Intel 4790K OC Report ~

[Charles Wirth]

Intel 4790K ES C0
Cascade and LN2 limit testing

http://valid.canardpc.com/w90q4y

Single core: Super Pi 32M
Threaded: Cinebench 11.5
Asus MVIIG and Gigabyte z97x-OC Force
Corsair Dominator Platinum 2666C10
Corsair 240GB Neutron
Enermax 1500W MAXRevo

Thermalright 120 heat sink
4 Stage phase change cooler
K|ngp|n F1 LN2 pot

VID 1.104v
Boot into windows 1.25v @ 4.4Ghz

Mhz Vcore Cooling Input Temp Threads
4500 1.25 air stock 35c
4600 1.25 air stock 35c
4700 1.25 air stock 35c
4800 1.3 air stock 35c
4900 1.35 air stock 35c
5000 1.4 air stock 35c
5100 1.42 phase stock 35c
5200 1.44 phase 1.85 -109
5300 1.46 phase 1.88 -109
5400 1.48 phase 1.9 -109
5500 1.5 phase 1.95 -109
5600 1.52 phase 1.975 -109
5700 1.55 phase 2 -109
5800 1.6 phase 2.1 -109 4c8t
5900 1.7 phase 2.2 -109 4c4t
6000 1.8 phase 2.3 -130 4c8t
6100 1.85 LN2 2.4 -135 4c8t
6200 1.9 LN2 2.5 -140 4c8t
6300 1.92 LN2 2.5 -142 4c8t

1.95vcore = instant crash @ -140c
2.6v input = instant crash

I need to test the memory controller more
Memory controller maxed out a little over 1.4v -130c 5.5Ghz, 1.4v @ 5.2Ghz seems to be a close max stable

Cold bug factory defaults in bios at -120c
Cold bug 5.5Ghz boot -140c
Cold bug 6Ghz 4c8t idle -141c ~ loaded -150c

You gotta get chip warmer to get it colder
It clocks better and goes colder with 4c8t, removal of HT and cores makes cold bug worse.

This chip is binned for review sample, it should be close to what you get from retail.
With the new chip you get more Mhz on air and water but you can see the wall trying to reach the last few multilples on the top end.
This chip is better than any retail 4770K that I have tested in the past year.

Two of the major changes help with getting all retail chips to 5ghz on air, The TIM and the added resistor packs on the back of the CPU allow more voltage and that pushes the chip a little harder to push past 6Ghz.

I could see a lot more going on with cold and voltages, what was making the chip go faster and take more cold. Running all cores with HT and stressing the memory controller increased the internal load thus moving the cold bug from -120c to -140c. Example: Hyperthreading on and off is a 10c swing on the cold bug.

I can imagine without this new TIM and resistor packs that this chip would be a 5.6Ghz chip, so yes the improvements help the chips OC on the top end a little better vs the retails we have now, there should be a lot more 6Ghz+ chips to be found. The trade off is that it will take more voltage to push that chip up hill.

Giving the chip 1.8v and 1.9v seemed to be easy but at the trade off of max operating Mhz on all cores, so finesse must be taken to meet the temp/volt steps. My steps above in 100Mhz increments, the voltages and temps needed to make each step is greater than before for last multiples. The 4770K usually wouldn't give up the multiples when pushed too hard. I am sure we will see 2 volts again real soon. Whoever hits the 7Ghz mark will be sporting some crazy voltages.

I really hope that single stage phase change coolers can make a come back with this new 4790K, I am sure my chip could do 5.8Ghz 4c8t 24/7 and pull off 6.1Ghz screen for bragging rights.

Now that I know to push the chip it needs vcore and cooling = more Mhz. When you reach past 5.5Ghz you will need to add input voltage above 2v

My chip will do 5Ghz on air with a decent heat sink. I am using this one:
http://www.newegg.com/Product/Produc...-125-_-Product

If you need to see 5Ghz from your 4790K just crank the vcore to 1.4v to 1.5v in XTU, increase multiples to 50x with cupid open. Set it back and you are done. Do not expect 5Ghz air machines, you need water cooling for 5Ghz+ 24/7 operation.
No one promised 5Ghz air cooling processors, be realistic and settle for 4.9Ghz
My chip is an ES and I have been to 1.7v on air and would have tried 1.8v if I made progress at 1.7v.

http://valid.canardpc.com/b8wwe5

For serious overclocking a higher capacity heat sinks are needed, the water cooling segment will benefit and after many hours of plugging at temps and voltages I am back to the same conclusion that phase change is the way to go on this generation of processor to open it up fully.

Looking forward, one thing that I would like to see fixed is the multiples and controlling cores independently. AMD did a great job with this, I would love to have 4 cores enabled and leave 3 cores parked at idle and one core maxed out that I can set affinity to.

Thanks for the unlocked anniversary edition, this will bring endless joy to many more overclockers than before.

http://fugger.ipage.com/32mair.PNG

Thanks

[Ace123]

Nice. I need one.

[BeepBeep2]

Good on air/water...average on LN2 :/.

[Buckeye]

Very nice. Thanks Charles for the info !

[nacho_arroyo]

Nice info!!

Thanks!!

[Calathea]

Thx for the test. Looks like DC is rather overhyped. I'll probably stay with my good 4770 instead of wasting money for "new" gen.

[Charles Wirth]

Added stuff after I got some sleep.

[schmuckley]

4-stage cascade AND F1?
DJ got 6.5 on his..but..you know..it's ES

[FlanK3r]

nice, seems very simillar as my 4770K from LAM. I hope, with Broadwell-K chips we will see lower coldbug or without coldbug...

[Splave]

Team msi removed the ihs at the Intel event. The tim and everything else looks the exact same as Haswell.

[FlanK3r]

Ah.. Its new info for me...

[Charles Wirth]

Was the IHS removal successful?

[TheGoat Eater]

I think that some of the MSI boards were coming with shims for the IHS removal people this time around

[zalbard]

Team msi removed the ihs at the Intel event. The tim and everything else looks the exact same as Haswell.

Pretty hard to judge thermal paste by looks... But yeah, it's obviously not solder.

[Splave]

Yep success

[Dumo]

@ Splave...1 in 20 ratio again or worse?

[Splave]

IDK man out of 130 ES maybe 10 5ghz 1.4v on air cooling 4c8t so far, hope they get better

[lutjens]

Pretty hard to judge thermal paste by looks... But yeah, it's obviously not solder.

Soldered chips will have a vent hole in the IHS, chips that aren't soldered don't have this hole. I'm guessing DC is using a ceramique-like paste instead of the normal generic stuff...better, but still not ideal.

[darkage]

so air / water results will be the same ?

[Charles Wirth]

Updated again, added air stuff.

darkage - Air and water would be much different.

[darkage]

i know its the English you know :-)
i was trying to ask if it will be the same results average as the last haswell revision for aircooling and watercooling because people started to hope for a 5ghz devil
sorry man sometimes it wont come out as i am thinking :-)

[Charles Wirth]

Once you get over the vcore increase the 4790K will end up being the faster 24/7 OC chip
5Ghz is possible on more chips with water cooling for sure.

[PatRaceTin]

I need one with low VCore @ 5.0Ghz on water.

[PedantOne]

your chip has to be one from that 5 GHz five chips form 130 pcs in intel tray, because in all reviews is maximum 4,7 GHz on air, and my chip is with max around 4,6 GHz

[Charles Wirth]

Don't expect to get 5Ghz on air with a gimpy cooling solution.

Pushed my chip to 5Ghz for benches, dialing in memory more.

http://fugger.ipage.com/32mair.PNG