I've never seen this and I work on laptops all the time. MX-4 performance never degrades. I don't know where all this talk about "pump-out effect" is coming from but I call complete BS.
Ok, go with that. TFX has been test and seems to perform very well if you want to spend the money. Examples;
Thermalright TFX $14
Buy Thermalright TFX Thermal Compound Paste 14.3 W/Mk, Carbon Based High Performance, Heatsink Paste, CPU for All Coolers, Interface Material, 2 g with Tool: Heatsinks - Amazon.com ✓ FREE DELIVERY possible on eligible purchases
www.amazon.com
MX-4 $7
Amazon.com: ARCTIC MX-4 (4 Grams) - Thermal Compound Paste, Carbon Based High Performance, Heatsink Paste, Thermal Compound CPU for All Coolers, Thermal Interface Material: Computers & Accessories
www.amazon.com
MX-5 $7
Amazon.com: ARCTIC MX-5 (4 g Incl. Spatula) - Quality Thermal Paste for All CPU Coolers Extremely high Thermal Conductivity Low Thermal Resistance Long Durability Metal-Free Non-Conductive Non-capacitive: Computers & Accessories
www.amazon.com
Nguyen is right on this. Check notebookreview forums. There are posts everywhere in almost all of the sections about degrading paste performance. Most are in the Alienware sections, and some discussion in random MSI BGAbook and some Clevo sections. The overwhelming complaints however are in the Dellienware sections, with the Tripod imbalanced heatsinks and warped cold plates and imperfect mounts (did anyone really think a heatsink with three screws was a good idea?).
I dealt with this pump out effect _extensively_ on my own MSI Titan. When I first used Kryonaut, I had to deal with high core temp deltas (core #1 being like 8C hotter than core #0), which I was only able to fix by using thinner VRM thermal pads and playing musical chairs with screwing down the heatsink a certain way. But after a few days, up to one week, sure enough, I noticed the core temp deltas doing the "Runaway temps" thing. Then I realized what the other users were finally dealing with. And sure enough, whenever I saw the core temp deltas reach 10C, I took off the heatsink and noticed most of the thermal paste was outside of the core. Yikes.
That's your pump out.
Some users have had MX-4 dry out--basically the paste separating from the oil due to extremely high heat on direct die (MX4 doesn't exactly like 99C temps). So you ended up with paste around the edge of the entire CPU, and swimmy liquidy material on the core.
I managed to get it somewhat stabilized with a LOT of work, but I still couldn't get the core temp deltas below 6C. So I started buying other pastes. I tried Phobya Nanogrease Extreme first, but that degraded *faster* than Kryonaut! Even though the initial temps were 1C cooler. And I'm very OCD about this stuff. Then I switched to Coolermaster Gel Maker Nano. That was more stable for a little while (although performed worse than Kryonaut) but even that pumped out.
So then I moved to liquid metal. Conductonaut.
I applied it, got nice temps at first, but after about one week, I saw the core deltas slowly getting away from each other. At that point I knew exactly what was going on. The temps were still better than Kryonaut. So I left it on for about a month and just dealt with it. After a month....10C core temp deltas and temps were WORSE than Kryonaut was on a fresh application.
I took off the heatsink and noticed ALL of the LM had hardened and was nothing but an oxidized layer. I never even realized something like that was possible. I scrubbed all of it off, applied a new layer and temps were back to normal, maybe 1C worse, but the same issue stared slowly happening again.
Thats when i got into heatsink modding. I ordered a spare heatsink from MSI (From Tom Ho, from their parts department) and sanded the heatsink as flat as possible. First I polished it to 5000 grit, however I remember people on the Clevo section of NBR saying that LM hates polished surfaces (I later found this out the hard way with my delidded 9900k and IHS underside) and again had some long term issues, so I roughened the heatsink up with 1500 grit, (this was after I found out what to do with my 9900k IHS and seeing how LM tends to "pool to itself slowly" on fully polished surfaces under compression--very poor wetting ability!)), and then spread the LM (without applying downwards pressure) for about 15 minutes around the entire HS until it stared becoming "thicker" (due to gallium absorption slowly into the roughened surface), applied a second layer on top, mounted it, and bam: perfect temps. 2C core temp delta max.
It's been almost 2 years now and still the same 2C core temp delta, with a perfectly flat heatsink and roughened surface and LM.
I even wrote a guide on notebookreview about how I helped address the LM on copper heatsink durability issue.
Papusan has his own method--he applies LM on IHS and heatsink, bakes it in an oven at 100C for two hours (as long as the heatsink is safe to bake), then removes the parts from the oven, wipes off the old LM, leaving the silver stain behind, then reapplies new LM on top. Gives 1+ year durability since the high heat caused early gallium absorption (thus limiting the battery effect (CU->GA differential) from a new LM layer).
Unfortunately I didn't test regular paste with the sanded heatsink, so I can't give you the results of that. But pumpout is very real, and it all depends on how well the laptop is designed and how balanced the heatsink is. Imperfect heatsinks + weak PSI pressure + convex cores=durability issues.
BTW I'm the person who 'cracked' the MSI power Limit problem with the 230W (GTX 1070) vs 330W (GTX 1080) limitations (to avoid premature CPU throttling and extremely high battery drain) on TDP modded GTX 1070's that exceeded 200W TDP, via the EC RAM (in RW Everything), to make the laptop think a 1080 was installed instead of a 1070. So I put in a hell of a lot of time and work in on this stuff.
Anyway I know I linked this thread before, but you should go through it carefully. You'll find this pump out and dry out information discussed at length.
Thank you for visiting the TechnologyGuide network. Unfortunately, these forums are no longer active. We extend a heartfelt thank you to the entire community for their steadfast support—it is really you, our readers, that drove
forum.notebookreview.com
www.techpowerup.com
, at much much lower TDP too.
