• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

*** Official Ryzen Owners Thread ***

AMDMatt said:
4Ghz @1.4V
VDD SOC manual voltage @1.175v (will test offset again and may change to that)
Load Line @3 for CPU and VDD SOC
DRAM 1.4V (will try 1.35V)
PLL 1.85V (will try 1.80v)
P State 0 Custom Settings (allows idle clocks and full speed clocks in balanced power profile within Windows!)
FID A0 (change this to adjust frequency, works in hexadecimal, A0 = 4Ghz 94=3.7Ghz 99=3.825Ghz)
DID 8 (don't touch this)
VID 3A ( don't touch this setting, if you need more/less voltage, use CPU Offset voltage in the first image)

M7swH7u.jpg

1hXnZWe.jpg

X4nou6U.jpg

JRw0vZV.jpg

Tlbeh5l.jpg
Click to expand...

Is it normal to have the 12V voltage hover between 11.7-11.8?

Nice overclock! did you try raising BCLK? Do you think you can squeeze any meaningful performance that way?
 
1700 + Gigabyte AB350 Gaming 3 here. A solid board it seems but what I didn't realise is that the vcore only goes up to +0.3V, and on my 1700 it means 1.34V max, I might struggle to get RB stable at 3.9 or even 3.8 with that limit. Apart from vcore the only other adjustable voltages are VSOC and VDIMM, for anyone who is considering this board.

A few issues I had so far, if anyone can shed a light on these issues I would be very grateful:

1. CPU-Z will get stuck halfway through hardware detection, the older versions work but they are useless for Rysen. Alsway at 65% when it's detecting storage. I have AMD RAID running so not sure if that's causing the issue.

2. Rysen Masters keep telling me to enable HEPT whenever I try to adjust the settings even though it's already on both in BIOS and in Windows. I ran the batch file and it said it was enabled.

The highest I've benched was 3.9ghz with ram running at 2933mhz CL15. cinebench score 1707.
 
AMDMatt said:
I am now I've overclocked via the P State 0. Here's the difference in performance in Cinebench between the two modes.

Balanced
PzpRsWd.jpg

High Performance
k51Sztz.jpg

High Performance is a little bit faster, so if you want absolute performance or are benching, this is the way to go.
Click to expand...

Mate aren't both screenies the same?

My testing so far.

I don't use Prime95, as when I got the i5 4690K I noted on ROG forum that Raja@Asus (Asus techie) stated it can pull so much power through CPU = degradation. I set my CPU 4.9GHz @ 1.255V CACHE 4.4GHz @ 1.10V, ~1yr+ ago. I used 48 loops of x264 (~8hrs), first as that only loads CPU and was highlighted as a cool running AVX stress test (I'm on air). Next 4hrs RealBench Stress mode and must have done several hundred hours of f@h plus gaming in past year or so. No issues to report, from my method and my uses of PC, it is not uncommon for me to do 48hrs+ continuous f@h runs.

So on the Ryzen 7 1700 I'm using similar method. Just like when binning the 2x i5 4690K I looked out for what made my rig hiccup quicker when OC'ing, as if I sorted that then others fell into place. I went for RB first on 3.7GHz profile, but noted x264 was making CPU fall over quicker and requiring more VCORE, so for 3.8GHz I went for x264 first.

CPU: R7 1700
Mobo: C6H (FW 0902)
RAM: 2x4GB Corsair LPX 2400MHz C14 (was supposed be using G.Skill 2x16GB 3200MHz C14 but turned out faulty, waiting on RMA)
HSF: TR Archon SB-E X2 with TY-143

1st 2hrs RB Stress mode @ 3.7GHz ACB @ 1.18750V in UEFI (DMM 1.19V)

pJa9W7l.jpg

2nd 10 loops x264 (~1hr) @ 3.7GHz ACB @ 1.20625V in UEFI

i4VnTCy.jpg

Next I did a tweak to UEFI as some owners of C6H had on OCN stated to Ellmor (Asus MB R&D) that temps differed between ROMs.

1. Temperature readings on 0902 and pre 5803. We tweaked it a bit too far and will be reversed in future updates, you can manually go back to stock/5803 behavior by going to Tweaker's paradise and set Sense MI skew = Enabled and Sense MI offset = 272 (0902 setting when overclocked is 283). Reading too low can affect your DRAM stability, so if your system is reading less than 30*C I suggest you set it.
Click to expand...

Quote extract link.

3rd 10 loops x264 (~1hr) @ 3.8GHz ACB @ 1.29375V in UEFI

RkXkzxp.jpg

4th 2hrs RB Stress mode @ 3.8GHz ACB @ 1.29375V in UEFI

8AOjMqv.jpg

Now the weird readings in HWiNFO are down to SMU (On die controller) requiring an update via UEFI, source of information is Martin Malik, author of HWiNFO, link to post.

I have also created a thread on OCN where in OP I place Rysen Essential info (when I get the time :P ).
 
gupsterg said:
Mate aren't both screenies the same?

My testing so far.

I don't use Prime95, as when I got the i5 4690K I noted on ROG forum that Raja@Asus (Asus techie) stated it can pull so much power through CPU = degradation. I set my CPU 4.9GHz @ 1.255V CACHE 4.4GHz @ 1.10V, ~1yr+ ago. I used 48 loops of x264 (~8hrs), first as that only loads CPU and was highlighted as a cool running AVX stress test (I'm on air). Next 4hrs RealBench Stress mode and must have done several hundred hours of f@h plus gaming in past year or so. No issues to report, from my method and my uses of PC, it is not uncommon for me to do 48hrs+ continuous f@h runs.

So on the Ryzen 7 1700 I'm using similar method. Just like when binning the 2x i5 4690K I looked out for what made my rig hiccup quicker when OC'ing, as if I sorted that then others fell into place. I went for RB first on 3.7GHz profile, but noted x264 was making CPU fall over quicker and requiring more VCORE, so for 3.8GHz I went for x264 first.

CPU: R7 1700
Mobo: C6H (FW 0902)
RAM: 2x4GB Corsair LPX 2400MHz C14 (was supposed be using G.Skill 2x16GB 3200MHz C14 but turned out faulty, waiting on RMA)
HSF: TR Archon SB-E X2 with TY-143

1st 2hrs RB Stress mode @ 3.7GHz ACB @ 1.18750V in UEFI (DMM 1.19V)

pJa9W7l.jpg

2nd 10 loops x264 (~1hr) @ 3.7GHz ACB @ 1.20625V in UEFI

i4VnTCy.jpg

Next I did a tweak to UEFI as some owners of C6H had on OCN stated to Ellmor (Asus MB R&D) that temps differed between ROMs.



Quote extract link.

3rd 10 loops x264 (~1hr) @ 3.8GHz ACB @ 1.29375V in UEFI

RkXkzxp.jpg

4th 2hrs RB Stress mode @ 3.8GHz ACB @ 1.29375V in UEFI

8AOjMqv.jpg

Now the weird readings in HWiNFO are down to SMU (On die controller) requiring an update via UEFI, source of information is Martin Malik, author of HWiNFO, link to post.

I have also created a thread on OCN where in OP I place Rysen Essential info (when I get the time :p ).
Click to expand...

Good info. :)

Yes i posted the same screenshot twice, I'll see if i still have it saved but it may be lost. Essentially performance mode was offering around 20 points more than balanced mode when overclocking the P 0 State.
 
steve258 said:
Is it normal to have the 12V voltage hover between 11.7-11.8?

Nice overclock! did you try raising BCLK? Do you think you can squeeze any meaningful performance that way?
Click to expand...

Yes i wondered that, hope it's an incorrect reading. Perhaps having all PCI-E lanes filled up with 2x Radeon Pro Duos and 1x Capture card is putting extra stress on the board. Hoping it's not my 8Pack PSU degrading. Will have to test some games using all four GPUs and see what's up. I'm also using extender cables on the PSU so it could be one of those playing up i guess.
 
**** IMPORTANT INFO FOR CROSSHAIR VI OWNERS ****

Fans and Bios Operation

For chassis fans set to manual mode, do not use any of the predefined modes as they are not working correctly. For stable chassis fan operation use manual settings.
 
awaybreaktoday said:
**** IMPORTANT INFO FOR CROSSHAIR VI OWNERS ****

Fans and Bios Operation

For chassis fans set to manual mode, do not use any of the predefined modes as they are not working correctly. For stable chassis fan operation use manual settings.
Click to expand...

Been using my C6H configured via UEFI for fan profiles and no issues. I used "standard" and "turbo". I have CPU header supplying PWM signal to a Gelid 4 in 1 cable so the 2x TY143 on front case intake and 2x TY143 on HS run at same profile. I use CHA1 for rear case lower exhaust fan (AC F12), CHA3 for 2x upper case exhaust fans (AC F9).

AMDMatt said:
Good info. 

Yes i posted the same screenshot twice, I'll see if i still have it saved but it may be lost. Essentially performance mode was offering around 20 points more than balanced mode when overclocking the P 0 State.
Click to expand...

Cheers :) .
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom