-Post your temps here- OMG! My rig is uber-cold!

My system temperatures(all at 100% load):

Temperatures taken either from onboard diagnostics or a thermal probe

Soulbox: P3 1ghz @ 1.13ghz, closed case, stock fan:

Case 38C, CPU 42C

Aera: Dual Celeron 400mhz @ 570mhz, closed case, trimmed volcano heatsink on cpu2, generic heatsink on cpu1

Case 50C, CPU1 58C, CPU2 51C

*** note: I have been having many heat issues with this box lately. I have been trying desperately to find an exhaust fan that will fit the case properly, as well as a better heatsink for the other CPU. The motherboard is an abit BP6 which has capacitors CROWDED around the socket, making it difficult to find a decent heatsink that fits.

Dunkle: Dual Pentium MMX 233mhz @ 233mhz, closed case, socket 7/370 intel HSF

Case: 20C, CPU1 25C, CPU2 25C

Lei: 233mhz Pentium MMX, closed desktop case, no fans at all (silent box)

Case: 25C, CPU no clue, can't get in to measure it properly

Eileen: Celeron 400mhz @ 540mhz (socket 370), closed case, volcano7 with thermaltake fan
Case: 43C, CPU 47C

Glazed: AMD 486DX4-SE 100mhz @ 160mhz, closed case, 486 extended style heatsink/fan

Case: ~20C, CPU unknown but heatsink is only "warm" to the touch

Profundus: Pentium II 300mhz @ 333mhz, closed case, dual fans (had a lot of "issues" with heat originally, is a Klamath core)

Case: 32C, CPU 38C

Chip: CeleronA 366mhz, closed case, no fan

Case: don't know, can't open case CPU: 36C

Phaedra: AMD k6/2 533mhz @ 550mhz, open case, not sure of the heatsink but I sanded it down EXTREMELY well, has a very large thermaltake fan

Case: Less then 20C, CPU: A bit tricky. The top of the heatsink is literally cool to the touch. Towards the bottom it gets warmer but is never "hot", which surprises (and scares) me, because the k6/2 is known to be quite an inferno. However, after more then 24 hours at 100% cpuload, the system is still stable and running fine. I'm not sure if the cooling is simply perfect (the heatsink is smooth as a babys bottom) or if it isn't cooling it properly at all, and the cpu is just surviving a ton of heat.


The rest of my boxes I do not use regularly or could not reach far enough back to open them up to test

Originally posted by Soul Harvester
My system temperatures(all at 100% load):

Temperatures taken either from onboard diagnostics or a thermal probe

Soulbox: P3 1ghz @ 1.13ghz, closed case, stock fan:

Case 38C, CPU 42C

Aera: Dual Celeron 400mhz @ 570mhz, closed case, trimmed volcano heatsink on cpu2, generic heatsink on cpu1

Case 50C, CPU1 58C, CPU2 51C

*** note: I have been having many heat issues with this box lately. I have been trying desperately to find an exhaust fan that will fit the case properly, as well as a better heatsink for the other CPU. The motherboard is an abit BP6 which has capacitors CROWDED around the socket, making it difficult to find a decent heatsink that fits.

Dunkle: Dual Pentium MMX 233mhz @ 233mhz, closed case, socket 7/370 intel HSF

Case: 20C, CPU1 25C, CPU2 25C

Lei: 233mhz Pentium MMX, closed desktop case, no fans at all (silent box)

Case: 25C, CPU no clue, can't get in to measure it properly

Eileen: Celeron 400mhz @ 540mhz (socket 370), closed case, volcano7 with thermaltake fan
Case: 43C, CPU 47C

Glazed: AMD 486DX4-SE 100mhz @ 160mhz, closed case, 486 extended style heatsink/fan

Case: ~20C, CPU unknown but heatsink is only "warm" to the touch

Profundus: Pentium II 300mhz @ 333mhz, closed case, dual fans (had a lot of "issues" with heat originally, is a Klamath core)

Case: 32C, CPU 38C

Chip: CeleronA 366mhz, closed case, no fan

Case: don't know, can't open case CPU: 36C

Phaedra: AMD k6/2 533mhz @ 550mhz, open case, not sure of the heatsink but I sanded it down EXTREMELY well, has a very large thermaltake fan

Case: Less then 20C, CPU: A bit tricky. The top of the heatsink is literally cool to the touch. Towards the bottom it gets warmer but is never "hot", which surprises (and scares) me, because the k6/2 is known to be quite an inferno. However, after more then 24 hours at 100% cpuload, the system is still stable and running fine. I'm not sure if the cooling is simply perfect (the heatsink is smooth as a babys bottom) or if it isn't cooling it properly at all, and the cpu is just surviving a ton of heat.


The rest of my boxes I do not use regularly or could not reach far enough back to open them up to test

Click to expand...

So many pcs....you can setup a mini cyber-cafe! lol....

anyone try nitrogen?

vehementi thats no lie, and with my amd xp 1600 oced to 1.6 ghz (152 mhz fsb.) I'll get a screen shot later of it, i have my case open with my cpu fan running at 7700 rpms...

and actually right now the cpu is running at 31*c and i just got done playing counter-strike...

Originally posted by palmboy5
anyone try nitrogen?

Click to expand...

LN² (Liquid Nitrogen) cooling is very expensive, hard & expensive to maintain, and VapoChill ($470) works alot better. Most people with LN² have crappy video cards and just overclock their CPU another GHz. To what end, I have no idea.

Originally posted by Vehementi
Most people with LN² have crappy video cards and just overclock their CPU another GHz. To what end, I have no idea.

Click to expand...

Another GHz? I'd have to see that to believe it. It seems to me that the substrate would break down completely due to the voltage increase. Transistors have a threshold, going beyond that will saturate it and it becomes a short circuit. This would either permanently turn it on or off depending on what its steadystate is.

Getting rid of the heat factor does push the limits of components much farther but they still have a limit. devices that are controlled by current and voltage have limits in the amount of current or voltage they can withstand.

A problem with supercooling is that if you get components too cold, they become super-conductive. This is a real problem with something which is supposed to be semi-conductive.

Originally posted by StormBringer
Another GHz? I'd have to see that to believe it.

Click to expand...

The Iceman Cometh: P4 at 4.1GHz

What's the fastest P4 out today?

I've read that article. I found it very hard to believe. I wonder if it still works. I've heard of aircraft that fly in the arctic and after being in the area for a few days or weeks, some of their systems malfunction because components go super-conductive(essentially becoming a short)

Originally posted by Vehementi
LN² (Liquid Nitrogen) cooling is very expensive, hard & expensive to maintain, and VapoChill ($470) works alot better. Most people with LN² have crappy video cards and just overclock their CPU another GHz. To what end, I have no idea.

Click to expand...

why do they have crappy video cards?

anyway, of course its hard to maintain, at start up the CPU isnt hot yet, there has to be a very good program to keep the nitrogen from flowing until teh CPU is hot enough, too early and the CPU will freeze(aww, no *freeze* smily...)

anyone have a pic of those PCs?

Originally posted by palmboy5
why do they have crappy video cards?

Click to expand...

They cash out on system board, CPU (e.g. $700 P4 3.06) and RAM and have no money left/use for a good video card.

Originally posted by StormBringer
I've read that article. I found it very hard to believe. I wonder if it still works. I've heard of aircraft that fly in the arctic and after being in the area for a few days or weeks, some of their systems malfunction because components go super-conductive(essentially becoming a short)

Click to expand...

Well yeah, I mean "overclocking their computer another GHz" was an overstatement. THG's article was probably incorrect in that it was "a stable computer enough for typical home usage," I mean I'm sure it wasn't that stable. I don't really think THG would lie in their articles however...

Oh, I wasn't saying it was a bold faced lie, but I'm sure they stretched some of the facts a little and probably got creative with some of their numbers. I also remember that they tried a few different CPUs all of the same speed and that one was the only one that made it to 4.1GHz.

1 out of 3(I think that is how many they had) isn't bad but who is to say that the next 100 would also fail to reach their goal.

http://service.futuremark.com/compare?2k1=5300380

still not believe it?

i also believe that was with a p4 2.8 ghz not the 3.06 ;/

Originally posted by Nick
http://service.futuremark.com/compare?2k1=5300380

still not believe it?

i also believe that was with a p4 2.8 ghz not the 3.06 ;/

Click to expand...

I didn't see anything in the form of proof there.

My point is that getting that kind of increase is very difficult and probably wouldn't be very stable for very long. Someone reaching it at all is amazing, but to be impressive, they need to keep track of how it runs on a daily basis for...say two months(just to pull a figure out of the air) If these machines can remain stable through everyday usage for an extended period then I would be impressed. If they can't then you shouldn't be impressed either. What is the use of having a machine running at 4+GHz if it is unstable or becomes unstable after a few hours or days?

Well it's stable enough to run a few CPU benchmarks and 3DMark, Storm. I suppose that's all that matters to these people.

It won't last long...

O/C-ing at such an extend will shorten CPU and GPU life... they just wanna break the 3DMark record to, you know, earn a "name" for themselves.

This is what I feel.

Keep in mind the Prometeia keeps CPU's at negative temperature, eddy. Any life shortening will not come from the heat side. I heard somewhere that for every 10 degrees you lower a component, you double it's life span.

Originally posted by Vehementi
Keep in mind the Prometeia keeps CPU's at negative temperature, eddy. Any life shortening will not come from the heat side. I heard somewhere that for every 10 degrees you lower a component, you double it's life span.

Click to expand...

When its too cool...the cpu will not be able to function properly and optimumly and it even may malfunction in long term running.

That's when it's too cool, we're only talking about -10C here.

-10*c = 12 fahrenheit (-10 + -10 + 32 = 12)

Originally posted by Nick
-10*c = 12 fahrenheit (-10 + -10 + 32 = 12)

Click to expand...

Actually it's -10 x 1.6+32 which equals 48*F...

I'll fear overstressing components. There one thing I heard before, that if you overclock a DDR266 to DDR333 it will NOT run as stable as a certified DDR333 will. In this case it's not about the heat right? It's about the speed in which the component is originally designed for.