"laurent.artaud[AT]free.fr" <"laurent.artaud[AT]free.fr"> wrote in message
news:40667a7e$1@news.povray.org...
> >   This is true for the *CPU*. However, it's not the CPU which handles
> > the floating point code, but the FPU.
> >   The FPU is completely separate from the CPU. The FPU has eight 80-bit
>
> Well, you seems to mix CPU and ALU:
> ALU stands for Arithmetic and Logic Unit. It is the part of the chip
> that compute on integers.
> The ALU and the FPU are separate, but are both inside the CPU.

However, when people refer to a CPU being 32bit, they usually are referring
to the ALU.  The FPU has been 80 bit since at least the '80s.

> Well, to know this for sure, we'll have to find an assembler programmer
> which will be able to say if it is possible to load one of the FPU
> register directly from the cache in the same number of cycles the other
> registers do.

Well, it's been a few years, but I used to do quite a bit of graphics
programming in assembly.  Did you know that, for instance, the FPU registers
load so fast that games used them to load bitmaps from regular memory to
video memory?  It worked about 30% faster than a standard transfer.  In
fact, this was the reason MMX used the FPU registers, and so you can't use
FPU and MMX at the same time.

> If so, then the speed difference between the Athlon-XP and the Athlon-64
> is partly (they can also have improved the FPU unit...) because the last

But they didn't.  Internally, the FPU is identical between the AthlonXP, 64
and FX models.

-- 
...Chambers
http://www.geocities.com/bdchambers79

Post a reply to this message