Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 24 Apr 1996 02:12:52 -0700 (PDT)
From:      "Rodney W. Grimes" <rgrimes@GndRsh.aac.dev.com>
To:        joerg_wunsch@uriah.heep.sax.de
Cc:        freebsd-current@FreeBSD.org, cat@ki.net, geoff@ki.net, scrappy@ki.net
Subject:   Re: MotherBoard Jumper Settings...
Message-ID:  <199604240912.CAA04144@GndRsh.aac.dev.com>
In-Reply-To: <199604240730.JAA26732@uriah.heep.sax.de> from J Wunsch at "Apr 24, 96 09:30:12 am"

next in thread | previous in thread | raw e-mail | index | archive | help
> As Marc G. Fournier wrote:
> 
> > 	Third, the Oscillator Frequency was set for 33Mhz instead
> > of 25Mhz...have fixed it...
> 
> (Others told you that a DX4 is 3x, so the 33 MHz was right.)
> 
> > 	As far as the cache is concerned, I have:
> > 
> > 	4x UM61512AK-15/95282/N52049's
> > 	1x W24129AK-15/95200 (Tag SRAM)
> > 
> > 	Am I correct in assuming that the -15 is the speed of the cache?
> > If so, and I haven't changed that one yet, my CMOS is set for 20ns...
> > would that produce any of the bugs I've been reporting?
> 
> I don't think so.  The tag RAM is IMHO normally faster than the cache,
> perhaps that's why they kept the cache timing slower than necessary in
> the setup.

This is only common when the data RAM is -20ns.  The more ``usuall''
conditions are that the TAG is 15nS (though some times it can be
safely 20nS) and the DATA is either 15 or 20nS.  A few rare boards use
12nS parts for TAG and either 12 or 15nS parts for DATA.

> > 	The jumpers for the cache are set for 256KB/64kbx4...how do I
> > determine the size of each of the cache chips to determine if *this* is
> > right?
> 
> The 61512 suggests a 512 Kbit cache.  Seems you are using only half of
> it with your jumper setting.

The 61512 is infact a 64k x 8 part, I looked it up as there had been
too much speculation on here about just what that chip was.  Cache RAM
on 486 and later machines is almost always x 8 parts for those who thought
this might have been a 128k x 4.

Your cache jumper settings are correct at 256K/64Kx8.  You may still have
a bad cache RAM though.  Do you ever get any signal 10's or 11's from the
gcc compiler, or do you just get panics?

-- 
Rod Grimes                                      rgrimes@gndrsh.aac.dev.com
Accurate Automation Company                 Reliable computers for FreeBSD



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199604240912.CAA04144>