Support NeoGAF

[Arulan]

So assuming his quote is wrong until he is prooven right is more logical and unbiased? Sorry but I don't get that.

I'm not asking anybody to blindly believe that statement. But you should give him at the very least the benefit of doubt that he may know what he is talking about. I'll call that unbiased.

I don't have a secret console agenda or something along those lines. If anything I try to defend the developer. I think it was a great interview where he spoke quite freely about the problems that occur when developing for the new consoles. He seems to be a pretty knowlegable guy that talks straight. Plainly dismissing his statements does not seem fair to me.

When all historical evidence shows that a PC with similar specs performs at least as good when running the same game, then yes, actual "evidence" is required to substantiate it. Console optimizations are important for specific aspects of its performance, but all empirical evidence shows no signs, or even hints at such as large gap (2x).

Do you not see how conveniently ignoring the empirical evidence, basing the pillar of your argument on a quote (appeal to authority), and then creating an unfalsifiable scenario in which it is impossible for your opponents to "prove" anything to you is the basis of a very poor argument?

 

[Zimbardo]

Oles Shishkovstov: Well obviously they aren't packing the bleeding edge hardware you can buy for PC (albeit for insane amounts of money) today.

another one of those gaming PC's cost too much money statements? feels like the year 2000 all over again.

anyway, its a good thing bleeding edge PC hardware isn't needed. my 4 year old cpu and 2 year old gpu still kick some large ass ...and looking at the modest hardware in consoles, probably will continue to do so for some time.

at this point, i'd almost love to have an excuse to upgrade ...but its just not needed for the resolution i'm gaming at, 1080p.

 

[IcyRhythms]

Man lol

People are truly sensitive over the Naughty Dog comment.

 

[ethomaz]

hm I don't know, Cry Engine is really impressive. I haven't seen Ryse in person, but I did play Crysis 3 maxed and that game had some amazing tech. I wasn't impressed by Killzone Shadow Fall at all to be honest.

Crysis 3 and Crysis (moded) both on PC looks better than these three games... so I think it is unfair to compare them with PC games maxed out.

I'm sure Ryse on PC will look completely different... I already saw they removed the blurry in the pics released.

 

[Inuhanyou]

People are looking at the comparison to PC hardware without looking at the rest of the sentence.

He was saying, PS4 and Xbox one aren't bleeding edge hardware, but they are positively acceptable for being budget machines, built at a average consumer price range(PS4 moreso IMO) without breaking the bank. That is a correct statement. How could anyone argue otherwise?

The most likely miniscule slim version of PS4 is going to kick ass in a few years, and we can finally get consoles that go sub 149 at the end of their console lifecycle

 

[SniperHunter]

Crysis 3 and Crysis (moded) both on PC looks better than these three games.

The original Crysis is an unoptimized mess of a game. Warhead was would be a better choice IMO. Crysis 3 has the most advance tech, when Yerli said no other game would beat Crysis 3 in the tech department for another two years, I totally believed him. The water simulation, the vegetation, the grass, yea Crysis 3's tech is far ahead of the pack. But of course the game is a complete piece of shit and the art sucks as well

I think The Order 1886 looks absolutely gorgeous, especially for a game coming from a small studio that has only made handheld titles in the past. They shouldn't be mocked.

 

[ethomaz]

The original Crysis is an unoptimized mess of a game. Warhead was would be a better choice IMO. Crysis 3 has the most advance tech, when Yerli said no other game would beat Crysis 3 in the tech department for another two years, I totally believed him. The water simulation, the vegetation, the grass, yea Crysis 3's tech is far ahead of the pack. But of course the game is a complete piss of shit and the art sucks as well

I think The Order 1886 looks absolutely gorgeous, especially for a game coming from a small studio that has only made handheld titles in the past. They shouldn't be mocked.

Yeap... I meant Crysis like both games (original and Warhead... both modded)

I think Ryse will take the graphic top spot when released on PC... it is not that impressive on the XB1... I guess the blurry image kills my ideia of good graphics (I hated the Killzone MP for example to the point I found it unplayable).

Crytek will show the game they wish to make at beginning.

 

[IcyRhythms]

I'm not impressed with a lot of the games mentioned. Metro beats TLoU, Infamous, KZ, and the Crysis games for me. The atmosphere and art are more appealing. Hell, Half Life 2 and Left 4 Dead look better to me. If we're talking about which is superior on a technical level, a lot my faves would lose out. I'm giving style points because I'm more of an art/atmosphere guy.

 

[jgf]

When all historical evidence shows that a PC with similar specs performs at least as good when running the same game, then yes, actual "evidence" is required to substantiate it. Console optimizations are important for specific aspects of its performance, but all empirical evidence shows no signs, or even hints at such as large gap (2x).

Do you not see how conveniently ignoring the empirical evidence, basing the pillar of your argument on a quote (appeal to authority), and then creating an unfalsifiable scenario in which it is impossible for your opponents to "prove" anything to you is the basis of a very poor argument?

So honestly asking. Is there any comparison of pretty late gen games (that were most likely quite well optimized) to their performance on similar PC hardware? Comparing to PS3 might be quite tough due to the different architecture, but against 360 could lead to better insight. At least its CPU is not as crazy as the cell...

Is there some article or something you would recommend that I look into it that you deem a fair comparsion?

 

[Respawn]

this seems confusing, because both have effectively the same CPU (Xbox even apparantly a little faster). So why would it be so much slower? Are they using the GPU somehow to generate draw calls for itself? main memory bandwidth limitation? shit driver from MS?

This was probably already answered but Cpu is not faster in X1. Actually the opposite. Dev stated this a while back. You get more performance out of PS4 Cpu.

 

[IncandescentPrecept]

I say the same thing about Star Citizen. Haven't seen a console game close to the graphics that game is pumping out. Plus Star Citizen is an open ended game with those crazy graphics which makes it more impressive.

Journey looks better than both.

 

[Mr Vast]

This was probably already answered but Cpu is not faster in X1. Actually the opposite. Dev stated this a while back. You get more performance out of PS4 Cpu.

both are the same cpu.

xbone clock 1.75 ghz and ps4 is 1.6ghz...

 

[IcyRhythms]

Journey looks better than both.

I would agree with that

 

[Dictator93]

Journey looks better than both.

I would agree with that

IN a technical thread, about tecnichal tings, usually screaming "art style" etc... "better art" is kinda inappropriate to the conversation. Sorry the thread turned this way and you posted that though... :/

 

[Respawn]

both are the same cpu.

xbone clock 1.75 ghz and ps4 is 1.6ghz...

Again. Matt has stated it's not so unless you can prove a Dev wrong then I'll take his word over yours.
Remember I said performance.

 

[Arulan]

So honestly asking. Is there any comparison of pretty late gen games (that were most likely quite well optimized) to their performance on similar PC hardware? Comparing to PS3 might be quite tough due to the different architecture, but against 360 could lead to better insight. At least its CPU is not as crazy as the cell...

Is there some article or something you would recommend that I look into it that you deem a fair comparsion?

There are plenty of examples in this thread, some were even directed at you earlier. To get more examples you need to find the equivalent in power because most of these video cards are so old they aren't even mentioned in modern benchmarks. You can also look at equivalent power benchmarks with PS4 and Xbox One titles. There are far more of these for instance.

 

[icecold1983]

Yeap... I meant Crysis like both games (original and Warhead... both modded)

I think Ryse will take the graphic top spot when released on PC... it is not that impressive on the XB1... I guess the blurry image kills my ideia of good graphics (I hated the Killzone MP for example to the point I found it unplayable).

Crytek will show the game they wish to make at beginning.

so youre saying that simply running ryse at a higher res takes it from "not impressive" to being the best looking game available?

 

[TheLaughingStock]

I don't see that comment as a dig at Naughty Dog. If anything he is acknowledging them to be one of the best.

People are being too touchy.

I mean have people seen the Uncharted articles? I get a very different interpretation because he's such a fan of what they were able to do on PS3 and doesn't mention many shortcomings. TLoU PS3 was a bit different because of the poor performance.

 

[aeolist]

another one of those gaming PC's cost too much money statements? feels like the year 2000 all over again.

anyway, its a good thing bleeding edge PC hardware isn't needed. my 4 year old cpu and 2 year old gpu still kick some large ass ...and looking at the modest hardware in consoles, probably will continue to do so for some time.

at this point, i'd almost love to have an excuse to upgrade ...but its just not needed for the resolution i'm gaming at, 1080p.

to be fair the guy lives in ukraine and it would not surprise me if PC hardware was a lot more expensive over there. lots of countries don't have it as good as the US/canada.

 

[jgf]

There are plenty of examples in this thread, some were even directed at you earlier. To get more examples you need to find the equivalent in power because most of these video cards are so old they aren't even mentioned in modern benchmarks. You can also look at equivalent power benchmarks with PS4 and Xbox One titles. There are far more of these for instance.

Actually I re-read the thread beforehand to make sure I'm not missing anything. Maybe I do. I found the following two links:

http://www.eurogamer.net/articles/digitalfoundry-2014-nvidia-geforce-gtx-750-ti-review
http://www.eurogamer.net/articles/digitalfoundry-2014-r7-260x-vs-next-gen-console

And some images/graphs about the performance of mass effect 3 and battlefield 4. Most of them seem pretty current gen (PS4/X1) focused or even use superior cpus like an i7.

Did I miss something?

As the quote quite literally said that it takes time to fully optimize for a given platform, I don't think it makes much sense to compare launchwindow titles to their performance on similar spec PCs. As I mentioned earlier I also don't think that even late gen multiplatform games translate 1:1, but I'm inclined to believe that they at least give some form of indication where the potential for optimization might meet its climax. Therefore I'm really interested in any comparsion about them.

 

[IncandescentPrecept]

IN a technical thread, about tecnichal tings, usually screaming "art style" etc... "better art" is kinda inappropriate to the conversation. Sorry the thread turned this way and you posted that though... :/

I'm not talking about artstyle. It just looks better.

 

[ethomaz]

so youre saying that simply running ryse at a higher res takes it from "not impressive" to being the best looking game available?

It is not only higher res from what I saw... I guess you will can use all the CryENGINE effects like Crysis 3... just the native resolution already will give it a better IQ plus the details on screen was increased and more advanced effects was mentioned by Crytek.

Think like running a game in Medium details vs Ultra details on PC... that I guess will be the difference.

 

[IcyRhythms]

IN a technical thread, about tecnichal tings, usually screaming "art style" etc... "better art" is kinda inappropriate to the conversation. Sorry the thread turned this way and you posted that though... :/

Art and atmosphere are all part of the over technical process. :-\

 

[Ambition]

Even crysis 1 with mods look better then the order and it was released years ago. It's also not rendered in a res lower then my phone's screen and at 30 fps. https://www.youtube.com/watch?v=tZApOsCc1oI

I mean, nothing special here. Characters lack details, textures are nice but flat and the breaking black bar while you watch a chain of QTE.

C'mon man

 

[Arulan]

Actually I re-read the thread beforehand to make sure I'm not missing anything. Maybe I do. I found the following two links:

http://www.eurogamer.net/articles/digitalfoundry-2014-nvidia-geforce-gtx-750-ti-review
http://www.eurogamer.net/articles/digitalfoundry-2014-r7-260x-vs-next-gen-console

And some images/graphs about the performance of mass effect 3 and battlefield 4. Most of them seem pretty current gen (PS4/X1) focused or even use superior cpus like an i7.

Did I miss something?

As the quote quite literally said that it takes time to fully optimize for a given platform, I don't think it makes much sense to compare launchwindow titles to their performance on similar spec PCs. As I mentioned earlier I also don't think that even late gen multiplatform games translate 1:1, but I'm inclined to believe that they at least give some form of indication where the potential for optimization might meet its climax. Therefore I'm really interested in any comparsion about them.

Did you ignore the Crysis 2 video on x1950 Pro running at higher than console settings while recording?

You should look at Intel HD 3000 benchmarks. It's very close to last-gen consoles and isn't so old that you will hardly find any benchmarks or videos on it. Here is Crysis 2 running on "High" on a HD 3000. https://www.youtube.com/watch?v=g00Y5K-SE-4

 

[bishoptl]

Its running on better hardware, its in production already longer than Ryse ever was [only 18 months]

Ryse was in production for 7 years, all told.

Carry on.

 

[jgf]

Did you ignore the Crysis 2 video on x1950 Pro running at higher than console settings while recording?

You should look at Intel HD 3000 benchmarks. It's very close to last-gen consoles and isn't so old that you will hardly find any benchmarks or videos on it. Here is Crysis 2 running on "High" on a HD 3000. https://www.youtube.com/watch?v=g00Y5K-SE-4

I really did miss that video. Unintentionally. While this is not an exact comparison of Crysis 2 on 360 vs PC I take your word for it that it looks similar to last gen. But still this PC is rocking an i5 and 4gb of ram. Thats not what I would call a fair comparison.

 

[icecold1983]

Did you ignore the Crysis 2 video on x1950 Pro running at higher than console settings while recording?

You should look at Intel HD 3000 benchmarks. It's very close to last-gen consoles and isn't so old that you will hardly find any benchmarks or videos on it. Here is Crysis 2 running on "High" on a HD 3000. https://www.youtube.com/watch?v=g00Y5K-SE-4

lower res and worse framerate than console versions

 

[Lonely1]

I really did miss that video. Unintentionally. While this is not an exact comparison of Crysis 2 on 360 vs PC I take your word for it that it looks similar to last gen. But still this PC is rocking an i5 and 4gb of ram. Thats not what I would call a fair comparison.

GPU bounded.

 

[Arulan]

lower res and worse framerate than console versions

Using the HD 3000 video as a reference:

A quick search of Crysis 2 Digital Foundry articles and videos show indeed the Xbox 360 version is slightly higher (829440 vs 786432 pixels) while the PS3 version is lower (737280 pixels). Judging from the HD 3000 video and the claimed average it is no different from consoles, which vary wildly between 20 and 30.

https://www.youtube.com/watch?v=FrTf2Katcd4

Additionally the console versions do not match the PC "High", for instance:

http://www.eurogamer.net/articles/digitalfoundry-tech-comparison-crysis2

Crytek says that the console versions operate at the High level, but in truth it seems that even at this setting the PC is running with more light sources producing more dynamic shadows, an effect that scales up still further as you move on up through the higher graphical levels.

Suffice to say, where is double the performance? Not even a little hint?

 

[SenjutsuSage]

Ryse was in production for 7 years, all told.

Carry on.

But actual development of the final game intended for Xbox One was much shorter than that, along the lines of what that guy said. It doesn't really matter how long the game was in the brainstorm or aspirational development phase. So saying 7 years is more than a stretch.

 

[grandwizard]

It's amazing how people continue to believe their console hardware is magical in the face of literally all evidence. Everyone knows you can squeeze some more out of a fixed spec with enough dev focus, but the idea that you can get 2x performance compared to equivalent PC hardware is insane. There is not one game that has shown that. However there are many examples of equivalent hardware running the same.

 

[icecold1983]

Suffice to say, where is double the performance? Not even a little hint?

compare the graphics that hd3000 is outputting at 20 to 25 fps to what gow, uncharted, killzone etc is outputting at 30 or more fps while also at a higher resolution

 

[bishoptl]

But actual development of the final game intended for Xbox One was much shorter than that, along the lines of what that guy said. It doesn't really matter how long the game was in the brainstorm or aspirational development phase. So saying 7 years is more than a stretch.

No. That's not how it shows up on the P&Ls, and that's not how it goes in development.

 

[Arulan]

compare the graphics that hd3000 is outputting at 20 to 25 fps to what gow, uncharted, killzone etc is outputting at 30 or more fps while also at a higher resolution

When the empirical evidence doesn't produce your desired results I'm given an unfalsifiable scenario instead? How am I supposed to objectively prove anything? Even if I thought or had reason to believe other games looked better, or were technically more impressive there doesn't exist any evidence to show whether or not those games are more or less demanding on the hardware. In other words, you've constructed a scenario in which your argument cannot "lose" because it is impossible to test for your "desired set of conditions", while you conveniently ignore all the other evidence.

 

[icecold1983]

When the empirical evidence doesn't produce your desired results I'm given an unfalsifiable scenario instead? How am I supposed to objectively prove anything? Even if I thought or had reason to believe other games looked better, or were technically more impressive there doesn't exist any evidence to show whether or not those games are more or less demanding on the hardware. In other words, you've constructed a scenario in which your argument cannot "lose" because it is impossible to test for your "desired set of conditions", while you conveniently ignore all the other evidence.

do you not have eyes? the moving images speak for themselves. and i said way back in the thread that exclusives should be used when trying to determine the benefits of console optimization

 

[KKRT00]

Ryse was in production for 7 years, all told.

Carry on.

To be fair, it was Project Kingdom. Ryse was restarted completely in Frankfurt 18 months before Xbone launch. But yeah, the whole project was in development for 7 years, assets and new CE features for Ryse however were not.

Same goes for The Order, they started officially in 2010, but probably next-gen tech and asset development started in second half od 2011, when they knew more about PS4.

----

so youre saying that simply running ryse at a higher res takes it from "not impressive" to being the best looking game available?

To be fair, Ryse on PC will run on higher settings, which means probably real-time caustics, definitely higher quality shadows and textures.

 

[djp6]

do you not have eyes? the moving images speak for themselves. and i said way back in the thread that exclusives should be used when trying to determine the benefits of console optimization

I feel like going out on a limb and saying those games would still perform better on pc in my eyes(see what i did there) a comparison to games that are not available on pc can not be used to prove that coding to the metal will give 2x performance gains. And the crysis 2 example you were given was that not coded to the "metal" or it was coded to something else??

 

[Sid]

Crysis 2 runs better on a PC with an 8800GT using which CPU and how much RAM?

 

[jellies_two]

It's amazing how people continue to believe their console hardware is magical in the face of literally all evidence. Everyone knows you can squeeze some more out of a fixed spec with enough dev focus, but the idea that you can get 2x performance compared to equivalent PC hardware is insane. There is not one game that has shown that. However there are many examples of equivalent hardware running the same.

You can't look to multi platform games as the standard bearer of platform optimisation advantages. You have to look at exclusives where the code base is not constrained by the needs to offer the same game on all platforms and the programmers are not spread thinly over multiple architectures and APIs. Just like cross-gen hobbles next-gen games, multi platform drags down platform optimisation.

Instead you can expect exclusives to tell the story. That tends to be more theoretical and open to argument so in the end you'll be able to deny the advantage.

Speaking as a programmer though I think it is insane to imagine that a fixed chip-set and fixed specification cannot be better optimised than the complete zoo that comprises possible PC specifications. And speaking of optimisation in code, 2x is a very conservative goal. Most things can be optimised by an order of magnitude once you wade in deep enough. Just is rarely worth it. But with an installed base of 10m identical consoles, optimisation becomes very attractive.

 

[djp6]

Crysis 2 runs better on a PC with an 8800GT using which CPU and how much RAM?

So are you saying you want a comparison to a pc that costs the same as a ps3 when it launched?? Or a comparison to a PC that has the same hardware as a ps3??

 

[Sid]

So are you saying you want a comparison to a pc that costs the same as a ps3 when it launched?? Or a comparison to a PC that has the same hardware as a ps3??

I'm just asking about the CPU and the RAM amount of that machine.

 

[KKRT00]

Crysis 2 runs better on a PC with an 8800GT using which CPU and how much RAM?

RAM is irrelevant. CPU is C2D and thats for 1080p and higher settings than consoles.

---

compare the graphics that hd3000 is outputting at 20 to 25 fps to what gow, uncharted, killzone etc is outputting at 30 or more fps while also at a higher resolution

HD3000 is less powerful than past-gen, so how is this a point? Check HD 4000 which is about the same.

 

[alexandros]

You can't look to multi platform games as the standard bearer of platform optimisation advantages. You have to look at exclusives where the code base is not constrained by the needs to offer the same game on all platforms and the programmers are not spread thinly over multiple architectures and APIs. Just like cross-gen hobbles next-gen games, multi platform drags down platform optimisation.

But... That's what many of us have been saying the whole time. That in the real world, in practical terms and in actual games, not high level game development theory, the mythical "coding to the metal" optimizations don't exist. They don't manifest into reality. In theory, noone should argue that you can't optimise the heck out of your code for a specific platform given enough time and money. In practice, noone does that. It's counter-productive. That's why there aren't any real world examples of these fabled optimisations.

 

[Sid]

RAM is irrelevant. CPU is C2D and thats for 1080p and higher settings than consoles.

---

HD3000 is less powerful than past-gen, so how is this a point? Check HD 4000 which is about the same.

Why is the RAM irrelevant?

 

[alexandros]

Why is the RAM irrelevant?

Because a PC is a general purpose machine, not a console.

 

[c0de]

I'm sure Ryse on PC will look completely different... I already saw they removed the blurry in the pics released.

What? Completely different? I doubt so and I don't know what your constant talk about blurryness means...

 

[KKRT00]

Why is the RAM irrelevant?

Because RAM doesnt affect performance and because many games do not even implement special streaming designed for consoles on PC, like for example Crysis 2.
Additionally PC version has different, bigger framebuffer layout for higher HDR precision.

 

[Sid]

I'd like if Leadbetter approaches Oles and Carmack to explain this '2x' comment.

 

[justsomeguy]

Again. Matt has stated it's not so unless you can prove a Dev wrong then I'll take his word over yours.
Remember I said performance.

Be interested to see if this had changed with kinect release and updates to the xdk as this article says there have been lots of improvements. Is matt still around? That statement was quite a while ago now, be interested in an updated view.