Lapping Vs Non-Lapping Vs Thermal Pastes - Temperatures Tested on GPU Die
# Improving GPU Cooling: A Hands-On Guide to Lapping and Thermal Paste Testing
When I recently got this 650 TI beast in for $70, I thought wow, what a bargain! Upon putting it into my computer, although it worked, the temperatures were a little bit high. So, upon taking off the cooler, I noticed how rough the heatsink base plate was. This led me to decide on doing a little guide on lapping and changing thermal paste, documenting the differences for you guys and letting you know what the best options are.
## Building a Workstation PC: A New Adventure
Recently, I decided to build a workstation PC for work. Who would have ever thought about building a workstation PC for work? Anyway, I decided to put in a 650 Ti boost in there, which I got for $70. Now, this price is honestly pretty damn good value for money considering it boosts—I mean, it boasts about the same performance as a 750 Ti, though using a little bit more power.
When I put it in, I noticed that the temperatures were a little bit high. So, I immediately pulled it out, started cleaning it down, and found that the heatsink paste they put on this was junk. Also, upon putting my fingernail and running it across the heatsink, I noticed that it was really rough with a lot of grooves in it.
## The Process: Changing Thermal Paste and Lapping
I decided not only to change the heatsink paste but also document the actual temperature change with just the heatsink paste alone. Additionally, I decided to lap the cooler as well and document the temperature change with that as well.
## Testing Different Thermal Pastes
So, let's get on to some results now. In conclusion, what can I say about the whole thing that just happened here? Well, the thermal piece made the biggest difference as we saw with the Arctic Silver 5. This stuff got us a 5-degree drop over the stock heatsink paste when we tested it. The Animax did pretty well as well, getting a 4-3-3 degree drop.
The Cool Board to his liquid pro I thought was slightly behind the Animax paste and it was a little bit behind the Arctic Silver 5. This surprised me because I read on the internet that this stuff was really good—like this was the best if you're putting it on a GPU die. Don't know where I read that, but this stuff didn't do the best. It did better than the stock thermal heatsink paste but it didn't do the best and especially since it costs the most here out of these contenders, I thought it would have done the best.
Also, considering that Cool Board to his liquid pro is really good on dies, I was surprised to see that this didn't do so well.
## The Results: Silicon Grease and Lapping
Last but not least—or should I say last but least—the silicon grease. This stuff did terrible on a GPU die, so don't use this stuff on dies. I think that's something to come out of this video. Even though I've been a bit of a fan of this stuff, just using our normal heat sinks, but it did terrible for what it's worth on a die.
Lastly, the lapping itself. You guys probably saw the results and they didn't do too well. Like, I was surprised. I thought we would have gotten a bigger drop out of lapping a cooler like this, especially one which I thought was sort of a worst-case scenario where the grooves were really rough and you could see with my fingernail are etched it on there. And then after I smoothed it down, it was really smooth where I sanded it back.
So, I was kind of surprised to see that the temperature drop was probably only half or a 1-degree at best over just the stock rough heat sink. So, I guess that's a compliment to the heatsink paste and how good they're getting and sort of how much of a slack they can take.
## Conclusion: The Importance of Heatsink Paste
So, that's time to come out of this video. Yeah, heatsink piece—they make a big difference. So, make sure you pick your heatsink paste carefully. I know a lot of guys are big fans of like on mx4—I'm a big fan of Arctic Silver 5—and I'm going to keep being a big fan after the video. This stuff's pretty damn good; it's consistent.
Anyway, keep in mind with lapping. I'm sure it would make a bigger difference though if we had a bigger heatsink on a bigger die—for instance, saying like my old GTX 780 lightning which had like a one-and-a-half kilogram cooler on it and the die would have been a lot bigger than a 650 TI boost. And I think would make more of a difference in those kind of circumstances.
Anyway, for what it's worth though, the average person on an average GPU would probably not need to go out and lap their CPU cooler. So, I think lapping is one of those things that that's for more high-performance like real the guy trying to get the last grain out of his arm out of his wheat or something like that.
## Final Thoughts
So anyway, that's a bad analogy. So anyway, guys, I'll catch you in another tech video very soon. If you have any questions yet, drop a comment below and I'll get back to you as soon as I can. If you liked the video, please give it a thumbs up, and I will catch you in another tech video very soon.
Bye!