Yeah seems like they dumped the engine on studios, who were then required to add the tools/features they needed all while trying to make their game at the same time. Like laying down the tracks in front of a moving train...
-
Ever wanted an RSS feed of all your favorite gaming news sites? Go check out our new Gaming Headlines feed! Read more about it here.
Amy Hennig on Frostbite challenges in Visceral's Star Wars production
- Thread starter EatChildren
- Start date
I read in there that the expected budget could have ballooned to 150 million?! Ok, so if it really was going to be that much, i would've canceled it too, even tho i don't think that would've actually happened. some of the biggest and best reviewed games didnt cost that much - even nearly that much.
I am just thinking here a loud, but what if the mistake is not that frostbite was forced, but that teams were not given enough time and programmers to make the systems they needed for their project milestones?
If FB did not have a feature or something, you would hope the team there should have been given the time and support to make it. Also you would hope that rnd dev peoriod would be graced in the project's total dev time. It sounds like maybe they did not always and it ended up eating into their other game dev time?
The technical issues ME: Andromeda suffered probably account for a significant portion of the 18 point MC difference between it and Dragon Age: Inquisition. A more polished product likely would've scored in the low 80's, sold better (not that it sold poorly) and probably left Bioware with its 3rd studio still.
After the third game? You don't blame the team, you start looking at other sources. Engine can be a source. There's a pattern here that third person games are wonky on Frostbite. The fact that Amy is even speaking of it very politically shows she doesn't want to piss off EA. Sadly, EA loves Frostbite because it can make great cinematics/cutscenes/trailers and not because of its ease of use or its diverse range of features. I don't doubt that Frostbite is a bitch, even their core team struggles with it.
"Excuse for failure" is so easy to say and so easy to dismiss the years put on a product.
because microtransactions suck or games should really cost over 100 dollars to be profitable with projected sell through? Are microtransactions greedy or simply a way to make ends meet.
https://www.raphkoster.com/2018/01/17/the-cost-of-games/
This was a fun look into the more real cost of games.
Always struck me as weird how they went for Frostbite when Renderware was already a proven engine for 3rd parties. I'd love to see where the engine would be now had EA not shitcanned it as it was great at the time.
I'd love to see EA remaster Hot Pursuit as it was such a great game.
I'd love to see EA remaster Hot Pursuit as it was such a great game.
Right but if you're creating say an Uncharted style 3rd person game, the tools that might be created by say BioWare from Dragons Age, those might not be the things this dev needs
Just cause Dragons Age is 3rd Person doesn't mean Uncharted will be easily replicated. Hell DA can be First Person or Isometric also if you go further into it, but games can differ, but tools can be really spread out
Just look at Frostbite, the hilarity of the engine when you try to do 3rd person or the glitches or bugs that arise in say Battlefield
Legs inverting, Necks getting long, the engine wasn't built to be used in that specific way
You have to have a dedicated team who job is to keep working on Frostbite
Create tools or assist devs in house about what they might need or help with
Well I stand corrected, then EA is doing a piss poor job helping their devs out then
I mean Sony has that Santa Monica dev tool squad that's with Sony Santa Monica Studio which go help all first party devs
Last edited:
Supposedly coming along fine and plays great. EA are happy and it launches this fall.
We should see gameplay in April.
Bioware has made the most progress on turning Frostbite into a engine for third person. But not without plenty of sacrifices on that journey.
EA, you suck.
EA, you suck.
Sounds like a lot of their recent problems with frostbite helped to hurt some bit EA IPs, like Mass Effect and probably several unannounced ones like this
I can't imagine how bad Frostbite Skate would have been
I can't imagine how bad Frostbite Skate would have been
Yep.
Guerilla games have ?270? staff. They wrote a new FPS engine for a game in 2013. They expanded that engine for use in an open-world RPG released in 2017. Both games were technically impressive.
Bioware have ?800? staff, working on 3+ projects, mostly using the same engine. All of their recent titles have had technical 'issues' and feel half-finished.
Even if the Bioware number is inflated, that's still ridiculous. How many people do they have working on their "game engine" rather than modelling?
From every report we have seen it seems making Frostbite the engine to use in all teams/games has costed EA much more than if they had used a middleware engine or if they let studios use whatever they wanted.
It wouldn't save every game but I bet the time used to implement that basic stuff could've been better spent...
It wouldn't save every game but I bet the time used to implement that basic stuff could've been better spent...
Well going by what we know on how EA is treating Anthem
They want that title out the door, get their fiscal year numbers bloated up a bit
Apex really helped them out on that front
You can see why Anthem still needs time even after 6+ years, but due to the way EA runs it's business with trying to get people to buy into their subscription model with EA Premier or EA Access, they just want to make shareholders always see gains
I mean that's great for a business short term, long term, they can either turn it up like how Ubisoft is or ho opposite on how Activision is flustering a bit by double downing on legacy titles
It was obvious to everyone that didn't have some bizarre motivation to stan for EA that BioWare didn't choose to use Frostbite for Anthem.
When will people learn?
It was the same thing with the Rockstar devs talking about crunch.
When will people learn?
It was the same thing with the Rockstar devs talking about crunch.
Yeah that really sucks, makes me glad i won't be buying EA games.
So they didn't take up into account adjusting to the engine?
Not to point finger at anyone but isn't that the CTO job to push for proper solutions both in engine choice and/or engine management/evolutions?
Apparently that job is being done by Ken Moss since 2014, I'm sure he had to comply with things beyond his power at some points but surely he could/should have seen things coming.
If the same team without this extra time and programmers which they were not given could have done better and had a more efficient workflow with a third party engine they were familiar with, then the mistake is still that Frostbite was forced in these circumstances. In the grand scheme of work reality, it doesn't matter how great an engine's potential is if there isn't the capability to properly support it. Given how many teams there are which struggled with this, is the benefit of Frostbite really worth adding all that extra staff and resources to support it for every single team even if that was done? Or could they have achieved similar or better results with a third party engine that can already do what they need?
It's like how KH3 ditched Luminous for UE4 early on once they realized it was going to cost too much extra time and effort to support that engine for their development.
It's entirely possible that Frostbite's documentation or feature set is lacking compared to Unreal Engine 4. As a developer I know how frustrating it can be to have to use someone else's hard-to-unravel codebase.
But the certainty with which people here are nailing Frostbite to the cross as the single blame for everything wrong within EA...wow.
But the certainty with which people here are nailing Frostbite to the cross as the single blame for everything wrong within EA...wow.
I get why big publishers don't want to give Unreal or Unity a cut but at some point it HAS to be a smaller cost than the lost dev time accrued from forcing people to use these in house engines no matter what. Especially when it seems to be affecting the final quality of games and ends up killing franchises and contributing to games getting cancelled. I'm glad Square Enix seemingly learned this lesson.
With so many recurring complaints about Frostbite from developers working with it what has the team that is solely there for developing the engine even been doing all those years?
Some random guy (i like to think he knew something about all this) was comparing game engines. Couldn't find that reddit post again so this is from memory.
Frostbite Engine - Like a fast F1 car. Does amazing stuff for what is designed to do, but you try to add anything new and things fall apart.
Unreal Engine - Like a Semi-Truck. Reasonably fast if you wants to, and can handle a ton of shit you add.
Thought it was interesting. By that logic extension I think Creation Engine would be like a Tractor. You can hitch almost anything on it but damn that thing will be slow.
And Unity could be a mini van then.
Frostbite Engine - Like a fast F1 car. Does amazing stuff for what is designed to do, but you try to add anything new and things fall apart.
Unreal Engine - Like a Semi-Truck. Reasonably fast if you wants to, and can handle a ton of shit you add.
Thought it was interesting. By that logic extension I think Creation Engine would be like a Tractor. You can hitch almost anything on it but damn that thing will be slow.
And Unity could be a mini van then.
Honestly it's not really surprising given both Dragon Age Inquisition & Andromeda issues.
EA has to improve Frostbite if it wants to be adopted by various teams with different games.
Regarding the business plan... There really isn't anything special to say other than yeah EA is transitioning to GAAS
EA has to improve Frostbite if it wants to be adopted by various teams with different games.
Regarding the business plan... There really isn't anything special to say other than yeah EA is transitioning to GAAS
Man, EA really fucked up there.
Frostbite not having the necessary tools and that taking some time to build isn't an issue by itself, it becomes an issue because EA made it an issue.
If buliding the tools for the engine EA wants them to use will take more time, then the sensible choices are 1) give them the time, 2) tell them to use something else instead. EA wanted them to use Frostbite but also didn't give them the time. What the hell did they expect other than the project no developing as expected?
If this really is the biggest issue in Visceral's Star Wars game, the cancelation is a very weird decision from EA. That time spent improving the engine isn't time wasted in Star Wars development, it's an investment to any future third person titles EA would end up producing.
Frostbite not having the necessary tools and that taking some time to build isn't an issue by itself, it becomes an issue because EA made it an issue.
If buliding the tools for the engine EA wants them to use will take more time, then the sensible choices are 1) give them the time, 2) tell them to use something else instead. EA wanted them to use Frostbite but also didn't give them the time. What the hell did they expect other than the project no developing as expected?
If this really is the biggest issue in Visceral's Star Wars game, the cancelation is a very weird decision from EA. That time spent improving the engine isn't time wasted in Star Wars development, it's an investment to any future third person titles EA would end up producing.
Come to think of it, I wonder if all the troubles that BF5 has also come from the shortcomings of the engine.
This kind of back of the envelope math cant really tell you anything productive about something as big and complex as an entire game engine used by hundreds or thousands of developers.
Projects become harder to work on as they scale, so it's very reasonable to imagine that when you take in coordinating with the frostbite engine team, internal ea regulations, and the size and complexity of the engine, training the team on changes, etc, that it might be vastly harder to update than an equivalent* update to a single team in house engine like (what's now called) decima. On the other hand, it might be much easier! Without knowing how the projects and teams were organized it's impossible to speculate.
*(there's no reason to actually assume they are whatsoever equivalent. there are a lot of ways to program an engine, and a lot of ways to run a team, and speed of doing one part of development vs another will vary profoundly from project to project)
Then how else do you explain that everything that isn't the actual gameplay is a mess?
"Opted" being the wrong word here.
From Amy Hennig's quote in the OP:
"because there was the internal initiative to make sure that everybody was on the same technology"
They're not, but artist friendly tools that allow cover zones to be placed dynamically and linked to geometry are. And thats the kind of thing that creates mountains and mountains of tedious work whenever small changes are made. when actual developers talk about implementing cover systems, you know theres a lot more to it than an animation?
Uncharted was well known for having the enemies aim kind of bad, but when you get in cover it focuses on a tighter ring around the player to make it feel more dramatic. there's a shitload going on under the hood in modern games. you cant code behaviors like that on a case by case basis - you need to give designers user friendly controls for it all.
I cant find the video right now, but theres a clip from horizon zero dawns development where they go over the engine feature which dynamically turns objects into walkable objects and determines how aloy behaves around them. the time that an engine feature like that saves pays off in manpower elsewhere.
That's pretty much equivalent to saying that, keeping staffing and development time equal, EA should not have forced Frostbite on all its developers.
Overall, these tidbits of information point to some really incompetent management that has been hurting many projects, employees, lost a lot of money for the company, and honestly it has irremediable damaged its reputation as well. EA has just become synonymous with wrecked studios, half-baked games, shady monetization, you name it.
DICE was asked to deliver a new game in a shorter window than Treyarch with BLOPS4. They also released and supported SWBF2 and from the sounds of it helped the rest of the company out with the Frostbite engine. The lack of modes and design decisions do not scream a bad engine to me. This same engine delivered a well reviewed and relatively well playing game in BF1.
I think that Henning's problem is that Visceral already had their own engine, one that they known very well, that was designed for third person linear action-adventure games, but EA forced them to use someone else engine, possibly not very well documented (speculation on my part), that was designed for large scale multiplayer FPS.
I imagine it was kinda like if you was using Windows entire life, but suddenly somebody forced you to work on Linux. Not that Linux is worse, but it takes time and frustration to learn how to use it efficiently if you are already used to something else.
Last edited:
There's a huge difference between
"gamers don't really understand how engines work, and the final result you play doesn't show near as much of the engine's features as the average gamer thinks it does"
and
"Quality of tools has an effect on productivity of development"
it's not at all contradictory to believe both.