2600+ barton at 2.2ghz, 1.85v, will it shorten life?

[Choppedliver]

Just bought a 2600+ barton, normal operation is 1.9 ghz or so. Ironed out my stability problems (bad geil stick of ram)

It would crash at 1.85v at 11x200 when running prime95 after about 4 hrs


At 1.875, it has been doing 11x200 all day, no problems, so i would consider it "stable"

I built this for my sister, and of course id love to be able to tell her i got her a 3200+

A) Would you worry that it wont be stable at that speed in the future, ie, is running a measly 2/10ths of a volt gonna shorten its life?

B) Also, the temperatures are 134F at full load... using a vantec aeroflow. Do you think thats an acceptable temperature? its not gonna be running prime95 all the time obviously and when its just idle it drops 10 degrees or so.

 

[DannyBoy]

dp

 

[DannyBoy]

overclocking it will shorten the life of the cpu maybe 5-7 years.

By the time the processor is dead it will be obsolete anyway, and you will be using an Athlon XP 8200, or whatever is available...

If overclocking really hurt your cpu that much, do you think everyone would do it?

 

[Corporate Thug]

1.85V is a bit on the high side, it MAY kill the cpu

you can really say OCing will shorten the CPU lifetime X amount of years because it depends on the OC. changing the vcore of a barton from a 1.6 to 1.65 will prolly do nothing to the lifespan but changing it to 2v will kill it very quickly

 

[Choppedliver]

man, dont tell me that, its nice having windows tell me ive got a 3200+

will it last 3 years reliably i guess is the question

 

[Richdog]

To be honest, I would worry about using that amount of voltage for that speed, using bog-standard air-cooling. If it's for your sister, i'd change the speed to either 2000Mhz (10*200) or 2100Mhz (10.5*200) and use a lower voltage like 1.7v or 1.75v. The extra 100Mhz you're getting out of that CPU isn't worth running it at 1.875v by any means, could lead to system instability, and could kill it inside 3 years. As long as she has a half-decent graphics card she's not going to realistically notice losing 100-200Mhz. You're running it too high, and too close for comfort in my opinion. If you need 1.875v to get to 2200Mhz on a Barton it must be one hell of a crappy overclocker, give it some slack.:beer:

 

[Choppedliver]

yeah its got a geforce4 ti 4200, not the latest but not bad, she doesnt even play games on it actually, just does video production/multimedia stuff.

Yeah i was thinking 2ghz wouldbe cool. its only an extra 100 mhz, but the bus will be 200mhz instead of 166 as well. There is something pscychological about 2ghz i think

 

[Richdog]

A Ti4200 is plenty fast enough that she won't notice the small decrease in speed dropping to 10*200 would provide. Her system will then be cooler and more stable. Everyone's happy.:beer:

 

[Choppedliver]

yeah, i agree.


u know what, this thing will do 9.5x202 (comes up as 1930mhz) at default voltage 1.65. Default is 1909 mhz at 166x11.5 at 1.65

BUT, i cant get it to run at 200x10 unless i bump the voltage to 1.80+ and at that voltage it will even do 2200mhz 11x200 at 1.85+

a measily 70mhz difference and requires a huge bump in voltage and then it will do 300mhz overclock. I guess im probably gonna leave it at 202x9.5, at least i got a good fsb and memory boost.

i must have got the all time worst overclocking athlon xp.

 

[jhbball]

Originally posted by: shady06
1.85V is a bit on the high side, it MAY kill the cpu

you can really say OCing will shorten the CPU lifetime X amount of years because it depends on the OC. changing the vcore of a barton from a 1.6 to 1.65 will prolly do nothing to the lifespan but changing it to 2v will kill it very quickly

Honestly, I've never seen or heard of *anyone* killing their processor with too much voltage. Many people claim that 2v will "kill" your cpu, but I've never seen any evidence to prove this.

 

[Richdog]

Originally posted by: jhbball

Originally posted by: shady06
1.85V is a bit on the high side, it MAY kill the cpu

you can really say OCing will shorten the CPU lifetime X amount of years because it depends on the OC. changing the vcore of a barton from a 1.6 to 1.65 will prolly do nothing to the lifespan but changing it to 2v will kill it very quickly

Honestly, I've never seen or heard of *anyone* killing their processor with too much voltage. Many people claim that 2v will "kill" your cpu, but I've never seen any evidence to prove this.

Err... well if you've never heard of it, doesn't it make sense when you think about it? Do you think a CPU is designed to handle an unlimited amount of voltage through it or something? This is so blatantly obvious i'm not even going to provide links. Have a good think about it...

 

[Choppedliver]

on the other hand, 1.85 vs 1.65 is hardly unlimited voltage... we are talking about 2 tenths of a volt. 12 percent

 

[ar47bl1]

Yeh it might shorten it from 10 years to 4-5. Nothing big to worry about. I think it is more temperature that kills more than heat.. But heat is a HUGE factor..IF you aren't running too hot I wouldn't worry. I have been running mine at 1.85 actual volts for over like 3 months now and no major problems..Mine is never over 40C either though so..

 

[Doh!]

1.8 is the highest I'll go but that's just me. If it's for your sister, I would drop down to more comfortable level. She won't notice any difference. By the way, I've seen people kill their cpus from too much voltage.

 

[Choppedliver]

well at 1.85 volts running at 2200, it runs at 132 F, thats not too bad...

 

[Jeff7181]

Originally posted by: Doh!
1.8 is the highest I'll go but that's just me. If it's for your sister, I would drop down to more comfortable level. She won't notice any difference. By the way, I've seen people kill their cpus from too much voltage.

... and it doesn't burn up in a cloud of smoke like some of you might think. It just stops working because (to put it simply) the inside of the processor gets beat to hell until it breaks.

 

[Jarhead]

I look back to my i233/MMX that has been overclocked since they were the hot item at the time, still runs fine, I look back at my Celeron 300a o/c at 450 since they first hit the shelves, both running 24/7 doing RC5/TSC ever since then, still no issues, both have decent core voltage bumps for sure....no issues yet.

Yeah voltage increases the 20-30 year rate of possible electromigration issues (I think this was stated for the maximum CPU temp of something like 80-90 degrees C). Of course I'm getting old, so the numbers may be off a bit.

Anyhow, the rate of electromigration is also due to the heat in the core....think about it...of course thats why I built a little heatsink of my own. On air, I get a C/W of 0.098 C/W (of course it is running a "3400+" now.

8 degree delta/81.4 watts = 0.098 C/W

(not bad for an -AIR- cooling heatsink, eh? Yeah, and in case you ask, that is based on the on-die diode, and the wattage is measured by measuring the voltage across the current sense resistors and multiplying by the core voltage) (of course I could turn the fan speed up to a 3500 rpm screamer....)

http://home.insightbb.com/~danzigrules/jarspage/welcome_to_jarhead.htm

I've always enjoyed homemade stuff, especially my mom's cookies!

/me tosses those silly *WATER* cooling setups in the dipsy-dumpster

 

[robcy]

Personally 1.85v is a little high in my book. When next summer comes along and the average room temparetures go up a good 10 degrees, will it be stable them. At 1.85v, and 2200mhz to many other variable that are moot now play more important roles when ambient temps are concerned. Tweak a couple of other timming to see it you can get 200x10 with 1.7 max 1.75.

 

[Choppedliver]

i just bought a 2500+ from newegg... im gonna try that one and hope i get a better overclocker. this one is pissing me off lol