From: "Alexei A. Frounze" <dummy_addressee AT hotmail DOT com>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Reading MSR (Athlon multiplier)
Date: Fri, 19 Jan 2001 10:27:44 -0500
Lines: 83
Message-ID: <949mbl$d3jno$1@ID-57378.news.dfncis.de>
References: <3a66161d DOT 226362160 AT news DOT sci DOT fi> <945a90$ckgq1$1 AT ID-57378 DOT news DOT dfncis DOT de> <945itn$3ai$1 AT nnrp1 DOT deja DOT com> <945mn8$cjroo$1 AT ID-57378 DOT news DOT dfncis DOT de> <946ofe$vlp$1 AT nnrp1 DOT deja DOT com> <946rg8$c61d6$1 AT ID-57378 DOT news DOT dfncis DOT de> <9471af$7r4$1 AT nnrp1 DOT deja DOT com> <947ao2$ca654$1 AT ID-57378 DOT news DOT dfncis DOT de> <947p8h$vij$1 AT nnrp1 DOT deja DOT com> <947rdf$ct4h3$1 AT ID-57378 DOT news DOT dfncis DOT de> <948d3i$gne$1 AT nnrp1 DOT deja DOT com> <83ae8om8xh DOT fsf AT mercury DOT st DOT hmc DOT edu> <9493kh$20s$1 AT nnrp1 DOT deja DOT com> <949ltm$d3onh$1 AT ID-57378 DOT news DOT dfncis DOT de>
NNTP-Posting-Host: pppa59-resalerochester3-5r7104.dialinx.net (4.4.209.248)
X-Trace: fu-berlin.de 979918007 13750008 4.4.209.248 (16 [57378])
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 5.50.4133.2400
X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

again, there is possibility that DX rounds the measured (if measured) value
towards closest known value (800MHz in this case). and this could be done in
order to satisfy customer... in M$ this is possible. Not many people are
clever/smart/whatever and they can flame because they see exactly the same
thing as you... If it's 807 instead of 800 they will probably not flame, but
if it's 793 instead of 800 very likely they will flame to the company they
purchased computer at. :) that's stupid, but guess what? this is what
happens. and they don't care about 1%. they've been told and guaranteed 800
and they don't what to have less than that.

--
Alexei A. Frounze
alexfru [AT] chat [DOT] ru
frounze [AT] ece [DOT] rochester [DOT] edu
http://alexfru.chat.ru
http://members.xoom.com/alexfru/
http://welcome.to/pmode/


"Alexei A. Frounze" <dummy_addressee AT hotmail DOT com> wrote in message
news:949ltm$d3onh$1 AT ID-57378 DOT news DOT dfncis DOT de...
> "Tom St Denis" <stdenis AT compmore DOT net> wrote in message
> news:9493kh$20s$1 AT nnrp1 DOT deja DOT com...
> > In article <83ae8om8xh DOT fsf AT mercury DOT st DOT hmc DOT edu>,
> >   Nate Eldredge <neldredge AT hmc DOT edu> wrote:
> > > Tom St Denis <stdenis AT compmore DOT net> writes:
> > >
> > > > Hmm loser:  I don't have a 807mhz processor. Your code is wrong.  A
> program
> > > > that says 1+1=3 is wrong despite being coded flawlessly in perfect
> ANSI C.
> > >
> > > How do you know?  800 MHz is only a nominal speed; the speed at which
> > > the processor actually runs can vary from that significantly.
> >
> > If that were true then I should get 793Mhz from his code once in a while
> to.
>
> It may not happen. Nobody said it should be a symmetrical fluctuation. It
> may be shifted towards higher or lower values due to various reasons.
>
>
> > Consistently biased is not biased, it's errored!
>
> Nope. You do not know fo sure.
>
>
> > Besides directx can say "800mhz" why can't his code?
>
> Because:
> 1. DX may not really measuer this frequency (I'm not expert here, but this
> is possible)
> 2. DX runs in privilege level 0 and thus can disable all those irrelevant
> things while measuring the frequency
> 3. Nobody said that 800MHz is the frequency you have on the CLOCK pin of
> your CPU (it may be both a little bit higher or lower). And if you'll try
to
> say that BIOS prints out this value correctly, I'll argue this. Because
BIOS
> definetely has access to resources and signals on the motherboard and thus
> it can acquire settings of switches and instead of measuring it may conver
> them into frequency by using a look-up table or something similar.
> 4. Again, nobody said, that the timer in your computer generates impulses
at
> exactly 1193180Hz rate - there can still be certain limited accuracy.
> Perhaps you may improve this by using RTC generator instead.
>
> Just try to run the program in DOS or boot from the "emergency diskette"
and
> run in this environment. Report what you get there.
>
> --
> Alexei A. Frounze
> alexfru [AT] chat [DOT] ru
> frounze [AT] ece [DOT] rochester [DOT] edu
> http://alexfru.chat.ru
> http://members.xoom.com/alexfru/
> http://welcome.to/pmode/
>
>
>