GP-Ultimate 15W-Thermal Pad 120x20x 1.5mm (2pcs). Excellent Heat Conduction, Ideal Gap Filler. Easy Installation

I used these thermal pads when I installed this heatsink: "JEYI M.2 HeatSink, Heavy Duty Aluminum Convective Heat Sink, Passive Heat Sinks with Fins, FinsCold iGlacier8 " replacing a smaller copper third party heatsink. I replaced the stock thermal pads with this: "Gelid Solutions GP-Ultimate 15W-Thermal Pad 120x20x 0.5mm (2pcs)." and I put a piece of thermal insulating tape under the SSD holder to protect the MB: "Tesa's Most Advanced High Heat Harness Tape 51036f" Not sure why the previous smaller but substantial copper heatsink did so poorly. It used much thicker thermal pads, and the short profile and lack of large gaps for airflow through the copper, rather than over it, may have been factors. Installation was easy but due to the weight of the heatsink, and the difficulty of reaching the SSD holding screw, I had to lay the tower case on its side so gravity would assist me in getting that screw in. I also had to remove the GPU from the adjacent PCIE slot. (Taichi X670E, RTX 4070S). That third party thermal pad has a high heat conductivity and its very thin, which is good, but it is also very fragile. The first time I applied it I ripped off a small piece and it was too fragile to put it back. So I cut off the end of the installed pad with an exacto knife and replaced that whole piece. With care and awareness of how fragile it is, it wasn't a problem to install nicely. Since the Samsung SSD already had some kind of sticker over the chips, the thinner pads still provided good contact. P.S. The design of the JEYI heatsink is great. Although I believe the length and width are standard, so it will fit on the MB, it is very tall, giving a lot of cooling surface. The thin, tall fins are all perpendicular to the motherboard and parallel to the length of the SSD, with large air gaps between them. In a tower case with airflow from front to back, or back to front, the air flows through the space tween the fins for optimum cooling and smooth air flow. I don't think it would work as well with airflow across the width rather than the length. The black aluminum looks great and keeps the weight down. It looks like the engineers actually used computer models to test their designs. The stock thermal pads are probably OK but replacing them with these thinner, better ones must have contributed significantly to the performance.

For the first time, I went all-out and bought the top-end GPU for the generation, the RTX 3090 FE. I was excited to install it. And then I learned about the VRAM. Even in light gaming, it was in the upper 90's. And mining? At 69% power limit, I was sitting at 94C. At 90% power limit it would quickly hit 110C. This took away all the fun of the card because it made me afraid to push it. Despite what anyone says, silicon should never hang out around 100C let alone 110C. Even IF you think that it is within spec, that heat is going into the PCB and into other components where it isn't OK. The last thing I want is a $1500 doorstopper in 3 years. It felt like owning a Lambo but only driving it at 45mph and afraid to take it on the highway. At least I assume it would feel something like that, my crypto gains have not yet been enough to acquire a Lambo. I decided to give these pads a shot since not many people have used the Gelid Ultimate, others told me these were too firm to work. But I went with the 1.5mm pads. Note you will need TWO 2-packs for the backplate VRAM. You COULD do front and back vram with two 2-packs if you don't mess up cutting at all - I did and was lucky I bought 3 of them. I did the backplate vram first since it is easier to get to. And was disappointed to say there was ZERO difference - at least in HWInfo. I didn't know if maybe I messed up, or if these pads weren't ideal, or what. I added a ton of backplate cooling (heatsinks and a 60mm noctua fan) and still no change. However, keep in mind that Memory Junction is the HOTTEST of all ram modules. If the back RAM was running cooler but front just as hot, you'd see no difference in Junction. And my theory is that by replacing only back pads, the heat from the front was preferring to go toward the back rather than the large front heatsink. Note that the stock putty/mesh "pads" are just 1.5w (yes that low). So finally I decided to go for the front vram. And it made a HUGE difference. 30 minutes temperatures: Before: 69% power limit, about 86 mh/s. Core -100, RAM+1000. 94C . After: 69% power limit. Same settings, 74C After: 90% power limit. 121mh/s. Core -100, RAM+1200. 86C. Yes, 8C lower when I INCREASED the power limit drastically. And 20C lower at same exact settings. I am no longer afraid that my VRAM is cooking itself and will give artifacts right after warranty is up and can push the card as hard as I want. Well worth the money for these pads., they are very easy to cut and also seem to retain their capability even if you open the card up again since I opened it like 4 times for various reasons and they are still giving me the same results.

There's better brands but i would say this does the job and doesnt break the bank

Boa qualidade.

A pasta é boa, porém no pacote enviado só veio uma tira e não duas como esta anunciado.

Usé los pads para reemplazar los de una rtcx 3060 y noté una mejora en las temperaturas (~4c).

The pads are thick as expected and work well, no complaints.