too high voltage on water?

[supercromp]

hey
just finished putin together my sys with a swiftec mcw5000A

on a 2400+ i can get 2.33 at 2.0 volts , 41 idle 47 load
i can get 2.4 on 2.15 volts, 51 idle 55 -57 load

my temps are in acceptable athlon range, what im askin is how bad is it to run at 2.15 volts?
the chip dont have to last that long, id like to use it for 2 years at most
thanks

edit : tbred-b ,1.65 default (?)

[schismspeak]

it probably will not last that long at 2.15volts. 2 years i doubt it with those temps on water. Look at my temps on air. and it is a crappy heatsink too

[I.M.O.G.]

WELCOME TO THE FORUMS! :)

The temperature rules are not the same when you overclock.

A safe temperature at rated settings is not necessarily a safe temperature at overclocked settings.

So just because you are under the recommended temps does not mean your temps are good and okay.

I'm just guessing completely, but I would say that a chip with over 2 volts going into it at a temp of 55C will not last 1 year let alone 2. If I were going to run at those kind of settings I would do it with the mindset of being prepared to buy a new chip anytime if this one failed.

I would not run with that many volts at all, but I only have one system and its gonna last me for another year atleast.

There is no clear cut answer to his question... it's bad, but there is no telling how bad really. I would say with that many volts at the temperature he reports I would wake up everyday wondering if my pc was dead though.

[schismspeak]

yeah i thought it was way to high and the temp being to high even though acceptable under normal settings.supercromp it is not going to last long at all but if you want to buy a new processor in a 'couple' of months i guess it is fine. it is up to you

[Maximus Nickus]

Another thing to take into account is the Default Voltage of a T-Bred is lower than the Default voltage of a Thunderbird.

Although it isn't by much, if you do the math a 10% increase in Voltage for a T-Bred is much less than for a Thunderbird, meaning that the general figure of a safe % in voltage increase is much lower.

Simple put an acceptable safe voltage for a Thunderbird isn't acceptably safe for a T-bred, even if you are running at the same temperatures.
The 0.13 Micron manufacture process also limits the amount of Votlage you can safely use, sn generally it has been established than these CPU's are less voltage tolerant than the 0.18 Micron manufactured CPU's.

M_N

[SkiFletch]

I think you guys are being a little conservative with that 2 year theory. at stock settings, chips can last 10-20+ years. yes, everything you guys mentioned about overclocked settings being different, and voltage increases being different than tbirds, but I still have to think his chip would last him at least 2 years, if not more like 5. just my oppinion

[ColdFlare]

Originally posted by Maximus Nickus
Another thing to take into account is the Default Voltage of a T-Bred is lower than the Default voltage of a Thunderbird.

Although it isn't by much, if you do the math a 10% increase in Voltage for a T-Bred is much less than for a Thunderbird, meaning that the general figure of a safe % in voltage increase is much lower.

Simple put an acceptable safe voltage for a Thunderbird isn't acceptably safe for a T-bred, even if you are running at the same temperatures.
The 0.13 Micron manufacture process also limits the amount of Votlage you can safely use, sn generally it has been established than these CPU's are less voltage tolerant than the 0.18 Micron manufactured CPU's.

M_N

Thats not completely true due to my experience and my friends experience. There are people constantly running 2.15 votls through their t-bred Bs, and they show no signs of dying.

I think you are referring to the certain high voltage death syndrome of the p4s due to gate oxidation.

[Mark Larson]

Originally posted by ColdFlare
Thats not completely true due to my experience and my friends experience. There are people constantly running 2.15 votls through their t-bred Bs, and they show no signs of dying.
how long have Tbred B's been out?

[PalominoBURN]

OK simply my opinon here. But why stress a CPU that much more to go from 2.33 to 2.4 GHZ? Makes no sense. Back it down to 2.0V and run it at 2.33. Be sensible, not crazy!

**SIDE NOTE** I ran 2.25VCORE on my palomino for 2 days running my 1900XP (1.6ghz@2.0GHZ) and guess what.. it's all but fried. I already ordered a new AIUHB T-BRED and my point is this, temp sensors are misleading BE CERTAIN you are reading ON-DIE temps not IN-SOCKET temps... (MBM 5 sensor LM90 REMOTE for on-die temps). I have good cooling also, I am not a complete idiot. :D :D Although 2.25VORE might make you think so.

[ColdFlare]

Originally posted by PalominoBURN
OK simply my opinon here. But why stress a CPU that much more to go from 2.33 to 2.4 GHZ? Makes no sense. Back it down to 2.0V and run it at 2.33. Be sensible, not crazy!

**SIDE NOTE** I ran 2.25VCORE on my palomino for 2 days running my 1900XP (1.6ghz@2.0GHZ) and guess what.. it's all but fried. I already ordered a new AIUHB T-BRED and my point is this, temp sensors are misleading BE CERTAIN you are reading ON-DIE temps not IN-SOCKET temps... (MBM 5 sensor LM90 REMOTE for on-die temps). I have good cooling also, I am not a complete idiot. :D :D Although 2.25VORE might make you think so.

Why overclock at all if you arent going to try to obtain the best performance under stable conditions?

I don't know how you can kill a processor with Voltage alone. Voltage alone doesnt kill a processor. And like you said, "Fried" referring to heat killing the thing.

AMD chips are more robust and open to voltage than P4s that dont like operating at high voltages at all.

[Mark Larson]

I believe the man has a point - 70Mhz isn't all that much.

One thing i believe many people are lacking is an education in cost-benefit analysis ;)

[schismspeak]

Originally posted by ColdFlare


Why overclock at all if you arent going to try to obtain the best performance under stable conditions?

I don't know how you can kill a processor with Voltage alone. Voltage alone doesnt kill a processor. voltage alone can and does kill a processor even if you have the best cooling they will not last as long. they can still die without being overheated he is nowhere near the max temp of the proc but it will be dead before the year is up im sure, because of the voltage it is at. so yes it is the voltage that kills it. And the northdridge might even burn up before the proc does if he dont have a HSF on it.