DFI UT X58-T3EH8 continued - Part 2

[Praz]

ci2kla

Your testing perfectly illustrates the issue at hand. When 8 instance of SuperPi will run with no problem but a single instance's success is down to a couple of hundreds of a volt something is wrong at the most fundamental level.

[Dami3n]

okay seems the freezing issue is pointed to the relation between VTT and vDIMM.
Here's the test.
vDIMM 1.65v VTT 1.28v CL8-8-8-20 2000Mhz -> 32M PASS
vDIMM 1.62v vTT 1.28v CL8-8-8-20 2000Mhz -> 32M FREEZE
vDIMM 1.65v VTT 1.34v CL8-8-8-20 2000Mhz -> 32M FREEZE
vDIMM 1.65v VTT 1.28v CL9-9-9-20 2000Mhz -> 32M PASS
vDIMM 1.62v VTT 1.28v CL9-9-9-20 2000Mhz -> 32M FREEZE

I'll say from this test my chip doesn't like VTT to be too close to vDIMM...

Same batch, and same findings. These xeon are very sensible to vtt/vdimm gap.

[MomijiTMO]

Is this problem on other boards as well?

[ci2kla]

ci2kla

Your testing perfectly illustrates the issue at hand. When 8 instance of SuperPi will run with no problem but a single instance's success is down to a couple of hundreds of a volt something is wrong at the most fundamental level.

Thanks. Will that be a MB BIOS issue that you're currently working on?
BTW I'm sure the 26 reboot issue is much improved in 0619 BIOS. Good for those who run for high uncore / dram clock testing.

[Praz]

Thanks. Will that be a MB BIOS issue that you're currently working on?
BTW I'm sure the 26 reboot issue is much improved in 0619 BIOS. Good for those who run for high uncore / dram clock testing.

The post code 26 looping is fixed with the 06/19 BIOS. The SuperPi thing is hard to pin down. Some users have no issues running it while others can work around it with settings as you found. And still some cannot successfully run it regardless what is done.

[ci2kla]

The post code 26 looping is fixed with the 06/19 BIOS. The SuperPi thing is hard to pin down. Some users have no issues running it while others can work around it with settings as you found. And still some cannot successfully run it regardless what is done.

Therefore I think the problem is on CPU itself rather than MB because I used an i7 920 D0 3847A951 before and it worked flawlessly. No freeze at all no matter what VTT / vDIMM applied. Or maybe just XEON has this problem more frequently? Anyway I just wanna help

[Praz]

Therefore I think the problem is on CPU itself rather than MB because I used an i7 920 D0 3847A951 before and it worked flawlessly. No freeze at all no matter what VTT / vDIMM applied. Or maybe just XEON has this problem more frequently? Anyway I just wanna help

The board I'm using now does it with 920 and 940 (both C0) and 920 D0 and 3520.

[ChaosMinionX]

Seems any voltage I keep getting BSOD "A clock interrupt was not received on a secondary processor within the allocated time interval." which I have read is related to Vcore, but it doesnt seem to be doing anything.

[p2501]

For me those kind of BSODs were related to VTT and not vcore. I got them when trying to find the right vcore/vtt balance for a given clock.

[ChaosMinionX]

For me those kind of BSODs were related to VTT and not vcore. I got them when trying to find the right vcore/vtt balance for a given clock.

I see.... how bout for Prime do you have certain cores that are faster than others? I have my 4th thread that always seems slower, I am not sure if its a chip defect or the program doing it wrong.

[Razer(x)]

For me those kind of BSODs were related to VTT and not vcore. I got them when trying to find the right vcore/vtt balance for a given clock.

Yep, same problem. Still no solution

I've emailed DFI support and they told me to contact OCZ it's a mobo's problem

[Tony]

Well when DFI supply me a board I will be able to sort out the issue. funny how the ram works fine in other boards

[p2501]

Well when DFI supply me a board I will be able to sort out the issue. funny how the ram works fine in other boards

That is what is puzzling me. I only heard good things from those sticks, but as it looks now I can't even run them on their stock timings. BIOS auto's them to 9-9-9-20 at 1600 and 1688, 8-8-8-20 doesn't work, 7-7-7-24 doesn't work. Not at 1600, not at 1688, varying volts doesn't help either, varying subtimings helps at 1688 8-8-8-20, but those timings aren't accessible via BIOS so I won't call this a workaround. My guess is that there's something wrong with the BIOS in this combination with OCZ 1600LV Platinum sticks. That or my W3520 has a really crappy IMC.

[SoulsCollective]

I wouldn't be surprised. The IMC in my i7 920 is completely arse. As I was outling a couple of pages back, I simply cannot boot above 185 BCLK no matter what I do. On top of that, running the OCZ 1866 HPC Reapers, I've noticed that the volts required to clock the RAM to 1,800 are ridiculous. I can boot, Prime, HyperPI and IBT at 180 BCLK with the 8* memory multi for 1,440 mem speed with vCore at 1.31 and vTT at 1.29. By simply raising the memory multi to 10* for 1,800 mem speed, my board will not even boot with those volts, and full stability requires vCore at 1.4V and vTT at 1.39V.

[ci2kla]

I wouldn't be surprised. The IMC in my i7 920 is completely arse. As I was outling a couple of pages back, I simply cannot boot above 185 BCLK no matter what I do. On top of that, running the OCZ 1866 HPC Reapers, I've noticed that the volts required to clock the RAM to 1,800 are ridiculous. I can boot, Prime, HyperPI and IBT at 180 BCLK with the 8* memory multi for 1,440 mem speed with vCore at 1.31 and vTT at 1.29. By simply raising the memory multi to 10* for 1,800 mem speed, my board will not even boot with those volts, and full stability requires vCore at 1.4V and vTT at 1.39V.

Is that your i7 920 in D0 stepping? I know some of them are crap but not as bad as expected...oh by the way, have you tried 8x multi of your ram but 20x multi of your uncore? that may test out if your IMC is the issue...

[Russ_64]

I wouldn't be surprised. The IMC in my i7 920 is completely arse. As I was outling a couple of pages back, I simply cannot boot above 185 BCLK no matter what I do. On top of that, running the OCZ 1866 HPC Reapers, I've noticed that the volts required to clock the RAM to 1,800 are ridiculous. I can boot, Prime, HyperPI and IBT at 180 BCLK with the 8* memory multi for 1,440 mem speed with vCore at 1.31 and vTT at 1.29. By simply raising the memory multi to 10* for 1,800 mem speed, my board will not even boot with those volts, and full stability requires vCore at 1.4V and vTT at 1.39V.

On my DK with 1866 Reapers, I am running them at 1800 (180x10) at stock / auto settings (9-9-9-24). I raised the freq to 185 and it booted into Windows but had BSOD bit later. At 1600 (200x8) it booted but Windows froze at start.
Need more time to play ...........

[darthvedor]

I wouldn't be surprised. The IMC in my i7 920 is completely arse. As I was outling a couple of pages back, I simply cannot boot above 185 BCLK no matter what I do. On top of that, running the OCZ 1866 HPC Reapers, I've noticed that the volts required to clock the RAM to 1,800 are ridiculous. I can boot, Prime, HyperPI and IBT at 180 BCLK with the 8* memory multi for 1,440 mem speed with vCore at 1.31 and vTT at 1.29. By simply raising the memory multi to 10* for 1,800 mem speed, my board will not even boot with those volts, and full stability requires vCore at 1.4V and vTT at 1.39V.

Exectly the same for me I can´t boot above 185 BCLK, the diference is that my i7 is a W3540.

[rioja]

I have a strange issue - I may run Prime several hours stable, but if I reboot PC speaker makes continious beeps and C1 error
What may it be?

[v_parrello]

According to Award BIOS Post Codes for C1(http://csd.dficlub.org/forum/showthread.php?t=8603) it could be:
- memory incompatible
- overclocking issue
- one of memory modules is defective

[ci2kla]

I have a strange issue - I may run Prime several hours stable, but if I reboot PC speaker makes continious beeps and C1 error
What may it be?

Are you running sky high bclk? Which BIOS version you're in ?

[rioja]

I just can't understand how it can be stable in Prime and not stable in reboot.
Or my stability tests are not correct
I run 210x21 CPU and 210x8 memory with CL9, bios 0508

After power off and on it boots sometimes, and sometimes I need to reset bios.

Edit: And one more, I don't remember if my previous DFI boards gave continous speaker beeps, not one or several beeps but endless

[ci2kla]

I just can't understand how it can be stable in Prime and not stable in reboot.
Or my stability tests are not correct
I run 210x21 CPU and 210x8 memory with CL9, bios 0508

After power off and on it boots sometimes, and sometimes I need to reset bios.

Try flash to 0619 and see what you get

[rioja]

there is no 0619 on official website, 0508 is latest

[eva2000]

there is no 0619 on official website, 0508 is latest

06/16 and 06/19 are beta bioses available from DFI site. I haven't tried any of the later bioses still on 4/28 beta with no problems but now I'm testing DFI JR X58 the little brother http://i4memory.com/f80/dfi-jr-x58-t...-photos-18537/ so DFI UT X58 is taking a break

[ci2kla]

06/16 and 06/19 are beta bioses available from DFI site. I haven't tried any of the later bioses still on 4/28 beta with no problems but now I'm testing DFI JR X58 the little brother http://i4memory.com/f80/dfi-jr-x58-t...-photos-18537/ so DFI UT X58 is taking a break

I envy you....so many toys to play with