• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

Epic knows PS4/NEXTXBOX specs - [Giving recommendations w/ commercial mindedness]

RooMHM

Member
I agree with this post entirely.

Graphics are not everything and some people need to learn that.
Then why did Sony and Microsoft focus exclusively on graphics for their current gen consoles? It's cool to say things like a word of mind but believing it is true and should be the norm is something else. Nothing was done to improve gameplay and other elements (maybe online? and even that) before Nintendo had released the wii for 4 years.
 
I don't care about fucking graphics. I just want stable frame rate and less loading times.

call of duty 4 modern warfare graphics is enough for me to feel content.
 

MrBig

Member
I don't care about fucking graphics. I just want stable frame rate and less loading times.

call of duty 4 modern warfare graphics is enough for me to feel content.

This is what most people are missing. All the extra resources doesn't mean you have to make the graphics on the cutting edge of technical developement, you can put all those resources to use improving IQ, framerate, and draw distance/count.
 

Clear

CliffyB's Cock Holster
BigTnaples said:
Most of the benefits of next gen wont cause devs costs to go up.

Sadly, completely wrong.

The more complex and feature-rich your program, the greater the chance some random combination of events/operations will cause it to malfunction. Validation by the team and test time within QA will inevitably increase.

Also, however much V/RAM a system has, the impetus will always be to fill it in order to provide the maximum return - developers will always be bumping their heads against the ceiling. The only difference a higher ceiling makes is that it takes more stuff to reach it (more complexity, more time, more cost) and there's further to fall when something goes wrong.

There are no short-cuts. A 1000 piece jigsaw is always going to take longer to complete than a 200 piece one if the picture is basically the same.

Time/space/feature budgets will always need to be observed and the higher the level of quality demanded as standard the more money it will cost to deliver that. You want movie-quality imagery, be prepared to pay movie-quality rates for your staff and equipment.

And ultimately that expense is always going to trickle down to the consumer.
 
Sadly, completely wrong.

The more complex and feature-rich your program, the greater the chance some random combination of events/operations will cause it to malfunction. Validation by the team and test time within QA will inevitably increase.


Also, however much V/RAM a system has, the impetus will always be to fill it in order to provide the maximum return - developers will always be bumping their heads against the ceiling. The only difference a higher ceiling makes is that it takes more stuff to reach it (more complexity, more time, more cost) and there's further to fall when something goes wrong.

There are no short-cuts. A 1000 piece jigsaw is always going to take longer to complete than a 200 piece one if the picture is basically the same.

Time/space/feature budgets will always need to be observed and the higher the level of quality demanded as standard the more money it will cost to deliver that. You want movie-quality imagery, be prepared to pay movie-quality rates for your staff and equipment.

And ultimately that expense is always going to trickle down to the consumer.
Thats not really how it works. Games(or programs) aren't really getting any more complex, the assets are. An executable for an next-gen game will be identical to one this year.

Whats getting more complex are assets and shaders. If Epic is anything to base off of, the next generation will be all about effects(shaders) and they generally require minimal troubleshooting. BigTnaples is correct in that most of these newer effects won't do much to raise budgets.

Assets are really gonna raise budgets.
 
Thats not really how it works. Games(or programs) aren't really getting any more complex, the assets are. An executable for an next-gen game will be identical to one this year.

Whats getting more complex are assets and shaders. If Epic is anything to base off of, the next generation will be all about effects(shaders) and they generally require minimal troubleshooting. BigTnaples is correct in that most of these newer effects won't do much to raise budgets.

Assets are really gonna raise budgets.

You've never dealt with parallel programming. Scaling across many processors, dead lock issues, etc are not exactly easy to program even for experienced developers.
 

Clear

CliffyB's Cock Holster
iamshadowlark said:
Thats not really how it works. Games(or programs) aren't really getting any more complex, the assets are. An executable for an next-gen game will be identical to one this year.

Assets aren't always passive, they have a direct influence on the complexity of the task at hand. If you have double the number of verts in collision mesh you have an increased number of possible points where your AI/NPC entities can get snagged. This needs validation within an implement->test->evaluation cycle.

Similarly, you double the number of entities operating within your new more complex mesh (because you have the CPU/GPU resources to handle them now) you have an exponential number of corner cases to deal with where object<->object collision resolution may or may not square with your new and more intricate environment mesh without ugly clipping or trapping.

The more detail you have, the closer you need to look in order to ensure that everything actually looks correct. There's more to it than simply upping the resolution/poly-counts.
 
You calling me a liar?

Just the shipment numbers, man. My PS360 number here is only updated through the end of last year, but their combined shipments at that time are about where PS2 shipments were at the end of 2007.
PS360_WW
 
Clear said:
Assets aren't always passive, they have a direct influence on the complexity of the task at hand. If you have double the number of verts in collision mesh you have an increased number of possible points where your AI/NPC entities can get snagged. This needs validation within an implement->test->evaluation cycle.

Thats a good example but that rarely happens these days. Well not to a point where it would break a scene. Simply increasing the vertices will have minimal impacts on QA outside of the norm.

Similarly, you double the number of entities operating within your new more complex mesh (because you have the CPU/GPU resources to handle them now) you have an exponential number of corner cases to deal with where object<->object collision resolution may or may not square with your new and more intricate environment mesh without ugly clipping or trapping.

Well there's prototyping at almost every level of game development. An increase in complexity doesn't automatically increase time spent debugging.
 

jman2050

Member
This is what most people are missing. All the extra resources doesn't mean you have to make the graphics on the cutting edge of technical developement, you can put all those resources to use improving IQ, framerate, and draw distance/count.

But they won't.
 

i-Lo

Member
This is what most people are missing. All the extra resources doesn't mean you have to make the graphics on the cutting edge of technical developement, you can put all those resources to use improving IQ, framerate, and draw distance/count.

Very true.

But they won't.

Actually, I think they will after the initial outbreak of utilizing power for better graphics only (because it's the first thing we see. Always). We have seen it happen this gen (better animation, AI, collision physics etc with new iterations of games in general) and if UE4's tools indeed live up to the promise of saving dev time and costs then I don't see why not esp. since new physics engine for cloth and collision are built into it.
 
I'm most looking forward to Epic's claim of UE4 getting rid of that "same engine" look. The read was brief but made me excited. But at the sametime I doubt you can get rid of something like that. Part of the reason many UE3 games had the same look was the fact that many developers used effects that came bundled with the SDK (Explosion, fire effects, texture usage etc). That and prior to Beast and Lightmass they all baked their lighting the exact same crude ugly as shit way and it stood out.

They don't have to worry about the lighting part any longer.
 
Epic... You have my sword. I want my consoles to be beasts. I don't want the Nintendo way. I'm willing to spend $500! Maybe even more if you really push the boundaries.
 
More powerful HW can also be easier to develop for, if better tools are provided.

then it's not easier to develop for, it's more expensive because developers either need to build the tools themselves or they need to buy middleware.

The more powerful the hardware, the more developers will be playing catch-up with the few studios that have the funds to create the real top of the line work, or they're paying to use someone elses work. I want cheaper, less powerful consoles so that developers can focus on making their games their way and not relying middleware or liscensed engines from crytek and epic.

Of course this means that companies like epic and crytek will take a huge hit, hence why they want them to be as powerful as possible. It's good business for them, bad for the industry, and bad for the gamer overall, but they're doing a really good job of selling it.

This is what most people are missing. All the extra resources doesn't mean you have to make the graphics on the cutting edge of technical developement, you can put all those resources to use improving IQ, framerate, and draw distance/count.

under the current metric of how games get reviewed and how metacritic plays a role in whether a studio gets funding, then they do have to make it cutting edge. They have to play catch-up to the big guns because reviewers are using games like uncharted and gears and crysis as the new demarcation for good graphics. Why aren't developers making wii-style games (like mario galaxy) for the xbox or ps3, then using the left over resources to deliver it to us at 60fps with 4xaa and all that other iq stuff. Because that's not how it works, it's never been how it works, and it's not going to work like that next generation.
 

MrBig

Member
under the current metric of how games get reviewed and how metacritic plays a role in whether a studio gets funding, then they do have to make it cutting edge. They have to play catch-up to the big guns because reviewers are using games like uncharted and gears and crysis as the new demarcation for good graphics. Why aren't developers making wii-style games (like mario galaxy) for the xbox or ps3, then using the left over resources to deliver it to us at 60fps with 4xaa and all that other iq stuff. Because that's not how it works, it's never been how it works, and it's not going to work like that next generation.

You can count the number of triple A studios that build tools for, and are developing the cutting edge of rendering tech, on one hand. That is not a metric you use to judge an industry as a whole. The vast majority of studios will see the new specs and see that they are able to extend their resource budget in ways they haven't been able to since 7 years ago. If their goal is to go bankrupt and sell an empty game with pretty graphics that's ok too i guess?
Look at Risen, Diablo III, WoW, DOTA 2, Portal, Minecraft, Trials, and many other games. Do they look like BF3, Crysis, Gears, or Halo? No. Do they still sell exceedingly well? Yes.
You have to have some hook to get people to buy and fund a game, it does not have to solely technical rendering based.
 
You can count the number of triple A studios that build tools for, and are developing the cutting edge of rendering tech, on one hand. That is not a metric you use to judge an industry as a whole. The vast majority of studios will see the new specs and see that they are able to extend their resource budget in ways they haven't been able to since 7 years ago. If their goal is to go bankrupt and sell an empty game with pretty graphics that's ok too i guess?
Look at Risen, Diablo III, WoW, DOTA 2, Portal, Minecraft, Trials, and many other games. Do they look like BF3, Crysis, Gears, or Halo? No. Do they still sell exceedingly well? Yes.
You have to have some hook to get people to buy and fund a game it does not have to solely technical rendering based.

yes, because every studio has the same reputation of Bliz, and Valve behind them? And xbla games have a completely different set of standards then retail games. Comparing apples to oranges here. How many XBLA games are made with ut3 and cryengine?
 

MrBig

Member
yes, because every studio has the same reputation of Bliz, and Valve behind them? And xbla games have a completely different set of standards then retail games. Comparing apples to oranges here. How many XBLA games are made with ut3 and cryengine?

They are retail games, played by humans. That's like saying any game sold on steam is junk, regardless of whether they were backed by some huge publishes. They still move a huge amount of copies. Plenty of games on XBLA use UE3, the most recent of which is Hybrid. UE3 is loved by smaller devs because many problems of dev are already solved for them, all they have to do is pay a fee, which may or may not be less than the money/time it would take to build an engine from scratch.
 
They are retail games, played by humans. That's like saying any game sold on steam is junk, regardless of whether they were backed by some huge publishes. They still move a huge amount of copies. Plenty of games on XBLA use UE3, the most recent of which is Hybrid. UE3 is loved by smaller devs because many problems of dev are already solved for them, all they have to do is pay a fee, which may or may not be less than the money/time it would take to build an engine from scratch.

give me a break, you know what I mean when I say a retail game($60 box title). And that link only shows 5 games for XBLA/psn as a platform.
 
Yes and some will argue that it's still in it's infancy.

Game have gotten considerably more complex this gen. That won't stop all of a sudden next gen.

There are only certain things that will get more complex. This past gen we saw games get more complex on a fundamental basis(think thread scheduling, splitting up work loads among units, synchronization, etc) because of a paradigm shift. The groundwork has been laid all generation.

This gen will see simply expansions of that work on a bigger scale. It'll be alot less work on developers to direct traffic amongst the computing elements compared to this generation. There is nothing in hardware today(i.e. next-gen consoles) that will cause a major paradigm shift similar to what happened last time. Its not "all of a sudden".
 

squidyj

Member
I agree with this post entirely.

Graphics are not everything and some people need to learn that.

Seems like you didn't read the post at all, maybe you should give it a try.


Sadly, completely wrong.

The more complex and feature-rich your program, the greater the chance some random combination of events/operations will cause it to malfunction. Validation by the team and test time within QA will inevitably increase.

Also, however much V/RAM a system has, the impetus will always be to fill it in order to provide the maximum return - developers will always be bumping their heads against the ceiling. The only difference a higher ceiling makes is that it takes more stuff to reach it (more complexity, more time, more cost) and there's further to fall when something goes wrong.

There are no short-cuts. A 1000 piece jigsaw is always going to take longer to complete than a 200 piece one if the picture is basically the same.

Time/space/feature budgets will always need to be observed and the higher the level of quality demanded as standard the more money it will cost to deliver that. You want movie-quality imagery, be prepared to pay movie-quality rates for your staff and equipment.

And ultimately that expense is always going to trickle down to the consumer.

So... it's more complex and challenging to make a game that runs in 1gb than it is to make a game that runs in 10mb?
 

SapientWolf

Trucker Sexologist
Information is not filled in for all of the titles on there. I couldn't find a more complete source.
From what I know:

UE 3.0
Shadow Complex
Undertow
Super Sonic Acrobatic Rocket Powered Battle Cars
Dungeon Defenders
Wheels of Destruction
Roboblitz
Section 8: Prejudice
Choplifter HD
Monday Night Combat
Rock of Ages
Warp
Alien Breed Evolution
Blacklight: Tango Down


Cryengine
Nexuiz

I'm sure there's more. Hybrid uses Source engine, BTW. Gotta say, it looks damn good running at 60fps.
 

KageMaru

Member
There are only certain things that will get more complex. This past gen we saw games get more complex on a fundamental basis(think thread scheduling, splitting up work loads among units, synchronization, etc) because of a paradigm shift. The groundwork has been laid all generation.

This gen will see simply expansions of that work on a bigger scale. It'll be alot less work on developers to direct traffic amongst the computing elements compared to this generation. There is nothing in hardware today(i.e. next-gen consoles) that will cause a major paradigm shift similar to what happened last time.

I agree we won't be seeing a paradigm shift like this gen, which was a big reason behind the jump in cost. I was just refuting your earlier claim that games won't get more complex when they most certainly will.
 
then it's not easier to develop for, it's more expensive because developers either need to build the tools themselves or they need to buy middleware.

The more powerful the hardware, the more developers will be playing catch-up with the few studios that have the funds to create the real top of the line work, or they're paying to use someone elses work. I want cheaper, less powerful consoles so that developers can focus on making their games their way and not relying middleware or liscensed engines from crytek and epic.

Although it is true that in order to take advantage of the hw you need to develop something that is specifically meant to run on the new HW, it is not always true that developing for a more powerful hardware is more expansive. There are situations in which developing on a less powerful hw can be more expansive than developing on a more powerful hw. For example, let's say that you have made an engine for the xbox360, and let's say that the visuals of the engine are amazing, but the engine struggles to run at a decent frame rate at 720p. So you would need to spend additional time to optimize the engine in order to obtain a solid frame rate, and this would mean spending a lot of extra money for opimization. Now let's say that the xbox720 is so much powerful that can run the same engine at 60fps at 1080p without any problem; then, you would save the money needed for optimization.
 

i-Lo

Member
I agree we won't be seeing a paradigm shift like this gen, which was a big reason behind the jump in cost. I was just refuting your earlier claim that games won't get more complex when they most certainly will.

And hence the necessity for dev tools to improve proportionally to ensure that the budgets for this gen games on average remain similar to what it is today across the board and be gradually reduced (to whatever end reasonably possible).
 
Although it is true that in order to take advantage of the hw you need to develop something that is specifically meant to run on the new HW, it is not always true that developing for a more powerful hardaware is more expansive. There are situations in which developing on a less powerful hw can be more expansive than developing on a more powerful hw. For example, let's say that you have made an engine for the xbox360, and let's say that the visuals of the engine are amazing, but the engine struggles to run at a decent frame rate at 720p. So you would need to spend additional time to optimize the engine in order to obtain a decent frame rate, and this would mean spending a lot of extra money for opimization. Now let's say that the xbox720 is so much powerful that can run the same engine at 60fps at 1080p without any problem. Then you could stop to worry about optimizing the engine, saving a lot of money.

yeah, and that's exactly epics problem, and why they're pushing for stronger hardware. I'm guessing they built middleware that's not effective for the next generation of consoles and are pushing MS and Sony, to increase specs so they don't have to optimize as much.
 
yeah, and that's exactly epics problem, and why they're pushing for stronger hardware. I'm guessing they built middleware that's not effective for the next generation of consoles and are pushing MS and Sony, to increase specs so they don't have to optimize as much.

It seems to me that Epic just wants a real jump in terms of power, which is what you would expect from a new generation of consoles. We saw a real jump with the xbox360, and it is normal that Epic wants to see a similar jump with the next xbox.
 
It seems to me that Epic just wants a real jump in terms of power, which is what you would expect from a new generation of consoles. We saw a real jump with the xbox360, and it is normal that Epic wants to see a similar jump with the next xbox.

no, what epic wants is money and they're fronting it as if they want what's best for gamers. The stronger the next console the more people will rely on middleware.
 

theBishop

Banned
no, what epic wants is money and they're fronting it as if they want what's best for gamers. The stronger the next console the more people will rely on middleware.

I don't see why it would be different from last Gen. This generation forced developers to cope with multithreading and programmable shaders. There's no equivalent sea changes on the horizon for next generation. Maybe general purpose gpu code, but that's avoidable, and can be implemented in manageable modules.

I expect epic to be slightly less influential next generation no matter how much ram is available.
 
I agree we won't be seeing a paradigm shift like this gen, which was a big reason behind the jump in cost. I was just refuting your earlier claim that games won't get more complex when they most certainly will.

I was only referring to one specific capacity of a game. I dont really know how to express it in layman's terms.
 
Top Bottom