Date: Sun, 20 Sep 1998 16:45:46 -0500 From: David Kelly <dkelly@hiwaay.net> To: Donald Burr <dburr@pobox.com> Cc: FreeBSD Hardware <freebsd-hardware@FreeBSD.ORG> Subject: Re: UNDER-clocking Cyrix 6x86MX-PR233 at 66x2.5 instead of 75x2.5? OK? Message-ID: <199809202145.QAA13708@nospam.hiwaay.net> In-Reply-To: Message from Donald Burr <dburr@pobox.com> of "Sun, 20 Sep 1998 04:21:29 PDT." <XFMail.980920042129.dburr@pobox.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Donald Burr writes:
> I have heard that some systems/boards/peripherals can't take the higher
> 75MHz bus speed, so I tried moving the bus speed down to 66 MHz, and left
> everything else as is (clock multiplier, etc.) Now the CPU is detected as
> a 6x86MX-PR200, and everything works perfectly.
>
> Since it seems to work OK now, I think I'll keep it set up this way. I
> just wanted to make sure if it's OK to "UNDER-clock" the CPU like this --
> will doing this damage the CPU or motherboard or anything?
"Its a PC so if it works then just do it."
It is *possible* the faster chip would not work right at a slower rate.
While its being clocked slower, the silicon still switches at a rate
that is supposed to be fast enough for a faster bus. If by some idiocy
the chip latched onto data before a clock edge and before it was stable
then it would do the wrong thing. I don't think this would be the case
as the final decision as to a bus state should be at clock transition.
Some CPU's have to run at some minimum clock rate in order to refresh
internal dynamic memory. Don't worry about that until you drop to 1 MHz
or so. Its a problem for the low power shutdown people and In-Circuit
Emulator designers.
With all that said, I'm running a PPro 166/512k. Supposed to be a 66 MHz
bus. Chip runs real nice at 200/66 MHz too but I've decided either my
ancient ATI Mach32 video card or my ancient non-ultra Adaptec 2940, or
my new Asus SC875 (Symbios) SCSI card don't like a 66 MHz bus on an Asus
P6NP5 MB. The NCR driver has reported some faults with PCI access. The
system either a couple of times per day, or it would wait months. Always
during heavy disk activity ("cvs -q update -P -d" was a favorite)
combinded with something graphic intensive, always Netscape. A mouse
click on Netscape which required disk activity (swap?) during the other
disk activity often resulted in a freeze.
There are BIOS parameters for the PCI bus I could mess with, I guess.
Things like "PCI Latency" and such. Don't understand them. Can't
reliably reproduce my problem. Ain't gonna touch them. Values are set
by the "factory default" option on the BIOS. That's another long story
where my floppy didn't work until a "factory default" altho I couldn't
see any visible BIOS parameter changes.
Dropped the bus divider down to 60 MHz, so now I run 180/60. Rock solid
the past 2 months. So I'm overclocking my CPU, underclocking my bus.
When I first played with overclocking, found the CPU ran at 233 without
problem for the 30 minutes I tested it (running rc5des and "make kernel"
several times.) 266 MHz wouldn't boot. Considering 233 ran well I used
200. Am thinking its time once again to see how 210 or 240 work.
--
David Kelly N4HHE, dkelly@nospam.hiwaay.net
=====================================================================
The human mind ordinarily operates at only ten percent of its
capacity -- the rest is overhead for the operating system.
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hardware" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199809202145.QAA13708>