[dinos22]

F1OC is over.... just noticed the site updated with scores. Some mixed emotions about this competition to be honest...it really lost it's "charm" for me after XS pulled the plug but with all the fighting and BS i dont blame Charles for doing that

Here is my final score for F1OC 8M SuperPi round and a W.R. for AMD CPUs

I disabled a couple of cores in CPUz....it went a bit funny lol! It didnt help me with frequency but i did get a slightly better wazza and gained another 0.3s so that worked out well i guess

This was my second session and i changed over to my gold pot......this was a first "proper" test for this pot hence i decided to install the probe etc. For those who haven't seen much of it i made a quick video a while back when good people from Thermolab made me and a few other overclockers a custom gold(plated)en LN2 pot heh > YOUTUBE VIDEO..........it held the cpu nicer than F1EE but i still need that Venom strapped to it, i reckon, to try and nail some 32M runs between 6555-6600......from what i saw, so far, it's going to be very nice fellas

Gigabyte 790FXTA-UD5 board is really very very good so far. It's held up to some serious torture and water all over it......efficiency was there i cant complain ...just gotta work out now how to setup Gigabyte EasyTune on my SuperPi OS as i had to revert to my "all round testing OS" for final results

R.I.P. to one of my Corsair Dominator GTX sticks...in fact my BEST stick was lost in first bench session...not realising until i tried on the second session :(:(. The ONLY thing i can think of is condensation but i still don't know how as i packed the shit out of vdimm slots with vaseline.....CPU pot is a bit too close to dimm slots on AMD boards so running -187C for hours will eventually freeze a large area around the board too as anyone benching AMD would know i guess

These Corsair Dominator GTX 2250 sticks are MONSTERS......For the final benching session with 936MHz 6-6-6-18-1T andn 4680MHz NB i only use 1.7vdimm (DMM measured)

[Chicken Patty]

Great clocking dinos J got tht same board and CPU heading over to me. I just need some LN2 :(

[Buckeye]

Great stuff Dino's !

I know what you mean about hours of benching at -190c and freezing the board up LOL.

To bad you live so far away, I bet you would love the Phantom. Just running it with single wall is really the best and its a beast for cooling.

Its really funny how you would normally think of pouring LN2 into it, just keeping it full of LN2 actually keeps it warmer. But once you get it down to max temp splashing LN2 in the pot is by far the best way to keep it the coldest while running.

In the pic's I had to use Vodka as we cannot get Everclear out here so it froze solid. But even Everclear would freeze solid given enough time.

[TheDarkN8]

look like a shave ice machine down on the North Shore. Can get one strawberry flavored?
That guy NeoForce look like Hannibal Lector.

[dinos22]

hi buckeye
there is a downside to what you are saying and pretty much any pot will work best when very little LN2 is left but what also happens is when the bottom gets exposed to air you start forming a layer of ice with reduces the efficiency of pots which this kinds of thing so i agree about keeping low levels of Ln2 but disagree about doing this for hours

[Buckeye]

Originally Posted by dinos22

hi buckeye
there is a downside to what you are saying and pretty much any pot will work best when very little LN2 is left but what also happens is when the bottom gets exposed to air you start forming a layer of ice with reduces the efficiency of pots which this kinds of thing so i agree about keeping low levels of Ln2 but disagree about doing this for hours

Well that is ture, after a long bench session things do get rather frozen and even the Gigabyte board starts to degrade but at least it doesn't show any cold bugging.

If you watch this video and skip to about the middle of it you can see after a few hours what it looked like. You can see the amount of LN2 I was using and watch it boil down. I think it was 6-8 hours there at the end and I just filled it up for effect for the video, it was very frozen at that time. But the idea there was when running the pot with it full like that temps were warmer than splashes.

http://www.youtube.com/watch?v=ZhwEaQ4gcX4

I like that gold pot you have and it looks to be very massive and I would say works very well for max temps.

One thing we noticed in testing was that huge surface area and a large number of sharp edges really transfers the heat well. I believe Chew had one of Vinces early proto types of Venom and tapped the holes inside plus a few other things to create more sharp edges and was able to get a bit more performance out of Venom.

I tried to show in the vid how the bottom of the Phantom circulates LN2 twards the center instead of having a pool of boiling LN2.

Its really hard to say tho at max LN2 temps with out a lot of testing equipment what really works the best. I suspect that a lot of these pot perform very close and really depends on personall preferance and pouring style.

[dinos22]

hi mate

i like your video...it's cool

for those who dont know what i am crapping on about have a look at 2:42 of this video and see how the ice forms in the holes

now your video and discussion makes me think that there might be a "fix" for this situation.........you are right about edges and mass at the bottom.....one thing that i am now thinking AMD LN2 pot design needs is this......the base need VERY tall craters/edges about 5cm high and it needs flow so instead of holes or example that F1EE has it would need the opposite.....it would need those old school matrix style grid at least 5cm high with about 20% of them going even higher in height to some 8cm....i would also leave the middle without any...just yank them out of middle...i reckon a mix of gold pot bottom and the one you showed me but both have drawbacks.......the bottom needs to be as wide as it possibly can be so that you can havea hell of a lot of these but there needs to be at least a cm between these matrices imo

the reason to have about 20% of matrices extending past 5cm to 8cm height is because thats where you want the LN2 to be in that level not allowing it to go higher or lower.....just leaving it around there when you need that little more....so the pot cannot be very tall at all for this so you can have a very good view of the base

i hope you understand what i am saying....hehehe...

there are a lot of ppl i would love to live close to damn it....you are one of them....mainly for that swimming pool thought lol j/k

[Buckeye]

Yup your right it does form some ice after awhile, both inside and outside LOL.

I think I have poured a 160L and 240L of LN2 into this pot, plus a few 50's and after awhile you get to know when its time to hang it up for the day, there are a few IC's on the Gigabyte Board that are close in to the CPU that really do not like the cold for really long periods. Chew even puts a heating pad on those <opps secret there> HAHA

This and the F1 are the only pots I have really used and F1 just is lacking for extreme stuff. I have seen many others that look pretty good for this also, like yours.

One thing that I have seen that is pretty cool is when I run this pot with out the center tube and start bringing down the temps to -180c ish or so the LN2 becomes very still and stops boiling, same thing I saw with the F1 at max temps. Then if I keep a large amount of LN2 in the pot and start jacking up the OC the sudden rise of temps will make it start boiling again rapidly, on F1 this may boil over and make a mess LOL.

Its just kind of a phase the LN2 goes through when it gets very cold, its interesting stuff and I am sure you have seen this with your big pot.

I have talked with Aaron from LN2cooling a great deal about all this and its still pretty much up in the air about which design is the best I guess, they all work, its just a matter of how well or which is the best performer.

When its above 100 out here in the valley that pool is so freaking nice. BBQ and swimming

Just reread what you were saying about pot design... that is very close to how the Phantom is, you would have to see it up close. If your going to CES look him up as I believe he will be at the XS event.

[dinos22]

UPDATE!!!




I had a few sessions with AMD now since F1OC thread > http://www.xtremesystems.org/forums/...d.php?t=242063 and throught i'd fire it up once more to use some left over LN2 and see what i can do about a new challenge from Gyrock-san :up:

tightened the subs, increased the clocks and she was flyin' to a new 1M SuperPi AMD W.R.

Test Configuration:

• GIGABYTE 790FXTA-UD5 F2 official bios
• AMD X4 965 Black Edition C3 week 44 retail
• Thermolab custom gold "dinos22" CPU pot
• Corsair Dominator GTX 2250 DDR3 2GB modules (940MHz 6-6-6-18T-1T 1.7vdimm)
• PCIe nvidia 8400GS
• 1200W Antec Quattro OC Edition PSU
• 300GB Velociraptor
• Windows 2003 Server 110MB ISO Superlight

32M will have to wait as both my pots for some reason cant hold the temps to even to 6.45GHz runs... beats me heh :shrug:....times look great even at 6.52GHz

i installed the second probe on CPU IHS.....very useful info and in fact thats all i was looking at throught the bench session.... ..........i've noticed a point at which i get the crashes..........my IHS temp was -173.9-174.4C idle and it drops to low -173C for Pi..............................at some point it will drop hard by to -172C to -171.5C and crash...........happened even in 1M at times.........but mainly 32M sometimes at start and sometimes down to loop 5.....bottom of pot was at -187.4C idle and -186.5C load in Pi

[Kal-EL]

Stellar!
Do you have pics of how you installed or can you explain how you are using the second probe on the ihs?