Michael S (already5chosen@yahoo.com) on 5/17/07 wrote:
---------------------------
>IntelUser2000 (Intel_user2000@yahoo.ca) on 5/16/07 wrote:
>---------------------------
>>
>>I always thought Conroe had fully pipelined FPU from looking at your article, but
>>its not true.
>
>80bit x87 multiplication (FMUL) on Merom* is indeed not fully pipelined. However
>64-bit SSE2 (MULSD/MULPD) multiplication is fully pipelined. So MULSD has two times
>higher throughput over Banias/Dothan/Yonah while MULPD has four times higher throughput.
>
>>It still has the same FPU latency for FPMUL as its P6 derivatives.
>
>Don't you mean "the same FPU throughput"?
>
>I don't like when people use the name "Conroe" to describe Intel's new microarchitecture.
>Conroe is just a name of desktop variant of Merom core.
>The use of the name "Barcelona" instead of Hound or Greyhound is only marginally better.

Here's the link: http://www.realworldtech.com/forums/index.cfm?action=detail&id=68284&threadid=68284&roomid=2

"-- NGMA x87 capabilities are very similar to the P6 descendants: 3l/1tp FADD, 5l/2tp FMUL."

I guess it isn't a problem anymore though, because most of the programs use SSE, right?