Asus ROG X570 Crosshair Dark Hero Release? *** NO Competitor Talk ***

Dupppex

Dupppex

Dupppex

Associate
Joined
22 Mar 2017
Posts
208
Location
London
michozara said:
Hello everyone,

I'm near the end of collecting all the parts for my new PC, in fact only the fans and cables left to arrive.
With that being the state of things I decided to dig into the motherboard BIOS a bit and noticed some people are having restart issues.

My question would simply be should I expect the restarts to happen with my particular build, and should I update to the latest BIOS (3801) and did it fix the problems? I couldn't find much about the 3801 out there.
Another thing I have a question about is BIOS update sequence for the Dark Hero, aka can I go to the latest BIOS update straight away or do I have to update it version by version?

Specs:
Asus ROG Crosshair VIII Dark Hero
AMD Ryzen 9 5950X
G.Skill Trident Z DDR4-3600MHz CL14-14-14-34 1.45V 46GB (4x16GB - F4-3600C14Q-64-GTZRA)
Seagate FireCuda 530 2 TB 4.0 3D TLC NVMe
Noctua NH-D15 Chromax Black
be quiet! Dark Power 12 1000W
ROG Strix RTX3080 O10G Gaming

P.s. that's a very clean build Dupppex!
Click to expand...

You have some killer PC Parts. I look forward to seeing your build :cool:
 

michozara

michozara

michozara

Associate
Joined
26 Nov 2021
Posts
3
Location
Wonderland
Dupppex said:
You have some killer PC Parts. I look forward to seeing your build :cool:
Click to expand...
Tnx :), just waiting for my cables now, its getting hard not just to build the thing but trying to keep my cool till they arrive :D.
Will be posting photos for sure once done!
Honestly I was and still am looking for the Noctua RTX 3070 but no luck so far, ebay guy won't ship to me and sold out in the few shops that had it, sob I grabbed the Strix I saw in a local shop.
 

Dupppex

Dupppex

Dupppex

Associate
Joined
22 Mar 2017
Posts
208
Location
London
michozara said:
Tnx :), just waiting for my cables now, its getting hard not just to build the thing but trying to keep my cool till they arrive :D.
Will be posting photos for sure once done!
Honestly I was and still am looking for the Noctua RTX 3070 but no luck so far, ebay guy won't ship to me and sold out in the few shops that had it, sob I grabbed the Strix I saw in a local shop.
Click to expand...

Dammm, Thanks for the heads up. I did not know this was a thing. If ASUS release a 3090 version it's going to be really hard for me not to get one :D
 

lovingbenji

lovingbenji

lovingbenji

Associate
Joined
5 Sep 2008
Posts
191
Location
Paris - France
Last edited:

ChocAndGray

ChocAndGray

ChocAndGray

Associate
Joined
4 Oct 2017
Posts
1,221
Does anyone with the motherboard have low boost clocks with a 5950x?

I had an MSi Unify and my cores at stock settings would easily boost to 5050Mhz.

With the dark hero they barely break 4900Mhz, and my second CCD hits 4700Mhz.

The only way I can get them to boost close to 5000Mhz is setting a ridiculous curve with massive negative offsets that cause instability.

I'm on the latest stable bios 3904.
 

GregI

GregI

GregI

Associate
Joined
10 Dec 2007
Posts
1,696
@Jay343434 I didn't do much testing but I feel like the latest BIOS is really bad for that. On 1.2.0.5 my Cinebench r20 single core would be 622 points and my multi was also down around 10,360.

I had this super weird system restart on 1.2.0.5 that happened after I changed to the high performance power plan so I decided to fresh install Windows while rolling back to 1.2.0.3c with manual driver installs. Cinebench r20 is back at 642 single and 10,486 multi.

Been running stability tests at stock now with core cycler for about ~70 hours and no issues so far. I get 5,049 Mhz on 1.2.0.3c. Definitely try 1.2.0.3c (3801) and see what your boosts do then.
 
Last edited:

ChocAndGray

ChocAndGray

ChocAndGray

Associate
Joined
4 Oct 2017
Posts
1,221
GregI said:
@Jay343434 I didn't do much testing but I feel like the latest BIOS is really bad for that. On 1.2.0.5 my Cinebench r20 single core would be 622 points and my multi was also down around 10,360.

I had this super weird system restart on 1.2.0.5 that happened after I changed to the high performance power plan so I decided to fresh install Windows while rolling back to 1.2.0.3c with manual driver installs. Cinebench r20 is back at 642 single and 10,486 multi.

Been running stability tests at stock now with core cycler for about ~70 hours and no issues so far. I get 5,049 Mhz on 1.2.0.3c. Definitely try 1.2.0.3c (3801) and see what your boosts do then.
Click to expand...

Are you able to do a screenshot showing what most of your cores boost to? I'm curious whether most of your cores boost to over 5000Mhz, the second chiplet in particular.

When I get a chance later I'll try and flash the older bios.

I even tried 4001 and 4002 bioses but they seem even worse for boosting.
 

GregI

GregI

GregI

Associate
Joined
10 Dec 2007
Posts
1,696
@Jay343434 Running ycruncher right now via core cycler. Don't wanna take screenshots or open up programs while that goes on to send screenshot but yeah only 2 cores reach 5000.3 Mhz. One core gets to 4,975 Mhz and the rest are all 4,900. Not absolutely amazing but could have been worse.

I bet if I flashed an even older BIOS I'd get even better boosts. There was some discussion here in which people claimed AMD saw all the crashes people got so just sliced the boost clocks down.

If your boosts are still not comparable to your old MSI board on 3801 BIOS I'd try the following things:

1: Make sure the chipset drivers you installed are the ones from Asus support page (3.10.22.706). Apparently they have some specific fixes in them over the general AMD website ones.
2: Likely won't do anything but as discussed in other thread try with fTPM disabled.
3: Did you disable the armoury crate stuff from installing automatically in the BIOS? I noticed when I had all that stuff installed it would not let my 5950x idle properly (31c vs 27c clean install). This also likely does something to the boost if I was to guess.
 

ChocAndGray

ChocAndGray

ChocAndGray

Associate
Joined
4 Oct 2017
Posts
1,221
@GregI

thanks for the response. As a comparison what sort of boosts did you get on 3904?

yeah I’ve disabled armoury crate in the bios, only just disabled fTPM last night.

I’ve also got the latest chipset drivers directly from amd.

you could be onto something, maybe amd have tweaked something in the agesea.

to clarify you’re running optimised defaults in bios? Just so when I flash the bios I get as close a comparison as possible
 

GregI

GregI

GregI

Associate
Joined
10 Dec 2007
Posts
1,696
Jay343434 said:
I’ve also got the latest chipset drivers directly from amd.
Click to expand...

Most likely your issue if 3801 also gives you similar boosts.

After BIOS flash I only used to just enable XMP but with the weirdness currently the 2 other steps are disabling fTPM and stopping armoury crate bloatware auto installing itself. Everything else default.

I didn't do any hwinfo checking of boosts on 3904. All I knew was that cinebench score was much lower. I should have looked at it though but then hwinfo would likely take processing power which might effect the score. Damned if you do moment.

Gonna run ycruncher for another 12 hours or so and if that passes I'll try 3904 again.
 

ChocAndGray

ChocAndGray

ChocAndGray

Associate
Joined
4 Oct 2017
Posts
1,221
lovingbenji said:
Jay have you seen my table comparing the 2 bios a few posts above ? seems like I experienced something similar
Click to expand...

Just seen it.

so it’s interesting that the clock frequencies show lower but better performance. I wonder if there was just more clock stretching in that bios?

based on the all core, did you notice single cores would boost lower too?

My personal experience has been lower core frequencies does result in lower benchmarks. Cpuz as a prime example.
 
Last edited:

ChocAndGray

ChocAndGray

ChocAndGray

Associate
Joined
4 Oct 2017
Posts
1,221
lovingbenji said:
Definitely yes

And yes, so surprised that lower VCore on Bios 3904 could lead to higher temperature and lower Frequencies .... hard to understand ... and what is even harder to understand is how lower frequencies could lead to better Cinebench scores !!!
Click to expand...
I could be wrong but the only way lower frequencies could result in lower scores is if its clock stretching.

So it could show 4600mhz but effectively be 4200mhz.

whereas another bios could show 4400mhz but actually be 4400mhz effective clock.
 
You must log in or register to reply here.
Back
Top Bottom