RSX = Stream Processor!?!
- Thread starter "Nerve-Damage"
- Start date
Titanio said:
Well, I never said G71 but left it open at G7x. No one seems to be sure on what G71 is. All these refreshes could be traced back to G70. But the subtle point here, without getting into semantics, is that given a future refresh of G70, RSX would be derived off that refresh. Implication being it would inherit any architectural changes, if any... in addition to any low level features exposed by the PS3 specific API...
I guess I'm more asking the question: couldn't RSX be a unique refresh of its own? Perhaps indeed sharing some insight gained in other refreshes (G71), but mostly an independent affair. Instead of thinking - first came G70, then G71, and then RSX as a little off-shoot - why not, first came G70, then G71 and RSX in parallel, as two seperate, if perhaps overlapping siblings?
It's a subtle difference, perhaps
It's a subtle difference, perhaps
!eVo!-X Ant UK
Banned
Titanio said:I guess I'm more asking the question: couldn't RSX be a unique refresh of its own? Perhaps indeed sharing some insight gained in other refreshes (G71), but mostly an independent affair. Instead of thinking - first came G70, then G71, and then RSX as a little off-shoot - why not, first came G70, then G71 and RSX in parallel, as two seperate, if perhaps overlapping siblings?
It's a subtle difference, perhaps
You have just said what i did on the 3rd post on this page
!eVo!-X Ant UK
Banned
Titanio said:Kind of, yes, lol. Sorry for the redundancy, that'll learn me for not reading threads more thoroughly
Its ok, you know great minds think alike
Eureka!
I think I've put the whole puzzle together:
I believe the RSX is this new streaming GPU that we've been seeing talked about in all of these patents, but it's streaming architecture arises not from spe-like units built into the GPU itself. Rather, it is a virtual streaming processor GPU in that the special low level inter-functionality between it and Cell, along with the flex-io bus, allow this specialized G 7.x variant to perform as if it had SPEs of its own.
It would make sense that RSX can lock SPEs (or even load balance them with the CPU/PPE core). This would make sense of the reusable architecture patent Nerve-Damage so kindly posted for us. True, the RSX probably does not have 4 SPEs of its own physically...but it can probably borrow and lock up to 4 (or 6 or whatever) of the CELL's SPEs as a virtual extension of itself.
The archticture Sony had originally imagined probably looked like so:
CPU/PPE core<--->[Pool of 8 SPEs, 1 reserved for redundancy (or something else)]<--->GPU/RSX core (specialized G 7.x variant)
The SPEs act as a processing bridge and can do pre and post graphics work.
I think I've put the whole puzzle together:
I believe the RSX is this new streaming GPU that we've been seeing talked about in all of these patents, but it's streaming architecture arises not from spe-like units built into the GPU itself. Rather, it is a virtual streaming processor GPU in that the special low level inter-functionality between it and Cell, along with the flex-io bus, allow this specialized G 7.x variant to perform as if it had SPEs of its own.
It would make sense that RSX can lock SPEs (or even load balance them with the CPU/PPE core). This would make sense of the reusable architecture patent Nerve-Damage so kindly posted for us. True, the RSX probably does not have 4 SPEs of its own physically...but it can probably borrow and lock up to 4 (or 6 or whatever) of the CELL's SPEs as a virtual extension of itself.
The archticture Sony had originally imagined probably looked like so:
CPU/PPE core<--->[Pool of 8 SPEs, 1 reserved for redundancy (or something else)]<--->GPU/RSX core (specialized G 7.x variant)
The SPEs act as a processing bridge and can do pre and post graphics work.
Last edited by a moderator:
This snippet stolen from the playstation boards inspired me to post this:
"The special characteristic of the PS3 is the connection between Cell and RSX
The special characteristic of PS3 Graphics is the connection between Cell and RSX. The RSX itself has a similar architecture to the G70, but the host interface for the G70 is meant for the PC and is completely different. The G70 uses PCI Express x16 to connect to the chipset as 8GB/sec (4GB/sec one-way), and it cannot directly access main memory. In contrast, the RSX has a 35GB/sec (20GB/sec down, 15GB/sec up) direct connection to the Cell, and can directly render from the main memory on the Cell side.
This is a big difference, because it allows a completely different way of using the GPU from PC architectures, SCEI explained. First of all, because the bus is wider, the Cell can perform a great amount of geometry operations, then send the vertex data [to the RSX]. Conversely, the RSX side can easily send data back to the Cell.
“The Cell processor can do both pre-processing and post-processing. For example, tessellation, dot filling, etc… Cell can perform physics processing like collision and motion calculations, and transform the vertex array.†said David B. Kirk, Chief Scientist of nVidia.
SCEI basically expects higher abstraction levels to be processed by Cell, and the details (like vertexes and pixels) to be processed by the GPU. The is reasonable – for example, in the case where the CPU side handles geometry transformation, collision detection, which is important in games, is not a problem. In the case where the GPU handles geometry transformation, if the data is not sent back to the CPU, clipping issues may occur. In the case of the PS3, the Cell side can perform transformations, and even if the GPU is used for transformation, it is comparatively easy to send the data back to the CPU side.
In architectures up to now, either the CPU or the GPU have been the bottleneck. It this is not resolved, we cannot go any further. To face this, in PS3 architecture, if the GPU becomes the bottleneck, it can shift work to the Cell, if the Cell becomes the bottleneck it can send work to the GPU, shifting the workload. For example, according to the software, the Cell side can perform more graphics processing, or, oppositely, or easily make an adjustment to leave the graphics work to the GPU, it was explained. In summary, between the CPU and GPU programmable processors, a flexible balance adjustment can be done.
In previous PC architectures, because they were limited by the CPU<->GPU pipe, geometry operations were held to a certain limit, and how rich an environment you can create within that limit became the main technical challenge. In contrast, the PlayStation2-type game consoles created large amounts of polygons, but after that it did not have the expressiveness of PCs. (Trans. note: probably means that PS2 is less capable in applying different effects to polygons than the PC, despite pumping out more polygons.) In the case of the PS3, both are possible, with the flexibility to balance the two.
However, in the case of the currently available PS3 Evaluation System, because of restrictions in the architecture, it is not possible to evaluate the balancing [of Cell and RSX]. This is a difficulty and a weakness, but, if we state it differently, software demos on the current systems still do not demonstrate the full potential of PS3. It is possible that the actual PS3 will have performance greater than current demos.""
"The special characteristic of the PS3 is the connection between Cell and RSX
The special characteristic of PS3 Graphics is the connection between Cell and RSX. The RSX itself has a similar architecture to the G70, but the host interface for the G70 is meant for the PC and is completely different. The G70 uses PCI Express x16 to connect to the chipset as 8GB/sec (4GB/sec one-way), and it cannot directly access main memory. In contrast, the RSX has a 35GB/sec (20GB/sec down, 15GB/sec up) direct connection to the Cell, and can directly render from the main memory on the Cell side.
This is a big difference, because it allows a completely different way of using the GPU from PC architectures, SCEI explained. First of all, because the bus is wider, the Cell can perform a great amount of geometry operations, then send the vertex data [to the RSX]. Conversely, the RSX side can easily send data back to the Cell.
“The Cell processor can do both pre-processing and post-processing. For example, tessellation, dot filling, etc… Cell can perform physics processing like collision and motion calculations, and transform the vertex array.†said David B. Kirk, Chief Scientist of nVidia.
SCEI basically expects higher abstraction levels to be processed by Cell, and the details (like vertexes and pixels) to be processed by the GPU. The is reasonable – for example, in the case where the CPU side handles geometry transformation, collision detection, which is important in games, is not a problem. In the case where the GPU handles geometry transformation, if the data is not sent back to the CPU, clipping issues may occur. In the case of the PS3, the Cell side can perform transformations, and even if the GPU is used for transformation, it is comparatively easy to send the data back to the CPU side.
In architectures up to now, either the CPU or the GPU have been the bottleneck. It this is not resolved, we cannot go any further. To face this, in PS3 architecture, if the GPU becomes the bottleneck, it can shift work to the Cell, if the Cell becomes the bottleneck it can send work to the GPU, shifting the workload. For example, according to the software, the Cell side can perform more graphics processing, or, oppositely, or easily make an adjustment to leave the graphics work to the GPU, it was explained. In summary, between the CPU and GPU programmable processors, a flexible balance adjustment can be done.
In previous PC architectures, because they were limited by the CPU<->GPU pipe, geometry operations were held to a certain limit, and how rich an environment you can create within that limit became the main technical challenge. In contrast, the PlayStation2-type game consoles created large amounts of polygons, but after that it did not have the expressiveness of PCs. (Trans. note: probably means that PS2 is less capable in applying different effects to polygons than the PC, despite pumping out more polygons.) In the case of the PS3, both are possible, with the flexibility to balance the two.
However, in the case of the currently available PS3 Evaluation System, because of restrictions in the architecture, it is not possible to evaluate the balancing [of Cell and RSX]. This is a difficulty and a weakness, but, if we state it differently, software demos on the current systems still do not demonstrate the full potential of PS3. It is possible that the actual PS3 will have performance greater than current demos.""
overclocked
Veteran
ROG27 said:
Thats the whole design, RSX has Cell and vice verse.
That's Right!
Well I always knew there was more to the RSX than simply a G70 complaint GPU. Sony always customized their GPU's and RSX would be no different. Just a larger amount of memory and a wider data bus are PCish GPU upgrades not Sony's way of thinking. Seems the smarty pants in this forum could not wrap there mind around the fact. In fact some are still stuck on RSX being completely based on some design primary based on the now out dated G70 GPU.
Well I always knew there was more to the RSX than simply a G70 complaint GPU. Sony always customized their GPU's and RSX would be no different. Just a larger amount of memory and a wider data bus are PCish GPU upgrades not Sony's way of thinking. Seems the smarty pants in this forum could not wrap there mind around the fact. In fact some are still stuck on RSX being completely based on some design primary based on the now out dated G70 GPU.
Last edited by a moderator:
leechan25 said:
Well I'm one of them, and I just think that the G70 is an awesome architecture that is not lacking for the PS3. I guess we always want more, but I really don't see more features, or more pixel processing capability is going to add much more. An evolutionary refinement is not going to bring a revolutionary difference in graphics quality to the screen. I would sooner see an extra 256 MB of memory over a DX10 capable GPU, as I think being able to hold more art assets in memory is much more useful for games, than some imperceptible rendering increase.
Getting back to CELL + RSX colaboration, how do you guys think load balancing will be done? In software, hardware, or programmer controlled?
Wait, now I just have to ask - what's the 'confirmation' people are reading that I'm not stating RSX is this new uber-architecture? I'm not sure, but I'm not seeing any debates as having been decisively won or lost yet.
Barbarian's comments - if taken at face-value - rather lead us to an in between place where perhaps we have something more novel than a G70 adaptation, but certainly his 'non-exotic' comments should moot the theory of multi-GPU chips or SPE's attached, which for some reason we see the proponents of claiming victory.
Barbarian's comments - if taken at face-value - rather lead us to an in between place where perhaps we have something more novel than a G70 adaptation, but certainly his 'non-exotic' comments should moot the theory of multi-GPU chips or SPE's attached, which for some reason we see the proponents of claiming victory.
xbdestroya said:Wait, now I just have to ask - what's the 'confirmation' people are reading that I'm not stating RSX is this new uber-architecture? I'm not sure, but I'm not seeing any debates as having been decisively won or lost yet.
Barbarian's comments - if taken at face-value - rather lead us to an in between place where perhaps we have something more novel than a G70 adaptation, but certainly his 'non-exotic' comments should moot the theory of multi-GPU chips or SPE's attached, which for some reason we see the proponents of claiming victory.
well most people have been saying that that could not be a multi-core GPU (when looking at the patents it could be) and it have nearly no Sony input (when Sony never allow a GPU in their product without their input)! Some of you guy even disregard the patents as proper evidence for a multicore GPU or anything other than a G70.
xbdestroya wrote...
"But those moves (and patents) are from like four/five years ago. Waaaay before any of this NVidia stuff. And that's nto to say that the NVidia thing was a plan B either *or* to say that they aren't using any of the Cell tech in RSX. But for you to feel so convinced of it based on some patents and discussions from 2001... I mean, don't you think your position lacks evidence here?"
Few of us have been crying out ( me, version, !eVo!-X Ant UK, "Nerve-Damage" and more) from the wilderness that we think the RSX in something more than just a beef up G70 GPU. from all sides (technical, sony's culture, Cell's partnership, and others) only the get dis on by the smarty pants. Now more info out and those smarty pants are scratching their head trying to think up some way to discredit these patent again.
Leechan here's my challenge to you: having read through the patents as I'm sure you have, provide for us the detailed version of what you think the RSX will resemble. What in your own words is the RSX going to comprise of architecturally?
A lot of patents and stuff get posted followed by what essentially amounts to "See!"
But see what? Everyone posting always assumes there's something there that someone else is going to verify or tell them about. But yet they're *sure* it validates their theories, either way. So now in your own words, with the massive evidence you have at your disposal, please tell us exactly what the RSX is, since apparently it is clear as day to you.
PS - I don't lump you Version Nerve and !eVo-X into the same category by the way. You come off as especially defensive and reactionary to any non-"faithful" views
A lot of patents and stuff get posted followed by what essentially amounts to "See!"
But see what? Everyone posting always assumes there's something there that someone else is going to verify or tell them about. But yet they're *sure* it validates their theories, either way. So now in your own words, with the massive evidence you have at your disposal, please tell us exactly what the RSX is, since apparently it is clear as day to you.
PS - I don't lump you Version Nerve and !eVo-X into the same category by the way. You come off as especially defensive and reactionary to any non-"faithful" views
Last edited by a moderator:
"Nerve-Damage"
Regular
xbdestroya said:Wait, now I just have to ask - what's the 'confirmation' people are reading that I'm not stating RSX is this new uber-architecture? I'm not sure, but I'm not seeing any debates as having been decisively won or lost yet.
Barbarian's comments - if taken at face-value - rather lead us to an in between place where perhaps we have something more novel than a G70 adaptation, but certainly his 'non-exotic' comments should moot the theory of multi-GPU chips or SPE's attached, which for some reason we see the proponents of claiming victory.
Why do people feel revolutionary means dual core or multi-core anything?
Don’t get me wrong…I like the Cell processor…but I don’t find it to be a revolutionary CPU.
Rather I find it to be something that was expected or time coming with multiple cores CPU. It’s more evolutionary than anything else……..
So why would the RSX be any different *if* it happens to be a multi-core GPU? I.E. comprised of a NVIDIA based core and a specific Sony IP sub-core.
Man evolving from a banana...that would be revolutionary (and shocking
). A multi-core GPU isn’t by no means exotic or revolutionary…IMO!!
