4.8ghz + Haswell

H

hywdx80

Gawd
Joined
Jun 28, 2006
Messages
549
Anyone know where to buy used / tested and stable 4.8ghz 4770k's (1.35v or less)? . I've been through 3 chip so far and none of them can even pass 4.4 @ 1.35v. I don't mind paying extra for one.
 
rfisher983 said:
Anyone know where to buy used / tested and stable 4.8ghz 4770k's (1.35v or less)? . I've been through 3 chip so far and none of them can even pass 4.4 @ 1.35v. I don't mind paying extra for one.
Click to expand...




Z87 chipset is a wreck and I think part of the problem with over clocking 4770ks is related to MB.


I just read that Asus has a new update for micro code 9h for mother board that made a 200mhz difference in over clock.
 
Could I request for your source? Rolled back my BIOS upgrade but this 9h is pretty interesting.
 
Getting to 4.8+ requires delidding to control temps unless you got lucky with a golden sample. Even after a delid, 4.8 is probably a top quartile sample. If dont want to chance it, you could always post a wtb thread and pay a little extra.
 
The newest bios did nothing for me.

3 chips / different batch # all max out at 43x at 1.25. 4x4 at 1.325v. So sad looking at other people hitting 4.8+ at 1.2.
 
4.8 at 1.2v must be rare.

I am "gaming stable" at 4.7Ghz and 1.335v, need to test further to see if I can get the voltage lower. Malay batch # L312B323, research indicates this is a pretty good batch. Tried pushing her to 4.8 @ 1.36v and BSOD, I won't go any further without a delid which I have no intention of doing at this point.
 
Last edited:
You're going to go through a lot more chips at this rate. A frequency almost no one gets at a voltage lower than what most need for 4.4GHz? Good luck
 
NukeDukem said:
4.8 at 1.2v must be rare.

I am "gaming stable" at 4.7Ghz and 1.335v, need to test further to see if I can get the voltage lower. Malay batch # L312B323, research indicates this is a pretty good batch. Tried pushing her to 4.8 @ 1.36v and BSOD, I won't go any further without a delid which I have no intention of doing at this point.
Click to expand...

Ha. I just picked up that batch number. Whats your mobo?
 
OP what motherboard are you running?

I'm running 4770k 4.5Ghz @ 1.2v on a Asus Z87-Pro. I cannot go any higher due to temps however.
 
rfisher983 said:
Anyone know where to buy used / tested and stable 4.8ghz 4770k's (1.35v or less)? . I've been through 3 chip so far and none of them can even pass 4.4 @ 1.35v. I don't mind paying extra for one.
Click to expand...

I feel this :( I read these reviews of 4.8 stable at 1.25 (this site I think actually) and I am currently at 1.35v and still not stable at 4.7. (worth mentioning ive had this pc for a few months and have yet to USE it due to spending every day all day so far trying to get a 24hr stable prime run. (passes everything else naturally, but since the community Is obessed w/ prime being the end all be all, I feel obligated to do the same)

This is delidded and ran naked with a beast of a watercooling setup and on an almost 500 dollar motherboard. (of course if what the prior poster said about the bios updates making that large of a difference, I hope gigabyte gets with the program. F7 has been the bios since I bought this board back in june/july)

I can post at 5.2 with 1.375v, I can post at like 4.6 with 1.25 np.... all the proper "guidelines" that are typically used to discern a good chip from bad. (stock vid 1.0 even) and its supposedly a "good" chip but nothing and no amount of voltage will get it prime stable (only blend, its stable on small fft) at anything past 4.6

when I say nothing I mean nothing. Ive upped the voltage to 1.55 in which I am still under 90c in IBT (that's how good of a watercooling setup im running) but still no prime stable.

I know its not temp, and its apparently not power... so im at a loss. If the chip was at its "wall" persay, why would I be able to post at much higher speeds, with much less voltage? Something is giving out and im not sure what. Its always 1 of 3 things, clock machine blue screen, whea blue screen, or freeze. Which are usually countered with vcore, vdram, or vring respectively.

Im super super close to dialing it down to 4.2 and saying screw it. Its a tough pill to swallow when I spent so much building this pc in order to have as good as possible a shot at overclocking.

/cry
 
NukeDukem said:
4.8 at 1.2v must be rare.

I am "gaming stable" at 4.7Ghz and 1.335v, need to test further to see if I can get the voltage lower. Malay batch # L312B323, research indicates this is a pretty good batch. Tried pushing her to 4.8 @ 1.36v and BSOD, I won't go any further without a delid which I have no intention of doing at this point.
Click to expand...

let us know what you get prime stable :p
 
What is regular stable? If you're failing prime, you're not stable, regular or otherwise. In fact, I've seen cases where prime can pass for hours on end, but the system still needs more voltage or lower clocks for some other things to be stable.
 
Funny story (not really, very saddening actually) - my 4770K was solid at 4.6 with 1.25V. I just attempted a delid with it and freakin broke it. Just 2hrs ago. Done. The end. Sad stuff.
I'm 90% sure I will not be able to find another one like it.

Going to Microcenter tomorrow and hoping for the best.
 
Brahmzy said:
Funny story (not really, very saddening actually) - my 4770K was solid at 4.6 with 1.25V. I just attempted a delid with it and freakin broke it. Just 2hrs ago. Done. The end. Sad stuff.
I'm 90% sure I will not be able to find another one like it.

Going to Microcenter tomorrow and hoping for the best.
Click to expand...

Details on how you managed to break it?
 
That sucks (about the failed delid attempt)... :(

Haswell has been a MAJOR disappointment for me as well. You cannot even rely on batch numbers anymore. It's way too random.
 
Too bad. I used the vice method on mine. I'll admit it took A LOT more force than I thought it would.

Once I crossed the threshold of not caring if I broke it, my backswing opened up and I was able to drive it pretty far. ;) Then I learned to put cushions around the vice.

I'm still experimenting with the clocks and voltages. (Damn adaptive crap; Grrr.)
 
what a lot of people here fail to know is that there are other factors to why the CPU wont reach high OC

it also has to do with MOBO, RAM, Power Supply, BIOS, OC Experience, etc.

it sucks to watch some of you here thinking its just the CPU.
 
Last edited:
That's why I won't delid. Fuggg dat.

I'm happy with my 4670k sitting at 4.6Ghz 1.295v. Prime temps hit just over 80c. Sure I could push it further if I delidded, but don't want to risk cracking the core.
 
TroyX said:
what a lot of people here fail to know is that there are other factors to why the CPU wont reach high OC

it also has to do with MOBO, RAM, Power Supply, BIOS, OC Experience, etc.

it sucks to watch some of you here thinking its just the CPU.
Click to expand...

When it comes to Haswell. It's almost always the CPU.
 
Just got back from MC a few hours ago. Found another 4770K in the exact same batch I had and I would say it's at least as good as the one I had. Seems to be a little bit better heat spread across the cores, VID is 1.008V according to the BIOS and it rocking 4.6 like the old one did. Delidding ain't for me - I'll gladly stick where I'm at!

I also got a pretty killer 4670K for my HTPC (replaced a 4570S). Real low VID and runs real cool.
Happy day. WHEW. That was an expensive lesson.
 
TroyX said:
what a lot of people here fail to know is that there are other factors to why the CPU wont reach high OC

it also has to do with MOBO, RAM, Power Supply, BIOS, OC Experience, etc.

it sucks to watch some of you here thinking its just the CPU.
Click to expand...

With 15 years on scene, an oc force, some corsair dram, a 1500w psu, the only bios for the board... Unfortunately I wish I could find something else to blame it on. Today with stock uncore I was able to run prime for about an hour at 1.5v (OCD to 4.8) but that's far more vcore than I care to run, plus if I upped the uncore ratio I'm sure it would become even more unstable . It's extremely likely it would have crashed a few hours later at best anyhow.

Fingers crossed, I think I may have 4.7 prime stable with 1.43v. Still high but ill run it 24/7. Temps stay at around 72 via IBT. Just a shame it takes that much and a shame we all still count on prime. And a shame haswell turned out as it did IMO. Great mobile chip or chip for the end user, for enthusiasts.... I think not.

To the guy who killed his, if I was him... I'd wait for the best 1150 chip early next year :(
 
RamonGTP said:
Why so secretive about how your delidding went wrong? lol
Click to expand...

Huh? No secrets here. I have no idea what went wrong. It just doesn't work. I got the dreaded solid-red CPU light after about 10 TIM reapplications and reseats. I took some close up macro shots of the PCB, die etc. Everything looked good, except a very, very small chip off a corner of the die. I've had far larger chips on die corners back in the Athlon days where everything was direct-die. So either the Haswell dies are very delicate/sensitive or the PCB got bent or something - no idea.

I thought I may have bent a pin on the mobo, but under magnification, all pins looked fine and the new chip obviously proved the pins were fine.
 
LOCO LAPTOP said:
OP what motherboard are you running?

I'm running 4770k 4.5Ghz @ 1.2v on a Asus Z87-Pro. I cannot go any higher due to temps however.
Click to expand...

What of you have for cooling? At 1.2v you should have no problems with temps.
 
Not delidded I assume? That could potentially afford u 4.6. Course u would lose your warranty but who has ever warrantied a cpu? (Generally speaking)
 
Bah! I knew I should have waited.
I delidded, amd was only able to manage 4.6.
I was at 4.5 before the delidding.

I scared the crap out of myself while I was delidding though. I used the razor blade method and put a scratch in the top, just enough to uncover some copper through the green. Got lucky and it still works.
 
ItsTooHot said:
Bah! I knew I should have waited.
I delidded, amd was only able to manage 4.6.
I was at 4.5 before the delidding.

I scared the crap out of myself while I was delidding though. I used the razor blade method and put a scratch in the top, just enough to uncover some copper through the green. Got lucky and it still works.
Click to expand...

I'm going to try the 40ft drop and sledgehammer method. :p
 
rfisher983 said:
The newest bios did nothing for me.

3 chips / different batch # all max out at 43x at 1.25. 4x4 at 1.325v. So sad looking at other people hitting 4.8+ at 1.2.
Click to expand...

You are looking at the 1%. A Haswell 4770k that can hit 4.8 ghz or higher stable is very very rare.
 
Brahmzy said:
Huh? No secrets here. I have no idea what went wrong. It just doesn't work. I got the dreaded solid-red CPU light after about 10 TIM reapplications and reseats. I took some close up macro shots of the PCB, die etc. Everything looked good, except a very, very small chip off a corner of the die. I've had far larger chips on die corners back in the Athlon days where everything was direct-die. So either the Haswell dies are very delicate/sensitive or the PCB got bent or something - no idea.

I thought I may have bent a pin on the mobo, but under magnification, all pins looked fine and the new chip obviously proved the pins were fine.
Click to expand...

He is asking you specifically the method you performed to delid which presumably caused the chip in the die...
 
You must log in or register to reply here.
Back
Top