5 GHz Project: CPU Cooling With Liquid Nitrogen

Quote:

---

Originally posted by Tedinde
Nice looking cooler, Is it all copper or just anodized?? I kinda stay away from anything MSI for about 3 years.

---

to tell you the truth, I don't know...
I had that card for about 3 days when I was considering getting a 5900 ultra, my friend lent me his, and it's a MSI, and I tell you, I have never been more frustrated with one card, then a 5900 ultra.
one second it was great.
2 seconds later, the screen "clicks" into 2d speeds and even when cooled with a R404 vapochill I had no idea how to stop it from doing that.
the 5900 NU is a complete 180, I totally love that card.

oh yeah, another thing, you want that heatsink, prepare to pay 150 extra, cause MSI's 5900's and such with that heatsink are insanely priced :rolleyes:

Quote:

---

I have never been more frustrated with one card, then a 5900 ultra.

---

A-freaking-men

yes, i agree , stay away from MSI parts. i have a FX5600 MSI video card and it can't o/c more than 30mhz on stock cooler. furthermore, my PC keeps restarting after each game. i'm using MSI 52.16 , and microsoft keeps saying my vc is causing device failure or soemthing. i found out, that it was the MSI card failure to toggle between 2d and 3d mode after playing a DX9 game causes BSOD. i have sent email to EAsports and microsoft regarding these issues and still no answer. MSI has a lot of bugs. haven't sent email to MSI yet , but i should.

MSI boards shouldnt be as bad as your saying.
the problem should lie in the driver your using maybe.

try a Nvidia reference driver.
then see what happens ;)

Quote:

---

Originally posted by Kanavit
yes, i agree , stay away from MSI parts. i have a FX5600 MSI video card and it can't o/c more than 30mhz on stock cooler. furthermore, my PC keeps restarting after each game. i'm using MSI 52.16 , and microsoft keeps saying my vc is causing device failure or soemthing. i found out, that it was the MSI card failure to toggle between 2d and 3d mode after playing a DX9 game causes BSOD. i have sent email to EAsports and microsoft regarding these issues and still no answer. MSI has a lot of bugs. haven't sent email to MSI yet , but i should.

---

Kanavit,

No more futuremark for you? What happened?

I had a problem with MSI kt133a turbo boards, they were dying like crazy in customers computers i was building a long time ago.

It put MSI on my black list. 5 or 6t boards came back.

Asus got there for the p4t 533 and "c" MB's i bought that said extreme overclocking options, Then only went to like 153fsb and weak Voltage options!! 2 boards in a row, that one review sites they reveiwed enginering sample boards that would do more than production!!!

WD got there for i keep track the 11th RMA on a HD in 2003!! Their new JB models if anyone has noticed, the houseing is cheaper and they just feel ligherter than the old JB's or SE.

for the record i RMA's 2 maxtors, and 0 seagates.

Back to topic.

They should've had tom there for the test, surely with all his skills they wouldve gotten more out of the Chip!!!

that video card is MSI FX5900 Non-Ultra. Not a bad clocker i might add.

Do you really think a guy who said 3DMark2001 is a DX7 bench while talking about Advanced Pixel Shader test would really be able to do that? Oh and he got even a modded Asus board for that test.

I don't like THG for there past photochop fiascos and strange looking AMD vs. Intel comparsions.

My opinion...the base on his container was sooooo thin that he couldn't bench. He'd have WILDLY fluctuating temps....which will kill stability so high.

C

wuahahaha, this is a big fake.

look this: http://www.kaltmacher.de/viewtopic/t-28279/start-0.html


THG suxx :banana:

wish I could, but no read German...
Ich verstehe nur sehr, sehr kleine Deutshe.

C

use this to translate. ;)

Quote:

---

Originally posted by pik-ard v1.1
use this to translate. ;)

---

yeah....

agreed with jason. MSI/PNY are very good o/cers. my ti4600 could do 375/770 on air cooling (Alpha socket A hsf). the FX5900NU i had could do over 500mhz core on stock sun-flower heatsink and that was one of them first revision boards.

and btw, i heard Tom had some benches but pulled them since in some cases his 5Ghz P4 was getting raped by an FX-51 chip.

you mean some FX-51 benches that he never published?

or benchmarks with the 5ghz?

meh, if he got 5ghz with a P4EE, might be another story... :rolleyes:

Quote:

---

and btw, i heard Tom had some benches but pulled them since in some cases his 5Ghz P4 was getting raped by an FX-51 chip.

---

Well of course.. he used some 3-4-4-8 slowass RAM at 3:2 divider LOL :D
4.4G with good RAM settings would be faster :P

Any proof of these '5ghz benches' ? they mention it was stable at 4.7G in the article..where did the 5ghz come from?

the whole thing is a fake i think :(
Dunno if it has been stated before, but please look @ http://www20.tomshardware.com/cpu/20...kondensat4.jpg

here you will see the mobo totally frozen, but you see the fanwire and the heatsink + piece of fan of the NB...

WHy the hell would they "change" their vapo for a stock NB cooler?
Iig thats not even possible, since either the vapo would have been virtually frozen on...

Theres several things that make the credibility drop...

macci, they said "completely stable". maybe they mean prime95 stable. and its got to be more then a few hundred mhz from p95 stable, to get-into-windows-for-a-cpuz-screenshot stable. ;)

serial, that link is to nothing... :P

Quote:

---

Originally posted by serialkiller
the whole thing is a fake i think :(
Dunno if it has been stated before, but please look @ http://www20.tomshardware.com/cpu/20...kondensat4.jpg

---

LOL, I think someone is watching this thread, seems the picture was pulled???

serialkillers link works fine for me. did you guys see the 13500mhz cpuz shot in the thread SKYNET posted. :scary:

could this stuff be intel backed/released 'spin'

I only want to know how his mobo didn't fry with all the condensation...I fried a P4C with a lot less than that!!!! And BTW, his mounting method for the CPU container sucks :D
And his condensation proofing is a joke....this is amateurish at best. He should have used his fat budget to fly Macci or Memesama or TAM to Germany to show him HOW to do it...
:rolleyes:
C

Quote:

---

macci, they said "completely stable". maybe they mean prime95 stable. and its got to be more then a few hundred mhz from p95 stable, to get-into-windows-for-a-cpuz-screenshot stable.

---

I could run a 8M SuperPi test (takes 5min34sec) at 4.8GHz and got a WCPUID at 5GHz. Thats 200MHz difference.
Believe me they didn't run Prime95 with LN2 :D
Also I'm quite sure that there system was not stable at any MHz with the LN2 cooling - looking at the way they pour ln2 it seems that they had no idea whatsoever how to use it :D

Remember a similar thing with [H] last year?? They couldn't give one verifiable benchie at speed. All they had to do was run 3DMark at default GPU clocks so the CPU speed would be recorded. Then!!! I'd give alllll PROPS to THG...if they posted a 4700-4800mHz run I'd be like :slobber:

C

Quote:

---

Originally posted by macci
I could run a 8M SuperPi test (takes 5min34sec) at 4.8GHz and got a WCPUID at 5GHz. Thats 200MHz difference.
Believe me they didn't run Prime95 with LN2 :D
Also I'm quite sure that there system was not stable at any MHz with the LN2 cooling - looking at the way they pour ln2 it seems that they had no idea whatsoever how to use it :D

---

AND Macci,
Can you believe NO THERMOMETERS???? He was probably shocked that the sensors on a digi-doc wouldn't work, lol. Macci, would it be possible to make a reservoir for LN2 with a copper spout dripping LN2 into a container?? Continuously??

C

like this actually?

@charlie,

dont see why not.:D