X360 Programming Bottleneck

aldo

Newcomer
Both are friendly, but the PS3 may actually be friendlier, reports one of out PS2 and Xbox 360 development sources, who wished to remain anonymous. People are finding that the [xbox] 360 has some... bottlenecks. - November 2005 issue PSM
ralexand said:
So what's this bottleneck between the frontside bus and the edram unit?
Some of us are here because of the technical expertise that many of the members have to offer. This post is not here to incite anyone, however, if it does then feel free to respond with an informed and civil technical response. No need to throw out offensive knee-jerk responses. If I wanted that I could visit just about every other console forum on the web.

Many of us are in here just biting on any news regarding the differences and capabilities of the next-gen consoles. If the latest news is pro-PS3, great. If it’s Pro-X360 then great as well. This last tidbit just happened to be PS3, and most of us know that there may be nothing to it. Fine. Some posters are even going way out of their way to be inoffensive by buffering their comments with "take this with a grain of salt", to the extent that I believe we are the first 100 Google results for the phrase. So there is no need to get riled up and post useless statements like, "More PS3/X360 fanpunk fodder". If your emotional connection to a specific console leads to uncontrollable outbursts then please avoid these threads. There are even forums that are better suited for that type of behavior.

So without further ado, can we discuss these possible programming bottlenecks in the X360 and feel free to touch on possible PS3 bottlenecks as well. (Would love to hear from DeanoC on any nagging PS3 bottlenecks he has encountered.)

Again my understanding/belief is that the X360 is much easier to program for, but is it possible that it may be easier when programming in single core on each console, but multi core might be easier on PS3. Just an ignorant's query tossed out here for discussion which is to be taken with a grain... ;)

-aldo
 
I don't think your supposed to access the edram by the frontside bus at all is it? even if you could I don't see any advantage of doing so
 
pegisys said:
I don't think your supposed to access the edram by the frontside bus at all is it? even if you could I don't see any advantage of doing so


The EDRAM is accessed internally inside the GPU to do the AA and a few more things important to game designers. its not connected to anything outside the GPU
 
We still don't know the GPU's are equal.

The one with the strongest GPU will probably have teh best gfx, the CELL will help out with some physics, maybe AI in some cases, but I don't think it's going to make a big difference in on screen visuals.

If the Xenos GPU proves to outpower the RSX then I would think the X360 will actually be the more powerful machine as far as visuals go, or vice versa if RSX proves to be superior to Xenos.

But even then I would think they're gonna be within spitting distance of eachother.
 
The thread originator respectfully requested that people did not reply with comments that are likely to incite heated arguments; this thread has been pruned to remove such comments. Please stick to discussing the actual points at hand.
 
i think you forgot to watch the videos of Carmack and Nevell saying the opposite?
They never said anything to the opposite. I think it's just bad wording referring to it as a "programming bottleneck." As far as bottlenecks at the hardware level, yeah, they both will have enough bottlenecks to talk on about for a lifetime.

In general, if you're taking an existing single-threaded codebase, and trying to bring it up to next-gen changing as little as possible, I'd probably have to think the 360 has an advantage over the PS3. A lot of the things that could be referred to as bottlenecks, could be more accurately referred to as "double-edged swords", I think.

e.g. shared cache in 360 can be nice and pretty when it comes to passing shared data between threads. However, it can be a hassle with unshared data and threads start evicting lines belonging to other threads. PS3 has a counter to the latter aspect in that each of the SPEs has its own local SRAMs, but then shared data between threads has to be moved around on a sub-par-speed ring bus.

If we're talking about double-edged swords, then I'd have to say that pretty much every single piece of hardware in both consoles is a double-edged sword.
 
c0_re said:
OK so were [sic] taking quotes from PSM now? I"m sure some [one] could post comments just like that from an Xbox magazine from an anonymous developer.
Why haven't they?
c0_re said:
The developers that have had the [testicular fortitude] to come out publicly have stated the 360 will be easier to develop for period.
Now that they are actually working with both systems do they still state this? Links? Sources?

-aldo
 
Neeyik said:
The thread originator respectfully requested that people did not reply with comments that are likely to incite heated arguments; this thread has been pruned to remove such comments. Please stick to discussing the actual points at hand.

There's an entire thread like 2 months odl dedicated to bottleknecks on both consoles.

Nothigns changed since then to speak of, except one comment from a PSM magazine without a source. For this we should start a new debate?
 
I just want to know the RSX specs....give me that and I would be happy....

What could Sony be hiding?

-Josh378
 
aldo said:
Some of us are here because of the technical expertise that many of the members have to offer. This post is not here to incite anyone, however, if it does then feel free to respond with an informed and civil technical response. No need to throw out offensive knee-jerk responses. If I wanted that I could visit just about every other console forum on the web.

Many of us are in here just biting on any news regarding the differences and capabilities of the next-gen consoles. If the latest news is pro-PS3, great. If it’s Pro-X360 then great as well. This last tidbit just happened to be PS3, and most of us know that there may be nothing to it. Fine. Some posters are even going way out of their way to be inoffensive by buffering their comments with "take this with a grain of salt", to the extent that I believe we are the first 100 Google results for the phrase. So there is no need to get riled up and post useless statements like, "More PS3/X360 fanpunk fodder". If your emotional connection to a specific console leads to uncontrollable outbursts then please avoid these threads. There are even forums that are better suited for that type of behavior.

So without further ado, can we discuss these possible programming bottlenecks in the X360 and feel free to touch on possible PS3 bottlenecks as well. (Would love to hear from DeanoC on any nagging PS3 bottlenecks he has encountered.)

Again my understanding/belief is that the X360 is much easier to program for, but is it possible that it may be easier when programming in single core on each console, but multi core might be easier on PS3. Just an ignorant's query tossed out here for discussion which is to be taken with a grain... ;)

-aldo


Why, what a shocker..coming from PSM lol
 
Josh378 said:
I just want to know the RSX specs....give me that and I would be happy....

What could Sony be hiding?

-Josh378


Has someone said theyre hiding anything? Havent all teh specs been published already?

J
 
expletive said:
The bandwidth numbers, frequency, transistor count, all that stuff at E3 was what?

J
The given information tells us nothing about what's different from G70.
 
aldo said:
Again my understanding/belief is that the X360 is much easier to program for, but is it possible that it may be easier when programming in single core on each console, but multi core might be easier on PS3. Just an ignorant's query tossed out here for discussion which is to be taken with a grain...

Since they do not conform to a common standard (like PCs), I doubt we can compare them decisively. They are apples and oranges. ;)

The real problem comes when the end system is programmable, rather than embedded. In this case, only generalizations can be made about the degree of parallelism available in either the data or the algorithms. The ideal of simply routing whatever might come along through one sufficiently fast CPU is no longer an option, so architects resort to another approach.

This scheme -- as illustrated in several papers on the IBM, Sony and Toshiba Cell architecture, and one paper on the Microsoft Xbox -- might be called virtualization. The architect tries to extract accurate generalizations about the data flows and tasks the system will face, and then build a network of processing elements, memories and buses than can serve as a medium in which to implement the transforms and connections required by each flow of data as it passes through the system.

Thus, in the system, one data flow might be processed by one set of processors under one operating system, while another flow takes a quite different path, under control of an entirely different OS. The underlying hardware is unchanged, but it is employed differently for different tasks and flows.

This emphasis on data flows makes both the Cell and Xbox processors resemble networking chips to a remarkable degree. Both architectures are described in terms of their general-purpose CPUs and specialized vector units. But both depend for their operation on extensive behind-the-scenes hardware and organization to, in IBM's words, orchestrate the movement of data. The idea is to provide a multiprocessing platform that will appear to the game developer to be a single CPU augmented with powerful vector-processing instructions and hardware threading.

This thinking played out very differently in the IBM and Microsoft camps. The Cell (see May 23, page 1) uses a single Power CPU to deal with control flows and a bank of eight rather specialized vector floating-point processors to process data flows. The Xbox, in contrast, does not separate control from data flows so firmly. Instead, it provides three PowerPC cores with extensive SIMD instruction enhancements. Both architectures depend upon external graphics processors and I/O processors.

The problem they will face was discussed at length in a keynote by David Kirk, chief scientist at graphics chip maker Nvidia Corp. (Santa Clara). Kirk warned that PC game performance was increasingly limited by the throughput of the CPU, which was unable to supply vertices fast enough to saturate the graphics chip. Rather than delivering faster CPUs, he said, Intel and AMD are fielding dual-core processors. "But virtually no games today can benefit from multicore processors," Kirk said. "Efforts to multithread them have not been successful. In fact, we have seen cases where these efforts . . . actually made the system run slower on two cores than on one."

Source: Electronic Engineering Times
 
Back
Top