Does a 2ghz+ oc really need sub-zero temps?

[Maximus Nickus]

I was just thinking why you can oc better @ exactly the same voltages at say a -5C temp than a 15C temp?

[RED Hot Machine]

To answer the header question, it all depends on the chip you are using, I would have thought you could get close, or even break the 2ghz barried with one of the new xp chips using air or water.

[Maximus Nickus]

Lets see what my 1900+ can do, its locked though so i'll need like 180fsb to break 2ghz....

[Ralphing]

WOOT!!:D

http://paran0id.homestead.com/files/2008.jpg

Hit it @ 167X12 w/ my spanking new KR7A-RAID. Vcore 2.24v and vmem 3v. Temps were in the mid thirties and dropping.:D

[Silver]

I suspect that in theory it can be done at higher temps however the variation in temps would have to held at a minimum. This means that the ability to move heat away from the core to the cooling medium would have to be much more efficient. I am playing with this some as the financial resources become available. Problem - small core - Problem - though the present heat sinks/water blocks I believe use good quality metals they are built for the conventional overclocker and standard board and are not designed for the extremist. This leaves but one avenue for them, overpower the chip. Thus we get pelts, kryo, etc. efficiency I believe could be increased. Take for example my first watersink. I built this to replace a Danger Den 2/1 water block. This was just playing so I took the heat sink off my kids e-machine (aluminum), plexied it, added three 1/4 inputs and three 1/4 out lines. Modified the socket to accept the larger block due to the plexi and the temps are 1c better at the same loads. Temps at stock settings never exceed 30c under torture tests. Alot of things could be improved on my next try. I.e. all copper, double the size of the waterblock, better use of multiple inputs and outs, increasing internal surface area (not convinced on advantages of water passages over internal fins and or fin directed passages). A water block should not get warm to the touch. When this occurs one wonders on the advantages of fan assisted water cooling. I have yet to see a water block that had an increased suface area I.E. was longer than the height of the board when atteached to the socket. Why? the block gets warm to the touch (not hot) yet the water temp is below ambient , indicating the saturation of the metal to me. This seems to indicate both the inability of the water to remove and the metal to absorb more heat. I have no problem with the rise in temp, I do have a problem when a chip trys to get into windows at 1.792 and then just before getting there falls out to reboot. The temps were good during the boot but increasing, at temp saturation the temps spiked or at least rose to the point of failure. Had the heat been absorbed during the whole chain of events then a full boot should have been achieved. If a small increase in temp can be achieved between the load no load condition then stability should be achieved without extreme cooling methodology. One can not deny the advantage of extreme cooling but is it not the brute force approach? Can not the same be achieved through controlled heat rise. Better designs seem to be achievable. Heck even when I get it in at 1.77 and bring up motherboard monitor I can watch the temps continue to climb at idle, to failure. Disconcerting. BTW, these ramblings are from a tenth grade drop out and do not indicate any real measure of scholastic achievement. But I'm having fun.

[Maximus Nickus]

Originally posted by Ralphing
WOOT!!:D

http://paran0id.homestead.com/files/2008.jpg

Hit it @ 167X12 w/ my spanking new KR7A-RAID. Vcore 2.24v and vmem 3v. Temps were in the mid thirties and dropping.:D

Benchmarks? Wat vid card you got?

What hsf/fan are you using and the temps (board/cpu/room) temps @ max load!

Thankyou!!

[Ralphing]

Originally posted by nick_cw


Benchmarks? Wat vid card you got?

What hsf/fan are you using and the temps (board/cpu/room) temps @ max load!

Thankyou!!

I'll try to find some time to post some benches tonight (right now I'm on a pee III). I'm using a Dangerden Maze 2 waterblock. Idle temps were about 34c and full load about 39c. These temps should be dropping significantly in the next week or so. I'm re-routing the cooling lines and it should get much colder here. I've got a 6 gallon res and radiator outside. The only fan I have are on the radiator... kinda weird getting used to a really quiet sys.

[Jeff Bolton]

sweet oc dude! you're my idol! just kidding...bebop bolton was my idol but i haven't seen him in a while...strange :D:D:D

jeff

[Maximus Nickus]

yeah it is.....2ghz wow!

Wonder wat my new 1800+ or 1900+ will do?

[SickBoy]

As Hoot used to say, higher overclocks require a lower operating temperature to be stable. For example, my 1 GHz Athlon starts getting unstable at 55-60C when overclocked to 1.2 GHz but when it's at 1.4, that threshold lowers to about 41-42 C....

SickBoy

[EclipseJP]

Hey Ralphing what kind of memory and other things for your computer? Please post specs.

[Ralphing]

Originally posted by EclipseJP
Hey Ralphing what kind of memory and other things for your computer? Please post specs.

KR7A-RAID
XP 1900+ AGKGA "Y"
256mb Corsair XMS2400
Visiontek TI500
SBLive Platinum
Linksys NIC
Antec PP403x
Dangerden Maze2, NB block, vid card block
2 Maxtor D740X-6L 40GB 7200rpm ata133 in RAID 0
Teac cd-rw
Samsung dvd

[EclipseJP]

Wow. I should get that kr7a mobo. This dragon one isnt doin it for me.