EVGA NF41 OPTY 165, Mach 1- Could it be a cold bug?

[Tomsawyer]

evga nf41 with the latest beta bios

2x1gb mushkin ddr4000 black

pc&p 510 asl power supply

Opty 165
OSA165DAA6CD
CCBWE 0551 UPMW
1392938L50037

evga 7800gt

winxp pro




Well heres the update (at work trying to recall the settings from memory)

I moved my mushkin ddr4000 series black over to the blue slots on the evga nf41 just to see if it would make a difference and tinkered until about 11pm last night.

Still working via air cooling for the testing purposes.

I can take the HT up to 280 with the cpu at 2580 and have no problems as I long as I do not allow the memory to get over 250. I have done this with both 1T and 2T settings. So for now its 3-8-3-4 stock voltage at 1T

Currently I am at 280ht, 2580cpu with the memory at 240 or there abouts. At that setting I have been able to game, surf, even burned a DVD to check. Ran Sandrasoft and ran some benchmarks, then I ran 2 instances of prime95 for an hour. At idle the temps with the stock air cooler are 33C, when I game or bench test they reached 49C, briefly going 50C and then back. (hence why I like phase cooling, never gets over zero lol)

Have not had a reboot yet as I did previously under cold, but have asked my wife to allow my son time to game on it and have her surf about.

Conclusion so far, the evga board is a decent mid range board, the limitations of the memory (that some major review sights pointed out) are still in effect.

If the system stays stable for 3 days at my current setting I will conclude that under air I have reached a decent and stable overclock point.

I shall then put her back under the Mach 1 and go supercold and using a new set of neoprene, seal string, and sealent make sure that I get a solid seal.

If the reboots restart at the settings the system can handle under air then I must conclude the chip just doesnt like cold or somehow its flucuating enough somehow in the temp that its causing the board to force a reboot?? Really puzzling me

If any guru's out there with experience dealing with cold bugs could give some feedback I would appreciate it. My limited understanding is that the cold bug prevents the cpu from booting up, not causing random reboots or bsod's

 

[marscay]

reboots and bsod's smell like condensation to me - how long until you get the problems usually?

not surprising as i find sealing with the mach's a pita.

coldbug will normally just prevent even booting to bios or freezing up within the o/s usually when idle or hitting it's coldbug trigger temp.

 

[Joe Camel]

agreed, although ive never got bit by the cold-bug, everything ive heard goes along the lines of cold-bug = lock-up not reboot.

GOOD LUCK!!

 

[Tomsawyer]

Thanks Guys, I too suspect somehow I must have had a micro leak or something around the pins or backside. However I got some brand new ac5, dielectric grease and tonight I plan on removing the air cooler I am currently using. Thanks to my finding this site:


http://www.ocforums.com/showthread.php?t=450123


I also have several sheets of closed cell foam and neoprene of various thicknesses with which I am going to beef up the rather poor sealing system the mach 1 and 939's use. This time around I will go ahead and use the heater in the back plate also. Never needed it on my old epox 8rda+ but perhaps the lower wattage on the opty is allowing even colder temps.

Still cant find a program that can even give me a guesstimate as to just how cold, short of getting a physical probe or something which defeats the seal the whole block in the first place

 

[Joe Camel]

tom, you might want to rethink the AS5 after you read this:
http://www.ocforums.com/showpost.php?p=4380404&postcount=13

edit:

here is a shot of my (all foam) insulation setup for my MachII (no back heater or "silly string" used and never a condensation problem ... knock on wood)

 

[El<(')>Maxi]

The cold bug can appear in several ways actually. A non-boot is just one of them and probably the most unlikely to occur to typical users. More often you will encounter the bug after booting and you are in Windows, the system will simply freeze and shortly therafter re-boot itself. This happens at varying temperatures depending on the chip itself. The common way to deal with this is always keeping the chip loaded inside Windows, this prevents the re-boot.

Another version of the bug is a HTT limitation, this usually goes along with the problem above but can be dealt with to some degree by adjusting Ram Driving Strength's. That why alot of xtremecooled FX-60's which have usually pretty bad cold bugs run low HTT with a ram updivider.

Basically, xtremecooling a .90nm A64 can be very tricky sometimes, and is also part of the reason why some people can go thru some many chips in their search for the primo one