|
Post by Bones on Jul 25, 2019 13:50:21 GMT -5
Needed a thread about this so here it is. rushed , what TIM were you using before? NT-H1 is good for anything from ambient to subzero.
|
|
|
Post by WhiteWulfe on Jul 25, 2019 14:19:31 GMT -5
I miss NT-H1... I never had an issue with it. Only reason I haven't picked up another tube of it is I have a... Handful of, uhm, blargh, forgot what it's called. The artifical diamond stuff. I think I have half a dozen tubes of the stuff.
Quick and dirty benching I use Ceramique 2.
|
|
|
Post by rushed on Jul 25, 2019 14:30:57 GMT -5
Been using MX-2 for most of my benching but I've gone through all but like 1/2 a gram now of everything I've had... Edit: i like the viscosity of mx-2 more than mx-4 for example, ordered some GD900 just for the lulz since it was 30g for 4 dollars that's in the mail now, WhiteWulfe I've heard Ceramique is good for sub ambient what's your opinion on it?
|
|
|
Post by Bones on Jul 25, 2019 14:44:21 GMT -5
You can use Ceramique for DICE but not LN2 for example. For SS it's fine, for cascade it won't work. MX2 and MX4 are better suited for lower temps overall, other TIM's like NT-H1 for example can be used for about anything with some doing better than others.
Cracking of the TIM is the biggest issue you'd face with LN2 use, I'm not sure if it happens with cascade but possible.
Whatever is cheapest to use and give good results is what I'd go for.
Kryonaut is what many use for whatever but it was developed for subzero/LN2 use and works well.
|
|
|
Post by WhiteWulfe on Jul 25, 2019 15:10:43 GMT -5
I've used Ceramique for subzero once or twice with dry ice... Not a fan, especially compared to NT-H1. After the longer sessions I tend to do (usually 6-8 hours), it becomes this really annoying chalky block... Making it rather hard to break contact between the cooling pot and the cpu when you're thawing things.
I've been tempted to try smurf paste (aka Kingpin Cooling's stuff) but ouch on the cost.
|
|
|
Post by georgekokovinis on Jul 25, 2019 15:11:18 GMT -5
There are only two ways to go :
1) Subzero - KPX blue paste or Thermal Grizzly Cryonaut 2) Ambient - Thermal Grizzly Condactonaut.
|
|
|
Post by ShrimpBrime on Jul 25, 2019 16:42:25 GMT -5
Been using Antec formula 6 and formula 7 for years now. Have not done quite as much sub-zero clocking as some other members here, but nothing but good experience. The paste doesn't seem to be an issue, not hardening and cracking under the extreme temps. Seems to spread very well and conduct fairly decent. Again haven't a need for any other paste. Have used and tested on very many of AMD de-lidded processors, ranging from socket A to Ryzen 7 2700x. edit: correction - On every single platform from socket A to AM4 1331 pin. Works very well against most top pastes (not tested other pastes on this theory) but generally seem to have nice temps. Ambient air and liquid cooling performs very well. Have set up for several years and haven't seen a dry out yet. Switched from AS5 paste years ago. Beats it by several degree.
|
|
|
Post by Macsbeach98 on Jul 25, 2019 17:10:32 GMT -5
For water and dry ice you can use most pastes there is only a 2 degree difference between good and bad I am using MX2 for water at the moment smooth and even mounting makes more of a difference. I am using MX2 at the moment
For full pot nitrogen. Kryonaught KPC Blue Gelid Extreme The rest will shrink and crack. I generally use Gelid because its easier and cheaper to get.
Rushed I knew a bloke that swore by GD900 he was even giving me tubes to try it worked Ok but dried out in the tube before you got to the end of it and didnt spread well.
Always check your mount when you take them apart if its thicker on one side of the mating face than the other side its not on there well.
|
|
|
Post by rushed on Jul 25, 2019 18:51:29 GMT -5
Ahh, Might end up going with more MX2 then I suppose. I'll grab kryo or gelid at some point to try them I suppose too. I've heard for full pot ln2 KPC is top dog - but we're not there yet
|
|
|
Post by Mr.Scott on Jul 25, 2019 18:55:21 GMT -5
MX2 is good. I use that on my 'long term' projects. I use Ceramique on the chiller. Have about 20 tubes of IC Diamond and hate it. Too freakin thick. Mount is more important than type of paste really.
|
|
|
Post by ShrimpBrime on Jul 25, 2019 20:18:17 GMT -5
Apologize for this dumb question, but what carat IC paste is it and would you be so kind to sell me a tube for a try out?
Agreed with mounting importance. A little lapping goes a long way too.
|
|
|
Post by Mr.Scott on Jul 25, 2019 20:23:55 GMT -5
It is 24 carat, and you can have a few tubes. They are the 4-5 gram size. Should last you a while.
|
|
|
Post by ShrimpBrime on Jul 25, 2019 21:41:42 GMT -5
Lasts a while depends on what cooling apparatus is used. TEC cooling... well it goes pretty quick. And I think for that particular type of cooling this IC 24K nano-D will do just fine. I'm gonna guess the temps will be very similar to the Antec Formula 6/7 within a degree or so. The difference will be the consistency where the way you describe it as pretty thick and difficult to spread.
|
|
|
Post by zila1 on Jul 25, 2019 22:58:59 GMT -5
I always put a blob on top of the IHS with IC Diamond and let mounting pressure spread it and that has worked well...………..but has anyone experienced pitting and scratching with it?? I have.
|
|
|
Post by WhiteWulfe on Jul 25, 2019 23:24:56 GMT -5
Ahhh, it was IC Diamond I couldn't think of the name of. And I'll definitely second that it's thick stuff... Also has a bad habit of scraping off any engraved details on the IHS too.
|
|
|
Post by Bones on Jul 26, 2019 1:06:48 GMT -5
I've heard some say IC Diamond is great if you use it a kind of paste for lapping the lid on a CPU. I won't use anything like it because of the fact it has diamond dust in it, that will also have an effect on a pot's mounting surface too.
I have a couple tubes of the new MX4 stuff that has carbon in it, don't know how well that would work but since I have it already I can try it.
|
|
|
Post by ShrimpBrime on Jul 26, 2019 6:12:12 GMT -5
People still using IHS plates??
Question on surface area... Wouldnt scratches increase surface area rather than cause an unseen issue.... Or is it just the etching on the plate becomes hard to read? These are serious questions, Cant recall this experience with dimaond nano pastes before or just may not have cared enough to notice.
|
|
|
Post by Bones on Jul 26, 2019 6:49:29 GMT -5
Not really but would eventually affect that mirror-finish on your favorite cooler or block.
|
|
|
Post by ShrimpBrime on Jul 26, 2019 15:44:13 GMT -5
OK. I was thinking maybe from re-seating the cooler to the cpu numerous times that would be an issue more so than applying the paste once and it does it's job for a length of time.
|
|
|
Post by dr4g00n on Jul 26, 2019 16:06:59 GMT -5
Most carbon based pastes are abrasive in some way (inc. MX-2/4), just to less of a degree than IC Diamond. When applied properly it works pretty good for air & dice. Just barely apply pressure to the HS/block at first and let it heat up, once it gets hot then cinch it down the rest of the way.
I wouldn't use it on the die though since it'll scratch it up quite a bit. LM is better for that anyway, at least down to dice temps.
|
|
|
Post by ShrimpBrime on Jul 26, 2019 16:35:41 GMT -5
I've been using diamond nano for many years. Firstly it's a fine dust. finer than 2000 grit wet sanding paper, and that's why it's so cheap. Been using it on processor die for many years (not IC diamond) and have lapped many cpu die after soldered de-lids. It really is ok to use Diamond pastes on naked chips. really it is. The dust is finer than cigarette ashes. lol.
BUT....
Nothing conducts better than diamonds. That's thermal conductive more than 2,000 watts per meter per Kelvin. That would be 5x better than copper which would be the second best. (your water block, LN2 pot base for example ?)
Now try mixing diamond dust with (thanks george) 1) Subzero - KPX blue paste or Thermal Grizzly Cryonaut 2) Ambient - Thermal Grizzly Condactonaut. And the results may make these pastes even better at thermal conductivity.
Let me edit: I do agree the diamond-nano paste would be considered abrasive. I'm only expressing why I use nano-diamond thermal paste.
|
|
|
Post by Mr.Scott on Jul 26, 2019 16:51:19 GMT -5
The IC Diamond works good. I just don't like the thickness. Yes, it's abrasive. 1 application and cleanup will make your IHS illegible. Works fine on bare die. Actually better on bare die than IHS.
|
|
|
Post by Aleslammer on Jul 29, 2019 10:20:35 GMT -5
First big brand name was AS5 maybe 4 or 5 mounts a year, then again probably all were there at one point in time. Early days on the Bot used OCZ Freeze after that was discontinued played with a few of the top tier pastes (barring Liquid Metal and the Diamond pastes), now use nothing but MX-2 big tube, cheap relative speaking and spreads well.
As pointed out most good pastes aren't that far apart (more or less) when it comes to ambient cooling, although the variance in the ambient area actually is a little more important than sub-zero, if all the small difference are ignored can easily leave 5 or 6c on the table possibly more if you throw in a sub par part, sorta like RAM to a gamer, makes a difference but better to spend the money on the video card.
|
|
|
Post by ShrimpBrime on Jul 29, 2019 16:08:08 GMT -5
My problem is reviews are reviews. Many very biased. Some even paid for. AS5 was super great. I remember release time, it was a better of most pastes, great for just any one. The problem we have is the lack of sustained testing environment. MX-4 tested by 2 people in totally different places could have a difference in 5-6c temps between the two systems even if identical.
Try LN2 pot, no paste and lapped surfaces. My experience was rather interesting to say the least.....
|
|
|
Post by zila1 on Jul 29, 2019 16:19:06 GMT -5
The IC Diamond works good. I just don't like the thickness. Yes, it's abrasive. 1 application and cleanup will make your IHS illegible. Works fine on bare die. Actually better on bare die than IHS. Yup. That shit screwed up one of my Big Typhoons. It was noticeably pitted, looked like a miniature version of the surface of the moon. I couldn't even lap out the pits. Edit: and it always leave a dark stain on the IHS. I had to use brake cleaner to get that stain off. That being said, it does perform cooling wise.
|
|
|
Post by Vinster on Jul 29, 2019 20:23:47 GMT -5
when I was doing dice and chilled I was running AS Ceramique
Vin
|
|
bilko
Regular Member
Posts: 21
|
Post by bilko on Aug 8, 2019 16:05:07 GMT -5
Personally for all my ambient stuff I use a cheap paste from Ebay called HY-510, it's like $1.50 for a 30g tube, I have used it sub zero before and seems to do ok on lightly threaded benches with soldered CPUs but if I'm actually serious about it then I'll use GC-Extreme or Kryonaut.
Oh and for GPUs I pretty much only use MX-4 for ambient and then GC-Extreme/Kryo for sub zero.
|
|
|
Post by Bones on Aug 8, 2019 17:41:53 GMT -5
Just ordered a 10g container of GC-Extreme to take up with me, already have some NT-H1 and the MX4 carbon stuff ready but wanted a TIM that's specifically stated to be good for extreme subzero use. As long as it gets here before I leave I'll have all my TIM needs taken care of for the meet and then some.
|
|
bilko
Regular Member
Posts: 21
|
Post by bilko on Aug 10, 2019 16:09:45 GMT -5
Just ordered a 10g container of GC-Extreme to take up with me, already have some NT-H1 and the MX4 carbon stuff ready but wanted a TIM that's specifically stated to be good for extreme subzero use. As long as it gets here before I leave I'll have all my TIM needs taken care of for the meet and then some. If it's goopy/runny then it's a great batch and you'll get good results, if it's a bit dry/thick then it won't do so good from my experience. Will still be better than most pastes for sure but it's a nice thing to know
|
|
|
Post by Bones on Aug 10, 2019 16:49:02 GMT -5
I'll see what I get and hope for the best, all I can do here.
|
|