PA-2007 and K6-2-300 misadventures
Rick Seiler
(Barracuda HD), ATI Xpert@Work PCI.
Here's the deal: I've tried underclocking, clocking, and overclocking this
CPU, at voltages varying from 2.1V to 2.4V, and no matter what I do, when
the system boots it never quite gets to the stage where it should display
the PCI Device Listing briefly and then run Win98. It stops right after the
listing of my SCSI devices, and I have to reboot. The CPU and its speed are
identified properly at the top of the screen.
I actually lied: It WILL boot past this and work fine if I disable the
EXTERNAL cache in the BIOS. No other setting that I played with in there
had an effect. Obviously, I can't leave the cache disabled, so what could
be wrong here? I can't think of anything else to try. I never had this
problem with my CyrixMMX 233.
Thanks
Robato Yao
You can try my misadventures with the PA2007. I had a Cyrix 686L PR200
on such a board for a year and decided to install a spare Cyrix 686MX on
it. The MX is a PR166 with a strange clock setting of 60x2.5 (150MHz.)
To be honest with you this chip runs reliably even if I overclock it to
83x2 (166MHz, PR233).
The settings for the 686L PR200 and the 686MX PR200 should be identical
(75x2) and that is how the board is set on the 686L, properly
recognizing it as a PR200. I removed the 686L and pop in the 686MX---no
jumpers changed. Boot up the machine and the BIOS says PR166, despite
unchanged 75x2 jumper settings. Maybe the BIOS needs updating.
I also have a FIC PT2007 (Intel TX chipset) that does the same thing on
the Cyrix. It says PR166 despite the PR200 chip is jumpered at 66x2.5
(166MHz).
Rgds,
Chris
>
>
(counting down from top 50 oxymorons...)
10. Tight slacks
9. Definite maybe
8. Pretty ugly
7. Twelve-ounce pound cake
6. Diet ice cream
5. Rap music
4. Working vacation
3. Exact estimate
2. Religious tolerance
And the NUMBER ONE top oxy-MORON
1. Microsoft Works
---From the Top 50 Oxymorons (thanks to Richard Kennedy)
George Macdonald
>This is a Rev 1.2 board with 113cd13, 64MB 60nS EDO, Adaptec 2940UW
>(Barracuda HD), ATI Xpert@Work PCI.
>
>Here's the deal: I've tried underclocking, clocking, and overclocking this
>CPU, at voltages varying from 2.1V to 2.4V, and no matter what I do, when
>the system boots it never quite gets to the stage where it should display
>the PCI Device Listing briefly and then run Win98. It stops right after the
>listing of my SCSI devices, and I have to reboot. The CPU and its speed are
>identified properly at the top of the screen.
>
>I actually lied: It WILL boot past this and work fine if I disable the
>EXTERNAL cache in the BIOS. No other setting that I played with in there
>had an effect. Obviously, I can't leave the cache disabled, so what could
>be wrong here? I can't think of anything else to try. I never had this
>problem with my CyrixMMX 233.
>
>Thanks
A few people have had trouble with 113CD13 and SCSI cards - in one case
Enabling Reset Configuration Data in PCI PnP Config helped but only till
the next boot, so not really a solution. Does it work OK with the 109CD12
BIOS? Seems like FIC buggered something in PnP with 113CD13 and they've
announced that it's the final BIOS for the PA-2007.<sigh>
Have you played with Manual setup in PCI PnP Config to try bumping IRQ
around a bit and maybe reserving the memory area for the SCSI BIOS?
Sorry I don't have better info.
Rgds, George Macdonald
"Just because they're paranoid doesn't mean you're not psychotic" - Who, me??
Michael P. Daly
I also have the Adaptec 2940UW hooked up to several peripherals. I've slightly
o'c my K6-300 (66x5) at 2.2v with no problems. If you haven't done so already,
go to the Via site (I think FIC site has it as well) and get the Via irq patch.
That may solve the problem. Also, make sure that usb is enabled in bios. Hope
this helps.
Mike
Stefan Huszics
109 to 113 bios series except it wouldn't boot with any bios tweaking (unless
protected mode was used). This was also present in the 113 bioses with the
VA-503+ and it didn't go away untill Diamond released a new BIOS for the card.
Now everything works again.
IOW FIC has made some change in the 113 biosseries that confilicts with some
cards biossetup.
Have you visited Adaptecs homepage and looked for a biosupgrade ?
George Macdonald wrote:
--
/ Stefan
Windows-95
A 32 bit extension and graphical shell, for a 16 bit patch to
an 8 bit operating system, originally coded for a 4 bit microprocessor,
written by a 2 bit company, that can't stand 1 bit of competition.
Rick Seiler
It's not the SCSI controller, since I pulled it and see the same problem
(it doesn't get past Award PnP BIOS Extension 1.0A and never shows the
ATAPI device listing or the PCI device listing). I even tried the
unofficial 135 "beta" BIOS that is supposed to solve the K6 "looping"
problem, and that didn't help either (note that I don't have the looping
problem, but I was desperate). The "Reset Configuration Data" setting
had no effect, not even for one boot. Only disabling the External Cache
fixes the issue. I've retreated to my CyrixMMX233, which continues to
work fine even with my external cache. Someone suggested the video card
(ATI Xpert@Work PCI), which I didn't think to swap out but will. I
could also pull my ISA sound card and PCI network card, but after that
there ain't a lot left but case and air molecules. <g>
Rick Seiler <rse...@usa.net> wrote in message
35d21...@news.cyberzone.net...
WNG
and not the AMD, and it's external cache....
Did you bother to disable Linear Burst mode for the board/BIOS before
using the K6?
Will
On Thu, 13 Aug 1998 15:47:48 -0400, "Rick Seiler" <rse...@usa.net>
wrote:
>Thanks for all the suggestions. Unfortunately, this is still an X-File.
Regards,
William Ng
========================================================
ATTENTION:
In order to stop the flood of SPAM,
Remove : <DON'T.SPAM.> from DON'T.SPA...@lynx.neu.edu
========================================================
*Anti Spam Notice*
By US Code Title 47, Sec.227(a)(2)(B), a computer/modem/printer
meets the definition of a telephone fax machine. By Sec.227(b)(1)(C),
it is unlawful to send any unsolicited advertisement to such equipment.
By Sec.227(b)(3)(C), a violation of the aforementioned Section is
punishable by action to recover actual monetary loss, or $500,
whichever is greater, for each violation. All incoming unsolicited
commercial traffic will therefore be billed at a rate of $500
per msg to compensate for loss of service.
Rick Seiler
Burst choice entirely once it knows that you have an AMD CPU installed.
WNG wrote in message <35d450ad...@nntp.neu.edu>...
George Macdonald
>No, and that's only because the BIOS is smart enough to remove the Linear
>Burst choice entirely once it knows that you have an AMD CPU installed.
But there *is* a jumper on the board as well.
Peter Jespersen
>No, and that's only because the BIOS is smart enough to remove the Linear
>Burst choice entirely once it knows that you have an AMD CPU installed.
>
>>Did you bother to disable Linear Burst mode for the board/BIOS before
>>using the K6?
Well did you disable the Linear burst mode on the MB???
Live long and prosper...
____________________________________________________________________
Me : Peter Jespersen, Denmark
E-Mail : fly_NOSP...@image.dk
WWW : http://www.image.dk/~flywheel/
The meek shall inherit the Earth, but not its mineral rights. (J. Paul Getty)
____________________________________________________________________
Stefan Huszics
Rick Seiler wrote:
> No, and that's only because the BIOS is smart enough to remove the Linear
> Burst choice entirely once it knows that you have an AMD CPU installed.
>
> WNG wrote in message <35d450ad...@nntp.neu.edu>...
> >I just started to read this thread....but since it works on your Cyrix
> >and not the AMD, and it's external cache....
> >
> >Did you bother to disable Linear Burst mode for the board/BIOS before
> >using the K6?
> >
> >Will
Rick Seiler
actually forgot that there was a jumper for it since it's also in the
BIOS. And the documentation for it is not near that of the voltage and
multiplier jumpers, so I was not reminded then either. Thanks! I think
I'm a bit spoiled too by my other motherboard, which is jumperless.
So, in the end, the K6-2-300 is working fine at 75x4.5 @ 2.1V (I thought
I'd have to make the jump to 2.3V, but so far it's OK). I'm sticking
with the 135 beta BIOS now that it's installed.
WNG <DON'T.SPA...@lynx.neu.edu> wrote in message
35d450ad...@nntp.neu.edu...
Duckman
wrote:
>No, and that's only because the BIOS is smart enough to remove the Linear
>Burst choice entirely once it knows that you have an AMD CPU installed.
>
>WNG wrote in message <35d450ad...@nntp.neu.edu>...
>>I just started to read this thread....but since it works on your Cyrix
>>and not the AMD, and it's external cache....
>>
>>Did you bother to disable Linear Burst mode for the board/BIOS before
>>using the K6?
>>
>>Will
>
So what about the jumper on the mother board ??
I guess it should move it self ??
try page 14 in the manual .
DDD
Calvin Harrigan
>No, and that's only because the BIOS is smart enough to remove the Linear
>Burst choice entirely once it knows that you have an AMD CPU installed.
>
>WNG wrote in message <35d450ad...@nntp.neu.edu>...
>>I just started to read this thread....but since it works on your Cyrix
>>and not the AMD, and it's external cache....
>>
>>Did you bother to disable Linear Burst mode for the board/BIOS before
>>using the K6?
>>
>>Will
>
I think that there is a jumper on the motherboard to disable that
linear burst mode.
Remove the question mark from the e-mail address to
reply.
Chris Papademetrious
wrote:
>Here's the deal: I've tried underclocking, clocking, and overclocking this
>CPU, at voltages varying from 2.1V to 2.4V, and no matter what I do, when
>the system boots it never quite gets to the stage where it should display
>the PCI Device Listing briefly and then run Win98. It stops right after the
>listing of my SCSI devices, and I have to reboot. The CPU and its speed are
>identified properly at the top of the screen.
>
>I actually lied: It WILL boot past this and work fine if I disable the
>EXTERNAL cache in the BIOS. No other setting that I played with in there
>had an effect. Obviously, I can't leave the cache disabled, so what could
>be wrong here? I can't think of anything else to try. I never had this
>problem with my CyrixMMX 233.
Rick,
I had this problem with a PA-2007 when building a machine for my
sister. No matter what I did, the board wouldn't make it past the
BIOS screen with the external cache enabled. Luckily, the system was
so much faster than her 486, I just left it disabled and she was
really happy anyway. :)
But believe me when I say I lost some sleep over that one. I thought
it must've just been me. I even exchanged the board for a new
board/CPU, with no luck.
My suggestion would be to try removing *everything* but the video card
and memory, and see if you have any luck with that. Then start adding
in the extra components if you troubleshoot it successfully this far.
- Chris
L.Angel
>The settings for the 686L PR200 and the 686MX PR200 should be identical
>(75x2) and that is how the board is set on the 686L, properly
>recognizing it as a PR200. I removed the 686L and pop in the 686MX---no
>jumpers changed. Boot up the machine and the BIOS says PR166, despite
>unchanged 75x2 jumper settings. Maybe the BIOS needs updating.
It probably does... though my PA2007 recognized the 686MX PR166
without a BIOS update, using the 6.18J90W one.
Overclocking the chip would also result in corresponding recognition
as MX PR200 then MX PR233(wouldn't run stable though)
The little lost angel & her featherhead's 2 cents of dreaminess. :)
Email : Figure out what to remove, I'm getting tired of spam
William Lane
but DOS NEVER WAS an 8-bit OS, either in op-code
size, operand size, or data size. This particular
BS sig has gone on long enuff' and people might
actually begin to beleive it. The 8086/88 processor
had a 20-bit address space and could work on 16-bits
of data at a time. Where the 4-bit microprocessor
comes from I don't know (maybe the ALtair?). Furthermore,
describing Win95 as a 32-bit extension is pure
ignorance. How can an OS that runs in protected
mode be merely an extension? Ever try coding an
OS that can switch from protected mode to real
mode? I haven't either but I'm sure it's a lot
more than just an "extension".
Stefan Huszics (med9...@student2.lu.se) wrote:
: / Stefan
:
:
:
Stefan Huszics
missinforming.
If you feel so strongly about it I will refrain from including it on posting
to this newsgroup.
BTW that protected mode stuff is just crap and makes your computer slower.
All you have to make sure of is that a program crashing doesn't take the OS
with it.
BEOS for instance seems to be able to let programs take direct controll of
the hardware and compairing the speed AND stability between it and Windoze ,
well...
Also, if it isn't just an extention to DOS, why the @#&! is it so unstable,
slow and generally buggy (read crappy)?
William Lane wrote:
--
/ Stefan
Once you pull the pin out of Mr Greanade
He is no longer your friend ...
(this one better ? ;)
L.Angel
>comes from I don't know (maybe the ALtair?). Furthermore,
>describing Win95 as a 32-bit extension is pure
>ignorance. How can an OS that runs in protected
If it isn't an extension, like Stefan ask, why is it so unstable and
crappy?
Real World Technologies
DOS was based upon CP/M, which was written for the 8008 (amongst others),
which was originally based upon the 4004. So, the 4 bit processor was the
basis for the 8 bit processor that ran CP/M, which was the basis for DOS,
which ran on a 16 bit processor. MS wrote a 16 bit shell, which later
became a 32-bit shell. Win95 'starts life' as a 16-bit program loader and
file system manager, but then 'replaces' it with it's own code to do much of
the device interaction in (mostly) 32-bit mode. Doesn't anyone remember
when Win95 was first introduced that MS admitted that there was still quite
a bit of 16-bit code in it? Geez - the public really *does* have only a
6-month memory (or is that 16-bit??)... <g>
The saying isn't 100% accurate, but it also isn't 100% inaccurate either.
That's what makes it funny...
Regards,
Dean
Stefan Huszics wrote in message <35D90F0F...@student2.lu.se>...
>Sorry, it's just for fun, but you have a point in it beeing a bit
>missinforming.
>
>If you feel so strongly about it I will refrain from including it on
posting
>to this newsgroup.
>
>BTW that protected mode stuff is just crap and makes your computer slower.
>All you have to make sure of is that a program crashing doesn't take the OS
>with it.
>BEOS for instance seems to be able to let programs take direct controll of
>the hardware and compairing the speed AND stability between it and Windoze
,
>well...
>
>Also, if it isn't just an extention to DOS, why the @#&! is it so unstable,
>slow and generally buggy (read crappy)?
>
>William Lane wrote:
>
>> You know this is a really neat saying,
>> but DOS NEVER WAS an 8-bit OS, either in op-code
>> size, operand size, or data size. This particular
>> BS sig has gone on long enuff' and people might
>> actually begin to beleive it. The 8086/88 processor
>> had a 20-bit address space and could work on 16-bits
>> of data at a time. Where the 4-bit microprocessor
>> comes from I don't know (maybe the ALtair?). Furthermore,
>> describing Win95 as a 32-bit extension is pure
>> ignorance. How can an OS that runs in protected
George Macdonald
<help...@realworldtech.com> wrote:
>Well...
>
>DOS was based upon CP/M
Conceptually anyway, but Microsoft bought QDOS (Quick & dirty OS - really!)
from Seattle Computer and "enhanced it". :-) Remember Xenix?.. a lot of
people thought they did that themselves too - nope, they bought that from
Human Computing Resources. Then there was their first "C" Compiler -
Lattice IIRC.
Of course they stopped paying for things a while ago - now they steal them.
:-)
Real World Technologies
I don't know how true the story is, but many, many years ago (before it was
'cool' to hate MS), I heard from some 'knowledgable' sources that Seattle
Computer had stolen their DOS from another source - which eventually became
DR DOS. DR DOS was based entirely on CP/M, because (as I understand it)
they were both written by the same person...
Allegedly (according to my source), one of the original programmers that
worked on DOS in MS had told someone that the original CP/M comments were
still in the code when they got it...
Anyone with more accurate information?
Regards,
Dean
George Macdonald <fammacd=!SPAM^noth...@garden.net> wrote in article
<35d9ca3...@news.garden.net>...
George Macdonald
<help...@realworldtech.com> wrote:
>Well,
>
>I don't know how true the story is, but many, many years ago (before it was
>'cool' to hate MS), I heard from some 'knowledgable' sources that Seattle
>Computer had stolen their DOS from another source - which eventually became
>DR DOS. DR DOS was based entirely on CP/M, because (as I understand it)
>they were both written by the same person...
All entirely possible Dean. I believe one of Seattle's contributions was
in making the "advanced" 8087 work right - they used to sell the 8087 and
some code to go with it at one time... and remember Intel was pushing the
8089 for the I/O at the time. I saw some weird machines which never worked
right back then.
From my personal standpoint I started to hate MS when they were about 12
people and they wouldn't work with Altos to make the 8087 work in a
multi-tasking (Xenix) environment. I lost some face over that. Mind you
Intel was a maverick company looking to overturn the mini-computer
industry... a bit different from now.
Russell May
<help...@realworldtech.com> wrote:
>Well...
>
>DOS was based upon CP/M, which was written for the 8008 (amongst others),
>which was originally based upon the 4004. So, the 4 bit processor was the
>basis for the 8 bit processor that ran CP/M, which was the basis for DOS,
>which ran on a 16 bit processor. MS wrote a 16 bit shell, which later
>became a 32-bit shell. Win95 'starts life' as a 16-bit program loader and
>file system manager, but then 'replaces' it with it's own code to do much of
>the device interaction in (mostly) 32-bit mode. Doesn't anyone remember
>when Win95 was first introduced that MS admitted that there was still quite
>a bit of 16-bit code in it? Geez - the public really *does* have only a
>6-month memory (or is that 16-bit??)... <g>
>
I remember that CP/M was written for the 8080. Before the 8080, there
was the 8008, before that was the 4040, and before that was the 4004
which was designed to be part of a programmable calculator. I was
never directly involved with writing code for the 8008, but I think it
was quite a bit different from the 8080, and pretty slow. I still have
a spec sheet which says the 8008 has a 20 usec instruction time.
slei...@sprintmail.com
Well,
I don't know how true the story is, but many, many years ago (before it was
'cool' to hate MS), I heard from some 'knowledgable' sources that Seattle
Computer had stolen their DOS from another source - which eventually became
DR DOS. DR DOS was based entirely on CP/M, because (as I understand it)
they were both written by the same person...
Allegedly (according to my source), one of the original programmers that
worked on DOS in MS had told someone that the original CP/M comments were
still in the code when they got it...
Anyone with more accurate information?
Tim Paterson from Seattle Computer wrote QDOS (quick and dirty DOS) in
1980 while working as a consultant to Microsoft. It was designed as a
CPM "clone", but quickly took on a life of it's own after the success of the
original IBM PC in 1981. He also designed and built the first Microsoft
hardware product, which was an 8080 plug in card for Apple II machines
to run MS BASIC. If Tim stole code, he surely wouldn't admit it, and later
versions of PC/MS DOS were virtually complete re-writes of the
original MS-DOS. The only "original programmer" of QDOS was
Tim Paterson (according to "Gates", by Stephen Manes and
Paul Andrews).
Steve L
Regards,
Dean
George Macdonald <fammacd=!SPAM^noth...@garden.net> wrote in article
<35d9ca3...@news.garden.net>...
> On Tue, 18 Aug 1998 00:19:21 -0700, "Real World Technologies"
> <help...@realworldtech.com> wrote:
>
> >Well...
> >
> >DOS was based upon CP/M
>
> Conceptually anyway, but Microsoft bought QDOS (Quick & dirty OS -
really!)
> from Seattle Computer and "enhanced it". :-) Remember Xenix?.. a lot of
> people thought they did that themselves too - nope, they bought that from
> Human Computing Resources. Then there was their first "C" Compiler -
> Lattice IIRC.
>
> Of course they stopped paying for things a while ago - now they steal
them.
> :-)
>
>
slei...@sprintmail.com
On Tue, 18 Aug 1998 00:19:21 -0700, "Real World Technologies"
<help...@realworldtech.com> wrote:
>Well...
>
I wrote assembly code for the 8008 in tech school. Completely different
instruction set than the 8080, and significantly different architecture as
well. The 8008 was an 8 bit chip with about 35 or so instructions
(sorry it's been a long,long time) compared to over a hundred
in the 8080. As I recall (again my memory may not be so
good), the input clock speed on the 8008 was about 90 Khz.
Steve L