Can only run AGP 2X

[Rdn]

Hello all.
1.2 Thunderbird, Ep-8K7A+ mobo, 512 RAM, W2k SP4, Radeon 9600XT.

When using AGP 4X, 3d games generally lock up after a few minutes or less.
What happens is that the screen locks, and the sound plays on or "loc-loc-loc" locks up as well. The keyboard doesn't seem to respond. (Pressing any of the lock keys yields no LED response.)
Occasionally, the screen just goes blank, or actually goes into sleep-mode.

I had this problem with a GeForce 2 card as well.

Anyone have any ideas?

 

[Koopa]

I also have 9600XT but no problems with 4x AGP, so I dont think its the gfx card, unless there's some problem with the manufacturer you bought it from or your individual card. More likely to be your mobo, but I wouldn't know much about that.

I spend my time here hoping that one day I might learn something.

 

[pIII_Man]

are you overclocking you're cpu or more importantly, what is you're current FSB speed?


If it isn't a P6 then it isn't a procesor
110% BX fanboy

 

[Crashman]

I used to have problems like that on some VIA chipset systems. Especially when I used VIA's drivers instead of the default Windows drivers.

Of course it could be a power problem, or a RAM problem.

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>

 

[Rdn]

Thanks for trying to help, folks.

piii_Man: Not OC'ing, FSB speed is 266MHz.

Crashman: The EP-8K7a+ uses the AMD761 chipset. The only AMD drivers I've installed are the AGP drivers.

Power or RAM problem? The only way I could find out if it was a RAM problem would be to borrow some DDR RAM, right?
What about finding out if it's a power problem?

 

[neveroldmilk]

The problem will probably be the power.U know 9600XT needs much power to work,so u mobos power supply may not be fit for it.Just find a Gerforce2mx(use little power)plug into u mobo,and if no problem u'd better mobo.

Truth is somewhere out there!
--Movie 'X Files'

 

[Rdn]

I've tried a GeForce 2 GTS, and I had the same problem there.
400W PSU, and I can't think of anything in my PC that draws alot of power.

 

[Crashman]

Motherboard Monitor 5 might work with your board to monitor voltages.

Another thought occured to me...what sound card are you using?

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>

 

[Rdn]

Soundblaster Live 5.1.

BTW, is a case-temp of 40C too warm?

 

[pIII_Man]

that is pretty hot. Blow a standard box fan into the case to see if the problem goes away. If so work on your case ventillation.


If it isn't a P6 then it isn't a procesor
110% BX fanboy

 

[Crashman]

I was afraid you were going to say that. It sounds like a symptom of the infamous 686B bug, that affects many boards using the VIA 686B southbridge and either a SB Live or Vortex 2 soundcard.

This is rather old stuff, and you're dealing with a lot of people who can't remember who was president before Bush, so finding any information might take a while.

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>

 

[Rdn]

piii_Man: Ok, wether it helps this problem or not, I guess I'll have to improve ventilation to prevent any other problems... thanks.

Crashman: Do you remember if the problem was solvable?

Of course, with todays prices, a decent Motherboard with integrated 5.1 and an AMD XP2500 won't cost more than what I gave for the 9600XT two days ago. Still, I hadn't planned on upgrading those two components for a while.

 

[phsstpok]

It could be the infamous AMD/WIN 2K AGP bug. Click link for registry patch.

<A HREF="http://www.amd.com/us-en/Processors/ProductInformation/0,,30_118_756_759^871^2367,00.html" target="_new">http://www.amd.com/us-en/Processors/ProductInformation/0,,30_118_756_759^871^2367,00.html</A>

The next link describes the issue.

<A HREF="http://support.microsoft.com/default.aspx?scid=KB;EN-US;Q270715&" target="_new">http://support.microsoft.com/default.aspx?scid=KB;EN-US;Q270715&</A>



<b>56K, slow and steady does not win the race on internet!</b>

 

[Rdn]

phsstpok: I've tried that before, but it didn't work.
BTW, perhaps I should delete that key, as it didn't solve the problem?

Again, thanks for taking your time to try to help me.

 

[Crashman]

Yes, it was a real ah heck though, I don't remember the details. Perhaps more information in the old graphics or motherboard/chipset posts.

To test for the 686B/Live problem, he could try taking out the card and unistalling the driver, if that fixes it he could start looking for details on workarounds.

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>

 

[phsstpok]

Too bad about it not fixing your current problem.

I would not remove the patch though as the bug is real issue, confirmed jointly by AMD and Microsoft. If you remove the patch the bug might get you later.

<b>56K, slow and steady does not win the race on internet!</b>

 

[Rdn]

I looked through the 686B/SB Live posts.
I have the sound distortions, don't know about the file transferring corruption since I only have one HD.

I didn't see anyone mention not being able to run AGP >2x though. I'll try to run motherboard monitor and check the voltages to see if it's a PSU problem (it's supposedly 400W, but not a brand name, so it could be erratic with the AGP voltages, I guess.)

Thanks, guys. Perhaps I will return to haunt you about this problem later.

Lastly, are KT600-based cards backward-compatible with Athlon Thunderbirds? If so, I could get something like the MSI KT6 Delta or an Abit KV7 and use the integrated 5.1 soundcard, and use the Athlon 1.2 with it until I can afford a Barton CPU.

 

[phsstpok]

The now defunct Viahardware.com (I think it was viahardware.com) did a pretty comprehensive test of the SB Live problem. They duplicated the sound problem (crackling and/or looping) on Via chipsets (KT133A & KT266), Intel chipsets (i815 I believe) and a third chipset (possibly AMD 761). That is, they produced the problem on 3 or 4 motherboards. The concluded it was the SB Live.

There never was a fix but the workaround for the "crackling" problem was to limit Windows volume to no more than 50% and use your speakers' volume control to makeup the difference. This kept the noise floor to a minimum and made the "crackling" inaudible (to most ears).

I think the article is still hosted somewhere at Sudhian.com but I'm not sure.

The infamous VIA 686B bug was the data corruption issue. Supposedly any mobo with the 686B Southbridge was subject to the problem. It wasn't limited to SB live owners but SB live may have made the problem more noticible.

The data corruption issue is easy to test. Copy a 100 MB file from one location to another (HD to HD will work as will CDROM to HD). If the new file isn't identical to the original then your mobo definitely has the corruption issue. If it does then your need a fix for PCI latency.

George Briese wrote one fix, one that actually worked. It supposedly reduced PCI performance but it wasn't major.

There were other fixes that were supposedly included in Via 4-in-1 updates, motherboard BIOS updates, and Windows Service Packs. However, I've heard claims that these other fixes work and claims that they don't work.

I would stick with George Briese's Latency fix for VIA chipsets. (I have no idea if the fix works with AMD 760 northbridge + VIA 786B southbridge, though).

I would not even try it unless you actually confirm the data corruption issue. Data corruption is serious but treat the fix as a last resort.

I don't have any experience with KT600 but I won't touch one. I'm tired of VIA 4-in-1 drivers on my old VIA boards. The drivers constantly "break". The complaints of VIA mobos don't seem to stop to this day. With nVidia nForce2 around (but not perfect) I don't see any reason to choose anything else (unless you need IDE RAID).

That's just my opinion, of course.



<b>56K, slow and steady does not win the race on internet!</b>

 

[Rdn]

phsstpok: Is checking the filesize enough to check if tow files are identical? Anyway, according to the readme for the PCI Latency patch, it only works for "a complete VIA chipset", so it won't help me if I do have the problem.

Used MBM5, and it turns out my +5V alternates between 4.76 and 4.66V. As this is outside the +/- 5% range, I guess I need a new PSU. Hopefully that will help the AGP problem. If not, I know I have to get a new motherboard.

After browsing through some of the 686B threads in the archives here, I didn't see any other possible solutions than the ones mentioned here.

Sincere thanks to all of you for taking your time to help me locate this problem.

 

[phsstpok]

I use DOS File Compare, FC.EXE, but I don't think you have that available with Win2K.

You could have found your problem as you say, with your PSU. You're voltage is in the same range where I have problems.

I'm an overclocker. My 330 watt PSU is normally fine but pushing my overclocking pushes my +5V rail down to around 4.7 volts. I run into stability problems when it drops a little lower than that, say 4.65 volts and down.



<b>56K, slow and steady does not win the race on internet!</b>