randomly came across this video and really liked the time you put into getting a nice spread across the chip. such a simple thing but shows just how much you love and respect your hardware . . . appreciated that as a viewer too, a little effort goes a long way in my book and getting an 88mhz oc on stock is testament to that. thank you for all the effort making this video
Agreed. Nicely done. I thought it was a really well established channel by the video quality. Then looked and only 2.4k subs so far. Well let me add +1 to that # :)
+zEus b thanks man! I love getting feedback like this, it makes all the effort worth it, it takes a lot of time to produce a video so I'm glad you enjoyed it!
I've heard the performance of the liquid metal will degrade a lot over time? Something that's not a real problem with the stock thermal paste? So you might need to redo this every year? That's what I thought I read somewhere once? No idea if it's really the case.
The liquid metal thermal paste is a niche market after all. You cannot expect a large scale factory manufacturer to paste with this expensive high end paste. Plus, when applying the paste, you have to worry about not splitting outside the die as well. All these are not so welcome to mass manufacturing.
Thank you! Very often when people does the temp tests, they never put the fans at fixed RPM. ..which is just stupid, because then improving the cooling does not perform cooler temps because fans just go lower RPM.. Keep it up!
I googled this because the founders 1080 i placed in my mini pc is not keeping cool. Everywhere online people are saying 'dont do it' 'you'll break it' 'the clamping pressure wont work'. This reassures that people dont know until they try. Thanks for confirming my question!
BangDroid wow thanks mate! I appreciate the feedback. I've got another liquid metal video coming this week.... On the evga 1080ti ftw3... ;) Make sure you catch that!
Super great and surprising test!! Subscribed. I did this to my Asus ROG 1080 OC when I installed my water block 2 days ago. Cant wait to see the results! Thanks for this test!
Just a tip when performing tests regarding temperatures. State the ambient temperature for each test so the metric of comparison is change in temperature AKA 'delta T' as opposed to absolute temperature.
mate you'll hate me, I've sold the guts of my computer and bought an intel Skull Canyon and an XPS laptop haha. still have the cables and my case but was waiting to see what happens with AMD and Ryzen / Vega.
Videos a 10/10 can’t wait to start doing videos of my own going to be using a pretty humble set up with good editing and great shots like this man big inspiration!
Haven't found many other youtubers go into such detail or explanation on liquid metal, not to mention as others have stated the production quality. I'm looking to swap the paste to conductonaut on my 2 1080ti Aorus Waterblock cards as soon as they get in. Also subscribed as well! :)
I subbed only because the video was great and you didn't miss a beat in speach. One thing I would have liked to see were the actual degree differences on the screen and maybe not from your word of mouth. I know some ppl would be skeptical but hey I subbed so it means I believed you man. Good job!
Such a shame that he replied to WHDii but not you... your comment was honest and upfront. WHDii 11 months ago Such a shame you've got so few subs man. Your videos are top notch. he loved that comment because it was a compliment. This guy is dodgy as fuck.
Should get some Artic Clean. I know most people just use the rubbing alcohol but it really does make a large difference. Don't need to use 5 coffee filters folded over 100 times and its just a lot quicker. Two-three drops and it dissolves the paste within 3 seconds. Wipe off, couple drops of alcohol and you're done.
If you ever do another like this replace the pads as well. Grizzly (and a few others) make aftermarket thermal pads. It would be interesting to see what you get out of those.
I had a gigabyte 980ti that I did this on. Some liquid metal managed to make its way to the pcb, in the cleaning process a tiny ball of liquid metal managed to get under the Graphics chip (Between the pcb and the gpu) She was a good GPU!
first of all it must not be put on both sides but must be applied in the same way as applying the thermal paste, then you must surround the whole die with the tg shield, an enamel that is applied around the die and then insert the metal while if you got ass you can buy the cpu guard for amd processors
I just ordered this card, very impressive little gtx 1080 with this liquid metal. So this convince me to order one. I will have it by the end of the week
This is a great video man. Keep it up. Even though the video is 10 minutes long, the transitions were great and you kept it moving along. Only thing I would change would be a recording of the actual performance and a side by side comparison. Otherwise, everything is looking great.
I've been thinking about changing the thermal paste on my card. This has encouraged me to stop thinking and actually do it. Might do the CPU whilst I am at it.
That off center badge on the fan would drive me nuts too! Plus it may even unbalance the fan as well. You should coat the SMD's with clear nail polish, then there will be much less risk of the LM making contact. And you can just use nail polish remover to remove it, if need be. Good video! Looks like you and the YES Man get your glasses at the same place!
So i finally did liquid metal on my gtx 1080...if left at at stock it reached 83c with the fans autoing to 85-94% so it sounded like a jet engine. After applying conductonaut, i got an unexpected crazy 14-15c degree drop. I now get 68-69c with autoing fan speed of 44-48% and it can overclock at 40 mhz higher than before. The only thing i worry about is the longevity. I've heard that the LM can last well over a year...but ive also seen and read that it does sometimes dry up eventually and does bond with nickel-copper at a very slow rate.
I also have a predator x34. It got rave reviews almost a couple years ago and lots of youtubers put their money where their mouth was and got one. Even Phillip deFranco has one I figured it must be special so I bit the $1400 bullet I don't think I'll replace thermal paste unless my CPU/GPU is thermal throttling. Good to see that liquid metal doesn't seem to be a meme
I´ve just bought an evga gtx 1080 sc, and now after watch this video, I am going to email to evga support for knowing if I put Thermal grizzly conductonaut may void my warranty. My sub and my like, thanks a lot for the video. Edit: evga answered me: Hi, Thank you for contacting Technical Support. We do not recommend using any Liquid Metal based thermal past on the GPU. We have observed these past damage GPU die eventually loosing warranty. Please feel free to contact for any further information.
It would have been very cool if you would have also included a comparison against something like the Thermal Grizzly Kryonaut. That way we would be able to compare between stock, a good paste ( Kryonaut ) and liquid metal ( Conductonaut ). Great video anyway but i would have been very curious to see the difference between the Kryonaut and the Conductonaut. Maybe something you can do in the future? Cheers
Any update on the Zotac 1080mini with Kryonaut! I just bought a Zotac mini 1080 this blackfriday - wondering If I should change the thermalepaste too - but with non-conductive though
i know Im asking randomly but does someone know of a way to get back into an Instagram account?? I somehow forgot my login password. I would love any assistance you can offer me.
I replaced the stock Thermal Paste on the CPU (no lid for some reason) in my 7 year old laptop with Arctic MX-4 Thermal Paste, and I also cleaned out all of the dust for once, and the fan got more force behind it, and it got a 9 degree improvement on CPU Core 1, a 12 degree improvement on CPU Core 2, and 9 degree improvement on the CPU Cache.
Did this on my Titan X Pascal with an EK Waterblock. Liquid Ultra dropped temps from 69C max to 52C max. That's 17C! Of course I shorted out my card a few times, but hey...it works now!
Great video. One suggestion would have been to note the fan speed (% or RPM) for all tests. If you dropped 10 degrees *and* were able to run a lower fan speed (quieter card), that's doubly impressive and well worth noting. Cheers!
Just gave this a try on the 1080 Ti Mini, but I used some nail polish also just in case. I also use a custom Bios (allowing zero rpm), running the card in OC all the time. The 1080 Ti still get's quite hot, but I got a more stable OC at a few degrees less.
Condutonaut stays liquid and apparently will move. Every other example of applying it to dies (admittedly nlt on de-lidded cpus) i have seen has first put a conformal coating (or nail varnish) over the tiny smd's around the die - just in case, it would only take a tiny blob of the stuff to short out one of those (presumed) capacitors.
I can't figure out why your channel is so small with the quality. Your proved a point that liquid metal helps but with normal article silver I achieved the same drop in temps. With my EVGA 1080. (80 to about 69-70. Would've been cool to see a comparison between stock, new paste, and the liquid metal. Otherwise great video
I just ordered some Gelid GC extreme (better than mx4 from arctic) I want to see if it can fix my max temp of 78 on my ftw 1080, I would like to get to 2100 mhz, the closest I got so far is 2050.
Arctic Silver contains an actual metal compound (silver) so it would probably be just as effective. I've been warned against using AS5 in laptops, but I still use it, and it hasn't harmed anything. Only helped.
My GTX 1080FE from Nvidia with an EK full cover waterblock hits 34c max under load and 1911MHz and 10010Mhz before it boosts no OC. It's not even factory rated to achive those clock speeds. I thinck I might try the LM and see what it can do. Great presentation by the way. Just subbed
I very much enjoyed the video! Sub granted as it is very informative and I also enjoyed your video and audio quality. Well done. As for liquid metal, my 6700K, currently running @4500MHz all cores goes first. It's gonna be delidded and I am looking forward to it. The Aorus GTX 1080 Ti will have a paste replacement to Conductonaut later on. Thank you for the very enlightening video!
Good result, but I still would have covered the small items along the core with clear nail polish or tape, since with "liquid" metal on a card may eventually run the conductonaut onto the small parts. Ounce of prevention costing only a dollar.
Brought my Alienware 17 R4 with a 7820 HK, and GTX 1080 to the following temps Before CPU MAX: 89 C GPU MAX: 92 C (massive thermal throttling) After CPU MAX: 69 C GPU MAX: 75 C Ambient room temperature is around 75 F I'm very impressed with the CPU, but I'm not sure if I should be with the GTX 1080. I've heard people getting temps like that for the gpu with the stock thermal paste (sometimes as low as 69). I know my GPU temps dropped about 15-17 C, but I also feel like that is what the temps should have been in the first place. Should I go in again and add more liquid metal? Thanks.
Hello. I think that's normal. I also put liquid metal on my CPU and GPU of my laptop. Max CPU temp before the liquid metal was 95 and after applying the liquid metal the max was 80, which is 15 drop. But I didn't check the GPU temps before applying the liquid metal so I can't say what's the drop there. Anyway, the laptop began loading faster thanks to the lower temps. ;)
You should have applied an acrylic conformal coating to the resistors on the GPU. You're really taking a big risk by not isolation them from the liquid metal.
would it be better if you applied a layer of clear nail polish around the GPU's shinny heat spreader and especial on those tiny electronic component in which you mentioned on your video "careful not to spill any liquid metal on it". :D great video anyways keep it up you earned my sub. :D
To be compairing temps, any temps you need to use the "scientific approach". The term Delta T (ΔT) is the difference of temperature between two measuring points. Which differ either in time and/or position. So to factor out influence like ambient room temps/Airing of room etc, you use the ΔT. Record the temps = let a hard or software tool autolog them incl time stamp. Record the ambient room temp, preferably in that same log. Then make list : ΔT Component temp (CPU for example) = ambient temp - component temp. Example1 (temps logged for 30 minutes on full system load): Average ambient temp is 32°C Av. cpu temp 59°C, ΔT CPU = 27°C Av. Motherboard temp 50°C, ΔT MOBO = 18°C Av. GPU temp 65°C, ΔT GPU = 40°C Example2 (temps logged for 30 minutes on full system load) for example after applying new paste like in this movie): Average ambient temp is 29°C Av. cpu temp 56°C, ΔT CPU = 27°C Av. Motherboard temp 47°C, ΔT MOBO = 18°C Av. GPU temp 52°C, ΔT GPU = 23°C Now you compare the delta temps (corrected for external factors), these findings are what you want to see.
I just saw another video that stated Intel says they use a paste that holds up much better over many years and that they specifically went with longevity as being of higher importance than absolute conductivity. I suspect the same is true for GPUs. So even though most people see a real improvement in peak temperature values when they delid a chip and use liquid metal, the question is how it will be a few years later. For an enthusiast that replaces their GPU every couple years it may not be as big a deal as for the average Joe who just plugs a card and runs it sometimes for 4-5 years or longer.
FYI, Tesla initially looked at using gold for their electronics connectors because it has higher conductivity and better corrosion resistance so they assumed it would be better even though the rest of the automotive industry used tin. They found out later that gold flakes off and doesn't handle vibration nearly as well as tin does so they went back to tin like the rest of the industry. All I'm suggesting is that if a premium GPU card manufacturers knew that they could run a faster boost speed by using a few pennies of a different product that they probably would. if you look at something like the Morpheus GPU heatsink, it works better than stock GPU air coolers, but you end up with a 3 and a 1/2 slot wide device, which is why you don't see companies producing that.
Today I applied Cool Laboratory Liquid Ultra on my Sapphire R9 390 Nitro with Backplate, but unfortunally my temps did not change a lot. I also used Unigine Heaven 4 to heat up the GPU and I also set a static fan curve with MSI Afterburner (35%), both time I looped for about half an hour. In both cases the graphics card was able to hold its 1040 MHz all the time, no throttling. In stock condition, the max temp measured by HWMonitor was at 85°C, after applying liquid metal the temp just went down to 82°C. Just a very marginal difference. The old layer of Thermal Compound was thin so the Liquid Metal should have a good contact with the cooler and Die. I dont know why that is since the r9 390 produces some heat and the cooler is not that bad either. A better heat transfer should have made a big difference, or am I wrong...?
The part of the heat sink touching the GPU die and liquid metal is copper, it'll be fine because it has to have direct contact with aluminium to react to it.
Nice! TH-cam did a great job recommending this Video! I subbed! I want to put grizzly kryonaut on my Zotac GTX 650 TI (Yes its old I know) and my CPU Cooler
This's more of dried thermal paste vs fresh liquid metal. If you replaced it with top tire thermal paste, it would've been a few degrees higher than liquid metal.
I used thermal Grizzly Conuctonaut with my intel i7 7700K and i have seen a 10°C decline in temprature vs the normale thermal paste. So yeah, that thermal paste is worth its money.
I like how you got the wire caught between the thermal pad and the heat sink at 6:42
randomly came across this video and really liked the time you put into getting a nice spread across the chip. such a simple thing but shows just how much you love and respect your hardware . . . appreciated that as a viewer too, a little effort goes a long way in my book and getting an 88mhz oc on stock is testament to that. thank you for all the effort making this video
Nice production quality for such a small channel. Good job!
nicklong27 thanks mate, the channel is only 5 months old, but that's no reason to not make an effort. Ha ha
Nice grab on the sound clip from paragon !
Just a matter of time he is going big.. Cool name :)
Agreed. Nicely done. I thought it was a really well established channel by the video quality. Then looked and only 2.4k subs so far. Well let me add +1 to that # :)
Thanks guys! :)
I appreciate the support, and I'll try to continue to improve as I go. :)
This educational experiment is so underrated! You just earned my thumbs-up, Sir!
Such a shame you've got so few subs man. Your videos are top notch.
WaRn00b it's only early days mate. Give it time :)
Time given. Disappointed.
I'm genuinely impressed man. Your productions quality is getting really good, you'll be a big techtuber in no time.
+zEus b thanks man! I love getting feedback like this, it makes all the effort worth it, it takes a lot of time to produce a video so I'm glad you enjoyed it!
The question is: Why didn`t the companies made this?? You buy a 700-1000$ GPU and you can`t expect such a simple thing? Thats a shame.
I've been asking that same question for ages now!
I've heard the performance of the liquid metal will degrade a lot over time? Something that's not a real problem with the stock thermal paste? So you might need to redo this every year?
That's what I thought I read somewhere once? No idea if it's really the case.
I read it here on a channel too. Why not to redo every year?
The liquid metal thermal paste is a niche market after all. You cannot expect a large scale factory manufacturer to paste with this expensive high end paste. Plus, when applying the paste, you have to worry about not splitting outside the die as well. All these are not so welcome to mass manufacturing.
saving 2 dollars on paste saves them millions.
Good quality video. Finally someone who loves liquid metas even on gpus. Also +1 for applying it on both sides.
Thank you! Very often when people does the temp tests, they never put the fans at fixed RPM. ..which is just stupid, because then improving the cooling does not perform cooler temps because fans just go lower RPM.. Keep it up!
I googled this because the founders 1080 i placed in my mini pc is not keeping cool. Everywhere online people are saying 'dont do it' 'you'll break it' 'the clamping pressure wont work'. This reassures that people dont know until they try. Thanks for confirming my question!
Just found your channel in a recommendation, this was a great video dude! Glad to see some Aussie making top content
BangDroid wow thanks mate! I appreciate the feedback. I've got another liquid metal video coming this week.... On the evga 1080ti ftw3... ;)
Make sure you catch that!
Super great and surprising test!! Subscribed. I did this to my Asus ROG 1080 OC when I installed my water block 2 days ago. Cant wait to see the results! Thanks for this test!
Regardless of your confidence, use clear enamel nail polish on those resistors, doesn't need a lot, just a careful coating will do just fine.
Great job being thorough (setting the fan curve and getting stock as well as OC info for both). I have subscribed for more and hope the best for you !
Thanks for the video tim , puts a bit of confidence for someone thinking of doing it for the first time.
Just a tip when performing tests regarding temperatures. State the ambient temperature for each test so the metric of comparison is change in temperature AKA 'delta T' as opposed to absolute temperature.
amazing quality content for such a small channel !
+Mr. Marmot thank you man!
good video mate, production value is getting so sweet now. good info too!
Nathan McCoy haha! Thanks mate! How are those cables going?
mate you'll hate me, I've sold the guts of my computer and bought an intel Skull Canyon and an XPS laptop haha. still have the cables and my case but was waiting to see what happens with AMD and Ryzen / Vega.
*gasp*....
Conductonaut is a must have. I re pasted all my cards always with it and it is really a day and night difference.
Its been almost a year so would love to see an update of temps now or in a few months and if any bonding happened in that time.
its super dried up and bad
@@notname3790 nah
@@kowismo ok
Wow now I am very interested in trying this on both my GPU and CPU. Great channel too! Subbed!
Whoa no nail polish or tape?! You are a madman!
Videos a 10/10 can’t wait to start doing videos of my own going to be using a pretty humble set up with good editing and great shots like this man big inspiration!
I can see this channel growing pretty big, nice video.
+Tyson SP well I hope so :)
Haven't found many other youtubers go into such detail or explanation on liquid metal, not to mention as others have stated the production quality. I'm looking to swap the paste to conductonaut on my 2 1080ti Aorus Waterblock cards as soon as they get in. Also subscribed as well! :)
rip stock paste
1987-2012
not for aluminium heatsinks
you really deserve a sub... Really high quality content.. for such a small channel! keep it up:)
That’s cool, good job man - high quality video with good, accurate info
You've done a much better job than all those AAA youtubers... Neat!
I watched 20 seconds and then subscribed - Love this squirrel man already!
+nanodrolone lol
How come you stopped making these awesome videos?
really nice quality content with so much perfection i have some questions about liquid metal
Great video man!!
Great quality video! Instantly subbed
I subbed only because the video was great and you didn't miss a beat in speach. One thing I would have liked to see were the actual degree differences on the screen and maybe not from your word of mouth. I know some ppl would be skeptical but hey I subbed so it means I believed you man.
Good job!
Such a shame that he replied to WHDii but not you... your comment was honest and upfront.
WHDii
11 months ago
Such a shame you've got so few subs man. Your videos are top notch.
he loved that comment because it was a compliment. This guy is dodgy as fuck.
Should get some Artic Clean. I know most people just use the rubbing alcohol but it really does make a large difference. Don't need to use 5 coffee filters folded over 100 times and its just a lot quicker.
Two-three drops and it dissolves the paste within 3 seconds. Wipe off, couple drops of alcohol and you're done.
If you ever do another like this replace the pads as well. Grizzly (and a few others) make aftermarket thermal pads. It would be interesting to see what you get out of those.
I had a gigabyte 980ti that I did this on. Some liquid metal managed to make its way to the pcb, in the cleaning process a tiny ball of liquid metal managed to get under the Graphics chip (Between the pcb and the gpu) She was a good GPU!
first of all it must not be put on both sides but must be applied in the same way as applying the thermal paste, then you must surround the whole die with the tg shield, an enamel that is applied around the die and then insert the metal while if you got ass you can buy the cpu guard for amd processors
I just ordered this card, very impressive little gtx 1080 with this liquid metal. So this convince me to order one. I will have it by the end of the week
Really good production on this.. new sub for sure!
one down side is that liquid metal degrades faster than thermal paste so you'll need to replace it more frequently than you would with paste
Amazing video, all respect and love from Dubai.
+Muhammed Saad thanks! And I appreciate the comment :)
This is a great video man. Keep it up. Even though the video is 10 minutes long, the transitions were great and you kept it moving along.
Only thing I would change would be a recording of the actual performance and a side by side comparison. Otherwise, everything is looking great.
Hey maaaaaaaaaate. Awesome video. Just what I needed to know. Looks like my 1080Ti will be coming apart!
I've been thinking about changing the thermal paste on my card. This has encouraged me to stop thinking and actually do it. Might do the CPU whilst I am at it.
Enjoyed the video and gave it a like
all the best
+Lee Harrison Thank you!
That off center badge on the fan would drive me nuts too! Plus it may even unbalance the fan as well. You should coat the SMD's with clear nail polish, then there will be much less risk of the LM making contact. And you can just use nail polish remover to remove it, if need be. Good video! Looks like you and the YES Man get your glasses at the same place!
you can cover those resistors with normal TIM an leave it there. So its always protected, even if you mess up while addin the cooler later on.
So i finally did liquid metal on my gtx 1080...if left at at stock it reached 83c with the fans autoing to 85-94% so it sounded like a jet engine. After applying conductonaut, i got an unexpected crazy 14-15c degree drop. I now get 68-69c with autoing fan speed of 44-48% and it can overclock at 40 mhz higher than before.
The only thing i worry about is the longevity. I've heard that the LM can last well over a year...but ive also seen and read that it does sometimes dry up eventually and does bond with nickel-copper at a very slow rate.
very professional! well done, i like the style of review too, thanks! just subscribed
+H. Roku thank you! :)
This is pro stuff you are doing! Got yourself a sub today!
Thank you!!! :-)
Great channel mate, I liked the mini montage. Subscribed already. :) Hope you make more great videos like this.
Great video! I used this liquid metal before on my gaming laptop with two GTX 980m and got similar results(about 10c drop)
Now I'm using custom water loop on my 980Ti and have it runs at low 40s with OC
You should post an update video on how the LM has affected the heatsink or what not. Please ADD A GRAPH makes the results much more simpler!
I also have a predator x34. It got rave reviews almost a couple years ago and lots of youtubers put their money where their mouth was and got one. Even Phillip deFranco has one I figured it must be special so I bit the $1400 bullet
I don't think I'll replace thermal paste unless my CPU/GPU is thermal throttling. Good to see that liquid metal doesn't seem to be a meme
I´ve just bought an evga gtx 1080 sc, and now after watch this video, I am going to email to evga support for knowing if I put Thermal grizzly conductonaut may void my warranty. My sub and my like, thanks a lot for the video.
Edit: evga answered me:
Hi,
Thank you for contacting Technical Support.
We do not recommend using any Liquid Metal based thermal past on the GPU.
We have observed these past damage GPU die eventually loosing warranty.
Please feel free to contact for any further information.
Great video man! You just got another subsriber! Looking forward to more vids!
+Levi Cantrill thanks Levi, that's kind of you to say and thanks for subscribing!
I'll be purchasing a Thermal Grizzly Hydronaut for myself soon. Hope it really is as good as everyone says!
A year has passed, how did it go?
Tip: if you're going to turn your head during your 'piece to cam', put your lapel mic on the side that your head is going to turn to.
Fantastic content and quality bro! Definitely Earned my sub!!
It would have been very cool if you would have also included a comparison against something like the Thermal Grizzly Kryonaut. That way we would be able to compare between stock, a good paste ( Kryonaut ) and liquid metal ( Conductonaut ). Great video anyway but i would have been very curious to see the difference between the Kryonaut and the Conductonaut. Maybe something you can do in the future? Cheers
+clockner I've got some of that on the way :)
*+Upgrade Addiction*
Nice, really looking forward for these.
Any update on the Zotac 1080mini with Kryonaut! I just bought a Zotac mini 1080 this blackfriday - wondering If I should change the thermalepaste too - but with non-conductive though
i know Im asking randomly but does someone know of a way to get back into an Instagram account??
I somehow forgot my login password. I would love any assistance you can offer me.
@Jadiel Hassan Instablaster :)
In order to avoid short circuit it was necessary to varnish the elements.
I replaced the stock Thermal Paste on the CPU (no lid for some reason) in my 7 year old laptop with Arctic MX-4 Thermal Paste, and I also cleaned out all of the dust for once, and the fan got more force behind it, and it got a 9 degree improvement on CPU Core 1, a 12 degree improvement on CPU Core 2, and 9 degree improvement on the CPU Cache.
Subbed. Very informative and well put together.
Thank you!
If your still running this mini card I suggest voltage curve threw afterburner as it would be the biggest change
good job man! really good video
AMDSITHLORD thanks, I'm glad you enjoyed it. :)
Did this on my Titan X Pascal with an EK Waterblock. Liquid Ultra dropped temps from 69C max to 52C max. That's 17C! Of course I shorted out my card a few times, but hey...it works now!
Great video. One suggestion would have been to note the fan speed (% or RPM) for all tests. If you dropped 10 degrees *and* were able to run a lower fan speed (quieter card), that's doubly impressive and well worth noting.
Cheers!
He clearly said he locked fan speed at 65 like four times
finally someone who shares the same OCD with the Fan Sticker xD
Nice video. Plenty of effort in the editing and such! :)
Just gave this a try on the 1080 Ti Mini, but I used some nail polish also just in case. I also use a custom Bios (allowing zero rpm), running the card in OC all the time. The 1080 Ti still get's quite hot, but I got a more stable OC at a few degrees less.
Condutonaut stays liquid and apparently will move. Every other example of applying it to dies (admittedly nlt on de-lidded cpus) i have seen has first put a conformal coating (or nail varnish) over the tiny smd's around the die - just in case, it would only take a tiny blob of the stuff to short out one of those (presumed) capacitors.
I can't figure out why your channel is so small with the quality. Your proved a point that liquid metal helps but with normal article silver I achieved the same drop in temps. With my EVGA 1080. (80 to about 69-70. Would've been cool to see a comparison between stock, new paste, and the liquid metal. Otherwise great video
+Saber Slayer it's only very young (the channel)
Yeah great idea. I'll keep that in mind!
I just ordered some Gelid GC extreme (better than mx4 from arctic) I want to see if it can fix my max temp of 78 on my ftw 1080, I would like to get to 2100 mhz, the closest I got so far is 2050.
@@krystian3797 Give Noctua NT-H1 a try. Dropped 7-9 on average from my CPU. Waiting another couple days to put new TIM on my Strix 1080.
Arctic Silver contains an actual metal compound (silver) so it would probably be just as effective. I've been warned against using AS5 in laptops, but I still use it, and it hasn't harmed anything. Only helped.
My GTX 1080FE from Nvidia with an EK full cover waterblock hits 34c max under load and 1911MHz and 10010Mhz before it boosts no OC. It's not even factory rated to achive those clock speeds. I thinck I might try the LM and see what it can do. Great presentation by the way. Just subbed
you confirm my choice about this product.
good video and simple keep this way :)
thx
Nice to see someone finally using the right amount of Liquid Metal. I want to use some on my Ref 1080ti but it looks like the heatsink is aluminium.
The heat fins are aluminum but the block touching the die is copper
I very much enjoyed the video! Sub granted as it is very informative and I also enjoyed your video and audio quality. Well done. As for liquid metal, my 6700K, currently running @4500MHz all cores goes first. It's gonna be delidded and I am looking forward to it. The Aorus GTX 1080 Ti will have a paste replacement to Conductonaut later on. Thank you for the very enlightening video!
Good result, but I still would have covered the small items along the core
with clear nail polish or tape,
since with "liquid" metal on a card
may eventually run the conductonaut onto the small parts.
Ounce of prevention costing only a dollar.
You could do that out of paranoia but not necessary if you use a thin layer of liquid metal.
Brought my Alienware 17 R4 with a 7820 HK, and GTX 1080 to the following temps
Before
CPU MAX: 89 C
GPU MAX: 92 C (massive thermal throttling)
After
CPU MAX: 69 C
GPU MAX: 75 C
Ambient room temperature is around 75 F
I'm very impressed with the CPU, but I'm not sure if I should be with the GTX 1080. I've heard people getting temps like that for the gpu with the stock thermal paste (sometimes as low as 69).
I know my GPU temps dropped about 15-17 C, but I also feel like that is what the temps should have been in the first place.
Should I go in again and add more liquid metal?
Thanks.
Hello. I think that's normal. I also put liquid metal on my CPU and GPU of my laptop. Max CPU temp before the liquid metal was 95 and after applying the liquid metal the max was 80, which is 15 drop. But I didn't check the GPU temps before applying the liquid metal so I can't say what's the drop there. Anyway, the laptop began loading faster thanks to the lower temps. ;)
Bro, just use kelvin, everyone will understand.
Very helpful video !!
Good job keep it up
You should have applied an acrylic conformal coating to the resistors on the GPU. You're really taking a big risk by not isolation them from the liquid metal.
You gonna burn your eyes out with that monitor that close :)
would it be better if you applied a layer of clear nail polish around the GPU's shinny heat spreader and especial on those tiny electronic component in which you mentioned on your video "careful not to spill any liquid metal on it". :D great video anyways keep it up you earned my sub. :D
we miss your videos plz come back
nice video man, it helped me a lot!
To be compairing temps, any temps you need to use the "scientific approach". The term Delta T (ΔT) is the difference of temperature between two measuring points. Which differ either in time and/or position. So to factor out influence like ambient room temps/Airing of room etc, you use the ΔT. Record the temps = let a hard or software tool autolog them incl time stamp. Record the ambient room temp, preferably in that same log. Then make list : ΔT Component temp (CPU for example) = ambient temp - component temp.
Example1 (temps logged for 30 minutes on full system load):
Average ambient temp is 32°C
Av. cpu temp 59°C, ΔT CPU = 27°C
Av. Motherboard temp 50°C, ΔT MOBO = 18°C
Av. GPU temp 65°C, ΔT GPU = 40°C
Example2 (temps logged for 30 minutes on full system load) for example after applying new paste like in this movie):
Average ambient temp is 29°C
Av. cpu temp 56°C, ΔT CPU = 27°C
Av. Motherboard temp 47°C, ΔT MOBO = 18°C
Av. GPU temp 52°C, ΔT GPU = 23°C
Now you compare the delta temps (corrected for external factors), these findings are what you want to see.
I just saw another video that stated Intel says they use a paste that holds up much better over many years and that they specifically went with longevity as being of higher importance than absolute conductivity. I suspect the same is true for GPUs. So even though most people see a real improvement in peak temperature values when they delid a chip and use liquid metal, the question is how it will be a few years later. For an enthusiast that replaces their GPU every couple years it may not be as big a deal as for the average Joe who just plugs a card and runs it sometimes for 4-5 years or longer.
FYI, Tesla initially looked at using gold for their electronics connectors because it has higher conductivity and better corrosion resistance so they assumed it would be better even though the rest of the automotive industry used tin. They found out later that gold flakes off and doesn't handle vibration nearly as well as tin does so they went back to tin like the rest of the industry. All I'm suggesting is that if a premium GPU card manufacturers knew that they could run a faster boost speed by using a few pennies of a different product that they probably would.
if you look at something like the Morpheus GPU heatsink, it works better than stock GPU air coolers, but you end up with a 3 and a 1/2 slot wide device, which is why you don't see companies producing that.
Today I applied Cool Laboratory Liquid Ultra on my Sapphire R9 390 Nitro with Backplate, but unfortunally my temps did not change a lot. I also used Unigine Heaven 4 to heat up the GPU and I also set a static fan curve with MSI Afterburner (35%), both time I looped for about half an hour. In both cases the graphics card was able to hold its 1040 MHz all the time, no throttling. In stock condition, the max temp measured by HWMonitor was at 85°C, after applying liquid metal the temp just went down to 82°C. Just a very marginal difference. The old layer of Thermal Compound was thin so the Liquid Metal should have a good contact with the cooler and Die. I dont know why that is since the r9 390 produces some heat and the cooler is not that bad either. A better heat transfer should have made a big difference, or am I wrong...?
I did the same thing on my Zotact gtx 1080 mini that I have in my NODE 202 HTPC build. Much better temps and performance.
cool test, did it on the hot HD 6970 huge temp drop of 20c. what i missed was a temp test 24hours later when the liquid was cured...
Hope this helps and works the same on both my RTX 2080ti cards!
WOW. I have a Zotac GTX 970 mini... I'm gonna have to do this.
You have a fan cable stuck between one of tbe VRAMs and its thermal pad. Look at 6:42.
10 Degree is literally a huge drop changing only the thermal pest alone!
Great video mate. Can't wait to give this a go myself ;)
+TechPlayground it's hugely beneficial! Let's do one together on Friday... Perhaps a 1080ti? ;)
Upgrade Addiction Sounds good to me :)
I hope there's no aluminum in your heatsink, if that gallium liquid metal touches it ....won't be pretty
jboydgolfer1 That heatsink is probably 90% aluminum. I’ve never seen people use Liquid Metal on off the shelf GPU’s for a reason.
The part of the heat sink touching the GPU die and liquid metal is copper, it'll be fine because it has to have direct contact with aluminium to react to it.
6:42 ground wire is between pad and vrm/ram
Nice work. Just subbed.
Great content and very well put together video!
But for long term use doesn't conductanaut and other liquid metals cause oxidisation on naked copper?
Nice! TH-cam did a great job recommending this Video! I subbed!
I want to put grizzly kryonaut on my Zotac GTX 650 TI (Yes its old I know) and my CPU Cooler
This's more of dried thermal paste vs fresh liquid metal. If you replaced it with top tire thermal paste, it would've been a few degrees higher than liquid metal.
I used thermal Grizzly Conuctonaut with my intel i7 7700K and i have seen a 10°C decline in temprature vs the normale thermal paste.
So yeah, that thermal paste is worth its money.