-
Hey Guest. Check out your NeoGAF Wrapped 2025 results here!
Anand's Xbox 360 and PS3 Hardware article
- Thread starter golem
- Start date
Graphics-wise the 360s Xenos GPU and the PS3s RSX are fairly different in implementation, but may end up being very similar in performance.
The 360 has more cache, while the Cell has a lower latency path to main memory
the performance difference from a CPU standpoint between the Xbox 360s Xenon and the PS3s Cell processor is basically a wash.
That's all that matters from the article.
The 360 has more cache, while the Cell has a lower latency path to main memory
the performance difference from a CPU standpoint between the Xbox 360s Xenon and the PS3s Cell processor is basically a wash.
That's all that matters from the article.
It means that they're close enough where it'll come down to the exclusive developers to put one group ahead of the other. And right now that's looking like Sony.
For the first gen or so, games will vary wildly between consoles, but as a single engine is made to work between all 3, the differences will even out into nothing.
For the first gen or so, games will vary wildly between consoles, but as a single engine is made to work between all 3, the differences will even out into nothing.
M3wThr33 said:
Thank you.
Slightly OT: Even though the CELL processor would be used in other appliances such as TVs e.t.c doesn't the money spent on R&D seem like a waste if the PS3 games look only marginally better than Xbox 360 games? Shouldn't the 200+ Gflops count for something or is the GPU holding everything back?
there has to be a diffrent. when you play madden on ps3 wont be the same as xbox 360. one has to be lower in terms of graphics even if its not very clear diffrence like ps2/GC/xbox but there has to be some... i think
anyway. seems like MS has lost the power advantage they used to have. wonder what will happen now.
anyway. seems like MS has lost the power advantage they used to have. wonder what will happen now.
Yeah it's pretty much what we, the more level headed folks, have expected. But this part is really something new to me.
For what it is worth, porting game code between the PC and the Xbox 360 will be a lot like Mac developers porting code between Mac OS X for Intel platforms and PowerPC platforms: theres an architecture switch, but the programming model doesnt change much.
The same cannot however be said for Cell and the PlayStation 3. The easiest way to port code from the Xbox 360 to the PS3 would be to run the code exclusively on the Cells single PPE, which obviously wouldnt offer very good performance for heavily multi-threaded titles. But with a some effort, the PlayStation 3 does have a lot of potential."
I always assumed, until now, that ports from the 360 to the PS3 should be pretty easy to do especially if we are to believe that the PS3 is slightly more powerfull. But this raises a serious problem. It's not going to be as easy porting games from the 360 to the PS3 as it was this generation porting games from the PS2 to GC/Xbox.
Could be a good or bad thing depending on which way you look at it. Could mean less multiplatform titles.
And this is HUGE:
"On the other hand, looking at all of the early demos weve seen of Xbox 360 and PS3 games, not a single one appears to offer better physics or AI than the best single threaded games on the PC today. At best, weve seen examples of ragdoll physics similar to that of Half Life 2, but nothing that is particularly amazing, earth shattering or shocking. Definitely nothing that appears to be leveraging the power of a multicore processor.
In fact, all of the demos weve seen look like nothing more than examples of what you can do on the latest generation of GPUs - not showcases of multi-core CPU power.
The majority of developers are doing things no differently than they have been on the PC. A single thread is used for all game code, physics and AI and in some cases, developers have split out physics into a separate thread, but for the most part you can expect all first generation and even some second generation titles to debut as basically single threaded games. The move to two hardware execution threads may in fact only be an attempt to bring performance up to par with what can be done on mid-range or high-end PCs today, since a single thread running on Xenon isnt going to be very competitive performance wise, especially executing code that is particularly well suited to OoO desktop processors."
But for the most part, on day 1, you shouldnt expect Xbox 360 games to be much more than the same type of single threaded titles weve had on the PC.
So it seems that 33% crap and that wait till second generation of games DOES INDEED hold a lot of truth to it. If the CPU in these systems are that weak when using single thread code, it explains quite a bit about some of the less impressive games we've seen so far, and also about the fact that this generation indeed will see some jumps in quality greater than what we've had over the past 4-5 years in graphics.
For what it is worth, porting game code between the PC and the Xbox 360 will be a lot like Mac developers porting code between Mac OS X for Intel platforms and PowerPC platforms: theres an architecture switch, but the programming model doesnt change much.
The same cannot however be said for Cell and the PlayStation 3. The easiest way to port code from the Xbox 360 to the PS3 would be to run the code exclusively on the Cells single PPE, which obviously wouldnt offer very good performance for heavily multi-threaded titles. But with a some effort, the PlayStation 3 does have a lot of potential."
I always assumed, until now, that ports from the 360 to the PS3 should be pretty easy to do especially if we are to believe that the PS3 is slightly more powerfull. But this raises a serious problem. It's not going to be as easy porting games from the 360 to the PS3 as it was this generation porting games from the PS2 to GC/Xbox.
Could be a good or bad thing depending on which way you look at it. Could mean less multiplatform titles.
And this is HUGE:
"On the other hand, looking at all of the early demos weve seen of Xbox 360 and PS3 games, not a single one appears to offer better physics or AI than the best single threaded games on the PC today. At best, weve seen examples of ragdoll physics similar to that of Half Life 2, but nothing that is particularly amazing, earth shattering or shocking. Definitely nothing that appears to be leveraging the power of a multicore processor.
In fact, all of the demos weve seen look like nothing more than examples of what you can do on the latest generation of GPUs - not showcases of multi-core CPU power.
The majority of developers are doing things no differently than they have been on the PC. A single thread is used for all game code, physics and AI and in some cases, developers have split out physics into a separate thread, but for the most part you can expect all first generation and even some second generation titles to debut as basically single threaded games. The move to two hardware execution threads may in fact only be an attempt to bring performance up to par with what can be done on mid-range or high-end PCs today, since a single thread running on Xenon isnt going to be very competitive performance wise, especially executing code that is particularly well suited to OoO desktop processors."
But for the most part, on day 1, you shouldnt expect Xbox 360 games to be much more than the same type of single threaded titles weve had on the PC.
So it seems that 33% crap and that wait till second generation of games DOES INDEED hold a lot of truth to it. If the CPU in these systems are that weak when using single thread code, it explains quite a bit about some of the less impressive games we've seen so far, and also about the fact that this generation indeed will see some jumps in quality greater than what we've had over the past 4-5 years in graphics.
sonycowboy
Member
Gotta love this article. Seems a bit premature until we find out more about both systems for them to be making a number of the declarations they've made.
Translation: Not that we really know anything about the PS3, but ...
Translation: Not that we really know anything about the PS3, but ...
M3wThr33 said:
Oh, good start....
More cache for the main core, less accessible memory. I'm not a developer, but I think I'd prefer CELLs 1.5MB of memory I can control + cache in my main core, over 1MB general purpose cache.
Now they are just twisting the comparisons. Its easy to port a single threaded PC app to X360, but difficult to port a multithreaded x360 app to PS3?
What about porting a single threaded PC app to PS3? What about porting a multithreaded app from PS3 to x360 (most likely scenario IMO)?
I'm really tempted to not read any more of these comparisons, or look at next gen screenshots anymore. Most are just dumb.
"What about porting a multithreaded app from PS3 to x360 (most likely scenario IMO)?"
What are you nuts? That's most deffinitely NOT going to be the most likely scenario. The 360 is coming out first. A LOT of the games and engines you will see over the next year on the 360 will be a basis for the first couple of generation of games, and these engines will be ported to the PS3. You think EA is going to developer brand new engines for ALL their sports games for next year? Or Activision...or any of the multiplatform Kings? Doubt it. Maybe in 3 years if and once the PS3 gains the lead in market share, but presently the reality is in the first 2 years expect the majority of ports to be 360 to PS3.
What are you nuts? That's most deffinitely NOT going to be the most likely scenario. The 360 is coming out first. A LOT of the games and engines you will see over the next year on the 360 will be a basis for the first couple of generation of games, and these engines will be ported to the PS3. You think EA is going to developer brand new engines for ALL their sports games for next year? Or Activision...or any of the multiplatform Kings? Doubt it. Maybe in 3 years if and once the PS3 gains the lead in market share, but presently the reality is in the first 2 years expect the majority of ports to be 360 to PS3.
sonycowboy
Member
jimbo said:
Both EA & Activision have been developing Next Gen software for over 2 years. It's a HUGE assumption that the basis of their development has been mostly on the 360.
From what we've seen, especially from EA & Activision, is that their initial 360 games look much more like modified console or PC ports than ground up next gen titles.
I understand your point, but I think it's a little bold to try and claim that the 360 currently is the reference platform. I don't think such a thing exists yet.
EDIT: Dammit. Darien beat me to it.
jimbo said:
Most games will likely be PC based ports initially. Not many publishers will risk a huge sum making custom X360 engines just yet.
In parallel, many developers will also work on their 'main' engines. I expect these to be based on PS3 as the lead platform. Simple economics.
So the devs will wait until the GAF NPD thread reveals PS3 has the lead before working on their engine? No. They'll base their decision on predictions. These are *very* likely to indicate PS3 as the largest platform and therefore the lead for development.
"I'll go you one further... Initially I think alot of the engines(sports titles) you're going to see from EA on the X360 and the PS3 are going to be based on their PC engines."
Which is still the same case...as it's easier to port to the 360 than PS3, right? According to these guys anyway.
"Both EA & Activision have been developing Next Gen software for over 2 years. It's a HUGE assumption that the basis of their development has been mostly on the 360."
I don't know, but I can already almost guarantee you some titles that will go from the 360 to PS3....Need for Speed, Test Drive, Tony Hawk, PC shooters LIKE GR3 & Call of Duty, etc....
"So the devs will wait until the GAF NPD thread reveals PS3 has the lead before working on their engine? No. They'll base their decision on predictions. These are *very* likely to indicate PS3 as the largest platform and therefore the lead for development."
That's pretty illogical if you ask me. The games they are developing now to come out over the next 2 years aren't going to be the games that they will care to sell 3 years from now. That will be a whole new generation of games. For the first 2 years I'm sure they're actually wanting sales NOW. It's not like EA is going hey let's develop Need For Speed with the PS3 in mind because it will probably be the market leader in 3 years, even though our game is coming out THIS FALL and no one's going to care about it 3 years from now. Think about it.
Anyway, I think this is a pretty good article, and it remains to be seen which will have the graphical edge, but one thing is becoming more and more clear....whichever console it may be, the difference will be minimal but the work required from developers both for exclusive and multiplatform titles won't . They have their work cut out for them this gen.
Oh and one more thing. It also seems as if this generation is at least in theory going to have long legs...as its predicted improvements over the following years is going to be more than what we've been used to. Single threading generations to multi threading generation is going to provide a nice jump. And then you have basic multi-threaded coding to expert/limit-pushing multi-threaded coding. Everybody should just be really excited about these systems as they will be pretty amazing to start off with and will only get better.
Next-Gen can't come soon enough. Good job MS and Sony as they both seem to have put together some excellent features.
Which is still the same case...as it's easier to port to the 360 than PS3, right? According to these guys anyway.
"Both EA & Activision have been developing Next Gen software for over 2 years. It's a HUGE assumption that the basis of their development has been mostly on the 360."
I don't know, but I can already almost guarantee you some titles that will go from the 360 to PS3....Need for Speed, Test Drive, Tony Hawk, PC shooters LIKE GR3 & Call of Duty, etc....
"So the devs will wait until the GAF NPD thread reveals PS3 has the lead before working on their engine? No. They'll base their decision on predictions. These are *very* likely to indicate PS3 as the largest platform and therefore the lead for development."
That's pretty illogical if you ask me. The games they are developing now to come out over the next 2 years aren't going to be the games that they will care to sell 3 years from now. That will be a whole new generation of games. For the first 2 years I'm sure they're actually wanting sales NOW. It's not like EA is going hey let's develop Need For Speed with the PS3 in mind because it will probably be the market leader in 3 years, even though our game is coming out THIS FALL and no one's going to care about it 3 years from now. Think about it.
Anyway, I think this is a pretty good article, and it remains to be seen which will have the graphical edge, but one thing is becoming more and more clear....whichever console it may be, the difference will be minimal but the work required from developers both for exclusive and multiplatform titles won't . They have their work cut out for them this gen.
Oh and one more thing. It also seems as if this generation is at least in theory going to have long legs...as its predicted improvements over the following years is going to be more than what we've been used to. Single threading generations to multi threading generation is going to provide a nice jump. And then you have basic multi-threaded coding to expert/limit-pushing multi-threaded coding. Everybody should just be really excited about these systems as they will be pretty amazing to start off with and will only get better.
Next-Gen can't come soon enough. Good job MS and Sony as they both seem to have put together some excellent features.
I agree that X360 launch games will be developed on X360 first. Obviously. I'm not sure how much effort will be put into 3rd party titles, but you're right on that count, as PS3 isn't out.
Well MS PR Spin is that 3 cores are better than one. But Anandtech say that most 1st gen games will be single thread. So they'd port just as well to PS3 as X360, just using one PPC core.
Well MS PR Spin is that 3 cores are better than one. But Anandtech say that most 1st gen games will be single thread. So they'd port just as well to PS3 as X360, just using one PPC core.
On the same topic, porting....I hate to say it because it's like beating a dead horse, but I think Nintendo is setting themselves up for a fall with the Revolution. Because from what we're seeing here, it already seems as if there's going to be a LOT of work from developers to port games back and froth from the 360 and PS3 and vice versa and these two systems WILL BE pretty similar in terms of power. If the Revolution doesn't at least come close to these two, I see a lot of developers just completely skipping out on it with their multiplatform games.
And that's not even taking into account the HD issue because you know 95% of all PS3 and 360 games will be designed with HD in mind where as on the Revolution games could benefit in completely different areas from the lack of it and hurt yet in others. Think about simple things as HUD's in games, text where you will be able to get much better visibility on HD monitors of those things as standard definition TV's. Now I realize this won't be so much in the first years of development as games will still have to run perfectly on regular TV's but later on, don't be surprized if you're playing a PS3 game on a regular TV and can't make out the map, or the writing, or the HUDs
And that's not even taking into account the HD issue because you know 95% of all PS3 and 360 games will be designed with HD in mind where as on the Revolution games could benefit in completely different areas from the lack of it and hurt yet in others. Think about simple things as HUD's in games, text where you will be able to get much better visibility on HD monitors of those things as standard definition TV's. Now I realize this won't be so much in the first years of development as games will still have to run perfectly on regular TV's but later on, don't be surprized if you're playing a PS3 game on a regular TV and can't make out the map, or the writing, or the HUDs
jimbo said:
Well no it's not in reference to what you were saying... if I understood correctly you were saying that X360 titles would be the basis for PS3 titles... saying that developers like EA will be creating game engines on the 360 and then porting them over... what I'm saying is that many of the initial game engines you're going to see are not going to have been created on the X360... they will be PC engines that have been modified... so the core development platform here is not the 360(as you were saying), but is the PC.
"Well no it's not in reference to what you were saying... if I understood correctly you were saying that X360 titles would be the basis for PS3 titles... saying that developers like EA will be creating game engines on the 360 and then porting them over... what I'm saying is that many of the initial game engines you're going to see are not going to have been created on the X360... they will be PC engines that have been modified... so the core development platform here is not the 360(as you were saying), but is the PC."
Possibly so to some extent, but do keep in mind these consoles ARE a bit of a departure from the PC. So I am not that ready to buy that EA is developing Need for Speed with the PC in mind when they will have to do some significant changes to get it to run even on the 360 in the first place. It still sounds as if porting from the 360 to the PS3(single threaded) will be easier than PC to the PS3(single threaded) just due to the nature of console developing. Therefore my point still stands.
Possibly so to some extent, but do keep in mind these consoles ARE a bit of a departure from the PC. So I am not that ready to buy that EA is developing Need for Speed with the PC in mind when they will have to do some significant changes to get it to run even on the 360 in the first place. It still sounds as if porting from the 360 to the PS3(single threaded) will be easier than PC to the PS3(single threaded) just due to the nature of console developing. Therefore my point still stands.
jimbo said:
I'm not sure where you get that... you yourself referenced the quote:
For what it is worth, porting game code between the PC and the Xbox 360 will be a lot like Mac developers porting code between Mac OS X for Intel platforms and PowerPC platforms: theres an architecture switch, but the programming model doesnt change much.
As I understand it(and if I'm incorrect since I don't pay much attention to EA's sports titles, then this whole line of reasoning falls through) EA's sports titles on the PC are very similar to their console cousins in terms of gameplay with higher end PC's look much better higher polygon counts, better textures, etc.... You aren't going to see any real delving in to the depths of the 360 from EA in the first gen on those sports titles.. since the sports titles already have the capability to look better, you're going to find that they are going so simply make the arch switch... and use the higher end textures and models that they would normally use on the PC side.
DarienA said:
That quote doesn't directly relate to what I'm saying. It's talking more about the difference between MULTI-CORE PC and Xbox 360 porting.
You have to read the article...there was a part that I didn't quite right before they said that. I'll quote the full paragraph:
"The most important selling point of the Xbox 360's Xenon core is the fact that all three cores are identical, and they are all general purpose microprocessors. The developer does not have to worry about multi-threading beyond the point of getting their code to be thread safe; once it is multi-threaded, it can easily be run on any of the cores. The other important thing to keep in mind here is that porting between multi-core PC platforms and the Xbox 360 will be fairly trivial. Anywhere any inline assembly is used there will obviously have to be changes, but with relatively minor code changes and some time optimizing, code portability between the PC and the Xbox 360 shouldn't be very difficult at all. For what it is worth, porting game code between the PC and the Xbox 360 will be a lot like Mac developers porting code between Mac OS X for Intel platforms and PowerPC platforms: there's an architecture switch, but the programming model doesn't change much. "
Also check out this part:
"Both will perform absolutely much slower than even mainstream desktop processors in single threaded game code, but the majority of games these days are far more GPU bound than CPU bound, so the performance decrease isnt a huge deal"
So I don't understand why developers would be concentrating on PC development. I mean they would have to take that PC code, PORT IT over to the 360, than take that 360 code and port it over to the PS3. Why do this? When they can just develop for the 360 and it would benefit BOTH the 360 AND PS3 while developing with the PC in mind would hurt both?
I don't know maybe I'm just mis-reading something, but this is what I'm getting anyway.
sly said:Thank you.
Slightly OT: Even though the CELL processor would be used in other appliances such as TVs e.t.c doesn't the money spent on R&D seem like a waste if the PS3 games look only marginally better than Xbox 360 games? Shouldn't the 200+ Gflops count for something or is the GPU holding everything back?
M3wThr33's "summary" isn't exactly giving you a lot of context, and is misleading in fact. I suggest you read the full article. That 200Gflops will count for something, and judging by early uses, things like physics and simulation should see the biggest benefit. Cell should buy PS3 better performance in those areas, I think. You'll see games adapt to the hardware their designed for..X360 games will take advantage of the things it is good at, PS3 games will take advantage of the things it is good at. So then it becomes a matter of which you would prioritise.
Tim Sweeney's comments about AI and game logic etc. and how they take up such a small percentage of CPU time is interesting given MS's claims about how the CPU is used too. It's not a direct response to that claim by MS, but I think if things like physics and so on mesh well with the SPEs, and "the rest" can run fine on a PPE, it'll have justified Sony's design decisions. Not all CPU tasks are created equally - I'd always thought, and had read, that physics took up a large proportion of CPU time - so optimising for the "big ones" first and foremost makes sense.
Anyway, I've yet to finish the whole article, reading bits as I can in work, but I'd agree that it seems a little early for full blown comparisons. But I guess this isn't, really..they acknowledge they'll have to revisit this later.
Last months Game Informer had a decent, in depth article on EA and next generation. The article stated EA has been getting ready for the next gen of games over the last 2 years. So I highly doubt EA sports games will be running off of PC versions. ESPECIALLY when it sounds like Madden 2006 for the PC is still based on the PS2 version while the X360 version is the true next gen one.
As for porting between PC and X360, well, PC processers are going to be going multi core like the X360 chip.
As for porting between PC and X360, well, PC processers are going to be going multi core like the X360 chip.
Right. When developers FULLY utilize all of the SPEs on the PS3 it SHOULD give the PS3 some slight advantages although I'm not sure if it will be in the area of graphics. And those games, will have a hard time running on the 360. Similar to what you're seeing with PS2 ports today that take full advantage of the PS2's superior particle effects processing over the Xbox. MGS, even GTA in some areas, are just some examples. Of course the same thing can be said about games using some of the features of the 360 fully, giving it advantages, even though it doesn't have a GFLOP advantage. There really is a wash here.
What you have to keep in mind is those extra GFLOPs on the PS3, would THEORETICALLY equal in those benefits only if the two systems' efficiency everywhere else or at least in those specific areas is equal. If the PS3 has better efficiency in that area it should actually allow for even more of a difference, if the 360, oth, is more efficient, it will nullify to some point the GFLOP advantage. In the end GFLOPs alone, don't equal more power. You have to look at the entire system and that's what I liked about Anandtech's approach. They're not comparing them like you do with different GPU's or CPU's on the PC today through simple benchmarks(I don't even know how they would do this to be honest). They're comparing them in their working environment as they should be, as best as they can because, like you said, it's still to early and even they admitted that. The PS3 is also clearly going to be more advantageous in areas of 1080p HD, which is not even supported by the 360.
What you have to keep in mind is those extra GFLOPs on the PS3, would THEORETICALLY equal in those benefits only if the two systems' efficiency everywhere else or at least in those specific areas is equal. If the PS3 has better efficiency in that area it should actually allow for even more of a difference, if the 360, oth, is more efficient, it will nullify to some point the GFLOP advantage. In the end GFLOPs alone, don't equal more power. You have to look at the entire system and that's what I liked about Anandtech's approach. They're not comparing them like you do with different GPU's or CPU's on the PC today through simple benchmarks(I don't even know how they would do this to be honest). They're comparing them in their working environment as they should be, as best as they can because, like you said, it's still to early and even they admitted that. The PS3 is also clearly going to be more advantageous in areas of 1080p HD, which is not even supported by the 360.
sonycowboy
Member
jimbo said:
Quick Question: If ~most developer will be relying on single threading ~mostly, and the PS3 has the more "PC" like GPU combined w/ OpenGL, whereas the 360 has the more radical unified share GPU combined w/ Direct X (XNA), why should the Xbox360 enjoy an easier transition from PC -> console?
Isn't the exactly what Tim Sweeney was talking about in the Sony PC @ E3?
I'm not saying that the XBox will be hard to port by any means, but to really take advantage of the unified shaders, the core graphics engines will have to be changed quite a bit relative to standard GPU design I would think.
In the end, it's all just hand wringing on both sides and we'll have to wait until developers get quality time on true dev kits for some period of time before we hear good word about which system is more powerful. I think we all expect that overall the 360 will be "easier" to develop for, although the PS3 seems to have made some decent strides in this area.
"Quick Question: If ~most developer will be relying on single threading ~mostly, and the PS3 has the more "PC" like GPU combined w/ OpenGL, whereas the 360 has the more radical unified share GPU combined w/ Direct X (XNA), why should the Xbox360 enjoy an easier transition from PC -> console?"
"the core graphics engines will have to be changed quite a bit relative to standard GPU design I would think."
Dude READ! They clearly stated PC games would be easier to port to 360 than PS3 so I don't know why you're trying to spin this around somehow. Also they never stated that they are relying MOSTLY on single-threaded code, but rather LIKE examples of single threaded coding out of what they've seen SO FAR, so I don't know where you're getting this from. They also clearly stated that SINGLE threaded PC code would SUCK hard on both, so single threading on either platform is a disadvantage. AND on top of that this is a CPU argument not a GPU argument. The GPU in the PS3 is more closely related to that in the PC, but ATI's USA architecture doesn't put the 360 at a disadvantage either over current GPUs simply because it CAN act as a traditional PC GPU. Developers don't have to recode anything for the GPU because the ATI GPU does the work AUTOMATICALLY or with very little programming on the part of the developers.
"In the end, it's all just hand wringing on both sides and we'll have to wait until developers get quality time on true dev kits for some period of time before we hear good word about which system is more powerful. I think we all expect that overall the 360 will be "easier" to develop for, although the PS3 seems to have made some decent strides in this area."
I agree with you here. I think everyone is just so eager to go ahead and pick a definite winner but it just won't be that cut and dry. Even when we do have full details on the processors it won't be until developers have had some seious time with both to get an answer.....or the answer may always just be...too close to call.
"the core graphics engines will have to be changed quite a bit relative to standard GPU design I would think."
Dude READ! They clearly stated PC games would be easier to port to 360 than PS3 so I don't know why you're trying to spin this around somehow. Also they never stated that they are relying MOSTLY on single-threaded code, but rather LIKE examples of single threaded coding out of what they've seen SO FAR, so I don't know where you're getting this from. They also clearly stated that SINGLE threaded PC code would SUCK hard on both, so single threading on either platform is a disadvantage. AND on top of that this is a CPU argument not a GPU argument. The GPU in the PS3 is more closely related to that in the PC, but ATI's USA architecture doesn't put the 360 at a disadvantage either over current GPUs simply because it CAN act as a traditional PC GPU. Developers don't have to recode anything for the GPU because the ATI GPU does the work AUTOMATICALLY or with very little programming on the part of the developers.
"In the end, it's all just hand wringing on both sides and we'll have to wait until developers get quality time on true dev kits for some period of time before we hear good word about which system is more powerful. I think we all expect that overall the 360 will be "easier" to develop for, although the PS3 seems to have made some decent strides in this area."
I agree with you here. I think everyone is just so eager to go ahead and pick a definite winner but it just won't be that cut and dry. Even when we do have full details on the processors it won't be until developers have had some seious time with both to get an answer.....or the answer may always just be...too close to call.
sonycowboy
Member
jimbo said:
Breathe.
Even at B3D, they're knocking this article for it's assumptions and ultimately it's conclusions. The point being, don't rely on the article to make your arguments. Make your own. I recognize the argument Anand was making and I think it was weak to try and say that porting would be easier "from the CPU side" and ignore the GPU.
If you think that the first batch of games developed for next gen are going to come anywhere close to being more CPU dedicated than GPU dedicated (both from a development side as well as the core engines), then I'd have to disagree. I do agree that the GPU can fly on it's own as the GPU can dynamically assign the shaders work, but that ignores the change in design that the graphics engine would need to have to really take advantage of that flexibility.
Calm down a little. People aren't attacking you or the 360. We're talking about this article . You just happen to be the counterpoint for a specific side of the argument.
sonycowboy said:
I don't think the devs really have to care. Basically they throw their shader code at the GPU and it figures out dynamically which ones need to be vector and which ones should be shader at the moment in order to keep the GPU busy close to 100% of the time.
That's at least what I have gotten out of all the tech discussions regarding the 360 GPU.
seismologist
Banned
jimbo said:
The biggest advantage will go to wichever system the games are natively designed for. Ports to the odd platform will suffer the most.
For example if MGS4 is made for PS3 and ported to Xbox360. Unless they put alot of time into it, the result will be 10x worse than any trouble they had porting from PS2 to Xbox.
sonycowboy said:
Sorry I'm just tired but it has nothing to do with attacking anything just that I just keep having to repeat myself and it's bugging me because people keep twisting my words or aren't reading the article very carefully. Like right here, I never even implied that I believed next gen is going to be more CPU dedicated than GPU dedicated. And neither did they. They mentioned that gaming is going more GPU bound.
"but the majority of games these days are far more GPU bound than CPU bound, so the performance decrease isnt a huge deal""
well he probably likes hdd more than wireles lan, several usb ports, memory card readers etc.. I would agree with him in that regard if MS telling developers to use the HDD to enhance the gameplay. Otherwise HDD is kinda useless and a memory card/stick can do the job.
He is alsio very upset about MS not promoting 1080P.
He is alsio very upset about MS not promoting 1080P.
seismologist
Banned
After reading that article both systems sound exciting to me.
I like the possibilities of huge amounts of geometries on Xbox 360.
I'm guessing X360 will be good at rendering huge virtual worlds.
I also like the possibilities of how the 7 SPEs might be used in Cell.
It seems like the SPE's will be good at processing parallel streams of hi-def video. That could produce some interesting gaming applications as well.
I like the possibilities of huge amounts of geometries on Xbox 360.
I'm guessing X360 will be good at rendering huge virtual worlds.
I also like the possibilities of how the 7 SPEs might be used in Cell.
It seems like the SPE's will be good at processing parallel streams of hi-def video. That could produce some interesting gaming applications as well.
Shompola said:
True. I just read that statement a little more like MS spent the money on the important stuff while Sony spent a little more on fluff. He is a little upset over 1080P, but in all honesty the number of 1080P games down the road will probably be the same as the number of 1080i games on Playstation 2. Yeah, it'll happen, but probably not until 3 or 4 years after the system is out. 720P/1080I is the sweet spot for this generation. 1080P will be more important with the future batch of consoles (X720, PS4). Also, getting a game to run at 1080P will take much more work than tweaking a game for 720P. 720P is definitely the sweet spot for this generation.
"After reading that article both systems sound exciting to me.
I like the possibilities of huge amounts of geometries on Xbox 360.
I'm guessing X360 will be good at rendering huge virtual worlds.
I also like the possibilities of how the 7 SPEs might be used in Cell.
It seems like the SPE's will be good at processing parallel streams of hi-def video. That could produce some interesting gaming applications as well."
I don't care about hi-def video that much, but the thoughts of the physics and particle effects running at full speed on those 7 SPEs has me pretty excited. Also if someone can pull off a 1080p game on the PS3 without having to drop too much detail and simplify the graphics it should make for a really juicy viewing experience. Perhaps a fighting game. Or indoor, corridor-type games.
I just hope all that geometry and physics power relates into some better water for games this gen, because honestly WATER currently REALLY sucks when it comes to physics and its been the most dissapointing aspect of games for me this generation. Wave Race is still one of the most impressive examples of water done right and it was on the freaking 64. I want next gen Wave Race waves.
Also I don't think they were saying Sony spent money on fluff and short-changed the other stuff. Sony spend money on the important stuff too, just also chose to spend money on fluff on top of all that and MS didn't because they didn't believe it was worth it. So Sony is basically giving you more hardware even if some don't consider the money spent justified compared to the features it got us. But as a gamer you should be happy because you're just getting better. Their loss our win. Although to be honest I'd trade those extra USB ports and other inputs for a standard hard drive in a heart beat.
I like the possibilities of huge amounts of geometries on Xbox 360.
I'm guessing X360 will be good at rendering huge virtual worlds.
I also like the possibilities of how the 7 SPEs might be used in Cell.
It seems like the SPE's will be good at processing parallel streams of hi-def video. That could produce some interesting gaming applications as well."
I don't care about hi-def video that much, but the thoughts of the physics and particle effects running at full speed on those 7 SPEs has me pretty excited. Also if someone can pull off a 1080p game on the PS3 without having to drop too much detail and simplify the graphics it should make for a really juicy viewing experience. Perhaps a fighting game. Or indoor, corridor-type games.
I just hope all that geometry and physics power relates into some better water for games this gen, because honestly WATER currently REALLY sucks when it comes to physics and its been the most dissapointing aspect of games for me this generation. Wave Race is still one of the most impressive examples of water done right and it was on the freaking 64. I want next gen Wave Race waves.
Also I don't think they were saying Sony spent money on fluff and short-changed the other stuff. Sony spend money on the important stuff too, just also chose to spend money on fluff on top of all that and MS didn't because they didn't believe it was worth it. So Sony is basically giving you more hardware even if some don't consider the money spent justified compared to the features it got us. But as a gamer you should be happy because you're just getting better. Their loss our win. Although to be honest I'd trade those extra USB ports and other inputs for a standard hard drive in a heart beat.
jimbo said:
Superficially though, Cowboy's argument does seem to have some merit. With Nvidia, Cg, and OpenGL you'd think that PC developers like Epic and ID (Carmack especially given his love of the OpenGL standard) would be ready to hop all over the PS3 whether PC ports will be harder or not.
jimbo said:
jimbo said:
I'd be very hesitant to say that, certainly on the CPU side.
I think with both CPUs, there'll be things each does better than the other. That doesn't mean it's a wash, or they're equal as far as games goes. Just because one CPU does X,Y and Z well, and another does A,B and C well doesn't mean they're as good as one another. What matters is how much processing time A,B,C,X,Y,Z take vs each other. If one CPU is better at things which take a long time, there's the bigger difference.
Looking at Sweeney's comments, they seem to weigh more in favour of Sony's choices. Specifically with regard to Cell, I think being better at physics/simulation would give it a disproportionate advantage given the already large, and growing, proportion of CPU time physics takes. Afterall, when you want to optimise something (e.g. games) you start with what takes longest (e.g. physics).
Tenacious-V
Thinks his PR is better than yours.
jimbo said:
I don't agree with the bolded part there. We aren't getting a better deal, we really are getting fluff in this case. A HDD would have been MUCH more welcomed and useful out of the box than extra USB, 3 basically redundant controller connections, and a bunch of storage ports for different devices.... Sure they can say they have a HDD, but with it not being standard most devs will treat it as niche unfortunately. Just like they did this get with PS2's HDD...... So in that assessment I don't think Sony made the right spending choices. We don't need extra USB, and all those extra storage slots... We needed 1 HDD built in. So I agree with MS's decision there, their priority was better thought out in that case.
seismologist
Banned
jimbo said:
Would there really be much advantage running physics and particle effects on SPE's. Vs. doing it the traditional way?
I'm thinking of something more non traditional uses for hi-def video. I mean GT4 already uses photo's in the background. Why not use hi-def video instead. (not even sure if this is possible). But you never know.