Ivy Bridge i5 3570K/i7 3770K batch and o/c results

[-JJ-]

I was not aware of the difference (AVX and non AVX versions). Now I got LinX from elsewhere, which seems to be right one and you were right about 10-15ºC more
Here's a quick run (cpu: 4,6ghz @1,18v; vccsa - 1,10v; vccio - 1,15v):
av.jpg

[andressergio]

#L209C345

decided to do it and man awesome almost 10C

see the after and before

almost 10C !!!





BEFORE

)



AFTER





here its what im more suprised before HS removal and change paste i launched prime at that volts and i got 94C to 96C look now



it also can do 2600 mem as shown above long live MIVE !!!

[LagunaX]

Do you just krazy glue it back on?

[andressergio]

Do you just krazy glue it back on?

no glue goes loose and retention mechanism holds it

amazing drop in temps

[C-N]

Very nice Sergio

[LagunaX]

Very nice - so bare delidded cores against the heatsink or did you redo the thermal paste and then put the lid back on?
Sorry to be a bug but just trying to see whether it is due to the lid or just the paste.

[babalouj]

Very nice - so bare delidded cores against the heatsink or did you redo the thermal paste and then put the lid back on?
Sorry to be a bug but just trying to see whether it is due to the lid or just the paste.

You replace the TIM under the lid and put the lid back on. The socket retention mechanism holds down the lid so there is no reason to glue it. It is Intel's thermal paste that is the problem.

[andressergio]

Very nice Sergio

Thanks mate !

Very nice - so bare delidded cores against the heatsink or did you redo the thermal paste and then put the lid back on?
Sorry to be a bug but just trying to see whether it is due to the lid or just the paste.

bug why mate ? im pleased to help

i took the IHS off just using what i showed gently on the corners and then moving around, then popped off clean that black thing over the PCB of the chip clean all with isopropyl alcohol and replaced the paste over the die, then just put over the IHS you will see it goes loose, so you put it on the retention mechanism it will move just be sure to align it good and thats it, and i did it with my mobo on lol i mean inside the chassis

cheers hope it helps !!!
Sergio

[Cecil]

Picked up another Costa chip. 3209B542. Much better then the Malaysia one I had, and at least as good as the other Costas Ive had. Will report back when I have more.
Also picked up a Z77E-ITX. Incredible board, regardless of size. Just wish there were more ram timing options in BIOS.

EDIT: Best Ive had yet. 4.6 LinX AVX with 1.2V. Still testing.

EDIT2: Got about 10 minutes in at 4.7 with 1.2V before getting an error, but not a BSOD. So Im thinking 1.22V or so for 4.7.

[eva2000]

Waves at everyone

Swapped over from 2700K to 3570K L208B104 cpu with 2x4GB Samsung 1600 30nm 1.35v memory on same motherboard Asus Maximus IV GeneZ.. looking good cpu clock wise

Just testing cpu first before clocking memory





First Ivy Bridge experience and first day with 3570K

[andressergio]

Waves at everyone

Swapped over from 2700K to 3570K L208B104 cpu with 2x4GB Samsung 1600 30nm 1.35v memory on same motherboard Asus Maximus IV GeneZ.. looking good cpu clock wise

Just testing cpu first before clocking memory





First Ivy Bridge experience and first day with 3570K

hey eva wellcome !!!

push the samsungs you will be surprised, and also if you can remove IHS will drop 10C !!!


im still on MIVE and pushed to 2600 CL12 no problems using what you see above

cheers !
Sergio

[andressergio]

Picked up another Costa chip. 3209B542. Much better then the Malaysia one I had, and at least as good as the other Costas Ive had. Will report back when I have more.
Also picked up a Z77E-ITX. Incredible board, regardless of size. Just wish there were more ram timing options in BIOS.

EDIT: Best Ive had yet. 4.6 LinX AVX with 1.2V. Still testing.

EDIT2: Got about 10 minutes in at 4.7 with 1.2V before getting an error, but not a BSOD. So Im thinking 1.22V or so for 4.7.

Cecil and 5GHz does it goes to game for ex ? at what volts

cheers and congrats hope you find the right one

[Cecil]

Havent tried 5, and not really worried about getting to 5 either. With the H100 and LinX, temps were at 85C already.
With gaming, you could under clock it and it would get the same FPS as 5 anyway.

Once under water Ill push more for a few benches, then probably settle in at 4.8 1.25V daily. Debating on removing the IHS.

[andressergio]

Havent tried 5, and not really worried about getting to 5 either. With the H100 and LinX, temps were at 85C already.
With gaming, you could under clock it and it would get the same FPS as 5 anyway.

Once under water Ill push more for a few benches, then probably settle in at 4.8 1.25V daily. Debating on removing the IHS.

well dont doubt it do it as 10C less you will reach easy what you want i even game at less vcore much less, im at your same 1,485 using LLC 75% stable prime at 5Ghz due to the IHS removal

[cab]

So... thermal paste from Costa Rica sucks?

ok I got an i5 3570k batch 3208C123, am I the only one?

[andressergio]

So... thermal paste from Costa Rica sucks?

ok I got an i5 3570k batch 3208C123, am I the only one?

all paste sucks i have a Malay and paste was horrible

[alex_tpc]

but dialed-down the rams to half the usual and used only 1 of 2 gpus. still too hot at max temp 89C.

using this instead 24x7: 4.9ghz @ 1.265V in bios:

[andressergio]

but dialed-down the rams to half the usual and used only 1 of 2 gpus. still too hot at max temp 89C.

using this instead 24x7: 4.9ghz @ 1.265V in bios:

great great chip mate congrats what batch is it ?

cheers
Sergio

[Splave]

this one on air.....will be getting ln2 later

3218B969

[alex_tpc]

great great chip mate congrats what batch is it ?

Thanks! Batch L215B620. Would love to get a further 10C off of the temps by removing the IHS and replacing the TIM like you did but don't want to risk it with this chip.

[trance565]

so i've gotten to 4.6 ghz at 2.65 stable(15 hrs of prime95, wish i took a SS lol), 4.7ghz takes 2.95 (1.304 load)to boot, and crashed after about 10 mins of prime.

Batch: 3218C290

anyways, I figure this is as good as any a place to check.

all I'm doing is setting my turbo multi to 46, and adjusting my cpu voltage, I'm not OCing anything else, my memory is running at stock 1600mhz.
Is that the correct way to do it? and then I read I can lower the PLL to help out with temps?

I lapped the ihs and my true 120(stock fan in pull orientation) and i haven't gone over 78* on core 2, and core 0 seems to run a good 10-15*(max 65*) cooler than 2 while 1 and 3 run about 5* cooler than 2. is that normal? I'm guessing they really botched the internal TIM job, so I was thinking of putting on some AS5.

anyways, back down to 4.6, i may just settle at 4.4, not like I'm doing anything crazy, but I do like to see what my chip is capable of.
Now to redo my prime at 4.6ghz and make sure it's still good and stable, i'll post a SS tmw morning probably.

EDIT: here's about 5 hrs or so of prime

[cab]

This 22nm technology should be running cooler but it's not, increasing fan speed has no effect, it seems that heat is not getting out of the cpu.

Idle no-load temps are jumping around by 10 degrees, 25-35+. Why are these temperature spikes happening?

[andressergio]

testing the Team X 2400 9-11-11-28 1T @ 2600 10-12-12-31 1T



Uploaded with ImageShack.us

i can do that but games crahes i think poor MIVE its on their limits lol either way its been doing a nice job

[alex_tpc]

bclock and IMC testing:
3770k bclock: 108.2 x 49 (5.301Ghz 2C2T)
ram: G.Skill Rip-X 2x4GB 2200c9 running almost 2600c10.
cpu-z validation


not quite sub-6 spi 32M (same rams at 2429c9)


and some more stability tests: this time with Linx@5.05Ghz 4C8T, 1.315v on bios:

[erickwok]

My best air 3770K chip


Attachment 127825

Attachment 127828

Attachment 127826

Attachment 127827