Advantages of 360 & PS3 Graphics
- Thread starter GreekDNA
- Start date
- Status
MonkeyLicker
Newcomer
Look up joker454's posts.
He's talked a few times about the different strengths of each platform.
He's talked a few times about the different strengths of each platform.
Let me summarize his view:
Xenos > RSX
SPU > Xenon Core
I think it was in RSX Vertex limited thread -very good one- he started posting,
OP, I recommend you to read the discussion in that thread first.
Additional views: B3D is Sony Land, one cannot openly criticize PS3, or something among those lines.
Anyway, it's unfortunate yet understandable, but developers don't reveal much details of both consoles, so these kind of threads tend to end up hosting fanboy chatter.
First because there are a ton of threads on this very topic. Second because the information is contained in many various threads in regards to specifics. Third because it is close to going out of bounds. And forth...
It all comes down to what you are trying to do, how you are doing it, and what other available resources your project allows you (resources from the CPU, bus, memory footprint and bandwidth, etc).
There is more than one way to skin a cat. Developer skill/experience, budgets, available tools and middleware, and design are all going to factor strongly into how the hardware is utilized and exposed. The graphic subsystems aren't abstractions apart from the systems; they are closed box designs.
For those reasons I don't see this thread getting far. Now if an insider posted some direct information that is very to the point, e.g. the performance they are seeing in specific areas, it becomes a valid tangent. e.g. Code speedup that Joker has seen on SPEs, or the troubles Ubi has had with AI on such, etc... and how their implimentations perform and how this impacts them going forward... now that is interesting. Noting specific issues, e.g. how fillrate may impacting particle systems on one system, that has some technical merits.
But a general solicitation of such is better grasped by reading the already available material and then observing the trends in the market and discussing them as they appear. There really isn't much to be gained right now that couldn't be gained from reading posts from ERP, DeanoC, Faf, MrFloopy, nAo, Shootmymonkey, cloudspace, repi, joker, Fran, Graham, and a number of the other industry people and tech folks here.
What would be interesting... and won't ever happen... would be some specific code benchmarks. Very limited ones... like specific shaders (not 3DMark stuff)... with some details about specific resource utilization and how it impacts the remainder of the system load. But it all gets really complicated because you may become bound in one item quicker on one system, but have a lot of other available resources for other things while the other platform has a higher peak boundary but once that is hit has fewer resources to dedicate in other areas.
Ultimately, as this generation is now underway, things are going to come down to art on the graphic side and then the ability for devs to tap the resources available.
Artistic mastery and software are much more limiting factors this generation than any disparity between the two platforms' performance in regards to how pretty the end product is on screen.
It all comes down to what you are trying to do, how you are doing it, and what other available resources your project allows you (resources from the CPU, bus, memory footprint and bandwidth, etc).
There is more than one way to skin a cat. Developer skill/experience, budgets, available tools and middleware, and design are all going to factor strongly into how the hardware is utilized and exposed. The graphic subsystems aren't abstractions apart from the systems; they are closed box designs.
For those reasons I don't see this thread getting far. Now if an insider posted some direct information that is very to the point, e.g. the performance they are seeing in specific areas, it becomes a valid tangent. e.g. Code speedup that Joker has seen on SPEs, or the troubles Ubi has had with AI on such, etc... and how their implimentations perform and how this impacts them going forward... now that is interesting. Noting specific issues, e.g. how fillrate may impacting particle systems on one system, that has some technical merits.
But a general solicitation of such is better grasped by reading the already available material and then observing the trends in the market and discussing them as they appear. There really isn't much to be gained right now that couldn't be gained from reading posts from ERP, DeanoC, Faf, MrFloopy, nAo, Shootmymonkey, cloudspace, repi, joker, Fran, Graham, and a number of the other industry people and tech folks here.
What would be interesting... and won't ever happen... would be some specific code benchmarks. Very limited ones... like specific shaders (not 3DMark stuff)... with some details about specific resource utilization and how it impacts the remainder of the system load. But it all gets really complicated because you may become bound in one item quicker on one system, but have a lot of other available resources for other things while the other platform has a higher peak boundary but once that is hit has fewer resources to dedicate in other areas.
Ultimately, as this generation is now underway, things are going to come down to art on the graphic side and then the ability for devs to tap the resources available.
Artistic mastery and software are much more limiting factors this generation than any disparity between the two platforms' performance in regards to how pretty the end product is on screen.
Neither do I...
A reader's digest of the best bits in the hardware just isn't practical or informative. Every hardware feature is also limited by software, and in terms of achieved results, by complexity and costs. And if you're only interested in graphical effects, it's also irrelevant, for there's "more than one way to skin a cat." The advantages of one platform, like Xenos being able to target all shaders on vertex work, are offset by advantages in the other platform. The only clear advantage any one platform has is XB360 has more available RAM. Though some could say that's offset with PS3 having an HDD as standard allowing for more streaming of content.
In summary, strengths of the consoles cannot be measured in terms of graphical effects. No platform is inherently better at HDR or shaders because to use those functions have costs, and it's always a trade-off. If RSX can handle 64 bit floating point buffers, does that make it better at HDR than Xenos if Xenos can't? Not if those 64 bit buffers can't be used in a practical situation.
Still, rather than just lock the thread down, it can stay open. If anyone has anything informative to contribute to different hardware solutions for graphical problems, that'll be cool. If not, this thread will just drop off the radar.
Man, give it up for the time being.
I'm guessing you are fighting a spam filter for registereds, a fight you don't seem to be winning
.The Beauty and the Beast
From: http://playstation-disorder.com/comm...?blog_id=795&
From: http://playstation-disorder.com/comm...?blog_id=795&
Update 2007/08/20: Added two paragraphs at the end talking about the latest EA Madden porting issues
Update 2007/08/19: From a reference to this article on vgchartz.com I found this link to a thread on beyond3d.com where the concepts refered here have been exemplified by various statements from PS3 developers on the usage of SPE's within their games.
What is one of the most heard prejudice when you ask people about the PS3? I guess number one is "The PLAYSTATION 3 is difficult to program, more difficult than other platforms". Well, this statement as such is of course stupid. Compared to let's say the PS2, the PS3 is a big improvement. While you had to use assembler code on the PS2 for almost all advanced stuff, the PS3 basically just needs C++ or C code in most cases. This is even true for the Cell SPEs, the main differenciator.
But before we even start let me already make a short disclaimer here as the article itself as well as the material that gets referenced are not easily digestable
The Beast
Of course what people could mean by that statement is the fact that the next generation consoles as well as PCs are now based on multi-core architectures, i.e. several CPUs can be used at any time to run all sorts of parallel code to achieve what we finally call next-gen gameplay. But then again is no different for the various platforms.
The real difference becomes obvious when we look at the guts, i.e. the different CPUs used within the various platforms / consoles. Starting with the 360 we find the so-called Xenon CPU which features 3 symetrical PPC cores, being able to run two parallel threads at a time. Each core features an additional vector processing unit that can be used to do some fast vector operations. Those three cores typical multi purpose CPUs and are all programmed the same way using the same memory resources which makes the development of multi-threaded relativly easy, as there is only one programming model used.
The PC on the other side is very similar in a sence that the latest Intel and AMD chips are all using symetrical multi-cores (usually dual core technology) which basically leads to the same pros and cons in terms of generating multi-threaded code.
Looking at the Cell, things start to look a little different: one of the design goals of the Cell chip was to produce the best performance / size ratio at the time of production of all the chips being out there. This was achieved by breaking with a lot of design principles that have been used at that time. First of all the Cell is not fully symetrical in the way the Xenon or Intel/AMD chips are.
[img=500x104, 39,0Kb]http://static.flickr.com/1222/1147256521_5d0678e078.jpg[/img]
Fig 1. The PS3 Hardware Architecture (click for larger view)
The Cell features one dual-threaded PowerPC Element (PPE), very similar to one of the Xenon cores, that acts as a conductor for what seems to be the orchestra, a set of 7 Synergistic Processing Elements (SPE), which are symetric cores that are once again very much simplified compared to the PPE. Think of those SPE's as very fast and flexible vector units. Even though we are talking about 7 SPU's here, keep in mind that the PS3 dedicates one single SPU to the PS3's OS that handles a couple of background tasks (e.g. content download, etc.). Looking at Fig. 1, The PPE as well as the SPE'S are connected via a very fast Element Interconnect Bus (EIB) which allows the PPE to talk to the SPE's, as well as the SPE's to talk to each other. Finally, all memory access is going thru the EIB and various memory and IO controllers.
What's interesting to note: even though the PS3 has destinct memory for the Cell CPU as well as the RSX GPU, this memory is somewhat "unified" in way that the RSX can access the different memory areas rather fast with a peak of 20 GB/s. In that sense, the PS3 architecture combines the best of both worlds, separated memory areas for CPU and GPU (no concurrent access issues), still being able for the GPU to use the respective other memory area for its own purposes. The 360 for instance uses a unified memory model where there is no distinction between different memory types, but with the problem of concurrent access to the same memory from CPU and GPU. Again, from a programming point of view, the 360 is easier to handle as you don't have to make a decision where certain stuff needs to reside.
One other complexity that needs to be handled by the programmer is the data feed to and from the SPE's. As the SPE are very simplistic processing units, they only feature are relativly small local storage (LS) of 256 KB. This storage is used for the SPE's program running at one point in time on a SPE. The data that gets processed by the SPE needs to come from the memory via DMA commands that need to be issued just in time for the data to be ready for processing. The same is true for writing back stuff. The other option of course is that data will be consumed by another SPE after it has been processed by the first one.
[img=400x235, 27,8Kb]http://playstation-disorder.com/uploads/ps2arch.gif[/img]
Fig 2: The PS2's Emotion Engine
Let me do one last comment on the PS3 architecture before we dive into the Beauty section
The Beauty
So far we have seen the beast part of the PS3, a platform that has some programming challenges in terms of keeping the various SPE's busy and managing the different memory types efficiently. Let's now come to the beauty aspects of it: its flexibility. As you can already see in Fig 1. the RSX GPU is the other major component within the PS3 architecture.
The RSX chip is very similar to any other standard PC based GPU with the exception, that it can access both memory types (main and graphics) almost at the same speed. Beside that it most likely features (no official specs have been released so far) a graphics pipeline consisting of 24 pixel shaders, 8 vertex shaders, 8 ROPS (raster operation units), and 550MHz clock speed.
Looking at those shaders as very specialized vector units similar to the Cells SPE's, again one could think of off-loading some of the tasks within the GPU's graphics pipeline to the Cell. In other words, you can use the Cell to seamlessly extend the GPU's capability and power. Let's take a look at two examples where this can be seen in real life in order to see that this is not just theory.
The first example is the already discussed "Deferred Rendering in Killzone 2" presentation that was given at the develop conference in Brighton this July. Check out my article on that one if you need more details.
The second example I just found on the SCEA research website a couple of days ago. The article "Deferred Pixel Shading on the PLAYSTATION®3" perfectly describes the benefits of what we just discussed: the Cell providing additional pixel shaders to the graphics pipeline that are even more powerful than the RSX build-in pixel shaders. In a nutshell the result of the exercise (the implementation of a certain soft shadowing alogorithm) described in the article is that 5 SPE's basically acting as pixel shaders achieve the same overall performance as the 24 pixel shaders used on a NVIDIA GeForce 7800 GTX running on Linux based system
Conclusion
Similar to the PS2, a platform that even for today's standards surprises us with visual qualities we wouldn't have expected from the console in the first place, the PS3 is a very flexible gaming platform. Almost nothing is carved in stone and it is really up to the developers to unleash the power of the architecture. This makes all sense as opposite to the PC world, hardware improvements are not an option for console makers. With a lifecycle of at least 10 years, the PS3 needs to be still capable of running new algorithms and methods which we don't even know by now.
Those are just two examples of how the combination of Cell and the RSX can achieve things that sometimes even the designing engineers havn't thought about. A great outlook for the future if you ask me.
This is not possible with a fixed hardware architecture that is designed to make life easier for developers. Sure, on the PC side you can easily extend your hardware, for consoles that's not an option. Or maybe we will see another Xbox just in couple more years, something that is most likely to happen for the Wii anyhow.
One last comment on those various statements regarding this let's say mediocre Madden port to the PS3. First of all, I don't think that this will have an impact on PS3 sales as one analyst has predicted. Second I think this is more of an EA issue, showing once again, that those guys don't get the necessary resources to address the respective platforms appropriately. Not knowing what the issues really are but it seems to be a perfect example where developers didn't have the knowledge or the skills to do a proper PS3 port of the game. I almost looks like as if the whole code is just sitting on the PPE whichI already said is similar to the Xbox Xenon cores. No need for them to do any adaption at all. Again, I don't know the real story. Is this going to happen more often in the future? Usually no. Again, similar to the PS2's VRAM issues (just 4MB graphic RAM) where developers had to learn that due to the PS2 architecture there is actually no need for a larger VRAM, best practices will be shared within the developer community.
This time things are also a little different: opposite to the PS2 days, where both the Emotion Engine & the Reality Synthesizer where very exotic devices only used within the PS2, both the Cell and the RSX (aka G70 based) are already well known chip and are supported otherwise by IBM and NVIDIA. In other words, knowledge will be spread a lot faster compared to the PS2. It just requires the EAs and Ubisofts of the world to apply those best practices within their own developments. What would be even more stupid, if those companies would not be able to share already available knowledge within their own company. Maybe the GRAW guys should talk to the SC Conviction team?
Additional Links
"Introduction to the graphics pipeline of the PS3" (PDF)
Presentation from Eurographics 07 by Cetric Perthuis, SCEA
"CELL: A New Platform for Digital Entertainment" (HTML)
Presentation by Dominic Mallinson & Marc DeLoura SCEA
Cell Software Model (PDF)
IBM programming course
"Deferred Pixel Shading on the PS3" (PDF)
Article by Alan Heirich and Louis Bavoil, SCEA
"Deferred Rendering in Killzone 2" (PDF)
Presentation from develop conference 07 by Michal Valient, Guerrilla
Last edited by a moderator:
This is quite an interesting paragraph:
Can anyone comment on the accuracy of this statement? Can cells SPU's really equal the pixel shading ability of a 7800GTX or are we talking about a very specific case here?
If this statement were true for pretty much any circumstance then the PS3 must be in 8800 territory in terms of raw pixel shading power.
archangelmorph
Veteran
Not sure how accurate this is..
However it DOES make an interesting proposition for someone looking to do PS3 linux based homebrew (meaning with a good Cell based software rendering system you probably don't need RSX access..)
While I'm sure you could do some nice things with the Cell for graphics, I'd still much prefer it if in homebrew I could go crazy with calculations on the Cell, and then visualise the results with the RSX. That just seems more convenient.
But I'm sure there are some interesting things we could do without the RSX.
But I'm sure there are some interesting things we could do without the RSX.
I'm one of the ones who think that thread won't have a long and peaceful life...lol
For the article, well I guess it's biased as hell.
It always the same, the ps3 is supposed to have more room tha n the 360.
And you know what nobody needs that kind of article because it's clear that the ps3 has more muscles.
But the wrong part is about is that that kind of articles (they come often with different arguments) always implie that the 360 as this point is almost topped witch is really wrong.
Joker454 who is one of the few devs here that works on the 2 platforms stated recently that making the most of 360 cpu is even less trivial than making the most of the cell, the spu swallow easily almost anything that you throw at them.
So far the xenon had maybe feel more 'familiar" to dev that doesn't mean that they used it to its best what could prove really tough.
More a lot less 360 games have used engine who aim at easy portability between systems, and on some upcoming title it hurts more than the ps3 extra muscles.
Bungy will come fine with Halo III but I've never feel that this team is on top when it comes to coding ability (not that they're bad ).
The only 360 team that could match some top first party or third party working for Sony is Rare who actually is really under used in regard to theirs coding prowess.
PS3 has extra muscles will see how much this weight in overall perception of both systems prowess.
Shall I tell it again mods close this thread we 've been through that a hell too much number of times, it's better to discuss strengh of each systems in seperate threads.
For the article, well I guess it's biased as hell.
It always the same, the ps3 is supposed to have more room tha n the 360.
And you know what nobody needs that kind of article because it's clear that the ps3 has more muscles.
But the wrong part is about is that that kind of articles (they come often with different arguments) always implie that the 360 as this point is almost topped witch is really wrong.
Joker454 who is one of the few devs here that works on the 2 platforms stated recently that making the most of 360 cpu is even less trivial than making the most of the cell, the spu swallow easily almost anything that you throw at them.
So far the xenon had maybe feel more 'familiar" to dev that doesn't mean that they used it to its best what could prove really tough.
More a lot less 360 games have used engine who aim at easy portability between systems, and on some upcoming title it hurts more than the ps3 extra muscles.
Bungy will come fine with Halo III but I've never feel that this team is on top when it comes to coding ability (not that they're bad
).The only 360 team that could match some top first party or third party working for Sony is Rare who actually is really under used in regard to theirs coding prowess.
PS3 has extra muscles will see how much this weight in overall perception of both systems prowess.
Shall I tell it again mods close this thread we 've been through that a hell too much number of times, it's better to discuss strengh of each systems in seperate threads.
- Status
Similar threads
