• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

OT: Help needed on Overclock

BofRA

Platinum Member
Well I have been pulling my hair out over the last two days. I got a refurbed 8RDA+ and a 1700+ from Newegg this week. Put it together in my system but ran into problems with my Ti 4200. To get it to work had to go back about 5 bios revs. Finally got Windows installed and started overclocking. Now here is the fun part. I have two 256MB K-byte pc2100 memory chips. I didn't expect much in overclock bu twent ahead and tried 10x166. Ran just fine. 10x200 ran fine. bumped vcore to 1.8 and ran Prime 95 for 12 hours with no errors at 11x200. It does SETI wu's in 2:20. I am thrilled. Then my horror starts. I installed a GAME!!!. As soon as I tried playing Everquest, Computer locked up. Same thing for UT2003!. so go back to 11x133 and games run fine.

So my question is: can the memory be causing this or is it something else? Sorry for the length of this but I am kind of frustrated right now and needed to vent. Thanks for any help.
 
To get it to work had to go back about 5 bios revs.

First suspect....... consider stock BIOS?

10x200 ran fine

Probably not really............. I'm guessing your PCI bus is way too far out of spec for the video to work correctly when acutally used

The obvious suggestions are always the best...... or so they say.

Consider getting your system back to stock.
then do your OC'ing in MUCH smaller increments testing with real applications.
when they quit working, you have gone too far on that particular setting.
No matter what you may read in various posts/articles...... there are limits to OC'ing


Best of luck

-Sid
 
Video card might be it, then again it might not be it.
What other test you try running? This and a couple other forum members tell me that prime95 is not the best stability test out there....
...
...
Then again... I am so drunk right now that I am about to hit the sack after checking my mail. I'll try being more help tommorrow when my mind is not on a teasing woman whom says all she wants is to be friends...
 
sounds like a video card issue to me.

Well the GF4 is a PNY 128MB 8x. I pulled that and went back to my radeon 64mb VIVO. Still same problem. LEt me clarify. The system doens't freeze just lags REALLY bad. i.e. takes about 2min to get to task manager and close out everquest. By the way EQ was using about 210mb of RAM just befor closing out.

Probably not really............. I'm guessing your PCI bus is way too far out of spec for the video to work correctly when acutally used

Well Epox claims the PCI bus is locked at 33 and I have tried the AGP bus locked at 66, 78, and 83. So I doubt that is it.

What other test you try running? This and a couple other forum members tell me that prime95 is not the best stability test out there....

I have also ran Sandra burn in test for 6 hours with no problems. What other tests are there to run?

I am about at the point where I am going to pull the motherboard and place it on my crack rack and go buy a A7n8x and some GOOD memory. At least that way I get a 2:20 SETI cruncher, and my kick a$$ system back.
 
Try Memtest86. This will test your memory, as if you have any memory problems, they probably won't start showing until you start addressing the end parts of the memory when you're loading game levels etc.


Confused
 
PCI is locked by default, but I do not think that the AGP is. Make sure you go in and manually lock the AGP to 66 and try again. And yes try memtest! This will let you know if it is ram. Let it run overnight.
 
Is running one stick of ram more stable then two ? Maybe you could try just one and see what happens, and i don't think you'll loose all that much on times, i think someone once said 5-7 min.
 
You need memtest86 as well as Prime95 to test for memory stability. Some people also prefer other torture tests (I am one of the ones who thinks Folding@Home with -advmethods is great 😉), but those first two (a few full repetitions for the former, and 12 hours - or overnight - for the latter) plus a rigorous test of whatever applications you use daily, should tell you whether or not your system is stable.
 
Originally posted by: MoFunk
PCI is locked by default, but I do not think that the AGP is...
IIRC, AGP was locked by default on my 8RDA (in addition to PCI, of cousre), but maybe it isn't locked by default in a different BIOS revision. 😕
 
Yes, do try Memtest86. The configuration you'll want is c-1-2-2-3-0-space
My feelings is 24 hours is too long. Let it run for ten loops.

I would not worry much about the AGP/PCI lock, that is just a precaution. Even if you lock the AGP @ 83, both your video cards will smile down at the speed and say "Is that all you got!"

Try lowering your CAS settings in the BIOS and see if that helps. It could work wonders... after all, your PC2100 is running WAY out of spec. 🙂 Memtest will detect if your memory is bad though...

If all else fails, raise the multiplier and lower the FSB down to 166. 200FSB will only give braging rights and not be all that much faster than 166.
 
Although I think it is your memory, I hate to say it... but your motherboard is a refurb after all. It was returned for a reason...
 
Well I gave up trying. The board is now in my shed still running at 11x200 doing 2:20 WU's. I put my trusty Dragon+ back in my system for now, and will save up for another motherboard/mem/cpu combo. Once I can afford it I will probably go with Twinmos 3200 mem and an A7N8X or an NF7-S. Thanks for the help guys.
 
FYI, the northbridge may be overheating. A stock 8RDA+ has a 1cm x 1cm square of thermal gum between the heatsink and the northbridge. From personal experience, I know that the first-gen nForce2 northbridges get very hot. I suggest removing the northbridge cooler, removing the thermal gum, and replacing it with thermal grease.

On my 8RDA+, it took a lot of thermal grease to fill the gap, because the northbridge is slightly recessed in the center. So do a test-fit and then pull the heatsink off to make sure it's getting full contact, and add more thermal grease if it isn't.

The southbridge can get pretty darn hot too, and some folks have put coolers on it too (note that there are pushpin holes so you can put on a heatsink like this one). This will keep you from using longer PCI cards in two or three PCI slots, but may help when the audio processor is going full steam ahead 😀
 
Originally posted by: mechBgon
FYI, the northbridge may be overheating. A stock 8RDA+ has a 1cm x 1cm square of thermal gum between the heatsink and the northbridge. From personal experience, I know that the first-gen nForce2 northbridges get very hot. I suggest removing the northbridge cooler, removing the thermal gum, and replacing it with thermal grease.

On my 8RDA+, it took a lot of thermal grease to fill the gap, because the northbridge is slightly recessed in the center. So do a test-fit and then pull the heatsink off to make sure it's getting full contact, and add more thermal grease if it isn't.

The southbridge can get pretty darn hot too, and some folks have put coolers on it too (note that there are pushpin holes so you can put on a heatsink like this one). This will keep you from using longer PCI cards in two or three PCI slots, but may help when the audio processor is going full steam ahead 😀

I am hoping I took this out of the equation by moving my crack rack to my shed and using a Windows AC unit to maintain temp around 55-60F in the shed. I will let it run for a week or so and see how stable it is.

And for those that were interested in the steppings on the chip here they are:

AXDA1700DUT3C
JIUHB0312VPMW
9468957281490

It was ordered on the 10th of May from Newegg.
 
Back
Top