• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

The Official Sudden Northwood Death Syndrome (S.N.D.S.) Thread[Retired sticky]

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Silversinksam

Moderator/ Silver Paste Taster©
Joined
Aug 8, 2001
Location
Sunshine State, USA
Retired Sticky

Lets make this thread more of a database of Northwoods that have died or degradated prematurely.

Please only post in this thread if your Northwood has suffered S.N.D.S

Please add the voltage that led to the chips demise(and/or Degradation factors) and roughly how long the chip performed before symptoms of S.N.D.S. appeared.



AGAIN PLEASE ONLY POST IN THIS THREAD IF YOU HAD A NORTHWOOD CHIP THAT SUFFERED SUDDEN NORTHWOOD DEATH SYNDROME.



*credit to Rasputin's Liver for aptly naming this Northwood disease
 
Last edited by a moderator:
MY 2.26b died after only a month and a half of use.
99% of the time @ 3011mhz @ 1,6v
5 min @3164mhz then lockup @ 1.7v

I didn't do the burn in thing I just jacked it up to 3011mhz from 150fsb to 160fsb then 170fsb and then to 177fsb.

Pack date 3-8-2 "Malay"
Can't remember the prod code, got rid of the box.
 
I had(and still have in the box) a Pentium4 1,6A GHz....The first thing I did when I got it was to vid-pinned it at 1,85VCore...I used to work with it at 2,4GHz STABLE as a rock...!!!...I have bougth it at 15th day of May...My temperatures first month (May 15-June 15)were low enough with Intel's stock cooler but moded with a 7000rpms 6x6 fan, like :
Abient temp=26C(A/C)
Idle temp=33C
Full load temp=45C

Second month and after(till it went down),till about the 1st day of August,I boutgh a Swiftech MCX 478 cooler with a Delta fan 5900rpms...My temps were like :
Abient temp=26C(A/C)
Idle temp=29C
Full load temp=37C

Sudenly one day(~1st of August) and when I was into Internet...buuufff...Blue screen....I said WTF...Anyway,these things happens!!!...In about not more than four days long,it crashed every day many times....I started to "download" the fsb from 150MHz to 145MHz to 140MHz to 133MHz.....Finally I "think" now that it is "stable"(for the time being) at 133MHz fsb - 2128MHz with 1,65VCore......

All I have to say is : I do not believe that people who say that they used to work their CPU's at 1,7VCore and lower,get crashed...I strongly believe that ALL of them used to work their CPU's above 1,75VCore not to say at the "limits" 1,85VCore....I also believe that even with "normal" VCore and LOW TEMP the CPU can be DAMAGED...

I overclock since there were no existance of fsb's in computers,about 8086 at 15MHz overclocked at 22MHz by moding the crystal oscilator(remove it from motherbord and placed another one)...The "older" guys may knew that...!!!...and I have seen A LOT...!!!

This S.N.D.S. will very SOON be S.A.D.S.(Sudden AMD Death Syndrome) as well and YOU WILL SEE IT....:(
My final words are :No more than 1,75VCore "MAX" at ALL CPU's made of 0,13ì (Intel - AMD)

P.S. I Edited because I wanted to say to everybody who reads this post that according to Intel,there is not only the factor Max VCore that can dammage the Northwoods but the Temperature factor as well...They claim that a CPU CAN BE DAMAGED by LOW TEMP as well(not below 5C)....See the Intel's datasheet below...:(





UPDATE - UPDATE - UPDATE ON 8-12-2003

After playing around about more than a year with extreme subzero temps, I found out that IF you MAX out your Vcore at 1.75Volts, you are NOT in danger of S.N.D.S. ...Pentium4's are NOT dieing with subzero temps...
 
Last edited:
data.gif
 
1.6A stable at 155fsb 1.65V for 3 days. Stable at 155 1.75V for 1 day. Stable at 150FSB 1.65V for 2 1/2 months. Stable at 145FSB 1.6 1.58V ever since.

I have an idea... burning in probably works by burning off a small amount of the process, thus making it smaller (maybe turning a .18 micron to a .17 micron.) And thus making it run cooler locally and therefore improving the overclock. With a .13 micron process, this may make the process too small to operate properly. And the smaller size may make this happen faster and at lower voltages. This problem should increase exponentially with smaller processes, so we'll see if this is right when smaller processes come out.
 
Caffinehog said:
I have an idea... burning in probably works by burning off a small amount of the process, thus making it smaller (maybe turning a .18 micron to a .17 micron.) And thus making it run cooler locally and therefore improving the overclock. With a .13 micron process, this may make the process too small to operate properly. And the smaller size may make this happen faster and at lower voltages. This problem should increase exponentially with smaller processes, so we'll see if this is right when smaller processes come out.

I think the jury's still out on 'burning in' silicon.
My Stanford EE buddies tell me 'no way', just a pipe dream. Also that the premature failures w/ higher voltages are the gates going.

JM
 
Are you running at those voltages all the time Johnny Mo?????
If so how long have you been running it at that 1.85????
 
I believe I have a Northwood on the verge of S.N.D.S...

It's a P4a 1.8Ghz. (not at home, so no identification batch numbers avail).


I had been running 110% stable at 133FSB (2.4Ghz) for four months. I Vid Pinned my default voltage for 1.7..

But just last week, my computer started wigging out on me.. artifacts on the screen, wierd pause/lock-ups.. and only when I was under a heavy CPU load. At first I thought it was my GF4 blowing up on me, but upon further testing I discovered that it wasn't.. I am 90% sure it's my CPU. I found it hard to believe since the thing had been soo stable and I had only been running it for four months... until I came here and saw this thread. My system hasn't fully crapped out on me yet, so maybe I still have some hope of at least running it at the default 1.8/100FSB (YUCK!) clock speed....

Going to do some more testing tonight, after work.

It was wierd though.. just started happening out of the blue.
 
Last edited:
WayneDolesman said:
I believe I have a Northwood on the verge of S.N.D.S...

It's a P4a 1.8Ghz. (not at home, so no identification batch numbers avail).


I had been running 110% stable at 133FSB (2.4Ghz) for four months. I Vid Pinned my default voltage for 1.7..

But just last week, my computer started wigging out on me.. artifacts on the screen, wierd pause/lock-ups.. and only when I was under a heavy CPU load. At first I thought it was my GF4 blowing up on me, but upon further testing I discovered that it wasn't.. I am 90% sure it's my CPU. I found it hard to believe since the thing had been soo stable and I had only been running it for four months... until I came here and saw this thread. My system hasn't fully crapped out on me yet, so maybe I still have some hope of at least running it at the default 1.8/100FSB (YUCK!) clock speed....

Going to do some more testing tonight, after work.

It was wierd though.. just started happening out of the blue.

I'm experiencing exactly the same on the same CPU with 133FSB and 1.7V :confused:
 
Yodums2:

So yours up and running fine for a while (months?), then just all of a sudden started crapping out on you to, eh?

(just clarifying)
 
WayneDolesman said:
Yodums2:

So yours up and running fine for a while (months?), then just all of a sudden started crapping out on you to, eh?

(just clarifying)

its been running at 133FSB for three/four months. now giving me errors, and its not the memory (i bought two more 256MB RDRAM sticks to make sure, its not the motherboard either cause i just replaced it). Now its giving me errors at lower FSB's too - including 100FSB!!! :mad:
 
I set my machine up for some stress testing when I went home for lunch today.. I am waiting in agony to get off work here so I can go home and check it!..

If it has screwed up, I am going to buy a new processor tomorrow. I'll keep you all posted.
 
My 1.8a ran 100% stable at 2.4 for 2 months. Then pooof, BSOD and even though it rebooted once or twice more after that if finally totally crapped out.

I had been using it with an Asus P4T533C with the Overvoltage pin and 1.625 V. That would yield about 1.75V no load at around 33ºC and 1.71 V under full load at 50ºC.


Noc
 
It seems that there is a pattern here. All of you who had their Northwoods crap out on you have done the wire wrapping trick. Has anyone that has not done the wire trick have their NWs die on them? I myself own an overclocked NW processor and this concerns me. I have not done the wrapped the vid pins on my CPU, but have downclocked my CPU back to default settings as a precaution.
 
I never had to resort to the pin trick. Just used the OV jumper to get enough voltage to be able to cold boot.

Noc
 
just a doubt

Are you guys talking about bios voltage or real voltage?
and another question....do you guys know were can i get a th7-2 raid but without the undervoltage problem?
 
P4 1.6A

I have had my 1.6A since May, vid pin modded to 1.85v - 2.61Ghz.

I haven't killed the CPU, but I have killed 3 motherboards in the last 2 months. 2 of them were MSI 845 Ultra-ARU's, and 1 was a Gigabyte 8IRX. The Gigabyte board didn't totally die, but I lost use of 2 of the 3 DDR slots (only slot 3 would detect ram).
The 2 MSI mobo's died because of a bad mosfet labeled "C5001" that is between the 1st DDR slot and CPU socket.

But, and any rate, the CPU is still alive and kicking.
 
My 1.8A has suddenly become unstable at any o/c speeds. I am not sure whether it is just a software issue, as it started happening after the installation of an Asus supplied chipset driver (1012). But this could be co-incidental.

I have since, however, re-installed XP (note: not reformatted disk, so there may still be some rubbish), and installed the Intel supplied 1013 chipset driver.

I have also tried everything else, and cannot get it stable at o/c speeds (even low o/c, like FSB=125).

The one thing that makes me think that there is still something funny in the software is that it is 100% stable with test utilities outside windows (eg memtest86, motherboard test utilities, etc). But I don't know how reliable these are - anyone know of a good system/cpu test utility that doesn't run under windows?

I was running the CPU for about two months at FSB=143 at 1.675v (bios, actual 1.725v at idle, 1.65v under load). Temperatures were a bit high, but never over 56 C.

I briefly (5 minutes) had the CPU at FSB=150 at 1.775v, but it wasn't stable so I didn't pursue this.

It seems that especially the 1.8A's can die even with modest o/c (if 43% is modest!!!) at modest voltages and temperatures.

Mine is still OK at standard setup, and I hope that it lives until I can get a 2.4B with C1 stepping.

Note: The chip is a 19 May 2002 Malaysian 1.8A.

Update: OK, my CPU suddenly started o/c again. So I guess that it was a software problem, and I don't know what caused that to be corrected again.

But I am running it conservatively now, at FSB=133, 1.65v, 4:5 memory at 2-3-3-7 (BY SPD). This gives me 2.4GHz and DDR333 and is OK for my requirements (and is fault-free and stable).
 
Last edited:
As i noted above, my 1.8a died.

I have since received a new one, a Philipine make and the first thing i noticed is that it takes a LOT less Vcore to run at 133 FSB.

While my previous 1.8a required the OV jumper to cold boot and about 1.75Vcore no load (1.70 loaded) this one now doesnt require the OV jumper and is running fine with 1.5 Vcore (default).

The funny thing is, it seems to run at the same temp as the old even though im running less Vcore.


I dont remember where the old one was made.


Noc
 
Last edited:
Back