Hi

Her you see the diff , made a mistake did the runs with HT on.

I have changed the Thermal compound to Non Conduktive Phobya NanoGrease Extreme works really great.

Have the 4770K FPU stable @ 4.9 ghz now if of some intrest a can show the fpu and some 3DM06 tests a did

the simforum prefer that before FSMark11 for testing.

Hi

This therminal compound was very good i want someting that work in subzero conditions and not Conduktive

Thermal compound : Phobya NanoGrease Extreme.

 

All runs with Custom loop , 2X D5 pumps in series and 560 radiator push pull config

Ambient (room temp) 21C

CPU 4770k delidd,Thermal compound replaced with Phobya NanoGrease Extreme under IHS and waterblock.

 

Remark: This CPU is not a very good , its just normal or slightly better Haswell

 

Phobya-NanoGrease-Extreme

Aida64 FPU test 4.8 HT on

Aida64 FPU test 4.9 HT off

3DMark06 4.9 HT off

AIDA CASH MEM test

Never heard of Phobya NanoGrease Extreme ... sounds really good ... I have my CLP at the house ready for me when I get back from FL on Tuesday so there is no need to try another.

 

I've used Liquid Electric Tape to cover the SMB components to the left of the thermal plate on the PCB of the i7-4770K ... should be good when I drop the IHS back onto the wafer.

 

After all this I fully expect the components to just sit there and refuse to power up ... :( ... but time will tell.

 

Hopefully Microcenter will get my new board to me early next week so I can put all of this behind me. ... definitely B4 the 777 comes out anyway.

Got nearly identical 3DMark06:

 

41214 Marks

16671 SM2

21530 HDR/SM3

9284 CPU

 

i7 4770K 4.9GHz HT-Off

GSkill Ripjaws 2133MHz cas8

EVGA 780 GTX sc 967/3004

 

State of the Art LC

intel thermal interface material

 

HLJAMES

The only diff after delidd was that i had no trottling in Aida FPU test 3dm06 score was

Identical. It runs cooler but its no gain in MHz its to much vcore each 100mhz over 4.9.

This CPU runs 5.1 with 1.48v cooled with the SS , Max vcore for the SS is aprox 1.52-1,55 witch give 5.3ghz

Its a beast

Oh gawd ... I hate you both  !!!  After all I've gone through I'm probably going to get 4.7 at a Vcore of 1.99 ... of course requiring an LN2 rig ... now that would be fun!

 

 

Oh gawd ... I hate you both !!! After all I've gone through I'm probably going to get 4.7 at a Vcore of 1.99 ... of course requiring an LN2 rig ... now that would be fun!

Follow NickN's advice to the letter then assemble four 4000 RPM Delta fans on the radiator....4.7 GHz.

 

HLJAMES

Follow NickN's advice to the letter then assemble four 4000 RPM Delta fans on the radiator....4.7 GHz.

 

HLJAMES

 

Man, 4.7 is sweet but sky ru that, 4000 RPM.

 

I'll take the 4.4 lidded and same Intel TC at 750 RPM any day over all that racket, makes about as much noise as my Ipad.  :lol:

Yea ... can't do the Deltas ... already sent two of them back ... I'm going to have to settle with four of these Corsair Air Series SP120 High Performance Edition CO-9050008-WW 120mm Twin Pack High Static Pressure Case Fan ... the original fans that came with the unit.  However, there is a compelling reason for the Delta's ... a Delta puts out 40 db ... the Corsairs put out 36 db ... for just 10 more db those fans move twice the amount of air.  Now one thing I'm not sure of is this:  Are dbs additive?  In other words, if you have two 45db fans are you looking at 90db or are you looking at 45db?  Then of course, the Deltas are twice the price.  For an extra 200 MHz I'm not sure I want to go there.  I'd be happy with 4.5 steady but am hunting for 4.7 steady for only three hours ... the amount of time it usually takes to complete the short flights that I fly.

 

I plan to create two profiles ... the first with the default settings of the board ... which will be the dominant profile to be used with software development ... then the MSFX profile to be used a couple of hours a day for flying.

That link should have been Corsair Air Series SP120 High Performance Edition ...

 

Anyway ... I have a question about the push/pull arrangements that can be set up with the H100i.  Does the pull side of the equation really set up a low pressure area that allows the push side of the equation to move more air through the radiator?  If so, I wonder how much more of the estimated 62.74 CFM actually sees an increase?  Also, I wonder what the actual CFM is without the push/pull mod?

Push / pull works best. Experiment and your CPU will reward you or punish you.

When you find yourself pushing up the voltages and it is stil not stable, regardless of core temps, you have reached the limit of your cooling.

 

HLJAMES

Now you have me a steppin' and a fetchin' ... that logic doesn't seem to work (in my limited logical brain) ... if you have an awesome cooling system that keeps the temps around 75C when running a stress test ... and Vcore is a 1.4V then why would the cooling system be the limiting factor?

 

Wouldn't that be the CPU being the limiting factor?

Now you have me a steppin' and a fetchin' ... that logic doesn't seem to work (in my limited logical brain) ... if you have an awesome cooling system that keeps the temps around 75C when running a stress test ... and Vcore is a 1.4V then why would the cooling system be the limiting factor?

 

Wouldn't that be the CPU being the limiting factor?

 

I think thats exactly what he is saying, in other words cooling will no longer help as your chip has hit the wall.

Oh man ... everything arrives tomorrow ... can you believe I actually had to spend the day doing billable work?  How boring!

Good luck!

 

Do an FSXMark11 run when you're up and running and post your results if you can.