@ ataya
This is what I am seeing, also, with my board. I've only seen 50 once and then it was not successful as it went into the loop with no post. I have even changed sticks and brands with no 50 code. Automatic mode may work, I do not remember looking if it did or not. I may need to use slower RAM and go from there.

I think it is BIOS related. Our memory is not supported yet, maybe? I run the G.Skill 4000 17-17-17. I am now able to run them at stock timings to 4300. MemTest shows errors on anything other than default and memory bench scores show no improvement from 4k-4.3k.

wwxww
Anyone know why my frequency BUS is not static? It's set to 100.00. I am trying to dial in this 86k same setup, just swapping boards to compare scores. Running 50 manual, target 1.95v -75%, offset 0, running 1.201-1.237v stable prime95
 
https://valid.x86.fr/51kn1w

XMP Profile maybe, but it is normal to go up and down .10 or .20.
99.93 is normal at least on EVGA Motherboards.

Ok, I just disabled XMP. So, does my vdroop have any relation to my Bclk frequency also?
 
XMP disabled running auto:    https://valid.x86.fr/ur6m72
 
 
edit: This is auto SAVED   https://valid.x86.fr/arubxu

That's going to be all for now with that 86K. The mail man just dropped off a package. In it is my other 86K. She is really nice. I only spent a few days with it all on the ASUS so I'll be able to do more comparisons vs The Dark. The IHS was messed with by the post office(I think) and it arrived in two pieces. Still, I slapped it together to play around with it before sending it back for re-lidding. The one above is a true 5.3G 86 running at 1.405v. It can do 5.4 but it takes 1.47+ and it doesn't like it very much. The chip runs it's best at 5.0,5.1, or 5.2. At 5.3 there are little gains and some losses. I think it's the IMC. The chip that came back today is a 5.7G on water, 86. It scored some ridiculous numbers on PassMark. It topped all 86's running at 5.3 or 5.4G and was only 13pts away for all processors record in single core. I hope it produces some good numbers.
Sorry for posting again and filling the thread up. I'm just really happy with this forum, board, and I just got my processor back. Thanks for listening and post em if you got em!
 
I want to thank my buddies over at OKC Overclocking for handing over these chips to me and for all the help.

Made the switch to the Dark. Amazing board, but is there any monitoring software that works well with it? Currently using hwinfo64 and it does not show vcore among a bunch of other things. e-leet works somewhat, but it'd be nice to have one program for it all. 
 
Did some vdroop testing with a volt meter; Delidded 9700k w/copper IHS LM @ 5.2ghz
Bios settings-- 1.330v -- -25% droop -- idle voltage @ 1.325v -- load voltage @ ~1.323v
Bios settings-- 1.330v -- -50% droop -- idle voltage @ 1.325v -- load voltage @ ~1.334v
 
So far I'm loving the board, just some kinks with the software. If I make any changes to the voltage or cpu multiplier in e-leet, it crashes pretty much immediately. Hopefully this gets works out.
 
*EDIT* Also to note, e-leet vcore was showing about .010v over what my meter was displaying. 
 
-Cheers 

Comparisons of Passmark scores ASUS XI EXTREME and EVGA DARK
 
ASUS    86K 5.4G   4 sticks Corsair 4000 cl19,  running 4000 cl19
https://imgur.com/YhnpMh4
https://imgur.com/n5v36Gu
https://imgur.com/GiSZcQq
https://imgur.com/l7KmPTu
https://imgur.com/vod9dNw
 
EVGA DARK   86K 5.4G   2 sticks G.Skill 4000 cl17,  running 4300 cl17
https://imgur.com/yqMRyan
https://imgur.com/QQk0onw
https://imgur.com/BRLCkEk
https://imgur.com/JLCUFGi
https://imgur.com/oOnRwbA
 
When using the ASUS, I was messing around. I had just received the processor. Using the EVGA, I put a lot of work or time into these scores. I use the PassMark for quickly testing changes made to the setting. This processor is not getting much love from the EVGA esp in the Physics and Prime # scores. I've tried everything. I now have my RAM running at 4400 17-17-17 and still get 45-6-7 for prime.
Are the scores low because of the board or because of 2 vs 4 sticks of RAM or other?
 
The processor is strong. I set it to 5.5. It was unstable and warm and late, I think. I ran this single thread. It was the fastest Single Thread I'd ever seen ran on PassMark.
 
https://imgur.com/qZ5fl1b
 
This was this morning. Last night I was successful at getting this RAM to run at 4400. Here is the first run of the morning. Cool and many things disabled.
https://imgur.com/UYG1gZZ
 
The ASUS was running Windows 10 PRO stock nothing disabled using full antivirus, glasswire, and etc, running all processes. The EVGA, I switched to W10 Home with no protection disabling a good bit of processes.
 
 
edit: this is a re-post pic of the EVGA with the first 8086 at 5.4 running 4300 GSkill before putting in the chip above. Soo, it does not to appear to be the chip as for the Physics and Prime.
https://imgur.com/QESfzG4
 
edit: Also, these are the very best scores from the DARK that I posted. I have many other really good scores from other programs, none to compare boards with. The ASUS had other PassMark scores that were higher than these posted.
 
edit: The Single Thread scores were much much more concistant and usually higher with the EVGA.

Delirious

ataya
So after testing for 2 days, my stable results; 5.3 ghz, 2 avx, 1.45 vcore, - 25% droop, my ram problem continues in a different situation. I am not expert in overclocking. But read a lot, try a lot. Learn from you guys how to push limits. So my situation evolved like this. At 5.0 ghz, 1.35 vcore, 1.4/1.4 sa/io, 1.6 dram voltage
 
I can work my rams stable at 3900, 15, 15, 15, 2T
 
Post but bsod on benching at 3900, 14, 14, 14, 1T
 
What ever I do system not posting at or above 4000 mhz. It is like there is a wall at 4000 mhz. 4000 mhz with 20 timings even not posting.
 
Could it be a bios issue or just my 9900k imc is weak? I pushed dram voltage to 2.0, sa/io 1.5/1.5, nothing changes. Same bootloop. System not even trying to boot :)
 

Read over kingpins article in the OC Lab area if you have not already.  What temps are you seeing at 5.3 on the CPU?

Could you link the article please?  I am not sure where to find it. Thank you big time

I'm in love with this board.  The voltage regulation has been so point and shoot for me.  I want 1.33v?  Enter 1.33v and -25% v-droop, and I get a steady 1.33v  +/-10mV regardless of load.  It has made dialing my overclocks so easy.  5.1GHz @ 1.33v with memory 4200 17-17-17-37 is 100% Prime95 and everything else stable.
 
12hrs of Prime95 w/ out AVX...

 
1hr Prime95 w/ AVX...

mattman657

TahoeDust
I'm in love with this board.  The voltage regulation has been so point and shoot for me.  I want 1.33v?  Enter 1.33v and -25% v-droop, and I get a steady 1.33v  +/-10mV regardless of load.  It has made dialing my overclocks so easy.  5.1GHz @ 1.33v with memory 4200 17-17-17-37 is 100% Prime95 and everything else stable.
 

Very nice. Same observations here with using -25% vdroop. Very stable vcore.

Ditto here with same and -25% vdroop, very nice - this Dark board is incredible!

Quick question from a relative noob. I used the OC Robot just to see what it would give me and to get a starting point. It spat out 5.2 GHz at 1.317 V the last time I tested. I ran 2.5 hours of xtu as a quick test and it passed just fine, with the package spiking to 77C max and the cores staying in the high 50's to low 60's most of the time. I suspect that Prime95 for 12-16 hours might be a different story though. I also suspect that since the vcore is still set to auto in the BIOS, the chip is being fed way more voltage than that under load, but I'm not sure where to look at that while testing. I guess my question is is the voltage the OC Robot spat out likely to be close to the right one, or should I expect to have to raise it and set the vdroop to -25% like everyone else has been once I set it to override? My issue is that I like the chip to clock down when idle, but my understanding is that override will keep the voltage and multiplier constant. It did on my z87 board; it ran at 4.5 GHz at 1.25V at all times. I sound totally ignorant, but I guess we all have to start somewhere. Any insight would be very much appreciated. Thanks.

Okay, using e-leet x, I was able to see that during prime95 small fft, vcore got up to ~1.37V. I feel like that is a bit too much. I guess I'll try to override it to with -25% vdroop and 1.33V to see if 5.2GHz is still stable. If not, I'll try 5.1GHz.

[image]file:///C:/Users/dad/Desktop/EVGA%20BIOS/clocked.png[/image]here is my socre, i know its not much .  But im trying to find a sweet spot with my set up, I have a EVGA Z390 FTW along with EVGA 1070 FTW, running 16GB ADATA 3000mhz, im not sure of what the model numbers is for my memory, never really looked at them.. 

https://hwbot.org/submission/4125330_

ddr4 4600