On Fri, 4 Aug 2000, Ookhoi wrote:
> Hi Stephen,
>
> > > I have two exactly the same Athlon systems at 750MHz and 512 meg memory
> > > (@ 133MHz). In the BIOS there is a option: "DRAM to CPU Frequency
> > > Ratio", which can be 3:3 or 4:3. The Help says: "Using this item to set
> > > the operating frequency of DRAM".
> > >
> > > I just installed Debian on the machines, rebooted, changed one machine
> > > to 3:3, the other was 4:3, and compiled a kernel on both of them. The
> > > time it took was almost exactly the same (about 5 min 32 sec). The 3:3
> > > one was 0.24 seconds faster.
> >
> > It is likely there are other limiting factors.
>
> Like? And should the faster ratio not make a little difference with a
> kernel compile?
>
> > > Does Linux ignore the DRAM to CPU Frequency Ratio (if at all possible),
> > > is it just a useless BIOS option (because of hardware limits or
> > > something), or is it not supposed to make the machine faster (if, what
> > > _does_ it do then ;-) ? (the book that came with the machine doesn't
> > > tell me much about the option).
> >
> > The question is what is the ratio between the speed of the CPU
> > bus and the speed the DRAM should run at. 3:3 means same speed, 4:3 means
> > the DRAM runs faster than the CPU. If you CPU runs at 100mhz for it's
> > communication to the chipset and you use '3:3' then your memory also runs
> > at 100mhz. If you use '4:3' then your memory runs at 133mhz while the
> > CPU runs at 100mhz. etc, etc.
>
> Oke, then I do understand the option. So if I have 133MHz capable
> memory, the 4:3 ratio _should_ make a noticeable difference. But it
> doesn't. Now I'm curious why. :-)
>
> Thanx for your thoughts!
>
> Ookhoi
I am kinda lost on why you expect a difference at all. Kernel compilation
is mostly crunching data that is in RAM and is limited by cpu speed and
speed of the cpu<->ram data path. So the speed of that data path has not
changed (it's still 100mhz).
My guess it that you can see an improvement in 3d stuff that uses agp, as
the extra bandwidth provided by the ram will be useful. (I.e. when there
is more than one consumer of the bandwidth).
Vladimir Dergachev
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.rutgers.edu
> Please read the FAQ at http://www.tux.org/lkml/
>
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Mon Aug 07 2000 - 21:00:13 EST