Good article regarding repasting CPU/GPU

Hello everyone!

I just stumbled with this new article from Tom's Hardware about repasting and thought about sharing it

A lot of us here already know how to, or have had experience repasting computer components, but new users or inexperienced ones always get scared about attempting to do so themselves. I thought maybe this article might help a bit

Part two is here!

Hope it helps a bit but at the very least it is an interesting read

 

Very nice info ryzeki and I am sure it will help a lot people.

 

Thanks!

I thought it was great info specially because it gives pics and measurements to get an idea about size of paste etc.

 

Lots of good info and objective advice! Thanks for posting, ryzeki.

 

The only beef I would have with this is that it only explores components with heatspreaders, and not bare die components like those found in notebooks. Anyone who sees my posts on the matter know I favor the spread method on bare dies.

Still nice to see nonetheless.

 

Thanks! I hope they put the second part soon!

I know what you mean. It is rather limited in that sense, but still the same overall knowledge applies. Spreading on notebooks works well, and one reason why some don't do it depends on the paste type, like IC diamond that unless you heat it first, it is a bit too thick.

Personally I favor the pea method, but that's because I tend to paste rather quickly haha

 

YES!!! Everyone should keep in mind that this is NOT the way to do it for our notebooks. Because of the nature of the bare die's, the paste does indeed need to cover then entire chip. Mind you a thin layer, but needs to cover the WHOLE thing due to lack of spreader.

 

Also to keep in mind the bare die of a notebook tends to be much much smaller than a desktop chip in general. It's actually less work to spread a notebook die than a desktop one.

 

Well, part two is finally up! It was a bit disappointing not seeing IC diamond paste to the test, but nonetheless a very interesting article. Here's the link!

Thermal Paste Comparison, Part Two: 39 Products Get Tested - 39 Thermal Compounds Get Benchmarked And Tested

 

For bare die, just look at the GPU results. I basically ignored all the CPU tests as they aren't relevant to notebooks. I use MX4 at the moment and it seemed to test pretty well. I hope with this article people stop recommending AS5. Old, conductive and not even that great anymore.

Now they need to test cleaning solutions. Arcticlean, Indigo Xtreme Clean, alcohol of various concentrations etc!

 

I agree! I do still see AS5 recommended but there are so many options right now, some which are better in every way. It was an interesting article suited for desktops, but we can still get some data in general

IT was also surprising how the metal compound made into thermal pad, had similar performance to regular pastes.