Motherboard Readings, what's wrong?

[FiL]

<A HREF="http://www.geocities.com/mol_gen/index.html" target="_new">http://www.geocities.com/mol_gen/index.html</A>

Motherboard Monitor always loses it's log file when i get a game to crash my computer, so this time i played for just a few minutes and left the games BEFORE a crash, anyhow, can anyone spot anything wrong in the readings above?
they're taken every 5 seconds by Mobo Monitor ver 5

(you can see the temp lower 5 C when i leave the game on the last 3 readings)

<font color=red>isit alan?</font color=red><P ID="edit"><FONT SIZE=-1><EM>Edited by FiL on 07/11/02 01:31 PM.</EM></FONT></P>

 

[Crashman]

Looks good to me.

<font color=blue>At least half of all problems are caused by an insufficient power supply!</font color=blue>

 

[FiL]

you don't think anything is wrong there?

is the 3.3V line dropping to 3.13V
or the 5V line dropping to 4.64V

significant?

<font color=red>isit alan?</font color=red>

 

[Crashman]

But that didn't happen right befor the lockup, did it? I mean, those are bad numbers, but if they don't happen at the time of the lockup, may be misleading. My old overclocked system used to lock up whenever 3.3v dipped to 3.15v, which only happened when I used Antec power supplies (my other power supplies of lesser wattage would simply shut off if overloaded, and my OEM power supplies of similar wattage gave better results than the Antecs).

<font color=blue>At least half of all problems are caused by an insufficient power supply!</font color=blue>

 

[FiL]

well those readings were while i was playing Operation Flashpoint for just a few minutes, Op Flash can usually lock the system up, on this occaision it didnt. but when it does lock up, the Motherboard Monitor log file erases itself, so i can't get a reading of an actual lock up

so instead i just monitored playing without a lock up to watch the voltage fluctuations. This is a very cheap 300W PSU. (the case and PSU together came to 25 pounds, that's only 37 US dollars)

59 C is quite warm too i've noticed, although i thought Palominos wouldn't lock up at that temp.

<font color=red>isit alan?</font color=red>

 

[Crashman]

59C doesn't usually cause instability, but can on some particular CPU's. Probably still a power problem.

<font color=blue>At least half of all problems are caused by an insufficient power supply!</font color=blue>

 

[FiL]

yeah, agreed, well i'm going to take heed of your sig an order a 430W Enermax, we'll see how it runs then

<font color=red>isit alan?</font color=red>

 

[Crashman]

Sounds like a plan, hope it solves your problems.

<font color=blue>At least half of all problems are caused by an insufficient power supply!</font color=blue>

 

[FiL]

hey there, just to let you know that I put the whole computer from the cheap case to a lovely CoolerMaster ATC201 Aluminium case. Also installed in the Enermax 430W PSU....Temps are down a few degrees, but the lock ups are still happening, even with the Enermax.

Specs are

AthlonXP 1900+
768Mb PC2100 DDR RAM
Soltek 75-DRV2 (on KT266A chipset)
WD Cavier 100Gb HD
Creative Geforce 2 Ultra AGP Card

So now I have 3 candidates for the problem


1. House Power Supply, i know the wiring is shaky and old, but, 3 other computers run fine in the house (P3 733Mhz, P2 400Mhz, Mac G4 900Mhz) but if it is this problem. I might need to get a UPS???

2. Motherboard, god bless Via

3. CPU (unlikely though don't you think?)




<A HREF="http://www.geocities.com/mol_gen" target="_new"><font color=red>My Computer</font color=red></A>

 

[Crashman]

I'm begining to suspect the board myself, but it could also be the memory. Try running just one stick at a time at the fastest timings available, to see which one is the least stable. Then take that one out and back the memory timing down a bit and see if the problem goes away.

<font color=blue>At least half of all problems are caused by an insufficient power supply!</font color=blue>

 

[FiL]

yep OK i've done some tests

I've got a Crucial 512Mb PC2100 CL2.5 stick and a Kingston 256Mb PC 2100 CL2.5 stick

I still get lock ups when using either one singly.

RAM is set to CL2.5 in motherboard. I will verify this and try again.....but surely i can't have 2 bad RAM sticks? that would be so unlucky.


I've tried 2 AGP cards, the Geforce 2 Ultra, and an old Matrox G200, got lock ups with both.

I think i tried Win98 and Win XP and got lock ups with both, but its been XP since March on the PC.


A mate of mine has a Thunderbird 900Mhz, i should put that in and see if the problems go away...if not then it's most likely the mobo, in which case i should get a Asus or Abit KT333 board yeah?

<A HREF="http://www.geocities.com/mol_gen" target="_new"><font color=red>My Computer</font color=red></A>

 

[FiL]

OK I don't really understand how to set RAM up so here is the BIOS page on RAM that i found, in <font color=red>red are the present day settings</font color=red>, and the other options are in <font color=blue>blue</font color=blue>. Let me know what needs changing if anything

FSB <font color=red>133Mhz</font color=red>
DRAM Frequency <font color=red>133Mhz</font color=red>
DRAM Clock <font color=red>By SPD</font color=red> <font color=blue>100, 133</font color=blue>
DRAM Timing <font color=red>Manual</font color=red> <font color=blue>By SPD</font color=blue>
CAS Latency <font color=red>2.5</font color=red> <font color=blue>2</font color=blue>
BANK Interleave <font color=red>DISABLED</font color=red> <font color=blue>2 bank, 4 bank</font color=blue>
DRAM Burst Level <font color=red>4</font color=red> <font color=blue>8</font color=blue>
DRAM Queue Depth <font color=red>2 Level</font color=red> <font color=blue>4 Level, 3 Level</font color=blue>
COMMAND RATE <font color=red>2T Command</font color=red> <font color=blue>1T Command</font color=blue>
System Performance <font color=red>Normal</font color=red> <font color=blue>Fast, Faster, Fastest</font color=blue>
DDSkew Level <font color=red>00</font color=red> <font color=blue>01, 02, 03, 04, 08, 0c</font color=blue>

if anyone can see something wrong in this set up then please let me know

<A HREF="http://www.geocities.com/mol_gen" target="_new"><font color=red>My Computer</font color=red></A><P ID="edit"><FONT SIZE=-1><EM>Edited by FiL on 07/12/02 01:29 PM.</EM></FONT></P>

 

[Crashman]

Er, red and green? Hmm, sorry, I'm one of the something like 1 in 20 males with red-green color blindness. How about red and blue?

<font color=blue>At least half of all problems are caused by an insufficient power supply!</font color=blue>

 

[phsstpok]

I would have bet money on the power supply but I guess that isn't your problem.

Since, you have a VIA chipset my second choice would be the VIA 4-in-1 drivers. I usually recommend a clean install of these before anything else just because, well because Via chipsets are so troubling. (I personally don't like the newest driver 4.40, preferring 4.38 but I have older hardware).

Download the driver <A HREF="http://www.viaarena.com/?PageID=2" target="_new">here</A>. Note that Via has been claiming Window XP compatibility for 4-in-1 drivers since version 4.38, or thereabouts. To do a clean install run the installer once but for each component select the uninstall option. Don't reboot when prompted but rather run the installer a second time. This time let all components load. This time Reboot when prompted.

Also which nVidia drivers are you using? I wouldn't use anything later than 23.xx for a Geforce2. Yes, the newer ones are bit faster but I found them less stable with my Geforce2. Use whichever driver you want but it's a good idea to reload nVidia drivers whenever you have updated 4-in-1 drivers.

Other things you can check are an overheating video card. You can check this two ways. 1) you can underclock the GPU making it produce less heat. 2) you can open the case and use a room fan to blow air into it. If either of these make your system system stable then you have a heat issue (Since you didn't change anything you know its heat). Item 1 points directly to the video card. Item 2 only means heat (but not necessarily the video card).

The only other thing I can think of is try bumping your DDR SDRAM voltage to 2.6 volts. Some people have found success with minor voltage changes on their DDR memory.

Good luck

<b>I have so many cookies I now have a FAT problem!</b>

 

[FiL]

hehe ok it's done, red and blue, so could you go and see if you can spot anything to change in my RAM settings : )

<A HREF="http://www.geocities.com/mol_gen" target="_new"><font color=red>My Computer</font color=red></A>

 

[FiL]

Since, you have a VIA chipset my second choice would be the VIA 4-in-1 drivers. I usually recommend a clean install of these before anything else just because, well because Via chipsets are so troubling. (I personally don't like the newest driver 4.40, preferring 4.38 but I have older hardware).

OK I've no idea what 4 in 1 driver i'm using but it was from February 2002....so i'll get the 4.38 version you suggested.

I'm using the very lastest Geforce DetonatorXP drivers, although i've had lock ups even with ver 21.xx

I've already tried putting the DDR RAM up to 2.6v and it doesn't make any difference.

1) you can underclock the GPU making it produce less heat.

I hope it's not a heat issue for the GF2 cause i've got the ATC201 Coolermaster case, Aluminium with 4/5 case fans...but i'll underclock the Ultra to Geforce 2 Pro/GTS levels to make sure.

<A HREF="http://www.geocities.com/mol_gen" target="_new"><font color=red>My Computer</font color=red></A>

 

[phsstpok]

It's the "clean" part of clean install that is most important for the 4-in-1 drivers. They have to be uninstalled for an update or a re-install to work properly. (Don't ask me why). nVidia drivers seem to need the 4-in-1's or at least the AGP driver to be preinstalled to work properly, meaning stable. Usually I recommend a clean install of the nVida drivers. For Windows 9x/ME Detonator Destoyer works well but for XP just installing the old driver will have to do. Think of all the above as a single procedure, 4-in-1's then nVidia each time you need to change either one. Also you should do the same after updating the BIOS since Windows often detects a new BIOS as a change in hardware.

<b>I have so many cookies I now have a FAT problem!</b>

 

[Crashman]

I would try changing your Cas Latency to 2.5 and your Command Rate to 2T.

<font color=blue>At least half of all problems are caused by an insufficient power supply!</font color=blue>

 

[FiL]

hehe i think you misunderstood the colour scheme : )


it's already set at CL2.5, and 2T

red is what is currently selected in the BIOS

<A HREF="http://www.geocities.com/mol_gen" target="_new"><font color=red>My Computer</font color=red></A>

 

[Crashman]

Sorry about your luck, you might have a bad board.

<font color=blue>At least half of all problems are caused by an insufficient power supply!</font color=blue>

 

[FiL]

even with the new Enermax supply, the 5V line is often dropping below 4.69V in normal windows operation.

so this is unlikely to be the PSU's fault, but can mobos actually have power distribution faults?



<A HREF="http://www.geocities.com/mol_gen" target="_new"><font color=red>My Computer</font color=red></A>

 

[Crashman]

It's possible, or it's also possible that you just were unlucky enough to get an Enermax that was a little off.

<font color=blue>At least half of all problems are caused by an insufficient power supply!</font color=blue>