NOOB - Have i pushed OC'ing too far?!

Unless you have very good cooling such as water cooling, limit OC to 10%. Most damage comes from heat.

CMOS error is a BIOS problem - since reload default failed, download new BIOS plus install tools and try a BIOS flash (MB site will have intructions).

Once flashed see if the BIOD detects your boot hard drive - as boot disk fail means the pc can't see your hard disk.

Good luck.

Did you try clearing the BIOS?
I think you would have to flash the bios. Take the official ones from the site.

Good luck..!

BS.

1. You can OC really high on air without doing any damage. I've OCed my AMD 64 50% on stock cooling.

2. Heat don't kill, voltage do.




BS again.

He doesn't need to download a new BIOS. He only got a message that since the current BIOS settings didn't work fail safe mode was enabled.

Make sure your harddrive settings in the BIOS are correct. If you are using a SATA-drive this is most likely the problem.

ok it seems to be working now. i just left it and tried the next day and it booted fine. but now i have (very slowly) got it running to 1.98ghz and acording to "motherboard monitor" the temps are 27, and 35 degrees. should i keep going and maybe get it to a 2.2ghz or is there really no point?

What motherboard and memory do you have?

My 3000+ Venice is running at 2.7GHz at stock cooling. You could probably reach somewhere between 2.5-2.7GHz.

The imporant things to change in the BIOS to get a good OC:

CPU clock
HT Frequency
Memory divider
CPU voltage

- Heat? do you have any idea how hot an Intel prescott runs compared to the AMD Venice core?

Damage is done with heat but its the extra VCore producing the heat and damaging the die.

There is no diffrence between two models with the same core, only the multiplier resulting in a higher clock speed (and maybe die quality) - overclocking half the time is not overclocking, just makes a cheaper cpu run at the speed of the more expensive one.

hey i have an Abit AN8 mobo with GeIL 1GB (2x512MB) PC3200 Value Dual Channel Kit CAS2.5 ram. my current temp is around 30 degrees and cpu is running and 1.98

I would just like to add that throwing a cooler on a cpu that is perfectly capable of freezing a mammoth is NOT good for it. CPUs are designed to run at the temperatures they do, so when you overclock, don't cool your cpu down lower than the temperature it normally runs at. You can get a modest cpu to run blindingly fast on just stock cooling and an extra fan somewhere else in the case. Just like the_guru says. And that is pretty impressive, i leave my CPU alone though, it was too expensive for me to be messing with any time soon.

Bah there not designed thermally, the HSFs are designed to cool down cpus at best as posible at a cheap price (eg slow/budget models might not have copper cooling), hotter cpus get bigger (more efficent) HSFs and when you overclock you get a bigger HSF to take care of the excess heat and keep temps lower (idle, load, max).

The biggest HSF posible installed on a cpu would only lower the temps at best to room temps (20 - 30c) and that would do the cpu better then sitting at 40 - 50c all its life.

Behold the Intel Prescott - im sure they designed it to run hot, its idle temp is something half the A64s will never see, if you want to overclock it its a bare minimum to get a huge cooler or water cooling.

May i ask what cpu you own?

AMD 64 3500+ 939 Socket

IF something is designed to run at say 72 degrees it was obviously designed to run AT 72 DEGREES, so why run it at a temperature it was NOT designed to run at? (72 is just an example) I have not checked my cpu temperature even once since i bought it. Take a look at a CPU chart next time you think buying an intel is a good idea. You can modify your stuff all you want but the bottom line is, if you run something too cool resitance increases and therefore voltages will be effected, so when you have your cpu at -5 celcius id like to see it go more than a second without an error. I personnaly won't put a heavy piece of copper directly above my graphics card, you say they make computer parts out of cheap material but you trust the motherboard's fan mounting enough to allow it to balance a brick of your nice EXPENSIVE graphics card. But then again each to his own, I can't stop you and if you are used to your setup then do what you want, this is not a flame, you disagree with me and i disagree with you, that is unlikely to change so God speed to your efforts.

Flame? nah this aint no flame - its pathetic.

Go onto google and type in "cpu too cold" and see what you see.

As for any big HSF's falling off and damaging components - most large hsf's dont use the original mounting bracket and are bolted directly to the board with its own retention bracket.

Colder the better, the only issue with say LN2 or DryIce (and the others) is condensation. What if i live in Antarctica - so now i cant use my cpu cause its "too cold"? Yeah pull the other one out. To my knolodge with overclocking - the colder the more stable and less vcore required.

As for Intels and the THG charts - did i say i like/want an Intel (currently)? There freaking hot and when you overclock with stock cooling you wont get far where as water cooling or better will see 80% of em hit over 4ghz.

Link to most CPU's max temps
http://www.heatsink-guide.com/maxtemp.shtml

CPUs are fine when they are o/ced as long as they don't go too high above its stock temp... if system won't boot or windows isnt stable add vcore, but it will also add heat.

A64 Venice cores are very overclockable I myself o/ced my A64 3000+ from 2ghz to 2.6ghz stable with temps not exceeding 45C+/- on my stock HSF.. and also a very helpful yet cheap duct mod
http://overclockers.com/tips1187/

I hate to tell you this, but you are way off; cooling something down actually decreases the resistance. The cooling systems on the market today have not in anyway been shown to damage the CPU. I have never read, heard, or seen a CPU that was damaged because it was run too cold. I have heard of countless CPUs that were damaged by heat.

Viper

You wouldnt' happened to have taken any physics classes would you?
HOTTER MATERIAL
LARGER SPACE BETWEEN MOLECULES
MORE ENERGY PER ELECTRON
LESS RESITANCE.

TEXT BOOK
FACT
LAW OF PHYSICS

Let me guess, you are one of those americans who thought that germany was one of the allies in ww2, right? ~40% of american children believed this.

:SCREAMS: :sweating:
Must
keep
anger
under
control

Rapidpeanut
must
have
failed
physics

The hotter any material gets, the more atomic motion is generated. More atomic motion means it's more difficult for electrons to move through that material. Think of it as crossing a busy freeway - is it easier to cross when the cars are going 5mph (cold) or 100mph (hot)?

With your logic, why would we use cpu coolers at all? You're saying they should get more efficient as their temperature rises...that's just retarded.

Yes, I took many physics classes...as well as chemistry, engineering and material science...all of which say YOU'RE WRONG.

If his logic was true compaines such as asetek or extreme cooling tech. wouldn't exist. As they use a extreme cooling liquids and specially made evaporators.
Have you never heard of
VapoChill lightspeed? or Prometeia Mach II Gt

Mach II Cooling Specs:
Cooling Performance: -65C at 0W/-30C at 200W

Vapochill Lightspeed Specs:
Cooling performance load/idle 200W @ -25,5'C / 0W @ -48'C

And if you haven't read any of the articles in TH on cooling... well this here 5ghz p4 and Liquid Nitrogen @ -190C should put some re-thought into your philosophy...
http://www.tomshardware.com/2003/12/30/5_ghz_project/

If his logic was true compaines such as asetek or extreme cooling tech. wouldn't exist. As they use a extreme cooling liquids and specially made evaporators.
Have you never heard of
VapoChill lightspeed? or Prometeia Mach II Gt

...


Are you addressing me? I agree with you...you kind of just restated what I said.

you can cool cpus very very very cold (-190 on ln2 -50 on phase change), so no, decreasing temps are not increasing resistance.

mind you, intels cpus are only rated for 5c, colder then that its no longer garunteed to work

I wasn't exactly addressing you i was quoting you because i agree with you.

As for Intel CPU's having a limit of 5C not true considering in that THG link it's an intel CPU running at -190C....

no i was saying intel only rates their cpus down to 5c. it is not garunteed to work, and the warranty is null and void.
mind you, if you are cooling it that low the warranty is sure to be voided another way (oc)

Oh okay, i misinterpreteted your post.

TWIMC

Allright you win, just checked giancoli. Sorry for flipping out i'm kind of pmsing due to the topic i made about being rather angry. And cooling companies only exist to stop your cpu fusing due to heat, even if what i said were true the companies would still exist to stop your stuff blwong up in your face.

 strich egg on face:

BUT TAKE A LOOK AT THIS IDIOTS

http://www.uel.co.jp/ptc/ptc-graph-01.htm

notice how the lower you go RESITANCE INCREASES, oh you stupid stupid people. IOW YOU ARE RUNNING YOU STUPID COMPUTER BELOW 200C arent you? therfore you are increasing resisantce just as i said you would.

'tard

VICTORY IS JUST SO SWEET.

to me that looks like the higher you go the mroe resistance you have




so in this case, rabid peanut is correct, because we are talking about a semi conducter. however, copper wire would increase resistance with a higher temperature. so in a way, you are both right, but rabidpeanut is slightly more right in this case by sheer and utter luck

::edit::
taken from this page
http://www.allaboutcircuits.com/vol_1/chpt_12/6.html

That graph is only valid for the PTC Thermister component on that page. It doesn't need to apply for any other component.

You are telling me that a law of physics only applies to certain objects? So then that would mean that it is possible that chemical reactions only happen in my brain and not yours. Look the fact remains you go too low, you get increased resitance, try passing a current through something at ->0 K and then pass it through that same object at a room temperature. 0 k precludes all movenet therefore at 0 k electons will collapse in on an atom and a black hole will form as other atoms collapse in on the collapsed atom. The lower you go the harder it is for objects to move, this applies to electrons too. So you people can cool your cpus down until the crack for all i care, it is a dumb idea, i am right, you are wrong.

"One can have no opinions about the truth"- TES-Morrowind.

Oh and uuuu.... Luck my rear end.

CPU's are chock full of semiconductor. I am going to genetically engineer an egg so big that i can drop it on your entire country.

(Yes i know that egg number is not heritable and therefore not modifyable but i think that size might be.)

You are making a fool out of yourself.

The only thing I said was that the graphs you referred to was only valid for that component. So please read my post again. Don't try and read in between the lines this time.

I went to the science program so I'm much aware of the laws of physics.

I don't even know if you can read the graphs right. According to the graphs resistance DECREASES to a certain point. Then it starts to increase slightly.

And different components have different curves. A thermistor and a transistor are very different components so the curve for a transistor is most likely very different.

Okay.. so you've proved that there is a greater resistance at lower tempatures, but your not proving that he is not right about that fact that the atomic motion would become extremely unstable at high tempatures as the electrons or whatever would be moving to fast.

Btw, have you heard a little saying called heat kills?

Car engines die when they get to hot, lifeforms die if they get to hot, cpus die when they get to hot, and so does everything else.


So in reality who gives a shit if there is a greater resistance, we want things to last.

Did you quote the right person?

i didn't quote anyone, the reply was aimed toward rabid.

Sorry I meant reply to.

Your post says "[in reply to: the_guru]".  

bah. wish it was a general reply button so it wouldnt do that, or am i blind an not seeing it?

Cold kills too. This is aimed at all of you who believe that resistance in a cpu will not increase when the temp drops, that is all i am arguing, i have stopped trying to convince you that freezing you computers won't help you.

Argue with Steven Hawkings, if you are lucky he might run over your toe with his wheelchair, maybe one of his cells will fall on you, that is the smartest you are ever going to get.

Ok then, if anyone has killed a cpu for "being too cold" please post..........


LN2 is the only way to get insane overclocking (P4 to 7+ghz, Pentium M to 3+ghz, A64 to 3.4+ghz), i dont see anyone posting "P4 3.8 @ 9ghz NO COOLING".

If you dont believe in good/extreme cooling then go back under the rock you came from - the more copper and LN2 cooling for me!

Go buy a prescott and find the best way to overclock to ~5ghz, and wich would get there easiest? Stock cooling? NO COOLING? Water cooling? LN2? Moron.

This winter make sure you knit your poor cold computer a warm jumper to keep warm and keep your poor machine out of the draft.

Lmfao. I too would like to see someone say they killed their cpu for making it too cold.

liquid hydrogen could potentially do that

It's the voltage that kills when you OC at low temps.

Yup, but the cpu still ends up dead doesn't it?

Yes, but it's not because of the low temp. You can run the CPU at low temp and standard voltage without any trouble. But at low temps you can increase the voltage a lot higher than at normal temps. So that is the reason to get low temps. But still the CPU comes to a limit where the voltage is to high. But the limit is higher at lower temps.

To high voltage is the most common that kills the CPU. Heat is not the most common thing... Cold killing a CPU has I never heard of.

I didn't ever see it happening personally but i have seen cracked chips that broke due to cold. What happend to the conductive paste/gel when you get to temperatures that low?

Cracked chips that broke due to cold? How then do you know the cooling element/plate wasnt mounted incorectly? and as for the thermal paste - it supposed to be a thin layer that fills the gaps betweem the cooler and the cpu not a thick gob so it wouldnt matter.

The only cpus that have the die exposed currently are the Athlon XP's and Pentium M's, other then them, you cannot see the die directly (P4/A64), so whats this "cracked chips" you are refering to?

There are more burnt out/dead chips killed due to overheating/overvolting then by overcooling (now thats a funny word) and i dont thing there has ever (or not many) that have been killed by overcooling or whats involved with it.

Your story changes every time you reply.

The point is that it does happen, and this thread needs to die.

3 days later

Yeah its a pitty no one posted details about a chip that died from underheating.

PFFFFF.

its looks to me that you need to retake physics. You ever heard of superconducting materials? They are cooled nearly to 1K. Now how do you explain that?

Have you ever in your life seen a super conductor in a PERSONAL !@#$%^ COMPUTER? So if you want to make worthless arguements go for it...

Theres only one person making worthless arguments.

Look, if you think if your cpu would run better off hot then why dont you just remove all the fans from your computer and get back to us with the results a few minutes after it dies so you can tell us all how much of a moron you were for thinking hotter is better.

While your at it explain why my GeForce 6600GT (with a dud fan) crashed at 93c and works now with a new Zalman cooler and overclocks further, Explain why my ancient AMD K6-2 500 @ 600 stays stable only with a larger cooler, explain why cpus even have coolers.[/b]

:roll: