Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : IDTI - an IC Play on Growth Markets -- Ignore unavailable to you. Want to Upgrade?

To: Cicero who wrote (7136)	3/3/1998 11:48:00 PM
From: Scumbria	Read Replies (2) \| Respond to of 11555

Cicero, RE:"I have seen NO indication that the P II is (RISC.)" Intel is of course not inclined to advertise the PII/Pentium Pro core as being a RISC core. However, if you look at the Pentium Pro technical document- intel.com you will see: "The FETCH/DECODE unit: An in-order unit that takes as input the user program instruction stream from the instruction cache, and decodes them into a series of micro-operations (uops) that represent the dataflow of that instruction stream." The uops are essentially RISC instructions. The decode unit generates load/store operations and register to register arithmetic operations. Make no mistake about it, those uops are textbook RISC instructions. This is essentially the same scheme used in C6 and K6. Cyrix processors on the other hand, decode and dispatch the full x86 instruction. The execution of these instructions is performed by traditional x86 microcode. The Media GX core is a simplified version of the M1/M2 core. Concerning power consumption, the simplicity and smaller transistor counts of C6 and Media GX are major factors in their lower power utilization. C6 performance is enhanced by a large Level 1 cache (64KB.) For comparison, Media GX only has a 16KB L1 cache. If Media GX had a 64KB cache, it's performance would be comparable to C6. Like C6, M2 also has a 64KB cache. However, at the same megahertz (not to be confused with PR rating,) M2 benchmarks significantly higher than C6. PII uses a smaller level 1 cache (32KB.) They make up for the small level 1 cache with a large level 2 cache. The Intel Covington processor (soon to be released) is a PII, with the level 2 cache stripped off. It seems reasonable to expect that the performance of Covington will be less than C6 at the same megahertz. RE:"Quake and Quake II both use FP numbers heavily" My sources (which are quite reliable) tell me that the Quake developers are beginning to shy away from floating point instructions. Scumbria