lkml.org 
[lkml]   [1997]   [Dec]   [31]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: Overclocking PCI / UNIX95 compliance
Date
From

psj@cgmlarson.com said:
> I'll pick a little nit here... If a motherboard claims to be PCI-2.1
> compliant, it is _SUPPOSED_ to be able to handle a PCI clock up to
> 66MHz, ditto for cards

Bzzzt! Wrong. PCI 2.1 maximum clock speed is 33MHz. 66MHz capability is an
extension, making 66MHz PCI a different spec. I quote from the PCI Local Bus
Specification, Chapter 7 -- 66MHz PCI Specification:

"The 66MHz PCI bus is a compatible superset of PCI designed to
operate up to a maximum clock speed of 66 MHz."

If a card does not pull up the M66EN signal, then it is an error for the
motherboard to clock it at anything greater then 33MHz, and a card can
depend on this.


psj@cgmlarson.com said:
> My cheap little TXPro (VIA Apollo-I) board claims PCI-2.1 compliance,
> but yet will not accept some PCI cards (CPU's too close to slot) and
> will not go above 33 MHz on the PCI bus.

PCI 2.1 has a short card specification, which is constrained to the
lengths 4.721" - 6.6". I bet the 6.6" will still clear your CPU.

The specification does not require the presence of full-size cards. I do
believe it is a shortcoming in the manufacturer's claim to not say which
of the PCI slots are short, but even if all the slots are short, it is
still fair to say it is PCI 2.1 compliant.

And is some cases, whether the slot is short depends on the installed CPU:-)

kwrohrer@enteract.com said:
> If *any* device on a PCI bus can't handle 66 MHz, none may use it.

There is such a thing as a bridge.
--
Steve Williams
steve@icarus.com
steve@picturel.com

"The woods are lovely, dark and deep. But I have promises to keep,
And lines to code before I sleep, And lines to code before I sleep."



\
 
 \ /
  Last update: 2005-03-22 13:40    [W:0.057 / U:0.824 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site