• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Delidded my 4670K What Thermal Compound

Lord_ZED said:
arad85 TRUE but with those ivys and haswells. If you got High end air cooler in room 20c you get 80c on cores and with high end watercooling u get 79c ........ And with cheap 30f air cooler you get 81c.
If your CPU got ****** applied thermal compound inside you are screwed. Or delid and lose warranty or deal with 80c on cores :(
Click to expand...
Which means the thermal resistance of the system as a whole is dominated by the thermal compound.

I haven't tried it, but I suspect your figures are a tad compressed though (i.e. you can get more of an overclock at higher temps with better heatsinks, just the max overclock is lower than you'd like and limited by temps)....
 
arad85 said:
Ambient does have an impact on temps. What the heatsink does is give you a delta between junction and ambient. Say you are running flat out and your die tems are 70 deg C in a room at 20 deg C. The heatsink is "dropping" the 50 deg C (70 - 20) to ambient.

If ambient goes up 10 deg C, core temps will go up 10 deg C since the heatsink is only capable of dropping 50 deg C....
Click to expand...

So you're saying I won't notice a difference in room temperature between the 2500k and 3770k?
 
wazza300 said:
idk,haswell seems to be different in overclocking and different again on gigabyte boards,sin is about to release an overclocking guide for haswell/gigabyte over on tweaktown forums so ill post a link up when its posted

it might help you understand what all the voltages do ect
Click to expand...

Thanks!. Just seems odd that the more voltage I added to CPU Offset the quicker the system BSOD when encoding
 
se_vietboi said:
So you're saying I won't notice a difference in room temperature between the 2500k and 3770k?
Click to expand...
No.

You have to think of this as a system. To get the 2500K to a particular frequency needs a certain power input. The heatsink (which includes the thermal junction between the chip and the lid as well as between the lid and heatsink) will get hotter by a certain amount for that power input. If you make the room 10 deg hotter, the on chip temps will get 10 degrees hotter for exactly the same input.

Moving to the 3770K, to get to the same frequency will require a different amount of power - it may be more, it may be less. Also, the thermal link between die and lid is not as good on Ivy, therefore the heatsink will be less effective. So, depending on what frequency you are trying to achieve (and hence what power input) your input power may be more or less than Sandy. Whether you will see a difference in tempos when comparing to Sandy depends entirely on what you are trying to achieve with Ivy vs Sandy.

What is true though, is, for a given thermal solution, your CPU temps will be higher the warmer the room is.
 
what batch # was your cpu lettuce?

from what im hearing asus/gigabyte seem to clock differently idk how true that is
 
arad85 said:
No.

You have to think of this as a system. To get the 2500K to a particular frequency needs a certain power input. The heatsink (which includes the thermal junction between the chip and the lid as well as between the lid and heatsink) will get hotter by a certain amount for that power input. If you make the room 10 deg hotter, the on chip temps will get 10 degrees hotter for exactly the same input.

Moving to the 3770K, to get to the same frequency will require a different amount of power - it may be more, it may be less. Also, the thermal link between die and lid is not as good on Ivy, therefore the heatsink will be less effective. So, depending on what frequency you are trying to achieve (and hence what power input) your input power may be more or less than Sandy. Whether you will see a difference in tempos when comparing to Sandy depends entirely on what you are trying to achieve with Ivy vs Sandy.

What is true though, is, for a given thermal solution, your CPU temps will be higher the warmer the room is.
Click to expand...

Oh yes I understand all that. But I wanted to know something else but found the answer to it anyways thanks.
 
it will be on the box by the barcode

they used to have a bunch of numbers with a letter a or b,not sure if that's changed

seems a good clocking batch your chip came from

ohh and tried with 1.10v cpu ring voltage? might need it with 16gb of ram

think cpu ring is same as cpu/vtt on ivy,stock 1.05v usually isn't enough for 16gb,8gb its fine
 
Last edited:
arad85 Thanks for writing it down in proper English im not that good in explaining technical things in English heh...

Anyhow in my case (watercooling) I am limited by Thermal Compound in IVY and Haswell not by my cooling eficiency. Only thing that coud get me lower temps is chilled water or upgrade from Rasa to Raystorm block :P

All in all i would not spend more than 30f on cooler for ivy or haswell. Cause i cant see a pint.

Anyone seen some cooler tests at [email protected] haswell ?? Just curious what results would we get...
 
se_vietboi said:
Ok my 3770k is here. Now off to the shops to buy a razor. I still think I'm too chicken to go through with it though :(
Click to expand...

It is not as scary as it first looks, when I got my 4770 I thought there is no I am going to do it but after a few days I gave it a try. Use the razor gently under the corners rocking it, I had trouble getting under the edges on the Haswell so after getting under the corners I finished it off in the vice, just 2 gentle taps with the hammer with a piece of solid wood, just take your time, end result is worth the sweat :-)
 
Last edited:
Ok so I bought a razor for nothing because I didn't go through with it. Well I tried to. With a mallet. LOL.

The edge of the IHS was placed alongside the edge of a desk and I proceeded to smash it off. After every few hits I'd inspect the chip to see if there were any physical damages. If not I'll proceed again. Needless to say after 30 hits each with progressively more force which resulted in the wooden edges of the desk breaking off, I thought I'd call it a day. Luckily, it appears the chip isn't only undelidable but also indestructible as it functions with no problems and is running at 4.2GHZ @ 1.19v 70-77c.
 
What did that cpu ever do to you?!!! Send it to me before ya attack it with a desk again :)

Like smerriman says once ya get the blade under the corners the rest is easy, ya can even use wazzas technique of cutting a thin piece of plastic and use that to finish it off, that way ya cant scratch the pcb.
 
I cut just one corner,rocking the blade back n forth till it made a wide enough gap,then I inserted a cut down plastic take away meal lid and used that the slice through the remaining sides,no nicks with plastic
 
se_vietboi said:
Ok so I bought a razor for nothing because I didn't go through with it. Well I tried to. With a mallet. LOL.

The edge of the IHS was placed alongside the edge of a desk and I proceeded to smash it off. After every few hits I'd inspect the chip to see if there were any physical damages. If not I'll proceed again. Needless to say after 30 hits each with progressively more force which resulted in the wooden edges of the desk breaking off, I thought I'd call it a day. Luckily, it appears the chip isn't only undelidable but also indestructible as it functions with no problems and is running at 4.2GHZ @ 1.19v 70-77c.
Click to expand...

I had this problem too!

I was using a vice, but it seemed no amount of hammer force would take the lid off. After breaking a chip out of the piece of wood I was using, I chickened out and was sure I'd ruined my chip.

Luckily I just installed it like normal and everything seems to be okay, too bad the temperatures suck though.
 
Ok I tried with the razor. I didn't even get under the IHS yet there was a little scratch on the PCB..at that moment I thought "WTF! The blade never once pointed down towards the PCB". I changed razors just in case it had imperfections and started working on a different corner because if things went wrong, I didn't want to make the nick worst. Surprise, the new corner got scratched! The sides of the razor was scratching the PCB as I was moving it left and right on its axis.

At this point I was pretty ****ed off and went full YOLO. I grabbed the mallet from the mornings' failed attempt and doubled the amount of hits and force from before. It obviously didn't come off and I was sure from all that hitting and scratching it was bound to be broken so I might as well dump the CPU in the bin. When I was calm I reinstalled the CPU and thankfully it still works. Never going to try delidding again.
 
I was thinking of trying again at another time, but judging from your experience I'm not sure if I want to now.

I noticed in some videos on Youtube that people rotate the processor 90 degrees every few hits when using the vice method. Did you try doing that?

Looking closely at my 3770k, the IHS is so close to the board that it's as if they're actually attached. I don't think there's any way I could get a razor in there.
 
You must log in or register to reply here.
Back
Top Bottom