Athlon & K7M quirks & solutions
Sales @ Bigfoot
I've been messing around with about 50 ASUS K7Ms v1.04, a bunch of Athlon
500Mhz cpus (running 700@1/2, 750@1/2+1/3 and 800@1/2+1/3), a wack of power
supplies and most importantly a POST board. I'm not playing with stock
speeds but I'm starting to see some reproducable patterns when overclocking
the Athlon.
I'm going to be making up a half-way decent FAQ when I'm done treasure
hunting. This motherboard is quirky to say the least. And believe it or
not the jumpers on the board make a _BIG_ difference when trying to boot,
nevermind post. I've also discovered that the nasty bug where you have to
hit reset after a cold boot in order to warm boot is a keyboard problem.
If anyone wants to contribute to the FAQ send email to
ne...@bigfootcomputers.com and I'll throw it in and credit you. Just please
make sure it's a valid solution or quirk, meaning no guesses :)
Hope this helps,
Neil
Bigfoot Computers
* * Chasman
Great, we need this resource.
The 1007 Bios seems to fix the K7M keyboard problem. I've been following
the NG and Forums for all of the Athlon mobos. I'm beginning to suspect
that the keyboard problem affect all of the mobos with AMD/VIA chipsets.
I had problems with a FIC SD11. After I updated the Bios it had the same
black screen death as the K7M when I first installed it! Once the power
button was pushed the system didn't respond to either the on/off button or
the reset button. I got my K7M working so I don't have time or interest in
messing with the SD11.... It's going back!
Chas.
Sales @ Bigfoot <sa...@bigfootcomputers.com> wrote in article
<u7J%3.69060$up3.1...@news21.bellglobal.com>...
Sales @ Bigfoot
advance for the length but there's some good stuff in here that many of you
might benefit from.
**snip**
I've isolated three bugs so far.
The first is the cold boot problem, where you power on and get a black
screen. Hitting reset will warm boot you successfully. I've reproduced it
with 4 keyboards. It's officially a keyboard problem =) A little cheap
Mitsumi PS2 keyboard works like a charm. I'd recommend buying off of the
AMD list for keyboards. The reset fix does not seem to be PSU related.
The second bug is another cold boot problem with no sound or picture. This
one is two parts. The first is obvious, inadequate power supply. The
second part of this is the Voltage I/O setting located in the upper right
corner of the mobo. It's set at 3.4v factory and the manual recommends
moving it to 3.56v (position 3) for overclocking. This jumper is wacky and
I'm a little confused with it. This jumper was causing all of my no-boots
(even the reset fix did not work). Because I had a post board installed I
could read where the system was
hanging and it was always memory related. With Micron original PC133 RAM,
the
jumper had to be moved to 3.1v, an underclock to boot. I don't get this
maybe you guys can fill me in. I've tried some other PC133 and PC100 RAM
and those both need the other setting (not factory) 3.56v. I can sometimes
get a couple pieces to boot at factory 3.4v but I'm letting that one go
right now because all the CPUs are overclocked. Bottom line here, use
original RAM, brand name, for the least probs.
The last bug that is kinda funny has to do with RAM sticks. 64M sticks seem
to work like a dream. However, I have 5 PC133 128MB sticks here, of various
configs. Some of you guys benefit being stateside with AZZO, Crucial and
the like.
Up here in Canada we often get crapola PCBs for our chips. My one original
Micron
PC133 128MB works just like it should as a 128MB. However various other 128
MB
pieces POST as 64MB pieces. But get this, when I fill all 3 DIMMs up with
RAM, no matter what size, they will all POST correctly. Two filled out of 3
get size errors, usually too big. This result may vary with quality of
DIMMs tho.
I've been working with the same motherboard for the last couple hours (I
have about 50). Once I started to be able to reproduce the errors I didn't
want to swap boards until I established some patterns. I managed to get my
431 watt power supply working. An In Win Powerman 235w supply got me
started with good boots. Then I was able to rule the 431 watt supply as ok.
If I'm annoying you guys with these posts just say so. I thought you might
want to stay in the loop. It certainly means that the K7M is looking very
promising with solid POSTs. Thank God I found my POST board and reference
codes :)
Neil Lightfoot
Bigfoot Computers
ne...@bigfootcomputers.com
www.BigfootComputers.com
Darin Genereux
<sa...@bigfootcomputers.com> wrote:
<good stuff snipped>
>If I'm annoying you guys with these posts just say so. I thought you might
>want to stay in the loop. It certainly means that the K7M is looking very
>promising with solid POSTs. Thank God I found my POST board and reference
>codes :)
>
>Neil Lightfoot
>Bigfoot Computers
>ne...@bigfootcomputers.com
>www.BigfootComputers.com
>
No don't stop, this is good stuff Neil! Many thanks for the info,
it's really hard for us end users to correctly diagnose these weird
problems without the necessary equipment.
Darin
Red Neck
wich is supposed to correct the booting problem, I get an error message
saying that the BIOS didn't load properly. The machine still posts after the
flash and has the correct BIOS number and date but the cold boot problem
still exists. Any suggestions? I'm using a new Microsoft keyboard.
"Sales @ Bigfoot" <sa...@bigfootcomputers.com> wrote in message
news:u7J%3.69060$up3.1...@news21.bellglobal.com...
> I've surfaced out of the trenches to make this post :)
>
> I've been messing around with about 50 ASUS K7Ms v1.04, a bunch of Athlon
> 500Mhz cpus (running 700@1/2, 750@1/2+1/3 and 800@1/2+1/3), a wack of
power
> supplies and most importantly a POST board. I'm not playing with stock
> speeds but I'm starting to see some reproducable patterns when
overclocking
> the Athlon.
>
> I'm going to be making up a half-way decent FAQ when I'm done treasure
> hunting. This motherboard is quirky to say the least. And believe it or
> not the jumpers on the board make a _BIG_ difference when trying to boot,
> nevermind post. I've also discovered that the nasty bug where you have to
> hit reset after a cold boot in order to warm boot is a keyboard problem.
>
Eppe
Red Neck <red...@home.com> schreef in berichtnieuws
81p3pf$sfa$0...@216.39.137.4...
Sales @ Bigfoot
> I have exactly the same problem. I have the Logitech Cordless Desktop.
>
>
>
Eppe
Sales @ Bigfoot <sa...@bigfootcomputers.com> schreef in berichtnieuws
Zf704.10902$sy5....@news20.bellglobal.com...
amdtweakster
have a 500 o/c'd to 700 but on an FIC SD-11 (very stable). I'm building
another Athlon 500 system for a friend and thinking about getting him
the K7M ($25 more) just to see if my o/c 700 will run at 770. Let me
know please.
Thanks :)
John Gill wrote:
>
> Has anyone any information on best BIOS settings re memory tinnings for the
> ASUS KM7 do they make any difference. Currently I have a 500 athlon clocked
> to 700 by overclockers.co.uk running at 735 with 105 frontside setting all
> very stable.
> >
> > A lot of problems in setting up went away when I installed windows 98
> version 2. Out of interest my V770 ultra nor runs at higher core and mem
> speeds than it did in a abit BH6 board.175/200. ASUS board is ok by me.
amdtweakster
John Gill
Red Neck
"Sales @ Bigfoot" <sa...@bigfootcomputers.com> wrote in message
news:Zf704.10902$sy5....@news20.bellglobal.com...
Phred
This is a copy of an earlier reply... I cannot determine if it posted, so
apologies in advance for the repeat, but I think this thread is better
suited for this anyway...
Anyway, look toward the back of the manual, and a leeetle Tiiiny picture
that attempts to describe the solder points you need to short to clear the
CMOS... (security section). By 'shorting' pins 186 you can avoid the
battery removal routine that I've seen mentioned many times... I use this
ALLL the time, as my K7M refuses to behave...
I can't begin to list the probs I'm having with mine...
First boot was fine, Trinity modified (and pre-tested) straight to 800MHz,
using the ill-advised matching 1.85v on mobo. Of course I got the dual-fan
alarm feature... I had to 'ignore' everything in the hardware monitor to
get it to stop... BUT... several times after shorting pins 186 to clear
the CMOS the system posted and booted to DOS fine with NO alarm at default
BIOS settings... <g> that is only the first of many multiple-faces of
this guy...
After an initial burn-in of 24hours running scandisk /surface with no
probs, I pressed the reset button... It SHUTS off the computer...
(enlight case w/ a PC Power/Cooling 350T). No jumper setting has addressed
this, I simply have no reset capability with this setup. When I tried to
power back on, power light blinks on, but no start-up. three hours later,
(having regained my temper) I pressed the on/off button, it started right
up... This IS a rev 1.04 board, with the P2T capability, but of course
did NOT include the P2T cable...
I'm thinking overtemp at this point, so I short 186, sure enuf, it starts
right up, albeit with the fan-alarm 'feature' beeping away...
run for a while, doing all I can come up with for DOS short of installing
Hexen or QuakeDOS... all is well. Shut down, cannot re-start... hmmm...
short 186, starts fine... BIOS mebbe? tried 3.31v as suggested by many,
system has not yet booted at that voltage, though I have the Micron memory
that that is specifically targetting...
I am now down to an S3 PCI video, quantum 18G and athlon 500 @ 800
128 MB Micron PC133 CAS2
KM1006, THEN KM1007 successfully applied, though I have the dreaded
FOO2NT BIOS <g>
Second fan of FFE moved off-kilter by 33% to get 'round ASus brilliant ATX
connector placement...
Processor board is wide-open to allow dipswitch mod from Trinity micro...
you can touch the heat-sink mounted to the processor, no indication that it
is overheating
P2T ready, but (surprise!) no P2T cable included
mb temp of 74deg F
scandisk /surface on 8G part for 24 hours as burn-in...
Enlight ATX case w/ PC Power & Cooling 350T ATX pwr supply
System is running DOS only, have not yet attempted the Window98 v1
install...
(now what have I forgotten to mention?)
knf
Sales @ Bigfoot <sa...@bigfootcomputers.com> wrote in article
<u7J%3.69060$up3.1...@news21.bellglobal.com>...
Darin Genereux
and get your system running at 500Mhz first, then increase your CPU.
I know you say it was tested at 800Mhz, but motherboards, memory, and
components are also a large factor in o/c success.
Also make sure your Reset switch is plugged into the correct jumpers.
<G> Good luck!
highli...@my-deja.com
card i was at 555 with a FSB of 111mhz , now i have my cpu at 700 , it
isnt stable at anything higher the default , but who cares ! I did
bencmarks on a system at 700 and 800 and its about a 10% overall
difference in my test ( Sisoft Sandra , 3D Mark2000 , 3D Mark99 , Final
Reality , Q3 testdemo , Wintune98 , Halflife timed demo )
Not enough for me to take my 700 to 800 , even though i cant get to
750 ! but im still at 1/2 cache multiplier , if i changed it to 1\3 i
can get it to 800 no sweat !
Sent via Deja.com http://www.deja.com/
Before you buy.