|Topic Review (Newest First)|
|08-22-2019 03:29 AM|
|b0uncyfr0||Well depending on which utility im using - its either a great success or it wasnt so good. Was it ever determined if Realtemp/others were better than Coretemp?|
|08-21-2019 12:10 PM|
Don't re-glue. The gap caused by the original glue is the main reason that delidding works so well in the first place.
that's why you don't put silicon/glue between the PCB and the IHS, you put the IHS on and then put a drop of glue on each corner. 0 added thickness = no gap.
|08-21-2019 01:16 AM|
|ttnuagmada||Don't re-glue. The gap caused by the original glue is the main reason that delidding works so well in the first place.|
|08-18-2019 03:03 PM|
|UltraMega||I had a 3770k that I delidded and use liquid metal and all that. In my case it actually didn't make a big difference. That said, I suggest you get a delid tool on ebay or amazon. You can get them for about $10.|
|08-18-2019 02:51 PM|
|The Pook||I'd probably tear it down again and redo it if it was me. the 14c delta between the hottest/coolest core is more worrying than the actual temp though.|
|08-17-2019 04:46 AM|
Edit: 45 minutes after P95 testing:
Core 1 : 75
Core 2 : 81
Core 3 : 86
Core 4 : 89
I forgot that CoreTemp wasn't the best way to check temperatures. RealTemp and HWInfo64 report the above numbers.
It seems abit high for 1.36v @ 4.8Ghz. Maybe contact between the IHS and chip itself isn't good. I am worried... perhaps i should reseat?
|08-16-2019 11:53 AM|
|The Pook||I'd probably draw the line at 1.45v for a 24/7 voltage on 22nm, 1.4v is plenty safe. I ran my 6400 and 6700 at ~1.425v for years without any issues.|
|08-16-2019 11:45 AM|
Results are in, I just run P95 for around an hour before and after:
1.32v at 4.7Ghz : 76-79 for Core 1 and 2. Core 3 and 4 are around 79-83.
1.32v at 4.7Ghz - 62-66 for Core 1 and Core 2. Core 3 and 4 are around 67-71. Last I remember, this unevenness is quite normal for 3770k's?
Its looking pretty good and i have to admit I'm impressed. I should done this wayy sooner. I think I have some headroom for possibly 4.9Ghz!
I am still worried about the voltage - what should be my cut off point at this rate? 1.35v seems pretty safe but would approaching 1.4v under 90 degree's be alright?
|07-25-2019 06:56 AM|
I am a big fan of Coollaboratory Liquid Ultra as a LM TIM.
Not that Conductonaut is bad but I prefer CLU.
Delidding is simple as can be with 3xxx 4xxx, 6xxx and 7xxx CPU's. I never even bothered buying a delid tool and just used the vice method with a block of wood and a towel to catch the PCB. Worked just fine and never damaged any CPU so far.
I didn't glue my IHS's back on either. Socket retention bracket will hold it just fine. Just be careful when closing the retention mechanism that the IHS doesn't shift when closing it. Don't forget to fully remove the black sealant from both the IHS and the PCB with a bank card or something similar. A knife will work but be careful not to hit the PCB.
For reference sake: I ran a 3770K @ 4.95Ghz 1.448v under a H100i and it stress loaded around 75-80c and games 60-65c ran for years like that, CLU never deteriorated.
|07-25-2019 06:19 AM|
|This thread has more than 10 replies. Click here to review the whole thread.|