[Liquid Metal Showdown] Thermal Grizzly Conductonaut vs Cool Laboratory Liquid Ultra / Pro

I capture all temp data on stock temp profiles first and then with Max so those are stock. The GPU is a lot less sensitive and less stubborn (tolerance wise) than the CPU so it has a wider margin for room for error, if that makes sense.

If you apply too little with liquid metal, it will make a much more dramatic difference vs stock paste. There is a certain amount of paste that is absolutely needed with LM...it's a pretty fine line...ie...if you apply too little the temps will have this strange behavior where under normal tasks it'll seem fine, then the moment you give it gas, it'll sky rocket in temps....if you apply too much LM, it won't hurt, but chances are you'll get some pump out.

 

buy bigger capacity charger


EDIT

Oh wow, a mighty small CPU heatsink fan you have there...

 

Thanks for sharing the screen shots! Always nice to see data...

There's nothing like Liquid Magic...it just dominates.

 

I just did my P170EM with Conductonaut. Full results are here: http://forum.notebookreview.com/thr...0-owners-lounge.655838/page-298#post-10395114

Basically took upwards of 15C off my temps under load and allowed full running as fast as the 3720QM can go (3.9GHz multicore under AIDA64 stress cpu+fpu+cache)

What interested me was how it worked with the CPU. After investigating and ultimately "repasting" (re-metalling?), the contact between heatsink and CPU is poor and the concave shape means direct contact is on three corners of the ivy bridge die, and is what I believe is responsible for a ~15C idle ~5C load difference between core 0 and core 1 and constantly fluctuating temps. When with AS5 temps were pretty much stable +/- 1C both per core over time, and between cores.

The 680M, well, that just can't heat up if it tries, it crashes from stock volts not handling 1GHz+ before it breaks 80C on anything I throw at it in Kombustor.

 

@iunlock

Why are you using two temp monitoring software (Hwmonitor & Hwinfo64)? I read that one or the other doesn't report as accurate temps (Hwmonitor)

 

What is the best way to spread the thermal paste, Im not a total noob I did put my pc all together 10 years ago no worries im not rusty so what is the best way to spread thermal paste and keep in mind I have no tools nor stores in my country sell some tool I only have GELID EXTREME spoon or whatever it is but I dont really think it does a good job so please tell me

 

I only use HWiNFO64 now. HWMonitor was just to compare, but that was in the past. I no longer use it.

 

The Gelid tool works just fine. But for Liquid Metal I like the brush that comes in the CLLU kit.

 

You painted very close to the edges of the die here. Given how wet Conductonaut runs, were you not concerned that it may drip onto the PCB even w/electrical tape as a safe guard?

I figured it'd be better to play it safe and give it a little room on the borders

 

Cool thanks for this thread

 

Definitely not concerned. That's what the "3M Insurance," is there for...

 

I will be honest tho, I am considering CLU, but really just leaning on the non metal Thermal Grizzly series atm... not sure if it worth the risk.

I have a 680M, with copper heatsinks, BUT sorrounded by aluminium, same goes for the CPU. So yeah.

 

The difference is huge. If you want the best temp drops, LM is the way to go. I've used LM on systems that also had copper plates with aluminum surrounds and never had an issue. If you do go traditional paste, Grizzly Kyronaut for sure.

Good luck!

 

Exactly!

I am stuck in between those.

But you are using Grizzly Conductonaut (not available from my local shops atm), so I will go with* CLU if I pull the plug on liquid metal

EDIT: Okay holy sh**, I am going to do it, lets go all in, nothing to lose!

 

on a desktop it makes zero difference between kryonaut and conducto



on laptops it's like magic

 

I can't figure out whether to get Conducto or CLU for my laptop.

 

Go with CLU. Not so watery. A safer choice if you are afraid for leakage. Almost same cooling capacity. I have used both Liquid metal. Not the big difference.

 

Wait ze minute.

Just found out Amazon sells Conducto cheaper.. ok, I have no choice, I save 3£, must pick dangerous choice.

 

Can't go wrong with either. Good choice all around, CLU or TGC.

 

Some info for you http://forum.notebookreview.com/threads/clevo-overclockers-lounge.788975/page-753#post-10411731

 

Too late.

I will figure something out.

 

Just a head up. Intended as info for all, not a warning not to use Grizzly or Phobya metal

 

Yes ofc.

I will do a extensive walk-through of everything I did, still feel like these liquid metals are unknown territory.

 

I am currently using both cond. and CLU on CPU and GPU respectively; cond. was a pain to apply compared to CLU but both perform good and so far non of them pumped out or caused any problems.

I take my heavy tank laptop to University in my backpack and I have a problem with one of my CPU heatsink screws so I have it screwed only with 3 screws.

I believe if you apply any of them good you won't have problems.

 

I've used TX-4 in the past but not for this application. I decided to go with MX-4 as it had a higher w/mk at 8.5

Here are my own results with some MX4 and fujipoly extreme/ultimate pads. CPU still throttles in synthetic settings but I have the 6820HK sitting at 4ghz which it stays at under 3dmark or just on OCCT by itself.

Here it's just at stock clock speeds.

 

Hey y'all! I plan to repaste my inbound Aorus X7 v6 which is arriving sometime next month with liquid metal. Which one would be better? I weighing between TG Conductonaut and Liquid Ultra. What worries me is that it will move around too much underneath the contact plate as this laptop will be my main machine when going for classes and such so that involves moving a lot (stationary mostly on weekends). Opinions?

 

Conductonaut has more conductivity compared to CLLU, but several people are reporting an immaterial difference to CLLU (less than 1 degree). Conductonaut also has a really runny consistency compared to CLLU, so if you plan to be on the move, CLLU is your choice for minimizing leakage risk. CLLU also comes w/a better brush to paint the liquid metal compared to conductonaut's tightly spun cotton swab.

 

you ever figure out of CLU dries out?

 

Hasn't dried out on me during my possession of the machine. I've already sold it so I won't be able to keep an eye on it for long term. If liquid metal does dry out I'm sure we'd be hearing a lot more people giving a disclaimer.

 

How badly does GC move around though? If it does move around a bit too much then is it assume to assume that tightening the heatsink just a fraction more help?

 

Yea, that is like how some people also apply it to the heatsink. I do not understand why anyone would do that because there should already be enough of the liquid on the die before you place the block on it. Sure it is not as viscous as paste but I can not see having it on both sides do anything other than increase risk of getting air stuck in there.

 

It is not about you not understanding.

It is matter of choice and what the actual Conductonaut manual says to do.

 

Ah, very interesting.

 

Unless you got a fat glob of GC on there, it doesn't really run. If you have so much that you can see it visibly move around you got too much. Most likely it'll come squeezed out the edges when you put that heatsink on. I'd know as I put way too much my first time using it.

 

Guys, tell me please, who is experienced with GC, is it safe enough to apply it on AW 17 R4 with which I'll travel sometimes? Of course I'm aware that all area around dies should be covered with the tape (i.e. as Pete did here https://s30.postimg.org/q1c29zwrl/20170101_222020.jpg).

 

I don't have personal experience with this stuff but, consider that, desktop users use liquid metal paste wherein the material is in constant vertical suspension and, it seems to say on the die just fine.

 

Excellent post. Never thought about this.

 

Indeed, it make sense. Though while road trips overall conditions can be more unpredictable, all that shaking and pitching..

 

Everything depends on whether the heatsink fit well enough over the die/ihs. Bad quality aka bad fit - warped or uneven heatsink may lead to leakage. As some have already experienced

 

you should cover with tape nonetheless

correct, a warped heatsink is problem not only to conducto/cllu but for all thermal paste

shame that clevo has a crapton of heatsink problem though

 

I.e. the amount of tape iunlock and Pete used should be enough?

 

yes

 

There shouldnt be an issue with travel. There are several people on here that not only go on road trips but they bring their systems with GC on airplanes etc.
There is also some extra precautions you can take on top of the 33+ method that may help if you are super worried.
However at the end of the day the biggest help is getting the amount of LM Juuuust right.

 

That is very nice to know. I'm aware of some extra tricks which help (i.e. bending of the heatsink arm).
But could you share these extra measures that can be applied?

Also another quick question. As your team already have one year's worth of experience with the GC, how often it should be reapplied? I believe it should be good at least for a year or even more.

 

Just an update. I pasted my 15 R2 with Grizzly Conductonaut almost 2 months ago. My temperatures haven't gone up and I haven't had any issues. I store my laptop in a bag that has it standing vertically a lot of the time and have also had the computer on several trips in the car. I haven't opened anything up to see how it looks on the inside (to see if there is run) but for right now I don't plan on it.

 

I got serious thermal performance degradation, not sure what it's all about, I will open and look when I'll have time.

 

What paste did you use and how long did it take to degrade?

 

2 months, Conductonaut. More information here

 

It did not degrade he just upgraded it to Conductonaut.

 

Now conductonaut was in for 2 months, and now I Constantly see 95c cpu spikes and huge differences between core temps. Whereas with fresh conductonaut even CPU burners couldn't push it over 70C. So yes, thermals have degraded with conductonaut in 2 months, but probably something else is wrong, I'll try to check it out tomorrow.