By using this site, you agree to our Privacy Policy and our Terms of Use. Close
NightAntilli said:
alephnull said:
NightAntilli said:


CPU:
If this is true, this CPU is gonna be VERY powerful.. People keep complaining the CPU of the X360 is not that strong, but that's simply not true. For games, general purpose computing is a big part of the calculations, while floating point is a small part, and the X360 CPU is based on general purpose, while the PS3 has only one general purpose core, and the other ones are floating point calculators, which means, if it wants to do general purpose, the calculations need to be emulated. And the 3 cores of the X360 can handle 6 threads in total, and if it's double, I guess they could do some wacky stuff with this...

Define "General purpose".

Any instruction/calculation regarding loading, integer, branch, store etc

So, an an SPE can execute 2 instructions simultaneously, but only if one is an "even" instruction and one is an "odd" instruction (assuming the instructions are aligned properly).

SP FP operations on an SPE have a 6 cycle latency on the even pipeline and Integer floating point (such as fma) operations have a 7 cycle latency on the even pipeline. Simple fixed point operations (such as addition, subtractionhave 2 cycle latency on the even pipeline. Compare operations have a 2 cycle latency (even) and branch operations have a 4 cycle latency (odd). Load/store operations have a 6 cycle latency (odd).

All of these instructions are fully pipeline-able without inducing any stalls as long as you double buffer loads.

When people say the SPEs are optimized for SPFP they are talking more about the fact that it's terrible at double precision. The primary inovation of the SPEs is their ability to hide memory access latency. This is the whole point of the crazy architecture and everyone seems to miss it.