• 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.

Odd issues with new build.

Saga

Saga

Banned
Going to recap as best as possible:

Running PC3200 Corsair ValueSelect - 4x512 for a total of 2GB. For diagnostic purposes I let the board default all settings and noticed that when running CPU-Z it displayed the memory clock at 166 (333 @ 3-5-5-9). Noticing something was definitely wrong with this I manually set the ram to 400mhz in the BIOS on an ASUS A7V880. CPU-Z now properly displays a 200mhz clock however it fails Prime95 within 15 minutes.

Set RAM voltage to 2.75 @ 2.5-3-3-8 (listed timings) and booted up. Prime still failed within minutes. Noticed that ASUS Probe was listing VCORE as fluxuating from 1.69v-1.71v so manually set VCORE to 1.75v. Prime ran stable for 4 hours 37 minutes, crashed with "4.9898... expected less than 4" or similar (screenshot at home, at work right now). Sort of an odd scenario and I can't tell if it is processor, memory, or board that is bad (this is the second of these boards I have used, both have had the same issue so I assume processor or RAM).

Tested each of the 4 512's individually over 24 hours with memtest, only one stick had two errors on test #7, removed that faulty stick and another and ran PC with just one pair of 512. Prime failed after 4 hours.

Is this memory controller related or power maybe? Also note that when running 2 sticks of memory dual channel watching the Windows Task Managers performance tab the CPU spikes between 5-20% constantly when running a blend test - when all 4 sticks are used CPU is at 100% constantly on the same blend test. What am I SUPPOSED to be seeing when running a Prime95 blend test?

Suggestions welcome this is the first time I've ever had this issue and I'm starting to think this is just a funky ASUS fluke - PSU being used is a trusty 480w Antec I have in about six systems in this house.

 
You have most likely reached (and passed) your processor's ceiling for a stable overclock. I had a Barton 2500+ that was not stable enough to run Windows at 2300 MHz or above, and was stable enough for all my apps/games at 2200 MHz but would crash Prime95 in less than an hour on average. Try decreasing your multiplier by 0.5 and run Prime95 again and see how you fare. You could also try nudging your VCORE a bit higher, but I would caution against approaching 1.775v-1.8v without extremely good cooling.

When you tested with MemTest, were you testing at stock clock speeds or at your overclocked speed? If you want to accurately determine if your RAM has physical errors, you should run diagnostics at stock speeds to rule out any overclock-related soft stability issues.

Let me know how it goes.
 
Forgot to mention, for the sake of troubleshooting the board's FSB was reduced to stock settings (thus the 2463.8 should be 2200). The scenario outlined above is all relative to stock FSB of 2200 on the 3200+ Barton. Even at stock would you suggest pushing the VCORE past 1.75? Running a copper thermaltake with a 80mm speed adjustable fan maxing out at 6,000rpm so I'm not worried about the heat that much. 1.75v @ 2200mhz is idle 35ish, 39 load with fan pushing full blast.
 
Well that is quite odd, then. Do you have another processor, or some other RAM you could try in that system? Conversely, have you got access to another system in which you can test your CPU & RAM?

Also perhaps try seeing if there's any BIOS updates for your board. If this is hardware you purchased recently, you may also just want to see if you can get it exchanged.
 
Yes. I have a 2600+ with a shuttle board that supports PC3200 which I used to test the RAM. Only one stick is bad over 24 hours. The shuttle board does not support 3200+ Barton's however, with the newest BIOS the max it supports is 3000+ so I cannot swap the processor.. the CPU is retail so I suppose I could send it back if it's actually a bad proc. This is the second of these boards (first one was RMA'd for power issues but unrelated to these problems - both boards have have this problem). The first thing I do with every build is upgrade the BIOS.

I'm hesitant to send the 3200+ back without testing it on another system but most of my boards are 939's (I'm still an oldschool fan of 462's, like a car guy likes the 68 Vettes, older doesn't mean they aren't as good in many cases). Will dismantle system in question and test the proc on another rig tonight, pretty sure I have a Soyo DragonII sitting around still. Will also throw in a 3000+ 333 and see how that ticks.

I still have hopes that it's not the processor, this Barton is my baby and I could get a mean overclock out of it. =(
 
Yeah I know the feeling, it is teh suck to have hardware that "ought to" be stellar act flaky on you. Good luck, hope you can narrow it down a bit more. I know I'd be (and have in the past been) pulling my hair out at this point.
 
Well, threw in a 3000+ Barton and flipped CMOS to let everything default to stock settings. Prime has been running for two hours now without a hiccup. Longest I've run is 4 hours with the 3200+. I just find it odd to think that the 3200+ is faulty, I've never had a processor go bad on me in that way, previously they've just flat out stopped working. Guess it's time to get in touch with AMD for a replacement.

Nobody answered my earlier question, with Prime95 running doing a blend torture test is the CPU useage supposed to spike between 5%-30%ish or is it supposed to stay stable at 100%? Testing the 3000+ right now and it's stable 100% with all 4 sticks of RAM in.
 
I am going to post a fairly generic message about this board as I am putting it on several forums.

THe ASUS A7V880 Mb is crap. I, along with many others have had nothing but trouble from it. The ViA 880 chipset is crap. Many people have trouble getting this board to work at 400 fsb (amd athlon 3200) me inclusive. It worked with one cpu (athlon 3200) but not another (the other CPU is working fine as a 2500 in another mb - this mb only support 333fsb - tried to get it to work at 333fsb in A7V880mb but still crashed). Many many many other people have had similar trouble with this mainboard

I have tried 450w and 500w psu but no difference. I use 4x 512mb (dual channel) Kingmax DDR400 RAM recommended by ASUS (exact model). I get error messages by memtest86, but none when I run it at 333. Have tried ramping up voltages and setting all bios manually. Other PC3200 RAM tests fine. I have seen other people have similar trouble with RAM on this MB (even the ones recommened by ASUS)

THe PC is running fairly stable now, but every now and then it freezes when left on for a while with nobody using it. I advise anybody wanting a 400fsb mainboard to go with a nforce 2 chipset. I have only heard good reports about them
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top