Warning: Problem w/new BIOS for Iwill KK266

[bw37]

Edit 3-24-01:

I've convinced myself that there is nothing wrong with the new kkr0305.bin bios except an ERRONEOUS "2.5V" voltage reported in the PC Health section of the bios. Read on if you want more background... MethylONE, you can ignore it!

What it appears to be doing is reporting the 3.3V I/O voltage twice, once as the "2.5V" measure and right below it , correctly as the "3.3V" value I "proved" this to myself by changing the I/O voltage with the JP10 jumper, to the +5% and +10% positions. Both ways, the "2.5V" reading goes up right with the "3.3V" reading. The difference in the two "identical" readings is a constant 0.01V which I chalk up to a to d conversion error. SOOOO: In my humble opinion, the new bios is safe, with just a minor bug. Sorry if I got anyone concerned (besides me) unnecessarily.

Original post:
The newest BIOS upgrades posted at Iwill's support site for KK266 and KK266-R mobo's appear to have a problem. I flashed to the kkr0305.bin version for my KK266-R to try to overcome the PCI bus multiplier problem I'm having when I run at 133 fsb. I now have a new problem IN ADDITION to the old one: the 2.5 Voltage reading for the CPU in "PC Health" in the bios now reads 3.3V, and my CPU temperature has gone up about 5 degrees C, which appears to confirm the Voltage reading.

Worse yet, while I did save the old BIOS when I flashed, the awdflash program won't flash the saved version to get back to the original . Somehow it was not saved correctly, so now I'm screwed. Back to the old P1-200 'til I somehow get a copy of the original BIOS or a new one that works properly!!

More info on this 2.5 V problem IS available at Iwill's Forum: <A HREF="http://www.iwillusa.com" target="_new">http://www.iwillusa.com</A>. Select Service/Support and then select "BBS". Naturally, I didn't find this until AFTER I'd tried the new BIOS posted on their support page for the KK266-R!

the more I learn, the less I'm sure I know... <P ID="edit"><FONT SIZE=-1><EM>Edited by bw37 on 03/24/01 12:04 PM.</EM></FONT></P>

 

[pvsurfer]

I highly recommend that you seek KK266 advice<A HREF="http://www.dslreports.com/forum/ocusa" target="_new"> HERE</A>.

A very helpful KK266 expert (who owns one) hangs out there under the user-name "toddblah".

Good luck!

 

[bw37]

bump

the more I learn, the less I'm sure I know...

 

[noko]

Have newest bios and noted abnormal 2.5 volt indication. This was also present with the original bios when I first booted up with the KK266. What is 2.5 volts used for anyways? The Abit KA7 had the exact same problem, their fix was to remove the 2.5volt reading from a modified Via Hardware Monitoring program. This didn't bother me since my computer was running fine.

 

[bw37]

Noko: I've investigated the "2.5V" "problem" and convinced myself that it is a bug in kkr0305 that reports the 3.3V I/O voltage twice in PC Health. With 0305, there is an additional voltage reported in PC Health. It appears as "2.5V" directly under "Vcore" in the PC Health Screen. It wasn't there at all in the original, Jan. 8, bios (for my board). The voltage it shows is about 3.3V. Thus the concern was that this might be the "VCCA Pin voltage" for the CPU which AMD specs. at from 2.25 to 2.75 Volts. To test, this morning, I changed the VIO with the JP10 jumper to see what changed in PC Health. The result: the "2.5V" reading and the "3.3V" reading both changed in synch. as follows:
VIO at std. (no jumper), "2.5V"=3.34V measured, "3.3V"=3.35V measured
VIO at +5% (pins 1-2), "2.5V"=3.40V, "3.3V"=3.41V
VIO at +10% (pins 2-3), "2.5V"=3.48V, "3.3V"=3.49V
So I think I can put that concern behind me.

I'm now using MBM 5.05 from <A HREF="http://mbm.livewiredev.com/
" target="_new">http://mbm.livewiredev.com/
</A>, based on recommendation from "toddblah" at dslreports forum <A HREF="http://www.dslreports.com/forum/ocusa" target="_new">http://www.dslreports.com/forum/ocusa</A>. Good stuff. It will save me sitting at the computer logging time/temp while I try higher speeds, stability tests, etc. It reports mosty the same info as VIA Hardware Monitor, but with much more power and flexibility. The actual voltage and temp values are about the same between the two programs. Interestingly, VIA HWM (V1.07) has always reported a "2.5V Value" but at 0.30 Volts with either bios. MBM also finds this 0.30 voltage as either "Core0" or "Core1" using the "VIA 686B Standard" settings (I changed the names and can't remember which was which originally).

(These are cut and pastes from my post at dsl. I hope it makes some sense.)

I can only guess that the bios is trying to read the CPU "VCCA Pin voltage" as suggested above. Is there another 2.5V possibility, maybe related to memory? I think I recall reading about it somewhere. Reading to much...getting too little done!

BW.......I need a new signature......


the more I learn, the less I'm sure I know...

 

[noko]

Looks like we will be getting another bios update in the future. Thanks for the explanation.

 

[Kronos]

I`ve flashed my bios to the 3/5/01 version also. I have noticed that my CPU temp using the VIA 1.07 monitor now reflects the exact same CPU temp as reported in the PC HEALTH section of the bios setup. Originally 1.07 always reported 15 degrees less at idle. That said...the extreme temp for heavy 3D use still stays the same for both as before the upgrade..about 48 to 49 celsius. Odd?

I want to die like my Grandfather...in my sleep...not screaming in terror like his passengers.