Overclock.net banner

1 - 20 of 301 Posts

·
Optimal Pessimist
Joined
·
3,047 Posts
Discussion Starter · #1 ·
I am running a 24x7 overclock at 4.7GHZ (4.3 uncore) @ 1.24v and memory overclocked from 1866 MHz to 2400 11-13-13-31-2T.

First to note, the 23h microcode has patches for Spectre (see the powershell reports at the end). I have been running 23h for a little while and I have noticed some stability issues. There is also a definite performance hit between 22 and 23.

I compared performance (Realbench 2.43) and stability (prime 95 28.10 @ 1344/1344 / 10GB memory) for 15 minutes. My method was:

1. I saved my OC BIOS profile to USB
2. Modified the BIOS for the microcode version I was testing
3. Flashbacked the BIOS
4. Set the BIOS to optimal defaults, rebooted
5. Loaded the saved OC BIOS and rebooted
6. Ran the tests

For both microcode versions I am running KB4056892 Windows 10 v1709 Build 16299.192 which is the Intel security patch. So performance differences are strictly due to the microcode revisions.

To cut to the chase, I am going to seriously be looking at a Ryzen 1800X for my next build, if I can find a suitable Motherboard.

Performance:

I ran 2 passes of Realbench and averaged them:

ucode 22h:139,125
ucode 23h 128,265

This is approx. a 9% drop on performance with microcode 23h. Note this is in addition to any Windows 10 performance drops.

For stability I had noticed I was getting a lot of WHEA errors with the 23h microcode. Here is the results of 15 minutes running 23h:




These are correctable hardware (internal CPU Northbridge) errors that also show up in the Windows event log. Each of these causees interrupts if nothing else.

For ucode 22h I get nada , zero, 0 of these errors !

To show that the 23h has patches for the security vulnerabilities, here are the powershell script results that test this:

22h no hardware BTI support:



and here is 23h with support:



Sigh.
 

·
Iconoclast
Joined
·
30,977 Posts
Very interesting results from the new microcode. Thanks.

I've got a few questions though:

Do the WHEA errors disappear at a less aggressive OC? Microcode changes often have unpredictable effects on stability, outside any specific fixes involved. Hopefully any issues here will be resolved in a future update.

What's the break down of the performance hit per test inside RealBench?
 

·
Registered
Joined
·
2,543 Posts
After injecting microcode 23 into my BIOS...


... I have stressed my 24/7 oc profile. Struggling...


Comparing it to a previous attempt (microcode 22 , Win 10 Pro patched), I do not observe much difference in fps...


...it is the first time, however, I see that "CPU Internal Errors" in HWiNFO64...

I do not dare to test at 4.9 now

EDIT: Wow, this SUCKS big time!! After rebooting I am now having 3 CPU Internal Errors already, showing in HWiNFO64... So, my "eternal" x47 OC settings are not stable, even though I've completed 5 loops of the x264, latest binaries, as shown in the screenshot. I will raise my VCore from 1.2V to 1.250V, Adaptive, in the BIOS, to see IF this will help. This sucks.................
mad.gif


I think this is the error in the Event Viewer
 

·
Registered
Joined
·
2,543 Posts
Quote:
Originally Posted by boombastik View Post

even in stock i take cpu internal errors.

When i Play BF1
What?? Really?! Jesus...................
mad.gif
mad.gif
 

·
Iconoclast
Joined
·
30,977 Posts
Yeah, starting to look like they botched something with this microcode update. Possibly an over aggressive fix missing certain timing windows.

Going to move my persistent VMs to an AMD box and wait for some tests with the next microcode revision, whenever that shows up.
 

·
Guru Meditation
Joined
·
2,068 Posts
Thanks for the post, i hope intel will manage to fix these CPU errors and mitigate the performance and stability loss with the next microcode...one can hope.
 

·
Registered
Joined
·
2,543 Posts
Well, here is my experience, sorry no screenshots this time.

I cleared CMOS + loaded Optimized Defaults. I rebooted and I loaded my XMP2 RAM profile + I disabled the usual in my BIOS (on-board Soundcard, Intel Virtualization Technology, ASMedia SATA controller, Intel Smart Connect) + All C-States Enabled. Then I boot into Windows.

I started the x264 v2.06, 5 loops. After 60 seconds I power on my plasma TV, HDMI connection with my system, and I started Heaven Benchmark 4.0 (on the TV). The two tests were running together, everything at STOCK, and I was happy because no "CPU Internal Errors" were appearing in HWiNFO64... After about 5 minutes I stopped Heaven, I then stopped the x264 and I shut down my TV. My monitors go blank for a second each time I do this, it is the way it is happening on my system. Well....after shutting down the TV one "CPU Internal Error" appeared in HWiNFO64!! With EVERYTHING AT STOCK!.....................

What the...................???

UPDATE:
And here, Ladies and Gentlemen, you can see the error appearing after leaving my system on idle, at stock. I am returning to microcode 22 right now! Thank you.
 

·
Registered
Joined
·
3,004 Posts
I was exspecting this as my 4790k was at 5ghz at 1.37v bios 1.41v load for 2 years, but the thing never BSOD with this garbage update, even playing ACO with it hamming the cpu to 90-100% for 10+ hours. Just got an 8700k today, 1.34v 5.1ghz AVX and yet to get any Internal errors.

Just crank that voltage up bro, I had mine for about 2 years running 80-90C LOL.
 

·
Registered
Joined
·
2,543 Posts
Back on MC 22 for me, in the BIOS that is, but at least on my Linux installation I run 23 (without any issues!)

 

·
Registered
Joined
·
135 Posts
I guess we have to wait perhaps a revised µCode comes out for the 306C3 Haswell CPUs...
Just unsure if this is likely going to happen or not... WE have to wait and see in a few weeks time I guess

In meantime I will monitor this thread closely on what happens
 

·
Optimal Pessimist
Joined
·
3,047 Posts
Discussion Starter · #15 ·
Quote:
Originally Posted by Blameless View Post

Very interesting results from the new microcode. Thanks.

I've got a few questions though:

Do the WHEA errors disappear at a less aggressive OC? Microcode changes often have unpredictable effects on stability, outside any specific fixes involved. Hopefully any issues here will be resolved in a future update.

What's the break down of the performance hit per test inside RealBench?
If I run it at system optimized default 4 GHz I did not see them. The more aggressive the higher the rate it seems, but I didn't thoroughly test that.

I didn't save the results of the Realbench (I know), but the most significant decreases were in the first photo editing and last multitasking tests.
 
  • Rep+
Reactions: Blameless

·
Optimal Pessimist
Joined
·
3,047 Posts
Discussion Starter · #16 ·
Quote:
Originally Posted by LostParticle View Post

After injecting microcode 23 into my BIOS...


... I have stressed my 24/7 oc profile. Struggling...


Comparing it to a previous attempt (microcode 22 , Win 10 Pro patched), I do not observe much difference in fps...


...it is the first time, however, I see that "CPU Internal Errors" in HWiNFO64...

I do not dare to test at 4.9 now

EDIT: Wow, this SUCKS big time!! After rebooting I am now having 3 CPU Internal Errors already, showing in HWiNFO64... So, my "eternal" x47 OC settings are not stable, even though I've completed 5 loops of the x264, latest binaries, as shown in the screenshot. I will raise my VCore from 1.2V to 1.250V, Adaptive, in the BIOS, to see IF this will help. This sucks.................
mad.gif


I think this is the error in the Event Viewer
CPU Internal Errors = WHEA errors.
 

·
Optimal Pessimist
Joined
·
3,047 Posts
Discussion Starter · #17 ·
Maybe the errors were just never reported before or are something nothing to do with stability? IDK. I am back to 22h though.
 

·
Registered
Joined
·
73 Posts
Maybe is has anything to do with asrock?
Because we take whea errors also in stock defaults.
And i have a 4790 non K.

Bf1 gaming:
22--->no whea errors
23----> whea errors
 

·
Registered
Joined
·
193 Posts
first time i overclocked with the new 22h uC installed to the bios, i got a Kernel Security Check BSOD right away after boot. Also had to bypass the normal bios update with powering down and using bios flashback.

Also I have undid all the powershell stuff, since it was slowing my system down and even gave a bsod.

https://www.ghacks.net/2018/01/05/find-out-if-your-windows-pc-is-affected-by-meltdown-spectre-vulnerabilities/

if you wan to undo the powershell commands, open powershell with admin and: Set-ExecutionPolicy Default

waiting for an official patch...
 

·
Registered
Joined
·
2,543 Posts
Quote:
Originally Posted by Technodox View Post

...
waiting for an official patch...
While you are waiting I would [strongly] suggest you to e-mail ASUS Tech Support and ask them when do they have in mind to release a BIOS for your motherboard. I did the same for my ASRock today but it takes them quite some time to respond from all the way, over there, in Taiwan..
 
1 - 20 of 301 Posts
Top