New Technical Information About RSX???...But In Japanese

Differences are more than just that. Stick a PS on RSX with dynamic branching and see how it's 'more shader power' fares against Xenos with the same shader...

Performance is but one metric. Features is another.

You can say that about anything on either system. They both have strengths and weaknesses.
Console developers play to the strengths of each console.
 
Acert is speaking though to the ability of R500 to more readily adapt and adopt some of the more 'sophisticated' pixel shading directions things will start to go in when DX10 hits... so it's not that he's saying Xenos is suddenly more powerful, simply that there are some shading instructions it may be able to handle that RSX simply won't be able to.

What difference would an abstraction like DX10 have on R500 development? More specifically, having been developing around the architecture for approximately a year, what has been the impediment to implementing those "sophisticated pixel shad[ers]"? Furthermore, even compensating for the learning curve, is there any indication that there will be such an increase attributable to the architecture?
 
You can't tap performance that isn't there.

RSX has more shader power and developers are using XDR and FlexIO to overcome bandwidth bottlenecks.

Xenos has the advantage of EDRAM for bandwidth (although it has less GPU to CPU bandwidth) but less shader power.


however, Xenos has more a modern shader level, closer to that of shader model 4.0 something RSX cannot boast. also, without the help of Xenon or CELL CPUs, the Xenos can produce more polygons than RSX can.
 
What difference would an abstraction like DX10 have on R500 development? More specifically, having been developing around the architecture for approximately a year, what has been the impediment to implementing those "sophisticated pixel shad[ers]"? Furthermore, even compensating for the learning curve, is there any indication that there will be such an increase attributable to the architecture?

Well Xenos isn't a DX10 part, I'm sure we can both agree on that, but you may be misunderstanding what I'm saying. I think it would be fair to call it a 'pre' DX10 part. The point is though, that it's not just a matter of devs having had a year to familiarize themselves with it... sure, this is the case when looking at first-party devs maybe, but the majority of the (Western) third/second-party support MS enjoys is coming from PC-familiar devs, and in that sense their approaches tend to reflect PC thinking.

As DX10 comes on the scene - and even now - the trend seems such that it should shift increasingly towards dynamic branching in shaders, which Xenos is well-equiped to handle. As those practices take hold, they will see use on Xenos where RSX is somewhat constrained in that respect.

Now, to what extent these practices will take hold, and what ultimately it will mean for visuals on the screen vs more 'conventional' methods, we just have to wait and see. But I was simply reasserting Acert's position that although RSX has the 'power,' there are aspects of Xenos' architecture we're not really seeing come into play yet.
 
Now, to what extent these practices will take hold, and what ultimately it will mean for visuals on the screen vs more 'conventional' methods, we just have to wait and see. But I was simply reasserting Acert's position that although RSX has the 'power,' there are aspects of Xenos' architecture we're not really seeing come into play yet.

XBD has it. It really is all up to devs and devs on both sides of the fence are gonna do some great stuff. Personally, I think games are more art constrained than anything else. Technical excellence is only one aspect of graphics. Pairing up great technology with the right art style is important, likewise asset quality. TF2 on Source is a great example of such, likewise Trusty Bell.

Anyhow, topic has been beaten to death. Time will tell one way or another, and if companies keep leveraging middleware it may be years before we get a solid indication of anything!
 
Megadrive,

Remember that Barbarian told us the RSX has had a few shader instructions added. Perhaps that will make it a tad closer to Shader Model 4.0.
 
Perhaps that will make it a tad closer to Shader Model 4.0.
No, not by any stretch. Well, I guess it depends on your definition of 'tad'. But there's no point thinking RSX's shaders are SM3.5 or the sort because of a few more instructions.
 
I dont think they say Xenos lacks shader power at all...

It says, *some* developers are having poor alu perfomance... Going by that could be a million factors causing this, including the concurrency in shaders, because if i remember correctly xenos is also a heavy threadly design...

I also remember a slide of an Epic presentation where they claimed they resolved concurrency issues in their shaders allowing to use much more alu ops.

Since its a japanese web site, i'm guessing thats japanese developers talking, and judging by the avarage 360's games coming from there I would say they are not using any shaders at all...

Thats of course a problem Ms needs to adress, it should be easy to develop for X360, and achieve at least a mininum graphical bar, at least its what she claims...
 
Yes, remember, Capcom haven't had any problems with shading capacity, but rather with fillrate and textures. Although you also have to admit that Lost Planet isn't making such a big use of shaders on most surfaces, and the wide open enviroments and many effects do usually stress the parts of the GPU they are having problems with. So it's more likely that bottlenecks depend on the design and goals of each particular engine.
 
Personally, if shading is such a special strength of the RSX then I would like to know what exactly about it's shading ability is enhanced or unique. We recently learned it had some extra texture lookup and fast vector normalize instructions added. But is this the extent of what makes it so good at shading?
 
Nothing about the RSXs shading capability is enhanced or unique - it is simply that by putting 24 NV47-class pixel pipes in there, it has the ALU advantage over Xenos in that regard.
 
I'm surprised they had problem with textures..

Not sure if it already has been posted here:

http://ps3forums.com/showthread.php?t=30662

Some stuff....

Hurdles involving the Performance of the Cell:

There are some challenges involving the architecture of the Cell, the cell consists of PPE and SPE cores. A developer states, "It is impossible to extract the full performance of the Cell on launch titles, it will take time get familiar with it". Another developer states that they are having difficulties with the 256KB memory of each SPE core. The actual useable area of the 256KB is closer to 128KB when buffering is considered with accessing external DRAM. "It would have been much different situation if there was 1MB of local memoryâ€￾. There is however a benefit for these restrictions on local memory, since latency can be reduced, and latency cycles are more easily read. This is an advantage for real-time gaming applications.

Till now, developers were stingy with programming and memory usage, and this will not change with the PS3, in fact, the Cell will reward developers that put more effort into programming. While that may not be a negative, it is a hurdle that will take larger developer resources and time. For the Cell it has changed from extracting performance from the hardware, but more towards multi-threaded performance and takes a different skill set then the previously.

RSX Memory Bottleneck

Developers were using 7800GTX for development, The RSX uses Nvidia’s G70 and performance programmable shader performance is very high. But the memory interface is 128-bit, in addition 8 ROP (Rasterizing Operation). It can be said that the RSX has a shader equivalent of a high-end PC with mid-range memory bandwidth. For that reason, due to the GPU high shader performance there is a bottleneck to the ROP memory and is causing a bottleneck. “For lower resolutions it is a fantastic GPU, but it gets difficult for high end HDTV resolutionsâ€￾, says a developer.

The biggest impact is the HDR and FSAA, the memory bottleneck becomes hard for PC levels of HDR and FSAA, to overcome this hurdle developers are using memory from the Cell for textures, and using FlexIO as a texture lead to reduce GPU bandwidth issues.

Developers are having exactly opposite problems then that of the PS3 and 360 as far as GPU performance is concerned. While there memory bandwidth issues with the PS3, they have great shader performance. For the 360, they have little memory bandwidth issues with the 10MB eDRAM, and less FSAA issues. But some developers are having issues with a lack of shader ALU performance and threading resources, however performance will increase as developers get more familiar with unified shader architecture.
 
If you don't have enough bandwidth for HDR+AA you can be more clever and use less bandwidth
If you don't have enough fillrate for your uber particles effects you can be more clever and use less fillrate.
If you don't have enough computational power for your shaders you can be more...no sorry, you're f&%£"d :)
 
If you don't have enough bandwidth for HDR+AA you can be more clever and use less bandwidth
If you don't have enough fillrate for your uber particles effects you can be more clever and use less fillrate.
If you don't have enough computational power for your shaders you can be more efficent and/or clever :)

Fixed it for you bud.
 
Back
Top