They have a thread at Hard OCP and From what I'm reading around the net the retail chips don't overclock well. People that own it make it sound like its pre OC 4770k clocked to 4.4 and they don't have much room to clock higher.

I hope it's just the first shipments.

rmorse27
that would suck after all the hype!

That's exactly what it was, nothing but hype. I have mine stable at stock clocks of 4.4 GHz and need a vCore of 1.250 I meant 1.125. I tried upping the multiplier to 45 this morning and it promptly froze up with code 124. I couldn't even get back into the BIOS to change it back. 5.0 on air? Not gonna happen.
 
EDIT: My mistake, my vCore is only 1.125 and vDroop takes it to 1.144 under load. My bad.

outhouse
4.7 is average once you go through all the threads. On water.
 
Not to many are willing to shove some voltage through it.
 
My 4770K would have stayed at 4.2 at 1.25v   but now its eating 4.5 at 1.4v temps low.
 
 

Yeah, 4.7 looks about the highest it will run stable. Scotty Mac sent me this pic from an overclocking thread somewhere unknown to me. He's getting 4.8 using higher voltages but look at his temps. OMG!!! That's hot! I haven't gone over 72° using XTU or IBT.
http://puu.sh/9NR41/b4cc4494b1.png
 
 
These are my scores and I'm pretty close to him in performance. BTW, my 2600k at 4.5 GHz only produced 107-108 GFlops.
 

 

Zuhl3156

outhouse
4.7 is average once you go through all the threads. On water.
 
Not to many are willing to shove some voltage through it.
 
My 4770K would have stayed at 4.2 at 1.25v   but now its eating 4.5 at 1.4v temps low.
 
 

Yeah, 4.7 looks about the highest it will run stable. Scotty Mac sent me this pic from an overclocking thread somewhere unknown to me. He's getting 4.8 using lower voltages but look at his temps. OMG!!! That's hot! I haven't gone over 72° using XTU or IBT.
http://puu.sh/9NR41/b4cc4494b1.png
 
 
These are my scores and I'm pretty close to him in performance. BTW, my 2600k at 4.5 GHz only produced 107-108 GFlops.
 

 

Run the same test but set to Very High, Then run at least 3 runs on Maximum *or 1GB less than the memory installed.
I never had or have seen an issue on Standard or 1024 MB

Zuhl3156
 I'm not a Sadistic CPU abuser.

 
I am :)
 
I have always gone a hair past what is considered really high.
 
To date, I have never lost a single CPU.   All the way back to overclocking P3 coppermines. 

I increased the multiplier to 47 this morning and everything is still stable at 1.260v. Max temps were 67°C at 100% load. If I remain stable for the rest of the day I will try a multiplier of 48 at the same voltage tomorrow.
 

 

4.8 GHz isn't happening without more voltage so it looks like 4.7 is my limit for now. 5.0 on air? LOLOLOL

Its literally impossible to get these passed 4.7
I've played with 2 4790k, 2 "pentium k"s and a 4690k. all of them literally behave the same way, up to 4.5 its a breeze low voltages, low heat. 
 
4.6 requires a good bump in voltage, but easily manageable. And 4.7 Ghz 24/7 stable has been impossible for me.
 
I've had the plesure to mess with MSI, gigabyte, and an asrock. Tried messing with  with all voltages, no luck. 4.6 with BLK of 101.2 is max I've gotten. with huge voltage increases in the PCH and Vin.
 
 
The decrease heat is a welcome, but something either physicaly stopping them from going higher or intel has put limit on these.
 
 
A friend is getting the gigabyte OC force mobo, I am hoping it has some extra voltage capabilities.
my 100 dollar msi PC mate has almost all voltage settings as anything else I've worked with tho, which is def. a nice touch.
 
 
I rem using 40W nedic fans on the EVGA cooler back in the day to join the 5ghz club, I was really hoping we can do this once more. Darn you entraphy! 
 
 

OK
 
Got a 4790K today myself  K41 batch
 
4.7 at 1.20 no problem.  Boots into windows at 5G but requires 1.425
 
Sitting at 4.8 at 1.325, and did a test run and 3rdmark 11 makes it all the way through.
 
Could not get a full run in due to temps at 4.9 BS at the end of the physics test next to last.
 
Once this break in after a week I will get it to 4.9 easy peasy.    After 2 hours 4.8 is a hair above average.
 
I swear I may have to put the old danger den RBX back in that kept full loads 10C cooler then this EK block that sort of sucks.
 
The GPI blocks hands down very sweet, CPU not so much. 

rmorse27
I overclocked my I7-3770k from 3.5 to 4.5 with no problems,so I figured I would be able to overclock my new i7- 4790k from 4.0 to 5.0 with ease wrong not even close 4.6 with 1.26 voltage is what I got tried 4.8 with 1.28 volts no go and dont really want to go over 1.30 volts very unhappy with results so far,I am on water by the way.

 
I think I can get to 5 on my 11 year old RBX, not this crappy EK block unless I mod it.
 
 
That voltage is low.  You can throw 1.45 at this all day long and not degrage the chip as long as temps are in check  
 
 
Im limited to 1.325 because that is all that poor EK waterblock will handle.  Idles at 28C and prime 95 or 3dmark jumps the load temp up to 90C if I go up in voltage.
 
 
I know its not my cooling system, because temps rise instantly.   
 
 

rmorse27
maybe I will try high voltage tonight thanks for insight

I was going through threads just now at overclockers.com
 
Some of the other seniors are saying they have seen degridation at 1.4 
 
oher searches have said on 4770K anything under 1.45 is oK
 
I think the key is temps.  If you can keep the cPU under 80C at load, you an get away with more voltage.
 
 
For 20 years I have beaten on CPU's always throwing more at them then limits state. To date, I have never lost one chip, usig murderous voltage. But always had the best cooling and never let the chips get too hot.  

Deidding is sort of a must on these chips.
 
What is limiting these is how hot they get. Mine running prime95 smalls will run it up to 99C at 1.325v. And that is on water 120 X 5 
 
The trick is using nail polish on the vrm's, and then using CLP or CLU on the die.  This can drop temps down 20C ish
 
Regular tim just doesn't cut it. It will make no change and thus delidding would be a waist of time.    

This guide is pretty detailed. It's what I used since I'm not familiar with all of the new BIOS settings for Haswell: http://www.overclock.net/t/1411077/haswell-overclocking-guide-with-statistics
I would set your Uncore to 'Manual' then the Multiplier to 40 at 1.15v to keep heat down. I increased mine, call 'Ring bus' by MSI, to 42 and I'm still stable.

I thought the whole reason they released the i7-4790k was because of better thermal compound they supposedly applied to help keep temps down?

Zuhl3156
The new thermal compound is very good.

 
It is very good.     It is above average but no where near the best.
 
 

  Delidding is not necessary.

 
 
Its not.   Overclocking is not necessary, but we all do it.
 
But the object here is to get the most bang for your buck.
 
Either you want lower temps or you don't, and delidding with one of two tims will lower temps dramatically.
 
 

  
 
  My temps at 4.7 and 1.260v are lower than my 2600k at 4.5 running 1.356v.
 

 
 
Run the latest batch of prime95 small fft's and report back how high they go to instantly. 
 

   Most are requiring 1.310v to reach 4.8 GHz and I'm really not seeing the increase in voltage and temps as justifiable to shorten my CPU lifespan for so little of a performance increase. 

 
 
You will never wear out a CPU with that low of temps and voltage.  Even at 1.45v if you could handle the heat would not degrade the CPU before it would be outdated.