4770k delid any advice?

[R4DIO4CT1VE]

planning on de lidding my 4770k, curious if anyone has any experience with it recently, most threads ive been going through are from 2014

 

[dudmont]

I've not delidded that chip, but I've delidded several haswells(celerons). I used a razor blade. I attacked the corners, until I was able to feel the blade get through the glue that holds the IHS on. once you get through one side, the other's come pretty easy. When the blade gets through, be careful, there's some connections on one side of the chip, right by the die. And you don't want to hit those, or the die. What you use for a new TIM is up to you, but if you choose LMU(liquid metal), be very careful. I tried it, created a nice little paper template, that fit the die pretty tight. I still managed to get some LMU on those connectors and fried the chip. If I were ever to try it again, I would do a bit more research on the best method(perhaps tape?) of protecting the chip.

 

[USAFRet]

Why delid?
IMHO, that should only be done as an absolute last resort.

For instance, when your current cooling situation is hovering around 95C, it is actually thermal throttling under whatever load you're doing, and you've done everything else.

So what temps are you seeing on this CPU?

 

[dudmont]

Why delid?
IMHO, that should only be done as an absolute last resort.

For instance, when your current cooling situation is hovering around 95C, it is actually thermal throttling under whatever load you're doing, and you've done everything else.

So what temps are you seeing on this CPU?

While I completely agree with you on this, this particular chip had some issues, maybe his overclock is being held back by temps rather than max voltage? *Shrug*
I certainly hope he's not doing it for funzies. I can assure that killing a perfectly good chip(even if it was merely a cheapy bottum of the stack celeron, in my case) with miss-application of TIM is not something that should be done lightly.

 

[USAFRet]

Why delid?
IMHO, that should only be done as an absolute last resort.

For instance, when your current cooling situation is hovering around 95C, it is actually thermal throttling under whatever load you're doing, and you've done everything else.

So what temps are you seeing on this CPU?

While I completely agree with you on this, this particular chip had some issues, maybe his overclock is being held back by temps rather than max voltage? *Shrug*
I certainly hope he's not doing it for funzies. I can assure that killing a perfectly good chip(even if it was merely a cheapy bottum of the stack celeron, in my case) with miss-application of TIM is not something that should be done lightly.

And that's why I asked 'Why'.

 

[iamacow]

If you are not sitting at 85c @ 1.4v than you don't need to delied. It is simply that your CPU cannot overclock any higher. Most Haswells are limited to 4.5ghz under safe voltage, which is 1.35v. Very few can do 4.8ghz and those almost always need 1.4v.

 

[R4DIO4CT1VE]

Why delid?
IMHO, that should only be done as an absolute last resort.

For instance, when your current cooling situation is hovering around 95C, it is actually thermal throttling under whatever load you're doing, and you've done everything else.

So what temps are you seeing on this CPU?

hitting 90c while stress testing, 4.1ghz at 1.101v

 

[USAFRet]

Why delid?
IMHO, that should only be done as an absolute last resort.

For instance, when your current cooling situation is hovering around 95C, it is actually thermal throttling under whatever load you're doing, and you've done everything else.

So what temps are you seeing on this CPU?

hitting 90c while stress testing, 4.1ghz at 1.101v

And your current cooling situation?

 

[dudmont]

Be as specific as you can, TIM, cpu cooler, case, extra fans, etc. etc.

 

[iamacow]

I think you need a better CPU cooler if you are at 90c @ 1.1v. Deliding won't help you until that is taken care of.