Yes, he said that on the WAN show. They did this so people could purchase the pads from a 'known vendor' instead of buying from some random shops that may or may not sell the real Honeywell pads. And even LMG had difficulties to obtain it, it only worked out because of a viewer who works at a redistributor for Honeywell.
@@Gift0rthat and Honeywell offers them in large sheets. For multiple applications. Not just a small amount needed for one install. So it’s way more expensive than just getting a small amount.
@@PTRRanger951yes and no buying a large amount is cheaper if your using alot but this shit you cut to size and it will last years and years but for a normal system you don't need much
@@0-B1which is what I said. No one is doing install after a needs sheets of. It’s cost effective in bulk. But to pay for sheets and only need a small square is a waste of money. Which is why LTT offers it. It’s cheaper to people to get it that way, even if it seems pricy. Because any user that is buying it from LTT is using it for a build and won’t ever touch the computer again until years later. If they plan on changing computers every year, or every few months they would be buying sheets, not a kit. The kit from LTT is only designed for single builder that is building one computer and plan on it touching it for years.
Some correction, you dont cook it to phase change but to spread the material around for better contact. Phase change happens at 45C and not after "cooking". PTM7950 works the best in very high pressure mountings.
@@MediaChanelle No that is incorrect as well because phase change happens at 45C. The “mysterious” burn-in process of PTM involves repeated heating and cooling of the material to achieve an optimal thermal bond. During burn-in, the pad reaches its phase change temperature, wetting the surfaces and achieving a minimum bondline thickness (BLT). A key aspect of the burn-in process is the change in the thermal interface resistances. At the beginning of the burn-in (i.e. before the first phase change), the resistances are typically significantly higher, as the PTM is not yet (completely) melted and wetted. During the burn-in process, the interface resistances decrease significantly as the material fills the microscopic gaps and unevenness of the contact surfaces and thus reduces the thermal barrier. Nobody reads the datasheet before commenting?
@@MediaChanelle that's completely backwards. It's already a solid; increasing the heat phase changes it to liquid (it melts it), and cooling will solidify it.
Yes but after several cooling cycles it will turn much more of a brittle material. Perminabtly. The cycling allows it to penetrate better, the cooling part lets it become a a more brittle and less liquid like material. This is why you can’t recycle PTM.
You're repeatedly suggesting that "cooking" the pad is going to turn it solid, except the pad is already visibly solid. Increasing the temperature induces the phase change to a liquid, not the other way around. Linus demonstrates this pretty clearly in his video when he heats up the surface it's attached to with a blowtorch. The phase change is occurring every time the TIM heats up, it's just that cycling it allows it to fill in air gaps more effectively over time.
Once the PTM does several heat cycles it turned it a true solid. It becomes brittle and unmeleable. Hearing it moves in to the cracks. Cooling it allows it to become a true brittle solid material. This is why removing the cooling solution after will ruin it, unlike paste. I recommend trying the stuff yourself. Run some heat and cool cycles the take off the cooler. You will find a drastically changed material inside
They've addressed the USD pricing on their site many times, something like 80% of their sales are to the US, so having to deal with multiple currencies isn't something they're willing to deal with. So for them, it makes more sense to only do USD.
It’s not just about where their sales come from, it’s about how complicated their accounting is even with one currency. Linus said at one point that when they tried to approach an external company about doing accounting for them, they looked at the range of things they’re doing and went “no thanks.” Which is why even at their size they need a dedicated accounting department.
I have nightmares of peeling that 2nd layer of plastic off. Putting the cut & measured sheet into the freezer for an hour or two helps it not stick as bad, but it's still a PITA.
I encountered the same issue and I followed these steps but encountered an issue: 1. Placed the CPU in the motherboard socket. 2. Applied the thermal pad on the CPU while keeping the plastic cover on the thermal pad. 3. Installed the heatsink and secured it by screwing it down. 4. Removed the heatsink. Afterward, the plastic cover on the thermal pad became easier to remove because the thermal pad stuck to the CPU.
Yup, what I found to help me out a lot is to insert a exacto knife in between the very edge of the plastic and PTM and slowly and carefully separate it and once enough is separated I hold the ptm edge with the side of the exacto knife and peel the rest for a good application and leaving the ptm pad intact, it's what I did on my GPU core.
After going back to air cooling on my 4090, I had a ton of issues with cpu and hotspot getting high after 2-3 months of apply paste. I did that twice before looking at alternatives. The PTM solution seemed like the best route, but was hard to find a reputable source. LTT ended up being the cheapest for the huge sheet.
Applied it to my reference 6950xt. I do have an undervolt to 1120mV and 2500MHz max, previous temps were 85C with 102C hotspot. After two weeks the temps are now 72C and 85C hotspot. When mounting on a GPU, re-tighten the screws after a few minutes of having it on. I'm thinking about re-pasting with this again, but using shims to improve the pressure, it could potentially be a 4C improvement which is pretty solid for summer
What were your initial temps like during the first few tests? I'm not seeing an improvement after running cinebench/3dmark around 10 times with a cooldown period. Temps used to be at 96-100 CPU - 85-86 GPU with a 91.8 hotspot and they are pretty much the same after using the pads.
@@pieuwpieuw6029 Sorry for the late reply. Thankfully I recorded the temps: 82 - 96 ºC first test same tuning. Delta down to almost 15 degrees vs the 25 degree difference before PTM7950 Ambient temperature has a lot to do with it though; we've had 40ºC consistently during the summer. You should have noticed a difference by now but make sure you have undervolted the GPU: the stock cooler really dislikes having the GPU set at 1.2v. A mild undervolt to 1170mv should take the edge off without the need to limit clock speeds. Also make sure to tweak the fan curve; should look logarithmic. Mine looks like: 35ºC 30%, 50ºC 50%, 65ºC 65%, 80ºC 75%, 90ºC 80%. Fan speeds higher than 75% make no difference If you do want to limit clock speeds, clocking down to 2400mhz and running 1130v should bring the temps and power down aggressively while not dropping a performance tier. As of today I've managed to see a stable 240w draw clocked down to 2400mhz 1130mv with the fan curve stated, edge and hostpot around 65 and 78 ºC now that ambient is lower
I bought an 80X80X0.2mm sheet from Ali Express for £3.85 and it works just as well as the LTT and Honeywell stuff for a fraction of the cost and you even get a bunch of tools with it too
I bought two big pads, and it lasts a lot longer than I expected. I repasted an AM5, GPU, Gaming laptop (CPU and GPU), an old laptop with a Intel U processor (barely got hot enough for phase change), Nintendo switch, mini-PC passively cooled switch, and an Xbox One S and I only used half of one... My graphics card (Sapphire) had been RMA replaced, and so I watched the new one's thermals and it was hitting ~120°C 💀. Decided voiding the warranty was worth it if it would keep me from having another 3 month $300 shipping RMA. PTM brought the temps down to ~60°C under furmark, the stuff that was on there was already "Pentium 4 dell, Never changed" hard. On my Dell G5, it works so well that I discovered that the Dell firmware ramps the fan to 100% before checking the temps which causes it to "sneeze" the fans as the temps are lower than it expects. The Dell heatsink has tons or toolmarks that need good TIM to fill in, this works great. Switch doesn't ramp up fan anymore, Xbox runs a lot cooler to the touch, it just works really really well.
I’ve been using mine and can honestly say my old computer has not messed up lol it’s also a desktop , so this is actually advantageous to use this in a desktop. I bought mine from amazon with a relatively large sheet for 16-20$ U.S
The best thing about it is that it does not seem to dry out, it does not lose performance over time. Thermal paste dries out and you need to replace it every few years if you want to keep the same performance. I don't personally know anyone who replaces thermal paste except me, so for 99% of users, these thermal pads should be way better.
Great vid! I would love to see a comparison with the Thermalright Heilos phase change pads to see if it's just rebranded PTM7950 or if it's a different material.
@@SuperBROKEN81 Good morning! Do you have any specific evidence as to why you believe it's different? I've been looking for quite a while and no one has anything concrete. I've been using the Heilos pads as they seem to work within margin of error of the PTM7950 but it would be nice to have some concrete info as to whether or not it's just a rebrand.
PTM isn't cooked to 'phase change' at all. It changes phases every single time it is heated and again once it cools. The thermal cycling is causing that PTM to liquify and settle into the microvalleys of the heatspreader and heatsink surfaces. Repeated cycling ensures it is settled as much as it can be, showing the maximum performance. Important for a performance comparison, but not so much for an end user, as normal use will provide these thermal cycles over the course of a few days. This is actually true for a lot of thermal greases too, where many users will notice 1-2 degrees drop once the compound has had enough thermal cycles to settle.
What I do is measure my CPU with a digital caliper. Then draw those dimensions in a vector program as a square. Then make the square like 2mm larger around. Print it out and cut the square out. Then hold the square over the PTM and cut it out. That way you get a slight overhang as I know it can mess up some when you start to peel it.
really, Laptops by default need Liquid Metal for their incredibly tight space to fit a cooler in there. but for those who didnt pay extra for liquid metal pre-applied on their space-saving gaming machines... and dosent have the hands of a surgeon to apply liquid metal, PTM7950 is the best option.
You could get this stuff for cheaper elsewhere but the quality may very. The best thing about getting it from him is quality control and his credibility is on the line.
PTM is better, but won’t last so long. But to be honest, most people (pc gamers) change their cpu in that time, so it’s fine to use ptm and not the sheet
@@checkoff8281 are you sure? according to the Linus video PTM7950 is designed to last years and years, while thermal grizzly has a reputation for only lasting 1-2 years at most.
I just bought some mystery PTM on ebay from a seller in the US for $13. I got a r5 3600 with a NH-L9x65 and feeling adventurous. Will report back if I can remember this comment so bump it in a few days if anyone's curious
I bought one from some shady sellers in shopee ( filipino version of taobao) and it surprisingly immediately dropped my 90c 12500h to 72c. Im honestly flabbergasted. Considering itll drop further in the future, im really optimistic about whats circulating in the online spaces.
@@xCheatah He can't answer a few days later, the fake PTM from ebay has caused @qbrt4050 some nasty untold distater that is preventing him from reporting back.
Thermal grizzly released a PTM pad for $13 on amazon, haven’t seen any video reviews on it but the comments on amazon make it seem like regular thermal paste with the benefits of regular ptm7950. Im probably gonna chuck it in an old build since i’d rather not have to re-paste anything. The application seems extremely easy so anyone complaining about it must have chubby fingers. All this to say, would you be willing to compare the two? If I really think about it they’re probably the same so it might be a waste of time but hey 🤷♂️
I'd love to see this on a laptop where the gpu wattage is increased. I know it's a pretty niche case. Most laptops don't have the thermal solution to backup more wattage. Most are better returning a low wattage variant and purchasing a full wattage variant if wanting more performance. 4000 series and older Nvidia vbios flashing is possible now. You flash to a same chip, same vendor, higher wattage vbios. This risks bricking or needing to have a tool to flash the vbios chip to recover. Some lower wattage Asus variants and cheap MSI variants get a good boost. 3060, 3070, 3080, 4080, 4090 variants have given a bump. RX 5000/6000 series are even better because you don't need to mess with vbios. Use morepowertool for RX 5000 and 6000 series. great for overclocking and undervolting as the power play table wasn't locked. RX 7000 series laptops are more rare and more locked down. Asus, MSI (even Lenovo thinkpads) come to mind as some success stories. You are going to get a bit more performance before hitting thermal throttling. They have a large product range with some that could take a bit more power. Some mod the heatsink to get around this. www.reddit.com/r/MSILaptops/comments/16tm1w9/heatsink_modification_to_fix_cpu_and_gpu_heat/
I repasted my asus vivobook s14 i7-12700h cpu machine. I tried to use Thermal Grizzly Kryonaut Extreme first and that was pumped out about in 3 weeks and temperatured went higher. I found only some thin liquid on the core when I disassembled it. Later I tried some chinese PTM and it is still ok after using it for a year. The original paste in the machnie was also some pretty thick stuff, not sure if it was also PTM or other stuff.
PTM is incredible. I’ve been promoting it for quite a while and using it long term. It has serious staying power. The laptop currently has the LTT variant in it, so I can see if it lasts
@@TechGuyBeaui applied it in my asus g14 (2023), which I bought a year ago, and it still performs the same. The Liquid Metal had moved off of the die, and heat was out of control. I also tried arctic mx6, but the performance wasn’t good enough. Replaced with PTM after a couple days, and it’s nearly as good as when the LM was new. Honeywell stuff btw.
I replaced the 12 month old mx4 on my water cooled 4090 with PTM7950 and dropped 8c, I’m changing out my cpu and motherboard and can’t wait to get the PTM onto my cpu.
my legion 5 using gelid gc extreme while gaming could reach 90c. then i replace it with thermalright heilos ptm, now it's drop to 70c while gaming, crazy peformance. i think any PTM with reputable brand on the market right now is the same, gey color with 8.5 W/m-K. i buy thermalright heilos the cheapest one for $5 and still has half to spare for next application.
i'd argue if you have a higher end cooling solution it makes even more sence to use PTM as yes your cooling itself can crush anything you throw at it but the main limiting factor is eventually always the thermal tranfer between the CPU/GPU and the cooling solution itself (it's why direct die cooling on the CPU provideds such a massive boost, liquid metal does a lot of heavy lifting but there's also no more transfer to the heatspreader and from the headspreader to the block/cooler it's more direct with less roadbumps to get there). next build i'm probably switching to PTM, currently have a carbon flap (yea it's not as good as really good thermal paste but it's still quite good). granted you showed in on a laptop which is more of a direct die scenario but those clock difference are quite impressive.
Not a bad argument. In the end PTM on everything. I bought three boxes of this stuff and already had a ton from Ebuy7. I covered myself in PTM so that now I run cooler too. :p Might as well use it up
LTT chose US dollars because the majority of their customers are USA residents. I'm sure it's a bit more complicated for taxes on their end as well with conversion and the value of fiat currency constantly changing.
The question I have, and I am curious about this stuff now, is can you change it. You say it turns into a "solid" and the word baking it on, you can still change these at later dates yes? or aswhen needed or even if you change your cooler on the cpu?
Well you can just clean it off after. Once it does temperature cycling it becomes hard and brittle. This is fine but if you remove the cooler it will make is basically useless. You then clean it off and reapply
@@TechGuyBeau awesome thank you for the response, I may have to try this stuff on my new laptop and my desktop. Been using kryonaut for years, but this looks interesting.
This is just wrong. It starts out as a gel-like consistent at room temperature. When put into the freezer it will become more solid, this should be done for cutting as it makes it much easier. Above 45C it will become more like a liquid and conduct heat very well. After it cools down from 45C, it will go back to its gel-like state. It does not harden with heat, quite the opposite. The "cooking" (which is just heat-cycling and repeatedly changing the phase of the pad) turns the pad from solid to liquid and back again. Doing this repeatedly allows the PTM to get into all the microscopic surface inconsistencies, as a gel-like solid (when it's cold) it has great difficulty getting into such small imperfections. In its liquid state (hot) it is much easier. Heat-cycling just ensures the best contact between the IHS, pad and cooler surfaces.
@@Splomf that is not correct.. It starts as a semi solid. You heat it, it liquifies and fills the gaps, spills out when not needed. After a long cooldown it turns into a brittle substance that cannot be re-used if taken off I have worked with this stuff for years.
Great video! I've been using this stuff on my GPU for about a month now and temps dropped by 2/3C between day 1 and day 30 of using PTM7950, safe to assume that's due to heat cycling like you talked about at @9:00. My GPU from ASUS has terrible mounting pressure which accelerated the pump out effect that normal thermal pastes have to deal with. This stuff really seems to shine in direct die/silicon applications like on GPUs and Laptops, I wonder how good PTM7950 is vs traditional paste on a CPU, maybe something to test for another video?
You mean a. Desktop cpu? It does help but it depends on the chip. For example I run a 7950x at 105w so it runs so dang cool anyways. But if you have a thermally constrained CPU it’s worth trying for sure
It’s amazing! I’m a person who’s 100% willing to pay for piece of mind. Just sit it and forget it! With the added benefit that when you do remember, you’ll be amazed at how well it’s running. Just dear god, think about having to take this off! You gotta remember to play a cpu heavy game then rush to take it off before it cools off so it doesn’t get physically and emotionally attached to the cooler. I didn’t know this the first time around so I said f it, it’s only a 10100kf. I’ll just buy another. “ and got to flat head and twisted. Popped off angrily but off non the less. Still happy buying more til something better comes out. I really wish that liquid diamond stuff actually worked. Still have some
Have to admit I've seen this stuff on moddiy and was always curious but thought it was more a gimmick than anything. Rebuilding some of my homelab servers, might give this a try even though those CPU's don't run very hot. If it works out maybe take apart my main machine and install it there too. Maybe a little fiddly to install but none of the possible problems you get with LM. Thanks for the video!
@@TechGuyBeau I'm like you, nearsighted. No problems with stuff as long as it's relatively close so that shouldn't be a problem. Kind of wishing I had seen this video a few days sooner now, just ordered a case for my new proxmox server and just hoping I don't have to wait on this to start building it.
Stupid question, what if you double up on it? Like, put two squares on top of each other before locking on your AIO or Air Cooler? Would the results be the same? Worse?
Excellent video. I don't understand the possible margin of error. The Linus PTM was in fact better, but you're saying it isn't a significant difference from the Honeywell such that it could be margin of error? I guess what I'm asking is, if it wasn't margin of error, what would the Cinebench figures look like? How different do those numbers have to be to say for sure that it's not margin of error?
There is no predefined margin of error, as it simply doesn’t exist. But if we’re looking at a few percentage points at most say two or three percentage difference, it’s not significant in any form of scientific test. To actually properly find significance I would have to rerun this test like 40 times with each compound. And look at the significant difference Realistically though on this test that I did it’s so minimal that it’s basically margin. I hope that makes sense
I recently repasted my 3080 Ti with Corsair branded paste (ran out of NTH-01) and noticed my fan hub, which runs my case fans, on my custom fan curve was locked at 100%. After applying the PTM7950 today my fan speeds dropped to 55% at the same temperature. Blew my expectations away, my PC is significantly quieter at the same load.
3 หลายเดือนก่อน +9
LTT source PTM from Honeywell you know that right It talked about WAN show and anoncement
The way I recommend thermal transfer options is based off who’s asking experience, and willingness to learn. I use water blocks for direct die for CPU and same on GPU w/Liquid Metal. Glad you did this comparison so I’ve got another option Vs what I use on my personal builds.
Cool comparison! so the ebay/ aliexpress stuff is "legit" :D fixed the overheating in my old msi laptop. Do you have a asus laptop with liquid metal installed? Would be nice to know how well ptm works in a laptop with factory lm?
Generally speaking if you remove liquid metal and apply PTM, you will decrease the performance of the cooling of course if the liquid metal is shoddily applied from the company then really anything will do better than that I have used many laptops with liquid metal, typically I just leave them as is, unless there’s some very serious issues with the thermal performance which would suggest sending it back for repair to the company
I just mentioned in another comment before seeing yours. I have an asus g14 2023, which I bought about a year ago. It worked fine for a couple months and then terrible heat. I tried arctic mx6 first, but it was not as good. Switched to PTM, and it’s only a few degrees warmer than with the LM. Some of the LM had moved off of the die, probably because I have to pack it up and immediately put it in my backpack all the time.
Great video, but the difference in actual performance when you stated you gamed it seems to be the same. When you go through the BG's temps and details it literally shows that Noctua had lower CPU avg watts with more clocks. Granted you get better temps, but I would expect a higher average clocks value considering the system is able to inject more power into it... GPU temperature difference is only 1,4ºC higher (which to me seems close to negligible). The memory junction is actually LOWER on the Noctua side and the hot spot temperature differential is 2ºC. Not enough for me to call PTM 7950 revolutionary. The last clocks were too similar for me to cast a vote on either of them. And then in the end, the "poorer" product (paste) actually gave you better framerate. Negligible, I know, but for a better product to post worse results it does not make PTM's case for me. At this point the only thing that might be left is longevity. And this is something I would rather like to find out. To see the differential on the performance with time degradation put into it. Problem is that this is a case where TIM easily lasts 2/3 years without steep loss of performance so I understand it would be demanding to take on such challenge.
I wish, LTT could sell them in smaller sizes for a little cheaper. Google tells me that 7950 has an estimated shelf-life of 12 months. So if I needed to build 2-3 PC's as once-off, it'd be a shame to just dispose/let it expire. But I guess the re-packaging cost wouldn't be worth it.
I'm pretty sure the PTM has a very long shelf life if any at all considering that it's intended for industrial use and they buy this stuff in kilograms. As long as it's air tight and not in rough temperature it shouldn't degrade.
Have you tried peeling the transparent sheet first, put it on the CPU. And then peeling the white sheet? Not sure if that works any better compared to peeling the white sheet first?
The Pad doesn't turn into a Solid after multiple Thermal Cycles... The Pad you are holding is it being a Solid, when it reaches a temperature of 45C then it goes into a Liquid state.
Rumor has it that these work exceptionally well on waterblocks that cool cpu and gpu at the same time. You could even cool the wrong gpu 😂😂😂 don't buy from bullies.
It lasts for an extremely long time. But if you remove the cooler ,it basically destroys the ptm. So after apply you basically leave it alone, if you change coolers you need to add more ptm
I might buy some to replace the corsair thermal paste that came with my aio. The 7900x doesn't run that hot lol, at least compared to that 13700k I got rid of but temps could be better
i have gotten used to replacing my cpu paste every 2-3 years, how long would one of these pads stay effective, on average?? and would this work as a replacement for gpu memory and vrm thermal pads??
You have a fundamental misunderstanding of how this phase-change material works. At room temperature it is a gel-like solid. If put into the freezer, it will become more solid (do this when you cut and apply it, it will make your life easier). When it is heated to ir above 45C, it becomes a liquid. It does not dry out or become solid unless FROZEN. Heat will always liquify it.
No…. It simply becomes much dryer after being properly heat cycles. Prior to its first application is a near gel like state. After heating (and then cooling), it becomes a brittle material. This is why once applied, if you remove a heat sink, the PTM is rendered useless and must be reapplied
I need this for my omen laptop with i7 9750h and 1660ti. This generation is known for running way hot. I have a custom power curve in msi afterburner but its still not enough. Im sure this will help at idle too since im not really gaming much. Its even too hot to use in lap on ecomode on idle lol
See if CPU & GPU share heatpipes, your CPU might needlessly boost & eat the thermal headroom of GPU. There used to be a tool called ThrottleStop that let you tweak boost behavoir. Basically find out if your usage is GPU bound, then stop the CPU from going into it's higher power states, so your GPU can use that extra headroom. Also helps temperature-frequency oscilation. Helped me loads on a laptop with underpowered heat dissipation capacity at 100% CPU&GPU usage.
That's brave, testing thermal paste/pads on a notebook PC. I wouldn't dare to 😎 Edit: I meant the risk of damaging and wear something down by doing this. My comment wasn't aimed at warranty.
Ordered some of the "Official" honeywell ptm, used it on my gaming laptop. After a month and a half, my temps went up, (idle, and gaming), my idle temps went from 48°c to 55°c, and gaming, went from 75°c to 98°c. I took my laptop down, to see what was going on, and found, the ptm had gotten dried out, and hard, and it had boiled out from between my heatsink, and cpu, and my cpu now had a small burn mark in the center. Not using that stuff again!😒
PTM drying isn’t a thing. Perhaps you had some knockoff PTM? PTM is supposed to change from a semi liquid to a liquid then to a solid and stay very dry.
The stuff you hold into the camera at 02:06 min. is a fake product. The original Honeywell PTM7950 is always 0,25 mm thick, not 0,2 mm. You can find that info also on the Honeywell homepage.
think that was clear in the video, but its important to test things to make sure they are legit. Any huge business like LTT can claim something, doesn't mean its true
Lmg/Ltt group has a well established relationship with many high quality foreign manufacturers so getting this ptm direct from manufacturer is not a far reach for their acquisition team.
Sure. And that isn’t important to why we test things as scientists. That’s how the scientific methods works. Test, test test. And re test. And then have others test your tests.
@@TechGuyBeau tell me you completely missed the point yet still chose to interject with a completely unrelated comment without telling me you completely missed the point yet still chose to interject with a completely unrelated comment.
I have just one question. I saw that the PTM is around 8.5 W/mK and some thermal paste can reach 12.5 W/mK. So I don't understand if it's better the paste or the pad in that case?!?
It's not just conductivity. It's also resistance to that same conductivity. I'm not an expert and only saw a 'for dummies' type of explanation. Plus, there was some click to the video because they were trying to sell their stuff. Anyway, from what I understand, the pads allow for less resistance which kinda evens the playing field. If you want my opinion, the real selling point is that PTM7950 is longer lasting between repastes, as well as no reported pump out. Of course, during that longer time you still have to keep your machine clean, and/or take it apart to clean fans...etc. Just not the heatsink/IHS assembly.
@@thedryparn1279 Yeah, I know. That's why I originally wrote,"...PTM7950 is longer lasting between repastes, as well as no reported pump out." Longer lasting is generally considered the same as _doesn't degrade over time,_ and no reported pump out is self explanatory.
what i dont get is the pkg says 8.5w/mk isnt the thermal paste i use which is thermal grizzly cryonaut extreme pink stuff 14.8w/mk ?? am i missing something here wouldn't the thermal paste i use be alot better then this stuff ?
@@TechGuyBeau ok ? so there lies?? wouldn't that be illegal ? im debating on buying this stuff for all my mining gpus but dont want to waste my money if its not going to be better i live in canada also so shipping will be cheap
The mess and risk of LM and the fact that the temp difference compared to PTM is at best 2-3 degrees makes the whole hassle and risk completely worthless!
No you misunderstood…. It melts at high temps, then fills in gaps. Then after cooling it turns significantly more brittle. And doesn’t melt anymore. Hence the hot cold cycling. This is also why you can’t reuse PTM when you remove the heatsink, as it is extremely brittle and will not melt anymore I know it can be a confusing concept for a lot of people, but basically it will “get better over time” after a few heat/cold cycles. In short, when you slap it on, don’t be surprised if you run a benchmark immediately and the results suck. Rather later on it will improve. I run a few heat cycles and the next day it’s usually at peak performance
Liquid metal will beat it. However you trade the ease of application and lack of conductivity for slightly lower performance. You really have to be careful with Liquid metal or you will damage the device.
Sure. a big business can say whatever they want. They’re also assuming that their source itself is legit. That’s why people need to test things. To simply assume that something is what it claims to be is just simply bad science
@@TechGuyBeauebuy7 is a shopping platform and the merchante is called Geek temperature control so how domyounknowmit an original Honeywell? Honeywell does not directly sell their product. You can only get those in very bih sheets and the ones in ebuy7 buys those, cut them into smaller pieces and sell them, which is exactly what LTT does and they much more reliable than some seller on ebuy7 in China, which does sell the original product as does LTT. Instill getbyour point and making sure it's the real thing is perefectly fine and justified. I'm a long time viewer of LTT and Linus is trustworthy but, as he learned witheir bag, not everyone knows him which is way he ended up providing a standard guaratee (even though they help anyone even if it's not covered by that guarantee) as you cannot count on the "trust me" guarantee (even though, in the end that's what counts) so checking and making sure thismproductbis legit was a good service for your viewers.
8:22 lol Funny how you misspoke and said its not thermally conductive instead of electrically. I wonder what effect it would have if that was actually true.
I’ve compared with TF-8 in the past (shown in video) which is one of the best pastes around. It has extremely high conductivity. Even in that case the PTM is way ahead
They sell them here, but for some reason say do not use on normal CPU with your usual heat spreader. What gives? In theory it should work normally, maybe they say don't do it because there is a chance to rip the lid off CPU with your cooler when PTM sticks to it hard after few years when you wnt an upgrade? how sticky is it after few months I wonder.
Good to hear this, so no system "integrator" could steal my heat pads. I've had worse temps upon "restoration" after a certain warranty claim they can't uphold.
@@ThePirateGodit … literally is Test using the exact same method with three different products under the same controlled environment. Even normalizing for room temperature? The only way to make to More accurate is to have another person replicate my tests with the same parameters. This is the scientific method my man. And I stated all the confounding variables openly and how I control for things.
From what I understand the cooling performance is just the icing on the cake, PTM7950 doesn't dry out or pump out like paste does. Longer times before reapply if at all with the PTM.
Also the longer you have an installed the more efficient and better it gets so after one year of usage it will have increase its thermal conductivity. Also it's meant for like industrial computer so you never have to disassemble to re thermal paste it ever again.@@Common_Cent
Could have sworn that LTT PTM 7950 is bulk purchased from Honeywell then repackaged to be resold.
Yes, he said that on the WAN show. They did this so people could purchase the pads from a 'known vendor' instead of buying from some random shops that may or may not sell the real Honeywell pads. And even LMG had difficulties to obtain it, it only worked out because of a viewer who works at a redistributor for Honeywell.
@@Gift0rthat and Honeywell offers them in large sheets. For multiple applications. Not just a small amount needed for one install. So it’s way more expensive than just getting a small amount.
@@PTRRanger951yes and no buying a large amount is cheaper if your using alot but this shit you cut to size and it will last years and years but for a normal system you don't need much
@@0-B1which is what I said. No one is doing install after a needs sheets of. It’s cost effective in bulk. But to pay for sheets and only need a small square is a waste of money. Which is why LTT offers it.
It’s cheaper to people to get it that way, even if it seems pricy.
Because any user that is buying it from LTT is using it for a build and won’t ever touch the computer again until years later.
If they plan on changing computers every year, or every few months they would be buying sheets, not a kit.
The kit from LTT is only designed for single builder that is building one computer and plan on it touching it for years.
Or they buy from whoever Honeywell buys from.
Some correction, you dont cook it to phase change but to spread the material around for better contact. Phase change happens at 45C and not after "cooking". PTM7950 works the best in very high pressure mountings.
That’s simply not true. It requires high high temperature to convert it from the liquid state to a solid.
@@MediaChanelle No that is incorrect as well because phase change happens at 45C. The “mysterious” burn-in process of PTM involves repeated heating and cooling of the material to achieve an optimal thermal bond. During burn-in, the pad reaches its phase change temperature, wetting the surfaces and achieving a minimum bondline thickness (BLT).
A key aspect of the burn-in process is the change in the thermal interface resistances. At the beginning of the burn-in (i.e. before the first phase change), the resistances are typically significantly higher, as the PTM is not yet (completely) melted and wetted. During the burn-in process, the interface resistances decrease significantly as the material fills the microscopic gaps and unevenness of the contact surfaces and thus reduces the thermal barrier.
Nobody reads the datasheet before commenting?
@@MediaChanelle that's completely backwards. It's already a solid; increasing the heat phase changes it to liquid (it melts it), and cooling will solidify it.
Yes but after several cooling cycles it will turn much more of a brittle material. Perminabtly. The cycling allows it to penetrate better, the cooling part lets it become a a more brittle and less liquid like material. This is why you can’t recycle PTM.
@@MediaChanelle you sure it goes from liquid to solid as it gets hotter? Might have that the wrong way roumd there pal.
Love this stuff. Have it running in some gnarly servers which can hang at 80-90 degrees 24/7 and I haven’t seen seemingly any degradation in months.
You're repeatedly suggesting that "cooking" the pad is going to turn it solid, except the pad is already visibly solid. Increasing the temperature induces the phase change to a liquid, not the other way around. Linus demonstrates this pretty clearly in his video when he heats up the surface it's attached to with a blowtorch. The phase change is occurring every time the TIM heats up, it's just that cycling it allows it to fill in air gaps more effectively over time.
Once the PTM does several heat cycles it turned it a true solid. It becomes brittle and unmeleable.
Hearing it moves in to the cracks. Cooling it allows it to become a true brittle solid material. This is why removing the cooling solution after will ruin it, unlike paste.
I recommend trying the stuff yourself. Run some heat and cool cycles the take off the cooler. You will find a drastically changed material inside
@@beau-urnsOf course it's always a solid at room temp, it will liquify over 45 C. It will not "cook" solid.
They've addressed the USD pricing on their site many times, something like 80% of their sales are to the US, so having to deal with multiple currencies isn't something they're willing to deal with.
So for them, it makes more sense to only do USD.
@@TechGuyBeau wtf, lol
It’s not just about where their sales come from, it’s about how complicated their accounting is even with one currency. Linus said at one point that when they tried to approach an external company about doing accounting for them, they looked at the range of things they’re doing and went “no thanks.” Which is why even at their size they need a dedicated accounting department.
@@Julian-xb4ikwhat did he say?
I have nightmares of peeling that 2nd layer of plastic off. Putting the cut & measured sheet into the freezer for an hour or two helps it not stick as bad, but it's still a PITA.
Not shown here. I messed up the official
Stuff on my firs application >.
I encountered the same issue and I followed these steps but encountered an issue:
1. Placed the CPU in the motherboard socket.
2. Applied the thermal pad on the CPU while keeping the plastic cover on the thermal pad.
3. Installed the heatsink and secured it by screwing it down.
4. Removed the heatsink.
Afterward, the plastic cover on the thermal pad became easier to remove because the thermal pad stuck to the CPU.
I repasted my 4080 and my 7600x. The 2nd layer of plastic was the hardest obstacle i've ever met building a PC.
@@hakz6777same
Yup, what I found to help me out a lot is to insert a exacto knife in between the very edge of the plastic and PTM and slowly and carefully separate it and once enough is separated I hold the ptm edge with the side of the exacto knife and peel the rest for a good application and leaving the ptm pad intact, it's what I did on my GPU core.
How is this not in every laptop or desktop consumer products? This thing awesome
After going back to air cooling on my 4090, I had a ton of issues with cpu and hotspot getting high after 2-3 months of apply paste. I did that twice before looking at alternatives. The PTM solution seemed like the best route, but was hard to find a reputable source. LTT ended up being the cheapest for the huge sheet.
Nice! An easier purchase solution is definitely helpful
Applied it to my reference 6950xt. I do have an undervolt to 1120mV and 2500MHz max, previous temps were 85C with 102C hotspot. After two weeks the temps are now 72C and 85C hotspot. When mounting on a GPU, re-tighten the screws after a few minutes of having it on. I'm thinking about re-pasting with this again, but using shims to improve the pressure, it could potentially be a 4C improvement which is pretty solid for summer
What were your initial temps like during the first few tests?
I'm not seeing an improvement after running cinebench/3dmark around 10 times with a cooldown period. Temps used to be at 96-100 CPU - 85-86 GPU with a 91.8 hotspot and they are pretty much the same after using the pads.
@@pieuwpieuw6029 Sorry for the late reply.
Thankfully I recorded the temps: 82 - 96 ºC first test same tuning. Delta down to almost 15 degrees vs the 25 degree difference before PTM7950
Ambient temperature has a lot to do with it though; we've had 40ºC consistently during the summer.
You should have noticed a difference by now but make sure you have undervolted the GPU: the stock cooler really dislikes having the GPU set at 1.2v. A mild undervolt to 1170mv should take the edge off without the need to limit clock speeds. Also make sure to tweak the fan curve; should look logarithmic. Mine looks like:
35ºC 30%, 50ºC 50%, 65ºC 65%, 80ºC 75%, 90ºC 80%. Fan speeds higher than 75% make no difference
If you do want to limit clock speeds, clocking down to 2400mhz and running 1130v should bring the temps and power down aggressively while not dropping a performance tier.
As of today I've managed to see a stable 240w draw clocked down to 2400mhz 1130mv with the fan curve stated, edge and hostpot around 65 and 78 ºC now that ambient is lower
I bought an 80X80X0.2mm sheet from Ali Express for £3.85 and it works just as well as the LTT and Honeywell stuff for a fraction of the cost and you even get a bunch of tools with it too
I bought two big pads, and it lasts a lot longer than I expected. I repasted an AM5, GPU, Gaming laptop (CPU and GPU), an old laptop with a Intel U processor (barely got hot enough for phase change), Nintendo switch, mini-PC passively cooled switch, and an Xbox One S and I only used half of one... My graphics card (Sapphire) had been RMA replaced, and so I watched the new one's thermals and it was hitting ~120°C 💀. Decided voiding the warranty was worth it if it would keep me from having another 3 month $300 shipping RMA. PTM brought the temps down to ~60°C under furmark, the stuff that was on there was already "Pentium 4 dell, Never changed" hard. On my Dell G5, it works so well that I discovered that the Dell firmware ramps the fan to 100% before checking the temps which causes it to "sneeze" the fans as the temps are lower than it expects. The Dell heatsink has tons or toolmarks that need good TIM to fill in, this works great. Switch doesn't ramp up fan anymore, Xbox runs a lot cooler to the touch, it just works really really well.
I’ve been using mine and can honestly say my old computer has not messed up lol it’s also a desktop , so this is actually advantageous to use this in a desktop. I bought mine from amazon with a relatively large sheet for 16-20$ U.S
Your audio is much improved with the DJI mic, great review!
The best thing about it is that it does not seem to dry out, it does not lose performance over time. Thermal paste dries out and you need to replace it every few years if you want to keep the same performance. I don't personally know anyone who replaces thermal paste except me, so for 99% of users, these thermal pads should be way better.
Great vid! I would love to see a comparison with the Thermalright Heilos phase change pads to see if it's just rebranded PTM7950 or if it's a different material.
It's different. Better to just use thermal paste than their stuff.
@@SuperBROKEN81 Good morning! Do you have any specific evidence as to why you believe it's different? I've been looking for quite a while and no one has anything concrete. I've been using the Heilos pads as they seem to work within margin of error of the PTM7950 but it would be nice to have some concrete info as to whether or not it's just a rebrand.
same results given 0.5c than with ptm 7950 theyre pretty much the same
PTM isn't cooked to 'phase change' at all. It changes phases every single time it is heated and again once it cools. The thermal cycling is causing that PTM to liquify and settle into the microvalleys of the heatspreader and heatsink surfaces. Repeated cycling ensures it is settled as much as it can be, showing the maximum performance. Important for a performance comparison, but not so much for an end user, as normal use will provide these thermal cycles over the course of a few days. This is actually true for a lot of thermal greases too, where many users will notice 1-2 degrees drop once the compound has had enough thermal cycles to settle.
What I do is measure my CPU with a digital caliper. Then draw those dimensions in a vector program as a square. Then make the square like 2mm larger around. Print it out and cut the square out. Then hold the square over the PTM and cut it out. That way you get a slight overhang as I know it can mess up some when you start to peel it.
If you put the sheet in the fridge for about 10 mins it makes it a bit easier to peel off the plastic.
i use this in my laptop dell precision 5530, after 1 year using ptm7950 safe to say i am impressed, the temp stay consistent
really, Laptops by default need Liquid Metal for their incredibly tight space to fit a cooler in there.
but for those who didnt pay extra for liquid metal pre-applied on their space-saving gaming machines...
and dosent have the hands of a surgeon to apply liquid metal, PTM7950 is the best option.
Store the PTM in the fridge before you apply it that way is easier to apply and get the plastics off.
smart. it should harden the sheet and provent it from ripping
You could get this stuff for cheaper elsewhere but the quality may very. The best thing about getting it from him is quality control and his credibility is on the line.
It would be cool to compare this to Thermal grizzly's high performance graphite thermal pad as well
PTM is better, but won’t last so long. But to be honest, most people (pc gamers) change their cpu in that time, so it’s fine to use ptm and not the sheet
@@checkoff8281ptm is made for server games that run 24/7 for years at a time so shouldn't have any issues with 5 years running 24/7
@@checkoff8281 are you sure? according to the Linus video PTM7950 is designed to last years and years, while thermal grizzly has a reputation for only lasting 1-2 years at most.
I just bought some mystery PTM on ebay from a seller in the US for $13. I got a r5 3600 with a NH-L9x65 and feeling adventurous. Will report back if I can remember this comment so bump it in a few days if anyone's curious
I bought one from some shady sellers in shopee ( filipino version of taobao) and it surprisingly immediately dropped my 90c 12500h to 72c. Im honestly flabbergasted. Considering itll drop further in the future, im really optimistic about whats circulating in the online spaces.
So what did you achieve?
how'd it go?
Few days later.
@@xCheatah He can't answer a few days later, the fake PTM from ebay has caused @qbrt4050 some nasty untold distater that is preventing him from reporting back.
Great vid man, very clear the difference it makes and good presentation on it
Thermal grizzly released a PTM pad for $13 on amazon, haven’t seen any video reviews on it but the comments on amazon make it seem like regular thermal paste with the benefits of regular ptm7950.
Im probably gonna chuck it in an old build since i’d rather not have to re-paste anything. The application seems extremely easy so anyone complaining about it must have chubby fingers.
All this to say, would you be willing to compare the two? If I really think about it they’re probably the same so it might be a waste of time but hey 🤷♂️
righteous laptop, man, take care of it.
I'd love to see this on a laptop where the gpu wattage is increased. I know it's a pretty niche case. Most laptops don't have the thermal solution to backup more wattage. Most are better returning a low wattage variant and purchasing a full wattage variant if wanting more performance.
4000 series and older Nvidia vbios flashing is possible now. You flash to a same chip, same vendor, higher wattage vbios. This risks bricking or needing to have a tool to flash the vbios chip to recover. Some lower wattage Asus variants and cheap MSI variants get a good boost. 3060, 3070, 3080, 4080, 4090 variants have given a bump.
RX 5000/6000 series are even better because you don't need to mess with vbios. Use morepowertool for RX 5000 and 6000 series. great for overclocking and undervolting as the power play table wasn't locked. RX 7000 series laptops are more rare and more locked down.
Asus, MSI (even Lenovo thinkpads) come to mind as some success stories. You are going to get a bit more performance before hitting thermal throttling. They have a large product range with some that could take a bit more power. Some mod the heatsink to get around this. www.reddit.com/r/MSILaptops/comments/16tm1w9/heatsink_modification_to_fix_cpu_and_gpu_heat/
I suggest using a little spraying of air duster to cool it before peeling the second layer off, it will cool it after handling and solidify it
PTM7950 and PTM7958 come in both pads and paste, and some Lenovo systems already utilize it
Would like to see tests going over PTM pump out. A test like that would take a very long time though.
I repasted my asus vivobook s14 i7-12700h cpu machine. I tried to use Thermal Grizzly Kryonaut Extreme first and that was pumped out about in 3 weeks and temperatured went higher. I found only some thin liquid on the core when I disassembled it.
Later I tried some chinese PTM and it is still ok after using it for a year. The original paste in the machnie was also some pretty thick stuff, not sure if it was also PTM or other stuff.
It’s on my 6950xt and it’s great.
Probably test it after 6months, 1 yr to see if there will be degradation. Surprised that it beat the Noctua paste by a long margin.
PTM is incredible. I’ve been promoting it for quite a while and using it long term. It has serious staying power. The laptop currently has the LTT variant in it, so I can see if it lasts
@@TechGuyBeaui applied it in my asus g14 (2023), which I bought a year ago, and it still performs the same. The Liquid Metal had moved off of the die, and heat was out of control. I also tried arctic mx6, but the performance wasn’t good enough. Replaced with PTM after a couple days, and it’s nearly as good as when the LM was new. Honeywell stuff btw.
I replaced the 12 month old mx4 on my water cooled 4090 with PTM7950 and dropped 8c, I’m changing out my cpu and motherboard and can’t wait to get the PTM onto my cpu.
my legion 5 using gelid gc extreme while gaming could reach 90c.
then i replace it with thermalright heilos ptm, now it's drop to 70c while gaming, crazy peformance.
i think any PTM with reputable brand on the market right now is the same, gey color with 8.5 W/m-K. i buy thermalright heilos the cheapest one for $5 and still has half to spare for next application.
Thanks for the insight, could you kindly provide the link site of the thermalright helios
i'd argue if you have a higher end cooling solution it makes even more sence to use PTM as yes your cooling itself can crush anything you throw at it but the main limiting factor is eventually always the thermal tranfer between the CPU/GPU and the cooling solution itself (it's why direct die cooling on the CPU provideds such a massive boost, liquid metal does a lot of heavy lifting but there's also no more transfer to the heatspreader and from the headspreader to the block/cooler it's more direct with less roadbumps to get there).
next build i'm probably switching to PTM, currently have a carbon flap (yea it's not as good as really good thermal paste but it's still quite good).
granted you showed in on a laptop which is more of a direct die scenario but those clock difference are quite impressive.
Not a bad argument. In the end PTM on everything. I bought three boxes of this stuff and already had a ton from Ebuy7.
I covered myself in PTM so that now I run cooler too. :p
Might as well use it up
LTT chose US dollars because the majority of their customers are USA residents. I'm sure it's a bit more complicated for taxes on their end as well with conversion and the value of fiat currency constantly changing.
It's also likely that all the overseas manufacturers they do business with is a factor as they tend to want payment in USD
The question I have, and I am curious about this stuff now, is can you change it. You say it turns into a "solid" and the word baking it on, you can still change these at later dates yes? or aswhen needed or even if you change your cooler on the cpu?
Well you can just clean it off after. Once it does temperature cycling it becomes hard and brittle. This is fine but if you remove the cooler it will make is basically useless. You then clean it off and reapply
@@TechGuyBeau awesome thank you for the response, I may have to try this stuff on my new laptop and my desktop. Been using kryonaut for years, but this looks interesting.
This is just wrong. It starts out as a gel-like consistent at room temperature. When put into the freezer it will become more solid, this should be done for cutting as it makes it much easier.
Above 45C it will become more like a liquid and conduct heat very well. After it cools down from 45C, it will go back to its gel-like state. It does not harden with heat, quite the opposite.
The "cooking" (which is just heat-cycling and repeatedly changing the phase of the pad) turns the pad from solid to liquid and back again. Doing this repeatedly allows the PTM to get into all the microscopic surface inconsistencies, as a gel-like solid (when it's cold) it has great difficulty getting into such small imperfections. In its liquid state (hot) it is much easier.
Heat-cycling just ensures the best contact between the IHS, pad and cooler surfaces.
@@Splomf that is not correct..
It starts as a semi solid. You heat it, it liquifies and fills the gaps, spills out when not needed. After a long cooldown it turns into a brittle substance that cannot be re-used if taken off
I have worked with this stuff for years.
Great video!
I've been using this stuff on my GPU for about a month now and temps dropped by 2/3C between day 1 and day 30 of using PTM7950, safe to assume that's due to heat cycling like you talked about at @9:00.
My GPU from ASUS has terrible mounting pressure which accelerated the pump out effect that normal thermal pastes have to deal with.
This stuff really seems to shine in direct die/silicon applications like on GPUs and Laptops, I wonder how good PTM7950 is vs traditional paste on a CPU, maybe something to test for another video?
You mean a. Desktop cpu? It does help but it depends on the chip. For example I run a 7950x at 105w so it runs so dang cool anyways. But if you have a thermally constrained CPU it’s worth trying for sure
It’s amazing! I’m a person who’s 100% willing to pay for piece of mind. Just sit it and forget it! With the added benefit that when you do remember, you’ll be amazed at how well it’s running.
Just dear god, think about having to take this off! You gotta remember to play a cpu heavy game then rush to take it off before it cools off so it doesn’t get physically and emotionally attached to the cooler. I didn’t know this the first time around so I said f it, it’s only a 10100kf. I’ll just buy another. “ and got to flat head and twisted. Popped off angrily but off non the less.
Still happy buying more til something better comes out.
I really wish that liquid diamond stuff actually worked. Still have some
With a good thermal paste you have the same. Just different application method, that's all.
Have to admit I've seen this stuff on moddiy and was always curious but thought it was more a gimmick than anything. Rebuilding some of my homelab servers, might give this a try even though those CPU's don't run very hot. If it works out maybe take apart my main machine and install it there too. Maybe a little fiddly to install but none of the possible problems you get with LM. Thanks for the video!
Yeah it’s great. Peeling the plastic can be tough for people with weak near vision, otherwise it’s not too too bad
@@TechGuyBeau I'm like you, nearsighted. No problems with stuff as long as it's relatively close so that shouldn't be a problem. Kind of wishing I had seen this video a few days sooner now, just ordered a case for my new proxmox server and just hoping I don't have to wait on this to start building it.
Stupid question, what if you double up on it? Like, put two squares on top of each other before locking on your AIO or Air Cooler? Would the results be the same? Worse?
Excellent video. I don't understand the possible margin of error. The Linus PTM was in fact better, but you're saying it isn't a significant difference from the Honeywell such that it could be margin of error? I guess what I'm asking is, if it wasn't margin of error, what would the Cinebench figures look like? How different do those numbers have to be to say for sure that it's not margin of error?
There is no predefined margin of error, as it simply doesn’t exist. But if we’re looking at a few percentage points at most say two or three percentage difference, it’s not significant in any form of scientific test. To actually properly find significance I would have to rerun this test like 40 times with each compound. And look at the significant difference
Realistically though on this test that I did it’s so minimal that it’s basically margin. I hope that makes sense
Nice. Getting this from overseas is a pita
Definitely much easier and the cost is close enough to make it worthwhile
Wonder how the thermal grizzly carbon and graphene pads compare.
Edit: after doing some research, the carbon pads are worse than paste.
Yeah for sure. They are more for rapid swapping in and out.
the common wisdom: Liquid metal -> Thermal paste -> most thermal pads.
but with PTM7950, Liquid metal -> PTM (really close) -> paste
I recently repasted my 3080 Ti with Corsair branded paste (ran out of NTH-01) and noticed my fan hub, which runs my case fans, on my custom fan curve was locked at 100%.
After applying the PTM7950 today my fan speeds dropped to 55% at the same temperature. Blew my expectations away, my PC is significantly quieter at the same load.
LTT source PTM from Honeywell you know that right It talked about WAN show and anoncement
Hence testing it versus the "also honeywell" you can buy on Ebuy7 and other sites. (aka the purpose of the video)
The way I recommend thermal transfer options is based off who’s asking experience, and willingness to learn.
I use water blocks for direct die for CPU and same on GPU w/Liquid Metal. Glad you did this comparison so I’ve got another option Vs what I use on my personal builds.
Cool comparison! so the ebay/ aliexpress stuff is "legit" :D fixed the overheating in my old msi laptop.
Do you have a asus laptop with liquid metal installed? Would be nice to know how well ptm works in a laptop with factory lm?
Generally speaking if you remove liquid metal and apply PTM, you will decrease the performance of the cooling of course if the liquid metal is shoddily applied from the company then really anything will do better than that
I have used many laptops with liquid metal, typically I just leave them as is, unless there’s some very serious issues with the thermal performance which would suggest sending it back for repair to the company
I just mentioned in another comment before seeing yours. I have an asus g14 2023, which I bought about a year ago. It worked fine for a couple months and then terrible heat. I tried arctic mx6 first, but it was not as good. Switched to PTM, and it’s only a few degrees warmer than with the LM. Some of the LM had moved off of the die, probably because I have to pack it up and immediately put it in my backpack all the time.
Great video, but the difference in actual performance when you stated you gamed it seems to be the same.
When you go through the BG's temps and details it literally shows that Noctua had lower CPU avg watts with more clocks. Granted you get better temps, but I would expect a higher average clocks value considering the system is able to inject more power into it...
GPU temperature difference is only 1,4ºC higher (which to me seems close to negligible). The memory junction is actually LOWER on the Noctua side and the hot spot temperature differential is 2ºC. Not enough for me to call PTM 7950 revolutionary.
The last clocks were too similar for me to cast a vote on either of them.
And then in the end, the "poorer" product (paste) actually gave you better framerate. Negligible, I know, but for a better product to post worse results it does not make PTM's case for me.
At this point the only thing that might be left is longevity. And this is something I would rather like to find out. To see the differential on the performance with time degradation put into it.
Problem is that this is a case where TIM easily lasts 2/3 years without steep loss of performance so I understand it would be demanding to take on such challenge.
I wish, LTT could sell them in smaller sizes for a little cheaper. Google tells me that 7950 has an estimated shelf-life of 12 months. So if I needed to build 2-3 PC's as once-off, it'd be a shame to just dispose/let it expire. But I guess the re-packaging cost wouldn't be worth it.
I'm pretty sure the PTM has a very long shelf life if any at all considering that it's intended for industrial use and they buy this stuff in kilograms.
As long as it's air tight and not in rough temperature it shouldn't degrade.
Have you tried peeling the transparent sheet first, put it on the CPU. And then peeling the white sheet? Not sure if that works any better compared to peeling the white sheet first?
The Pad doesn't turn into a Solid after multiple Thermal Cycles...
The Pad you are holding is it being a Solid, when it reaches a temperature of 45C then it goes into a Liquid state.
FIRST
Love your videos!
And I love the audience. Happy you enjoy 🫡
Where do you import it from? You mentioned overseas, so I am guessing AliExpress, but what seller?
ebuy7
Aren't you meant to put it in the fridge for a lil' bit so cutting & applying it becomes easier?
That’s more for the liquid. The “pads” are easy to apply anyways
Rumor has it that these work exceptionally well on waterblocks that cool cpu and gpu at the same time. You could even cool the wrong gpu 😂😂😂 don't buy from bullies.
Is this a apply and forget solution like the kryosheet or you need to eventually replace it like thermal paste?
It lasts for an extremely long time. But if you remove the cooler ,it basically destroys the ptm. So after apply you basically leave it alone, if you change coolers you need to add more ptm
I might buy some to replace the corsair thermal paste that came with my aio. The 7900x doesn't run that hot lol, at least compared to that 13700k I got rid of but temps could be better
7900x can also be turned to balance mode to get like 95% performance at half the wattages. I have a 7950x and it works wonderfully
@TechGuyBeau I have PBO on, -30 offset with a 95c temp limit
i have gotten used to replacing my cpu paste every 2-3 years, how long would one of these pads stay effective, on average??
and would this work as a replacement for gpu memory and vrm thermal pads??
Laptop CPU = Estimated 5 Year's
Dekstop PC CPU? Idk
GPU Die and GPU memory? Yes
VRM? I don't think so probably too thick
You have a fundamental misunderstanding of how this phase-change material works.
At room temperature it is a gel-like solid. If put into the freezer, it will become more solid (do this when you cut and apply it, it will make your life easier). When it is heated to ir above 45C, it becomes a liquid.
It does not dry out or become solid unless FROZEN. Heat will always liquify it.
No…. It simply becomes much dryer after being properly heat cycles. Prior to its first application is a near gel like state. After heating (and then cooling), it becomes a brittle material.
This is why once applied, if you remove a heat sink, the PTM is rendered useless and must be reapplied
I need this for my omen laptop with i7 9750h and 1660ti. This generation is known for running way hot. I have a custom power curve in msi afterburner but its still not enough. Im sure this will help at idle too since im not really gaming much. Its even too hot to use in lap on ecomode on idle lol
See if CPU & GPU share heatpipes, your CPU might needlessly boost & eat the thermal headroom of GPU.
There used to be a tool called ThrottleStop that let you tweak boost behavoir. Basically find out if your usage is GPU bound, then stop the CPU from going into it's higher power states, so your GPU can use that extra headroom. Also helps temperature-frequency oscilation.
Helped me loads on a laptop with underpowered heat dissipation capacity at 100% CPU&GPU usage.
Its always sold out on ltt store..... I really want this since I wouldnt have to worry about reapplying ever then :/
That's brave, testing thermal paste/pads on a notebook PC. I wouldn't dare to 😎
Edit: I meant the risk of damaging and wear something down by doing this. My comment wasn't aimed at warranty.
I replace all the paste on my laptops with PTM, silly not to. I got a 8 degree drop on my Framework laptop alone.
If you are in Canada or the USA you have right to service your laptops. Other countries not sure
2:05 On the package of the honeywell paste you bought from China, it says in Chinese "Origin: USA"
I watch LTT, they don't push this. Respect but I went and bought the thermal grizzly direct ❤if I knew it would be hard to choose😢
This is not a phase change material it´s a phase transition material (PTM).
liquid metal is eating aluminum so that is a other risk so you need a cooler that have coper on the cpu
Thinking of using this instead of thermal paste for an Xbox 360 .
Ordered some of the "Official" honeywell ptm, used it on my gaming laptop.
After a month and a half, my temps went up, (idle, and gaming), my idle temps went from 48°c to 55°c, and gaming, went from 75°c to 98°c.
I took my laptop down, to see what was going on, and found, the ptm had gotten dried out, and hard, and it had boiled out from between my heatsink, and cpu, and my cpu now had a small burn mark in the center. Not using that stuff again!😒
Supposedly they're both "official"
PTM drying isn’t a thing. Perhaps you had some knockoff PTM?
PTM is supposed to change from a semi liquid to a liquid then to a solid and stay very dry.
Yeah I have an official bridge to sell you if you would like to put in an offer
Heh, you may have been scammed. That doesn't happen with ptm.
How does the Honeywell PTM7950 SP pad compare to the Honeywell PTM7950 SP Paste?
The paste is the same thing uncured. Use the pad.
The stuff you hold into the camera at 02:06 min. is a fake product. The original Honeywell PTM7950 is always 0,25 mm thick, not 0,2 mm. You can find that info also on the Honeywell homepage.
Its not made by LTT, they just reliably sourced it from Honeywell
think that was clear in the video, but its important to test things to make sure they are legit. Any huge business like LTT can claim something, doesn't mean its true
Lmg/Ltt group has a well established relationship with many high quality foreign manufacturers so getting this ptm direct from manufacturer is not a far reach for their acquisition team.
Sure. And that isn’t important to why we test things as scientists.
That’s how the scientific methods works. Test, test test. And re test. And then have others test your tests.
@@TechGuyBeau tell me you completely missed the point yet still chose to interject with a completely unrelated comment without telling me you completely missed the point yet still chose to interject with a completely unrelated comment.
I have just one question. I saw that the PTM is around 8.5 W/mK and some thermal paste can reach 12.5 W/mK. So I don't understand if it's better the paste or the pad in that case?!?
Which is just proof that those numbers don’t mean much. Since paste is much less effective
It's not just conductivity. It's also resistance to that same conductivity. I'm not an expert and only saw a 'for dummies' type of explanation. Plus, there was some click to the video because they were trying to sell their stuff. Anyway, from what I understand, the pads allow for less resistance which kinda evens the playing field. If you want my opinion, the real selling point is that PTM7950 is longer lasting between repastes, as well as no reported pump out. Of course, during that longer time you still have to keep your machine clean, and/or take it apart to clean fans...etc. Just not the heatsink/IHS assembly.
@@poorplayer9249 thx
@@poorplayer9249also PTM doesn't pump out or degrade over time.
@@thedryparn1279 Yeah, I know. That's why I originally wrote,"...PTM7950 is longer lasting between repastes, as well as no reported pump out." Longer lasting is generally considered the same as _doesn't degrade over time,_ and no reported pump out is self explanatory.
what i dont get is the pkg says 8.5w/mk isnt the thermal paste i use which is thermal grizzly cryonaut extreme pink stuff 14.8w/mk ?? am i missing something here wouldn't the thermal paste i use be alot better then this stuff ?
Those numbers don’t mean much.
@@TechGuyBeau ok ? so there lies?? wouldn't that be illegal ? im debating on buying this stuff for all my mining gpus but dont want to waste my money if its not going to be better i live in canada also so shipping will be cheap
Can you provide a link to your chinese vendor of Honeywell one?
The mess and risk of LM and the fact that the temp difference compared to PTM is at best 2-3 degrees makes the whole hassle and risk completely worthless!
Does this solve the "pump out" effects on laptops?
In theory.
which effect is that?
@@SkegAudio Probably means that the paste will get pushed out to the sides over time because of the mounting pressure.
All the Exchange Rates and Everything are all dealt with by the Banks and since physical Cash isnt Involved its a Simple Process
4:36 Cooling chamber 🤩😍
im in the uk where can i buy this most are 1 week shipping from china ? ?
seems pretty clear from the product info that it’s viscosity decreases as temperature increases. it melts at high temps. This guy mixed it up.
No you misunderstood….
It melts at high temps, then fills in gaps.
Then after cooling it turns significantly more brittle. And doesn’t melt anymore.
Hence the hot cold cycling. This is also why you can’t reuse PTM when you remove the heatsink, as it is extremely brittle and will not melt anymore
I know it can be a confusing concept for a lot of people, but basically it will “get better over time” after a few heat/cold cycles.
In short, when you slap it on, don’t be surprised if you run a benchmark immediately and the results suck. Rather later on it will improve. I run a few heat cycles and the next day it’s usually at peak performance
I wonder how the performance is compared to liquid metal
Liquid metal will beat it. However you trade the ease of application and lack of conductivity for slightly lower performance. You really have to be careful with Liquid metal or you will damage the device.
Im not sure if that stuff likes to be touched with oily hands....
Ltt ptm is honeywell ptm, even says it on the store listing
Sure. a big business can say whatever they want. They’re also assuming that their source itself is legit. That’s why people need to test things. To simply assume that something is what it claims to be is just simply bad science
@@TechGuyBeauebuy7 is a shopping platform and the merchante is called Geek temperature control so how domyounknowmit an original Honeywell?
Honeywell does not directly sell their product. You can only get those in very bih sheets and the ones in ebuy7 buys those, cut them into smaller pieces and sell them, which is exactly what LTT does and they much more reliable than some seller on ebuy7 in China, which does sell the original product as does LTT.
Instill getbyour point and making sure it's the real thing is perefectly fine and justified. I'm a long time viewer of LTT and Linus is trustworthy but, as he learned witheir bag, not everyone knows him which is way he ended up providing a standard guaratee (even though they help anyone even if it's not covered by that guarantee) as you cannot count on the "trust me" guarantee (even though, in the end that's what counts) so checking and making sure thismproductbis legit was a good service for your viewers.
I will recommend FEHONDA 15w pad, grease buy PTM7950.Everything is very good in the hot summer
Heat cycling sounds like so much work. If it changes from solid to liquid to solid wouldnt that shift the heatsink also
8:22 lol Funny how you misspoke and said its not thermally conductive instead of electrically. I wonder what effect it would have if that was actually true.
The cpu melts lol
What about Thermal Grizzly or Kingpin paste? That's way better than the Noctua paste.
I’ve compared with TF-8 in the past (shown in video) which is one of the best pastes around. It has extremely high conductivity.
Even in that case the PTM is way ahead
I didn't cook it like you said... Is that concerning?
It will cook itself anyways. That’s only when you want peak results asap. Just using it for a few days will have the same results
@@TechGuyBeau alr!
They sell them here, but for some reason say do not use on normal CPU with your usual heat spreader. What gives? In theory it should work normally, maybe they say don't do it because there is a chance to rip the lid off CPU with your cooler when PTM sticks to it hard after few years when you wnt an upgrade? how sticky is it after few months I wonder.
Can confirm. It’s been in my old laptop for 5 months and now the heatsink is permanently on there 💀
@@Nekudon Good to know if you want to upgrade down the line, could break something.
It’s not sticky at all. PTM will turn it a somewhat brittle solid after a few heat cycles. It’s easy to take off.
Have you tried PCM-1 yet?
I have not
my question is it reusable like i want to clean the aircooler or laptap fins due to clog with debris
no. once you remove it, then it is useless. it turns from a semi solid, to a more fluid when heated, then it hardens permanently. after cooling
thanks i guess ill go for thermal grizzly for their reusable sheet pads
Good to hear this, so no system "integrator" could steal my heat pads. I've had worse temps upon "restoration" after a certain warranty claim they can't uphold.
Would you put this in a PS5?
The ps5 already has Liquid Metal. I put it in my ps4 and Xbox one x thoufh
@@TechGuyBeau I dont like the fact that I need to take care of dry spots
Can you apply this to the GPU also?
I do yes
What about PTM vs LM?
Lm works better but is quite risky
Well ones a long term solution and ones a short term solution…. So there’s that.
@@TechGuyBeau Not to forget LM has its own problems especially with non plated copper surfaces.
I think personally I wouldnt trust knock off honeywell...
its depend your cooler
LTT had higher temps but also higher wattage and score? Quality tests bruh
Higher Watts = higher temps.
More watts = more fps.
LTT said they buy straight from Honeywell
Sure. People can claim anything. That’s why we test things. Scientific method
@@TechGuyBeautesting on a laptop like that is not very scientific.
@@ThePirateGodit … literally is
Test using the exact same method with three different products under the same controlled environment. Even normalizing for room temperature?
The only way to make to More accurate is to have another person replicate my tests with the same parameters.
This is the scientific method my man. And I stated all the confounding variables openly and how I control for things.
So that ptm7950 is an 8.5w/mk the paste I use is 13.7-14w/mk so I'm not sure if this would be a benefit for me
From what I understand the cooling performance is just the icing on the cake, PTM7950 doesn't dry out or pump out like paste does. Longer times before reapply if at all with the PTM.
@@Common_Cent kingpin extreme kpx thermal paste.
Also the longer you have an installed the more efficient and better it gets so after one year of usage it will have increase its thermal conductivity. Also it's meant for like industrial computer so you never have to disassemble to re thermal paste it ever again.@@Common_Cent
W/mk is something to be ignored because there is no official way to measure it so 5w/mk can be better than 10w/mk in some cases