• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

FRONTPAGE EVGA Frostbite Thermal Grease for Sub-Zero Overclocking

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
At a given load, the cascade will remove X watts of heat, the amount of heat removed doesn't change especially.
If the evap temp is colder, less heat is getting to it. So either the wattage of the CPU went down, or the evap isn't getting as much heat from the CPU.

The better the contact and TIM, the warmer the evap temps will be for a given CPU at a given load.
 
Then how do I know when the CPU temp is optimized? Warmer evap isn't better indefinitely, 0C evap has to be worse than -100C evap. Obviously the unit would be overloaded if at 0C, but how do I know what temp is best if it's not the lowest possible?

You said the watts removed doesn't change, if that's the case, then the colder the better. Removing 300W while at -118C is better than removing 300W at -109C, right?
 
Matt, when you run something like 1024M, does the temp stay exactly the same, does it rise, then level out, or does it continually rise? If the temp raises continually, then the load output is greater than the cooling capacity. If the load output is less than the cooling capacity, it will stay at minimal temps.

So I don't know how you'd find the best point for a cascade to be at, but I would assume the best point would be at or below the heat removal ability of the machine. Being colder means one of two things: 1. The wattage applied by the cpu was less and the evap could easily cope with the heat or 2. the evap wasn't receiving any wattage to remove.
 
Matt, when you run something like 1024M, does the temp stay exactly the same, does it rise, then level out, or does it continually rise? If the temp raises continually, then the load output is greater than the cooling capacity. If the load output is less than the cooling capacity, it will stay at minimal temps.

So I don't know how you'd find the best point for a cascade to be at, but I would assume the best point would be at or below the heat removal ability of the machine. Being colder means one of two things: 1. The wattage applied by the cpu was less and the evap could easily cope with the heat or 2. the evap wasn't receiving any wattage to remove.

Yeah, temps would continually rise until the compressor fails if I ran too much wattage for way too long. Temps level off after a 4-5C increase in wPrime, none of my CPUs can produce a ton of heat. If the cascade can handle the heat in both cases (temps level off), which is better, a colder evap or warmer evap?
 
Yeah, temps would continually rise until the compressor fails if I ran too much wattage for way too long. Temps level off after a 4-5C increase in wPrime, none of my CPUs can produce a ton of heat. If the cascade can handle the heat in both cases (temps level off), which is better, a colder evap or warmer evap?

I would say warmer evap. warmer evap means more heat dissipated from your cpu. With my SS I know I have a bad mount if I am getting -40 under load. a good mount is -30 under load. the evap is warmer because it is taking the heat away from the cpu.

EDIT: but if the evap is too warm then that means need to upgrade cooling solution
 
Yeah, temps would continually rise until the compressor fails if I ran too much wattage for way too long. Temps level off after a 4-5C increase in wPrime, none of my CPUs can produce a ton of heat. If the cascade can handle the heat in both cases (temps level off), which is better, a colder evap or warmer evap?

I would say warmer evap. warmer evap means more heat dissipated from your cpu. With my SS I know I have a bad mount if I am getting -40 under load. a good mount is -30 under load. the evap is warmer because it is taking the heat away from the cpu.

EDIT: but if the evap is too warm then that means need to upgrade cooling solution

That's a good question... I would say colder evap... IF YOU ARE 100% SURE IT'S STILL HANDLING THE CPU'S HEAT OUTPUT. ;)
 
I would say.......get someone that knows about this stuff in here so you guys can stop guessing. :p
 
I would say.......get someone that knows about this stuff in here so you guys can stop guessing. :p

I don't want to waste money on it until someone else tests it.
:)

Too much guesswork for sure. It could be worth it for some, but for a n00b like me I'll stick with the cheap stuff for now.

Matt
 
FTR I was addressing the slight aside on the evap stuff not the article. :)

As far as the thermal paste, its too expensive to be worth it to many here Im afraid...but we all look for the best of the best so Im sure someone will grab it and test it...*anxiously waits*
 
My bad :chair:

I agree it's too expensive for what little you get, especially for benching, where you make tons of mounts. For a set it and forget it daily system, I don't mind spending a little more on TIM.
 
Questions....so what was used before EVGA came up with this magic subzero TIM? How come ceramique is all of a sudden crap if it was used all these years by the world's top benchers?:shrug: Iirc icebob still bought ten big tubes of ceramique for use on LN2, that alone says enough for me.
As Ed said, marketing is the only thing that makes the new kid on the block great and the old crap.:rolleyes:
 
Last edited:
Questions....so what was used before EVGA came up with this magic subzero TIM? How come ceramique is all of a sudden crap if it was used all these years by the world's top benchers?:shrug: Iirc icebob still bought ten big tubes of ceramique for use on LN2, that alone says enough for me.
As Ed said, marketing is the only thing that makes the new kid on the block great and the old crap.:rolleyes:

EVGA didn't say Ceramique was crap. Gautam did.

EVGA went through many different pastes before coming up with this one. I know a bunch of sub-zeo testing was done, which Vince was personally involved in.

Until speaking with Gautam and doing research for this article I hadn't realized Ceramique was no good for sub-zero. I think when it comes down to it, many of us have not considered TIM all that much when doing LN2. Thermal transfer, as I mentioned in the article, is not that important because the pot is so damn cold usually. Only way to really test it is by using a die-simulator while managing the flow of LN2 very very carefully in order to get results that make any sense. I think TIM is something overlooked since blips in temperature of 3-5 degrees C don't really matter when you are sub-zero.

Matt
 
Until speaking with Gautam

Oh, did someone hijack IMOG's account? :D Matt is right though. In almost all situations, Ceramique is good enough, because you can simply lower the temperature of the pot to compensate. It's only now with coldbugless Gulftowns and GTX580's and the like that you actually need the best contact that you can get.

edit:

The better the contact and TIM, the warmer the evap temps will be for a given CPU at a given load.
This is correct.
 
Last edited:
so here is the million dollar question. Has anyone HERE got their hands on this stuff yet? I know it is available on the egg and EVGA has been selling it from their site so has anyone tried it yet? i am placing a big order from newegg and was wondering if it would be worth trying this over IC diamond (since i do not want to pay double shipping from another site)
 
Do you extreme cool? Thats where this product supposedly shines... with that said, get MX-2, AS5, IC Diamond, whatever is cheapest out of those.
 
Back