Well, well ... the new motherboard and new memory seem to be playing nicely with one another ... a big sigh of relief that the CPU didn't crap out on me during all of this stuff.  I can now attest to the fact that Liquid Electrical Tape, if used to coat the SMB components to the left of the thermal plate on the CPU wafer, does not cause any problems at all.  As a matter of fact it may have saved my butt as there was a slight mishap during the mounting of the CPU into the motherboard.

 

To all of you who think you might want to buy the components to put together your own rig ... I can tell you right now that the components are pretty robust (with the exception of the pins in the CPU of the motherboard socket) ... I've tried on three occasions to be aggressively over stupid and the thing kept working anyway.

 

In any event, I now have a delidded and lapped CPU with Coollaboratory Liquid Pro underneath the IHS and on top of it.

 

To anyone who is interested I have an almost full tube of CLP here that I'll sell for $10 plus a $1.50 for shipping and handling.  This is one of the good packages.  The first one that I received from Frozen CPU did not have the little needle at the end of the syringe.  I can attest to the fact that the needle makes it infinitely easier to apply just the exact amount of TIM required.

 

I also have an almost full tube of MX-4 TIM (same price ... same deal) ... this stuff worked really well on top of the IHS ... I believe it will also perform just as admirably under the hood  ... and being non-conductive and non-capacitive and non-curing and non-corrosive and non-bleeding is a big relief when dealing with electronic components if a mistake is made.

As requested my 3DMark

 

Ice Storm:  164,946

Cloud Gate:  26,161

Fire Strike: 9,079 ... which is the overall score (better than 91% of all results.

 

Note, this is with no overclocks at all ... no mobo ... no video and no memory.

 

When I'm done messing around with all overclocks I'll redo the test and make a comparison.

I just did a test with the NGX and the 777 at KHTS (active runway) with no weather and the only scenery being FTXGlobal ... and a relatively mild overclock of 4.2GHz.

 

NGX - 28-29 fps

777 - 29-30 fps

 

This was very good news indeed.

As requested my 3DMark

 

Ice Storm:  164,946

Cloud Gate:  26,161

Fire Strike: 9,079 ... which is the overall score (better than 91% of all results.

 

Note, this is with no overclocks at all ... no mobo ... no video and no memory.

 

When I'm done messing around with all overclocks I'll redo the test and make a comparison.

 

 

Holy Smokes CT, pulling that off with no MB, VC and no memory! You just hold the CPU in your hand?

 

Hah! I know what you said though, it just sounded funny.

 

Glad its working out, looking forward to your final tweaks!

 

Cheers!

LOL ... that would be a feat wouldn't it :lol: .  I'm not sure if 9079 is a good score ... I think I remember people getting 11,000+ ... have to research that.

This is odd ... I went into the UEFI bios and simply changed the multiplier to 44 ... and Vcore set by the motherboard/BIOS is only 1.26 ... wonder how high this thing will go with no special tweaking?

Well, after a two hour Prime95 test the temps never went above 63C and the Vcore went up to 1.36V.  It looks like the IHS delidding and lapping did their job. 

 

Over the next three to four weeks I'm gonna push the limit as Nick completes his Bible.

 

The really nice thing about the ASUS Overclocking BIOS option is that I save a profile of stock parameters (@3.5) and use that for software development and program installation ... and then save another profile for overclocking.

CT 100mhz week , you really like this kind of work.

You seems to have a just normal Haswell retail CPU , 4.7ghz your Goal shall not be impossible if do the prime at 60 you add aprox 7C each 100mhz and 10c each 0.03v you probaly end up close to 90C at 4.7 then it depending un uncore ( cash) voltage if

You go for 4.7/4.4 or 4.7/4.5 if your lucky 4.8/4.5.

I run on water 4.8/4.5 at ramspeed 2400 with 2133 its possible uc4.6.

With the SS its a differnt ballgame the sweetspot for me is 5.1/4.8 2400 cl 8

I just ran OCCT for 5 hours at the max temps were 69C ... up 6C for that 100MHz gain (so your estimate for 7C / 100MHz is spot on)  ... Vcore stayed @ 1.36V (I have not yet optimized as per NickN's bible) ... going to try to up the NB to 4.1 today and see how that goes.  If all is well, I'll run this configuration for a couple of weeks and see how stable it really is with regular use.  I don't perceive this to be a problem as I'll only be running the overclocked profile for a couple of times a week ... all the rest of the time the CPU will run the stock 3.5/3.9.

 

Maybe I'll actually get a flight in ... haven't done that in a while ... which is ironic since that is why I went through all of this effort.

Charles you are scaring me! I am way behind you. I am still at the Non FSX programs installation of Nick's Guide. I haven't even Overclocked yet. For God sake go fly your sim. Enjoy it before it kills you :lol:

I have to say I have learned a lot about the inside of a computer ... and the BIOS ... I would have lost out on sooo much if I had kept that piece of crap from CyberPowerPC.  I have a feeling that this new computer will outlast the warranties on its component's parts.

 

The neat thing about this is that when setting idle at 4.5GHz ... without any load ... the VCore is at .736V.  I initially thought I would have to reboot to a 3.5GHz profile except when running FSX.  Upon further investigation the voltages being pushed to the CPU are minimal in this situation.

 

Hopefully, the future will show that either Prepar3d or XPlane (more likely due to licensing issues) will gain the support of the first PMDG product outside of FSX ... and let me take advantage of this hardware.

 

I can tell you this ... I won't be doing this again in the foreseeable future. 

I got to 4.4 ghz stable at 1.20v Is that considered a good chip?

I would say you are in the normal bell curve.  HLJames posted a range of voltages and frequencies on one of my threads ... it was this one ... here is what he posted.

 

Haswell FIVR  (manual above 4.6GHz) stable FSX:

 

4.3 - 1.168

4.4 - 1.264

4.5 - 1.296

4.6 - 1.312

4.7 - 1.348

4.8 - 1.392

4.9 - 1.44

5.0 - 1.49

 

On a completely different note ... I also fly with SouthWest Virtual which uses the wonderful PMDG 737NGX ... I just jumped into the cockpit for a short trip from MCO to EYW and bringing her up from cold and dark took about 5 minutes.  I believe I know that bird now ^_^ ... now I can't wait to understand the T7 like that.  But you can't put a 777 down at Key West though :huh:.

I had a crash in FSX using the stock settings on the motherboard and NickN's initial configuration ... now I've added his additional settings (see below) and am going to find out if it crashes again:

 

MIN CPU CACHE: 39

MAX CPU CACHE: 39

CPU CORE VOLTAGE: Manual Mode

CPU CORE VOLTAGE OVERRIDE: 1.29 (I used the upper range to minimize crashes ... will reduce incrementally as stability proves itself)

CPU CACHE VOLTAGE: Manual Mode

CPU CACHE VOLTAGE OVERRIDE: 1.28

CPU SYSTEM AGENT OFFSET MODE SIGN: +

CPU SYSTEM AGENT VOLTAGE OFFSET: .280

CPU ANALOG I/O VOLTAGE OFFSET SIGN: +

CPU ANALOG I/O VOLTAGE OFFSET: .050

CPU DIGITAL I/O VOLTAGE OFFSET SIGN: +

CPU DIGITAL I/O VOLTAGE OFFSET:  .100

INTERNAL CPU INPUT VOLTAGE: 1.82

EVENTUAL CPU INPUT VOLTAGE: 1.88

 

... off to Key West ... and arrived at Key West ... more stable this time.

What "initial settings" by Nick Needham did you use for your "stock" BIOS settings?