

"Indie" stands for "Independent". AAA indicates budget. Titles like Witcher and Skyrim could probably be called an Indie AAA games.

 

IDK, Bethesda now has its own stage at E3. I wouldn't call that "independent" from the industry giants.

 

I doubt you can call UE4 a true "AAA Engine" (whatever that means)... I see Unreal being used more by bigger Indies and "AA Games" (Whatever that means).

Like some japanese fighting games built by a big(ger) studio, with 3D cell shaded graphics using Unreal as their engine. For me a typical "AA Game". Not pushing technical boundaries, still a way bigger project than even a big Indie Studio probably could push out in a reasonable amount of time (unless they go Hello Games and put all their cards in the ProcGen basket).

I am sure somebody will know some, but I cannot remember one of the big AAA games of the last few years using Unreal engines... the Call of Duties, Battlefields and EA/Ubisoft bugfests used Frostbite and similar inhouse engines AFAIK. Then there are some bigger, less technically advanced games like Armoured Warfare, developed by studios without access to these or less mola invested using CryEngine (and most probably paying quite some people to modify those awful tools).

The only AAA games I can think which are... well... almost AAA are Epics own productions, and they are few and far between. And of course Epic HAS to use their own engine, as most of the justification for Paragon and a new Unreal is selling their own engine in the AA, and hopefully AAA space.


What IS true is that Unreal Engine 4 is more focussed on PC games, that at least push SOME of the boundaries of the platform. Which is for me, one of the big shortcomings of Unity. Its a fantastic engine... just not very otimized for PC or Console games. Too many other build targets to cater for.
Unreal Engine 4 does a better job to cater to Indies and middle sized studios that are creating games for PC and Console.
And of course, yes you could use the engine for AAA games. You could also use Unity for AAA games. At the amount of money these studios can invest into tools production and engine optimization, any engine is fair game really.

Question is just if its viable to do so without access to the source. One big plus for the open source of UE4... but inhouse engines also come with full source access, so hardly making Unreal a competition for Frostbite.

As far as I got it the problem wasn't the engine... maybe in part the money (everyone today tries to shortchange their projects and companies, far beyond just the games sector).

The most important factor though was the new team most probably. Seems like most of the old left after the fallout to ME3, and the Ending-gate. That couldn't have gone over well unless management made some brave decisions, invested SOMETHING ("Oh noes, ma moola wasted on lowly game devs") in Team building and then allowed the new Team to go in a wildly new direction with a new Mass Effect... which in turn might have angered some of the old fans which wanted more of the same.

Given that some of the ME3 BS seemed to have been caused by upper management in Bioware (or EA, maybe the Bioware Management is also being forced into it by higher powers), I guess they weren't willing to do that. They just treated their Team as an exposable resource that should push out the next conveyer belt built cookie cutter product to continue raking in the cash from the ME fans.

I really, really doubt that Andromeda is the fault of the engine. The ME3 Team most probably would have built a better, bigger ME without the bugs and bullshit, if they were given enough time and money and not be blocked by corporate bullshittery.
The newbie Team brought in most probably never had the time and money, given Bioware had to build up a team from the ground and most money might have been spent on that, as well as time. And you can bet the new Team was expieriencing more corporate bullshittery than the old, given the old left because it was fed up with it, and big corporations seldom do the sane thing to subject its employees to LESS bullshittery if they get fed up with it. They set up a new set of bullshittery just to make sure their employees suck up the existing bullshittery.


Oh, and lets not forget EA now has its greedy mittens in the cookie jar that is Bioware. EA manages to taint everything it touches. Sure, SOME games published by EA manage to be good. Most of them, as soon as they become cash cows, suffern from EAs bullshit business decisions. Maybe EA forced Biowares hands to work faster and cheaper?

 

If you're talking about Guilty Gear Xrd Sign, this one is made using Unreal 3 and not Unreal 4.

Gears of War 4, Batman: Return to Arkham Asylum, Dragon Quest XI...
Basically:
https://en.wikipedia.org/wiki/List_of_Unreal_Engine_games#Unreal_Engine_4

Probably. But the quality drop was significant enough to deter me from pruchasing it.

Still haven't watched the modified ending after all that time...

If I remember correctly, there are now multiple bioware studios, and andromeda supposedly was made by a newer studio, which did not work on original mass effect series. See:
https://en.wikipedia.org/wiki/BioWare#Structure

 

I wasn't around for KotOR's release, but I assure you ME3 was bog-standard for the time. There were no significant visual upgrades from ME2.

"The same artists should be able to provide better quality visual output."

I agree with you there. The problem is, it wasn't the same artists. It was the newbie EA studio that was renamed as a Bioware studio. All the people who worked on the previous games are off working on a mysterious new IP. Andromeda was developed by a largely untested group. That's why the game has issues.

http://www.neogaf.com/forum/showthread.php?t=1353243&page=1

This the same EA that let Bioware delay DA Inquisition by a year, and let the team spend 5 years on Andromeda?

Blaming EA for any of Bioware's failings is ridiculous.

Edit: And as for EA tainting games, by and large that's nonsense. Titanfall 2 and Battlefield 1, two EA first person shooters that came out last year are largely considered excellent games. Battlefront is considered a solid casual experience, and DICE looks to be incorporating almost every complaint in the development of Battlefront 2 (single player campaign, offline play against bots, space battles, etc.). Mirror's Edge, which came out last year, was a niche first person parkour game handled quite well (not a fantastic game, but with no egregious sins). I can't speak to their sports games.

The irrational gamer hatred against EA is largely just that--irrational.

 

Certainly not. Not_Sure already addressed Bethesda. The Witcher 3 had a budget of 81 million dollars, including marketing. Consider as well the fact that it was made in Poland, with a significantly lower wage level than places like America.

 

We should probably count everything Blizzard does too.

Plus many of the tiny studios we consider to be 'indie' aren't really independent. Many of them have relationships with publishers that fund the games, and take on much of the marketing side of the business.



Okay, so we've clearly established that the literal meaning of the word indie has nothing to do with the meaning we assign it in day to day conversation. Can we all move on now?

 

Sorry, I just noticed this.

1. That isn't true at all. You could maybe say Rockstar and Bethesda do it and are successful at it. You could maybe say Ubisoft tried to do it with AC and it failed. Outside of that, it really isn't happening.

Minecraft? I said world-building and detail. Does Minecraft have hundreds of years of in-world history, carefully detailed through dozens of in-universe stories and evident in the interaction of half a dozen different races? There's not even a comparison there.

Tell-tale? A complete season is at best half that (in hours played), there's almost no interaction (certainly no setpieces), and the games are running on an awful, unoptimized engine because the guy who started the company insists on using the one he made. There's barely a comparison there.

Are you trying to say The Witcher 3 was an indie game? Anything but, see above. I'd be dubious of calling TW2 an indie game, considering that CDP was around for a while before CDPR.

It seemed to me that your point was "there's no value to AAA." I don't agree at all, which is what my examples were about. Was your point something else? If it was "AAAs grow bigger until they explode" that's not in any way true at all, and I'd love to see some proof of that claim (outside of AC which I mentioned).

 

While bethesda case is arguable - because they're apparently split into two companies, meaning software is dependent on publisher, Witcher's dev, last time I checked is still Independent. They're developer and the publisher.

Spoiler

^_^

In all honesty, though.. the talk about "Indies" vs "AAA" starts resembling the mythical idea of "Average User". Everybody has a different idea of what "Average user" is. Might be a good idea to determine studio size, project size, budget size. At least this kind of info/differentiation will be more useful. IMO.

 

If we're going to start defining the term "independent," obviously there's no consensus, but you could probably say it means "independent from a publisher." Which CDPR is not--they ARE a publisher (or CDP is).

If a company being a publisher doesn't disqualify them from being indie, then heck, Ghost Recon Wildlands is indie.

Furthermore, if we want to talk things like team size--The Witcher 3 had a team size of 240 in house with over 1000 more around the world.

For reference, DA Inquisition was less than 200.

 

Which throws out me. Because I publish my own games. I'm a publisher and a developer, and I'm just one person

A better way to get meaningful results would be to look at specific studio sizes. Which is what the question of indie vs AAA is normally aimed at anyway.

 

I wonder if "distributor" might be a better term. Of course in this digital age that means less and less...

I think neginfinity is right though, with project and studio size, and budget of course.

 

This is not a useful definition of indie, IMO. For example, spiderwebgames self-publish games on their site. They're as small as it gets. Dwarf Fortress self-publish. It is two guys. Minecraft used to be self-published as well.

I think it would be more useful to talk about budgets and team sizes instead of vaguely defined "Indie" and 'AAA".

 

Agreed. Still disqualifies TW3 though.

 

Is it just that Frostbite is pushing at the limitations, the forefront of game engine design, look at the number of technical papers they produce http://www.frostbite.com/topics/publications/

Tiled Light Tree (faster lower gpu lighting) > http://www.frostbite.com/2017/02/tiled-light-trees/
Photogrammetry and Star Wars Battlefront > http://www.frostbite.com/2016/03/photogrammetry-and-star-wars-battlefront/
4K Checkerboard in Battlefield 1 and Mass Effect Andromeda > http://www.frostbite.com/2017/03/4k-checkerboard-in-battlefield-1-and-mass-effect-andromeda/

I'm not aware of any publications on the forefront of technology that Unity have produced.

Is Unity pushing the technology with high end R&D or being pulled and is that the difference between a AAA engine like Frostbite and Unity?

On the flip side this could be a good business strategy as Unity can copy ideas that work and bring them into the engine when mobile technology catches up with PC/Console. When the mobile tech catches up with previous aging current generation consoles Unity will be AAA right?

This would make Unity about 6-8 years behind AAA engines does that sound about right?

And in 2008 Resistance 2 was a top title...

What do you think?

 

For an engine there are 3 things that matter.

Capability: Graphics, Physics, Streaming, Platforms, Features etc.
Scalability: Workflow, Version control compatibility (branching, merging, handling team size)
Stability( In a business sense ): Will it get bought, changing license agreements? Will features I need change during development? What are the business risks?

I've listed these in order of how I see indies and most developers I know think about engines, when in reality the reverse order is more like how we should think about things and is probably how big studios and publishers assess if they should use an engine or build their own.

I feel Unity has Stability covered. They don't look like they're going to sell and for as long as I've used Unity they've been pretty good at respecting agreements and not changing things on a dime.

Scalability is probably the place Unity is most lacking. As soon as you scale team size Unity starts to fall over in terms of workflow. There are workarounds like making every single thing a prefab etc... but that only mitigates the problem.

As for capability, yeah Unity has some short comings here and there, but it's also very good at what it does well. It's very flexible and for that reason I give it a good score here. For the most part you can write a system to accomplish things Unity can't do out of the box and the cost of building that system in Unity will be far cheaper than writing your own of licensing something else.

 

Well, as a personal opinion (currently working on a small open-world game) I could say that Unity is not getting close to anything AAA in the near future. They, as you already have mentioned, have no publications, the source code is closed. Except that some huge lacks are:
- A decent vegetation system
- A decent terrain system

And those aren't the worst, the occlusion works kinda bad, enlighten doesn't bake correctly for LODs, your are forced to use only 1 LOD for buildings (at least at the moment, if you require light-baking on them), the performance is kinda bad you always have to help the engine out with custom code, you can't edit shaders easily and the list can go on...

It is kinda ok for small games, and mobile games, you can also prototype quite fast but if you want to do something at a decent level it is a lot of work... And it is absolutely impossible to fix some issues, since the source code is closed. Absolutely impossible.

You can have games like 'Ethan carter', 'Ark' or 'Kingdom Come' (which are not fully AAA) in Unreal/Cry, but it is absolutely impossible to have them in Unity. Not even with a huge team.

Maybe if you buy the source code and re-write it you might have something )

That's why they show off on having 30%+ of the free mobile market. They don't have a lot to compete on the Desktop/Console...