[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: off topic, intel processors



On Mon, 22 Mar 1999, Ira Abramov - Scalable Solutions wrote:

> 
> 
> Today, Ariel Biener blurbed:
> 
> > I would assume that the calibration loop wasn't properly adjusted for pII.
> > If anyone can recall, 486DX4-100 gave better performance that some Pentium
> > chips, because of the same thing. Browsing back through the Linux archives
> > of that time will reveal an explanation by Linus on this subject.
> >
> > Of course, I might also be mistaken.


Indeed a long answer, stating exactly what my MUCH shorter answer did :)

What Linus said in that ancient post was that the delay loop (aka
calibration loop) works differently for different processors, and non
comparable between families of processors, although it is usable within
the same processor family (i.e. within pII's for example). (note that this
is not the exact technical description, but it should be clear enough).

That means that comparisons between families of chips based on Bogomips
is bogus, at best, or completely wrong in the worst cases.


--Ariel

> 
   +---------------------------------------------------------------+
   | Ariel Biener                                                  |
   | e-mail: ariel@post.tau.ac.il           Work phone: 03-6406086 |
   | fingerprint = 07 D1 E5 3E EF 6D E5 82 0B E9 21 D4 3C 7D 8B BC |
   +---------------------------------------------------------------+