Epic: 1/3 as many new gen games coming, each at 3X development cost

[Rangers]

http://www.edge-online.com/features...real-engine-looks-to-define-a-new-generation/

At E3 and later this year, however, we’ll see plenty more games built on Epic’s Unreal Engine 4, says Sweeney. “There’s a lot [of games] on the way and a large number of them haven’t been announced,” he tells us. “You’ll see lots of triple-A stuff coming out over time. The industry’s changing – this generation it seems like there are about a third of the number of triple-A titles in development across the industry as there was last time around – and each one seems to have about three times the budget of the previous generation. I think we’re heading towards a future where triple-A is the minority.”

Explains the fairly bare release schedule...

Not sure what I think, as usual gaffers picked it apart with a few possibly relevant points such as "how would Epic know what the rest of developers who aren't using their engine are doing" and "isn't Epic the one touting how low UE can make dev costs and didn't they say dev cost needn't increase this gen? Hypocrites!". Others noted while the cost of entry of development is lower than ever for indies, with often free dev kits and such nowdays, huge pubs still ramp up the AAA costs with content chasing the huge franchises.

Myself, I sure hope it's not possible for AAA to go away, as it's what I game for. Personally, I dont think less games it's the horror it's made out to be, I think there's definitely a point where there is too many games being released and we need a balance. However, of course I dont want the number to go too low!

But I definitely have noticed fewer releases. For example, Watch_Dogs being practically the only big budget title for a few months here.

 

[Xbat]

Sony just had a big aaa year last year its unrealistic to expect them to have one every year. I also call bullshit on the trippeling of budgets, I dont believe games that cost 50 million to make will cost a 150 million now.

 

[djskribbles]

I thought development costs would be the same, if not lower this gen. I think the PS4/XB1 have had pretty solid first years software wise and I don't see it slowing down.

 

[Shifty Geezer]

AAA's supposed to be the minority. Like movies, there are a few big budget titles and the rest are lighter entertainment.

 

[I.S.T.]

I thought development costs would be the same, if not lower this gen. I think the PS4/XB1 have had pretty solid first years software wise and I don't see it slowing down.

Art assets are getting more and more expensive...

 

[Wynix]

I don't believe either of these remarks, also I've never cared much for Epic.

 

[BoardBonobo]

Tim Sweeney's trying to scare starry eyed devs into buying into the UE4 subscription. Because Epic is your family now...

 

[ArcticCircle]

I dont believe games that cost 50 million to make will cost a 150 million now.

GTA V and Star Wars Old Republic are in league of their own, just like Final Fantasy 7 and Segas sailor simulator was in 90's. I'd guess Tim means 20 million games will cost 50-60 million, that may or may not include advertising budget.

IIRC average top tier game (that we call AAA now) used cost something like this in past generations.
Gen 4: 0,25m
Gen 5: 1m
Gen 6: 5m
Gen 7: 20m

Modern indie games are in that gen 4 range and interestingly enough, often look and play like games from that era. Some kickstarter games like Pillars of Eternity is in Gen 6 range and it's design is from early 2000's. It's like Infinity Engine games never stopped and budgets got frozen to that 5 million level. Sure, models are more vivid, pre-rendered backgrounds more detailed, lighting way above anything possible in early 2000 games and resolution about 7 times higher...

Actually, that's how I always imagined games would move forward and not how they do now. Not every game needs to be cinematic action adventure with ever increasing budget (and I want to say that I love cinematic action adventure games, especially open world games). Here is 5 million dollars - create best game you can with that budget. Don't create it if you don't think it can sell over 150k full priced units. Do a million dollar game and sell it for 30 dollars. Maybe you'll hit over 50k sales and make profit. (I used Mike Pachters numbers how much publisher / developer would get)

AAA will be minority in this generation. Indies will be plentiful. There will be some success and maybe more failure in 1-5 million range. I don't expect many AA or A games to be successful (at 5-20 million range), unless they change design from previous generation.

 

[Xbat]

GTA V and Star Wars Old Republic are in league of their own, just like Final Fantasy 7 and Segas sailor simulator was in 90's. I'd guess Tim means 20 million games will cost 50-60 million, that may or may not include advertising budget.

IIRC average top tier game (that we call AAA now) used cost something like this in past generations.
Gen 4: 0,25m
Gen 5: 1m
Gen 6: 5m
Gen 7: 20m

Modern indie games are in that gen 4 range and interestingly enough, often look and play like games from that era. Some kickstarter games like Pillars of Eternity is in Gen 6 range and it's design is from early 2000's. It's like Infinity Engine games never stopped and budgets got frozen to that 5 million level. Sure, models are more vivid, pre-rendered backgrounds more detailed, lighting way above anything possible in early 2000 games and resolution about 7 times higher...

Actually, that's how I always imagined games would move forward and not how they do now. Not every game needs to be cinematic action adventure with ever increasing budget (and I want to say that I love cinematic action adventure games, especially open world games). Here is 5 million dollars - create best game you can with that budget. Don't create it if you don't think it can sell over 150k full priced units. Do a million dollar game and sell it for 30 dollars. Maybe you'll hit over 50k sales and make profit. (I used Mike Pachters numbers how much publisher / developer would get)

AAA will be minority in this generation. Indies will be plentiful. There will be some success and maybe more failure in 1-5 million range. I don't expect many AA or A games to be successful (at 5-20 million range), unless they change design from previous generation.

The AAA games that came out at the end of this last generation I think would of cost 30 million plus but even if they cost 20 million why would those same games need a three times bigger dev team. IMO most dev teams only would have to expand there art department. From my undertstanding the art created for games is already of high quality and then downgraded to work on the consoles.

 

[AzBat]

I guess it's a good thing Unreal Engine 4 is now available for id@Xbox game development...

http://news.xbox.com/2014/04/games-unreal-engine-id-xbox

Tommy McClain

 

[Sigfried1977]

I still think that the whole death of the mid-tier game thing is a stupid, self-fulfilling prophecy. Of course your slightly smaller game isn't going to do well if you simply release it without telling anyone while blowing your entire marketing budget on that singular mega-budgeted, one-size-fits-all kinda game in your portfolio. The Souls games are perfect proof of how profitable catering to a niche can be if you're being smart about it.

 

[Scott_Arm]

Even if the cost to produce assets stayed constant, the games that are coming out are bigger and more detailed than ever. More and more games are open-world, or have larger play areas. And even the areas that aren't playable are considerably more detailed. Look the new Killzone compared to previous gen versions. I imagine it's a fairly linear shooter, but you get semi-sandbox environments with incredibly detailed backdrops. There's just more work for artists, even if the games aren't necessarily longer to play through. It seems like just about every title Ubisoft is pushing is an open world game now.

 

[Laa-Yosh]

The up side of an open world game could be that any additional content can be produced with a minimal amount of assets, you can just set an infinite number of missions in the already existing areas. Maybe that's their approach here.

As for Epic's predictions - while they're pretty good on the technology and tools front, they've been kinda wrong with most of them.

 

[Rangers]

Tim Sweeney's trying to scare starry eyed devs into buying into the UE4 subscription. Because Epic is your family now...

He's also putting his money where his mouth is, though. Fortnight is designed to be more like a Valve game, open ended, MMO-like, probably full of microtransactions, not based on graphics. It looks like Minecraft crossed with TF2.

They're doing one high end graphics pushing game, which is unannounced (the one that is possibly Samaritan). But they've said their goal is to be more like Valve or the creator of League of Legends going forward.

 

[nightshade]

Art assets are getting more and more expensive...

Yea but didn't they already used to create high res art for PS360 and then downgrade them for ingame use? Like they'd create a high res mesh for the purpose of creating normal maps and such and then use them in the downgraded version for the game.

 

[Laa-Yosh]

High res meshes for PS4/X1 games need to be considerably more detailed and complex. More accessories, more geometry resolution for the sculpt, more realistic quality and so on. It's not that uncommon to have 50+ million poly source assets for a single character now. That's about twice the detail level compared to Gears games.

Also, while textures may have been authored at 2x the size, that workflow means they'll have to double-quadruple the res because of the higher ingame resolution, too.

Assets are becoming more expensive, no other way around it.

 

[Cyan]

I still think that the whole death of the mid-tier game thing is a stupid, self-fulfilling prophecy. Of course your slightly smaller game isn't going to do well if you simply release it without telling anyone while blowing your entire marketing budget on that singular mega-budgeted, one-size-fits-all kinda game in your portfolio. The Souls games are perfect proof of how profitable catering to a niche can be if you're being smart about it.

Thing is that it is usually scary for other developers without resorting to well-known formulas that work, which is sad at the same time.

Sure there are worse things, as you can't buy happiness, but increasing the price of games wouldn't be a solution either.

Say you want the top artists and hire Laa Yosh... What kind of AAA game would you create?

 

[AlNom]

It's not that uncommon to have 50+ million poly source assets for a single character now. That's about twice the detail level compared to Gears games.

What kind of a computer do you need to even load that up?

 

[Laa-Yosh]

Funnily enough, in zbrush it can be done on relatively modest systems. My workstation is somehting like a 3ghz core i7 I think, 16gigs of ram (even though zbrush is still 32 bit...) and it can load the file all right.

I wouldn't like to work with it though, must be a stability and file operation nightmare. Maybe it was put together into a single file at the developer just for us. Although it's a pretty common technique used at multiple studios - just add a new subtool for every small object. We prefer to apply the divide and conquer principle...

Anyway, the point is that assets are getting very complex, close to movie VFX levels of detail and workflows. More on this later...