De-lidded 7820x - Method Comparison

[mithiral67]

So, doing a overkill build and wanted to delidd my 7820x. If you take a look at my build guild . . .

https://forums.overclockers.co.uk/t...t-singularity-computer-spectre-case.18815756/

.. . you will see that I am waiting on my case and currently have an EVO 212 as a temporary cooler atm until i can build my custom loop. That said, i want the system running rock solid before i build the loop and have to try and trouble shoot.

So with the evo and the goal of not going over 90 degrees, I was able to get pretty good results. I didnt stress test this other then cinebench runs because that cooler is not meant for hours of running, so this is just a gauge of sorts. https://valid.x86.fr/ilxn4z

Anyways, these positive results motivated me to order a delid kit and some LM and a week later I was ready to go. I had dialed my system down to 4.4g until the custom loop and probably should have gone back to 4.85 for the test but oh well.

Before the delid, the spread between ambient and average peak temps (78c) were 53c and after the delid (69c) were 45c. Honestly i was hoping for better. My question is, was using the tiny EVO the problem and if I had been on water I would have seen a larger spread?

Kinda like you dont see a huge spread at idle?

Thanks,
James

 

[LuckyBenski]

Edit: derp, didn't read post properly

 

[james_2k]

So, doing a overkill build and wanted to delidd my 7820x. If you take a look at my build guild . . .

https://forums.overclockers.co.uk/t...t-singularity-computer-spectre-case.18815756/

.. . you will see that I am waiting on my case and currently have an EVO 212 as a temporary cooler atm until i can build my custom loop. That said, i want the system running rock solid before i build the loop and have to try and trouble shoot.

So with the evo and the goal of not going over 90 degrees, I was able to get pretty good results. I didnt stress test this other then cinebench runs because that cooler is not meant for hours of running, so this is just a gauge of sorts. https://valid.x86.fr/ilxn4z

Anyways, these positive results motivated me to order a delid kit and some LM and a week later I was ready to go. I had dialed my system down to 4.4g until the custom loop and probably should have gone back to 4.85 for the test but oh well.

Before the delid, the spread between ambient and average peak temps (78c) were 53c and after the delid (69c) were 45c. Honestly i was hoping for better. My question is, was using the tiny EVO the problem and if I had been on water I would have seen a larger spread?

Kinda like you dont see a huge spread at idle?

Thanks,
James

yes it would be a bigger gap if you upgraded your cooler. the delid makes your cooler choice more significant.

 

[mithiral67]

Yep, found an article by Gamer nexus that made me think the same thing. Most of the big temp changes are those pushing the cpu to the limit where the improved heat transfer will have the most benefit. Comparing at relatively stock speeds is likely to show lower deltas. Once, i got the room to the same ambient, the delta was 10c. At the upper end of OC, i am estimating it would be a 13-15c change, which is great imo.

Thanks all!!