"Sudden Northwood Death Syndrome"

[JSang]

link to inquirer.net article

what do you guys think of this....ive had a few lockups the past couple of weeks but just thought it was my mobo. ill have to swap my cpu out and check now =/

 

[bgeh]

i saw it on overclockers.com
looks bad🙁

 

[nealh]

I read many of the posts seems as long as you stay near 1.65 v maybe 1.7v ...all seems probably ok....it would be strange to see processors so sesnitive a small boost in voltage killing them..look at the varying degrees of voltages supplied by mobo...makes a good case for retail chips


I think many have said on this forum to stay 1.7v or less...maybe 1.65v is best

 

[Wingznut]

Originally posted by: tehinquirer
Intel only supplies warranties for particular frequencies at particular voltages, but doesn't mind end users killing CPUs worth hundreds of dollars, because, after all, it means it will sell more chips at the end of the day.

If that's not the most ridiculous thing I've ever read on theinquirer, I'm not sure what is.

I wonder what Mike Magee suggests... That Intel should just have an unlimited warranty? I suppose he thinks car manufacturers should adopt a similar unlimited warranty.

What a dork!

 

[redpriest_]

Amusingly enough, Mike Magee is only repeating what an Intel executive said about overclocking. He's not suggesting unlimited warranties, he's saying that Intel doesn't mind overclockers -- because the inherent suggestion is if they fry their chip, they'll buy a new one from them. Makes sense no?

Besides which, the situation is really only dire if the user is operating out of spec sheet parameters. You can't have a free lunch and expect to get away with it, which is why *all* mission critical stuff will remain clocked at spec'd speeds.

 

[THUGSROOK]

DUH! 😛

 

[Duvie]

NOtice at the overclockers forum the number of users using asus mobos which have atendency to overvolt as well as are the mobos that offer a overvolting jumper, which from my experience will not read the vcore correctly in the bios.

I am now at 1.71v for 6 months plus.....

Speaking from a user who about 8 months ago had a SNDS chip I can tell my epxerience of vcore and errors that happened.....

1) vcore was 1.83v with overvolting jumper on...was at 1.65 (1.68v actual) for 3 months with no issues...sis645 asus mobo overvolted .03v from bios setting...1.83v (1.8@2.52) lasted one week then I had to back down to 1.68v (1.8@2.4) again and it ran fine for 1 week then I had to go down to 1.58v (1.8@2.25) where it ran for 2 months and still runs as far as I know it....

2)errors would happen in cpu intensive programs like divx encoding and would be illegal errors with stack dumps...was running win98...winxp was verysensitive to it and would usually blue screen out and then not get past splash screen on next go around....


3) Tried multiple boards, ps, ram...basically replaced whole system and it continued to happen until I put my current 1.6@2.74ghz in now....


I am interested in what the longevity of amd's .13micron chips running at 1.85-2v as I have been seeing...

Early on ppl looked only at heat and felt that if the temps looked good then the vcore was fine...Yeah that may seem to work with xp chips but many who ran water and vapor cooling northwoods early on were running 1.85v to 2v and manhy of those have gone along the wayside.

 

[THUGSROOK]

of course part of the problem is that noone knows for sure wth their voltage is!

pair that with 12 hours of prime95 and you can easily kill your cpu overnight.

i highly suggest NOT using more then 1.7v (as seen in windows) and dont use prime95 so often! 😉

 

[First]

Yes, that's why we run our overclocking tests on default voltage. Besides, increasing voltage by that much really doesn't yield significantly more MHz.

 

[Duvie]

Agreed...good points...On average I have usually only seen 180-200mhz more over default vcore to 1.7v with my setup....My 1.6 does 2.53 at default vcore or closest I can get (1.52v)....

 

[THUGSROOK]

*Default Vcore

heh ~ thats another problem subject in of itself.
my default vcore is 1.44v ~ on an asus board it would be closer to 1.58v.

i think well know the truth about Roswell NM before we learn the truth about P4 vcore