Xenon "thread' vs Cell "SPE"

[expletive]

How do these 2 compare to each other in various gaming applicaitons (grfx, AI, physics calcualtions) and general computing overall?

I'm not really sure how they differ and in what cases one might outperform, or be prefereable) to the other.

If "difficulty in being able to utilize it" is a ocnsideration, i'd be interested in that too.

Thanks in advance for any insight.

J

 

[overclocked]

Use the search function and you can find lots of info on this.
Or rather mixed with info and LOTS of wild speculation.

 

[expletive]

Use the search function and you can find lots of info on this.
Or rather mixed with info and LOTS of wild speculation.

Will do, thanks.

J

 

[Shifty Geezer]

Well in rough summary...

An XeCPU thread is like a standard PC processor thread, and can do what a standard PC processor does, only with some extra vector power and with less efficient memory accessing that benefits from developer guidance. There's two threads to a core and they share resources, so having 6 threads is not the same as having 6 cores. They allow a core to be used more efficiently with less waste, which results in something like (rought ballpark figure) a 20% boost in performance.

A Cell SPE is a processor that works quite differently to standard PC processor in that it works entirely in it's own local storage. It's optimized for floating point/vector work and lacks branch prediction and convenient memory access. To work on data you have to specifically fetch it. Being small means lots can be fit onto a chip, giving 8 on the current version of Cell (one inactive in PS3). The SPE forgoes a lot of developer nicities to try and overcome the principle bottleneck of modern processors which is that RAM can't supply data as fast as processors can process it. Working in the SPE's local storage and using efficient memory prefetching, Cell should avoid slow main memory accesses and so keep running with less stalls.

That's an overview. For ideas on difficulties and advantages/disadvantages proceed with your search

 

[blakjedi]

From IBM's website:

"The most distinguishing feature of the Cell Broadband Engine is that although all processors share main storage (the effective-address space that includes main memory), their function is specialized into two types -- the Power Processor Element (PPE) and the Synergistic Processor Element (SPE). The Cell Broadband Engine has one PPE and eight SPEs.

• The first type of processor, the PPE, is a 64-bit Power Architecture core. It is fully compliant with the 64-bit Power Architecture specification and can run 32-bit and 64-bit operating systems and applications.
• The second type of processor, the SPE, is optimized for running compute-intensive applications, and it is not optimized for running an operating system.

The SPEs are independent processors, each running its own individual application programs. Each SPE has full access to coherent shared memory, including the memory-mapped I/O space. The designation synergistic for this processor was chosen carefully -- there is a mutual dependence between the PPE and the SPEs. The combination of the two working in harmony produces a greater effect than each working alone. The SPEs depend on the PPE to run the operating system and, in many cases, the top-level control thread of an application. The PPE depends on the SPEs to provide the bulk of the application performance. "

In a nutshell.