Cell is not so fast as you think!

Status
Not open for further replies.

gamepower

Banned
Look i have found that:
http://www.realworldtech.com/page.cfm?ArticleID=RWT021005084318&p=4

"The estimate given by IBM at ISSCC 2005 was that the DP FP computation in the SPE has an approximate 10:1 disadvantage in terms of throughput compared to SP FP computation. Given this estimate, the peak DP FP throughput of an 8 SPE CELL processor is approximately 25~30 GFlops when the DP FP capability of the PPE is also taken into consideration. In comparison, Earth Simulator, the machine that previously held the honor as the world’s fastest supercomputer, uses a variant of NEC’s SX-5 CPU (0.15um, 500 MHz) and achieves a rating of 8 GFlops per CPU. Clearly, the CELL processor contains enough compute power to present itself as a serious competitor not only in the multimedia-entertainment industry, but also in the scientific community that covets DP FP performance."

http://www.realworldtech.com/page.cfm?ArticleID=RWT021005084318&p=8

"After some discussion (and more wine), it was determined that the ATO unit is most likely the Atomic (memory) unit responsible for coherency observation/interaction with dataflow on the EIB. Then, after the injection of more liquid refreshments (CH3CH2OH), it was theorized that the RTB most likely stood for some sort of Register Translation Block whose precise functionality was unknown to those outside of the SPE. However, this theory would turn out to be incorrect.

Finally, after sufficient numbers of hydrocarbon bonds have been broken down into H-OH on Wednesday, a member of the CELL processor team member tracked down the relevant information and he writes:

The R in RTB is an internal 1 character identifier that denotes that the RTB block is a unit in the SPE. The TB in RTB stands for "Test Block". It contains the ABIST (Array Built In Self Test) engines for the Local Store and other arrays in the SPE, as well as other test related control functions for the SPE.
"
What do you think about this??
 
Double Precision FP being significantly slower was to be expected, i'd say. It is great they included it anyways (so you can employ Cell for say CAD/CAM workloads...). That however has little significance to its use in playstation 3 because most (if not all) data will be SP. e.g. EE's VUs only do SP (as do SM3 shaders)...
 
This is something we already know. And yes it is still just as fast as we thought. It is however not as fast as you thought. See, games don't use double precision floating point operations, they use single precision.

Where it would not be as fast as you thought, would be in supercomputer-driven, highly accurate simulations. In games and in PCs, it's still basically god-force. In terms of double precision floats, it's still a pretty tough contender too, 25-30 billion double precision floating point operations per second is still incredibly high performance for any CPU.


Later


Added: I use this in the context of a game because this is the console forum. If you wanted to argue that it's not as fast as everyone thought for supercomputer applications, then I'm sure you'd hear the same defence in the hardware discussion forum too. As mean as that all may have sounded, it's not, it's just how things are.
 
Status
Not open for further replies.
Back
Top