Overclock.net banner

801 - 820 of 12401 Posts

·
Registered
Joined
·
33 Posts
Manual voltage is broken LLC wise in 0601 bios.
Or not broken, but the control doesn't work as it should.
This is sad. Have someone performed LLC tests/confirmations too? Can someone with voltmeter test difference between several LLC levels, like Buildzoid did in his latest video?
@VPII when you have spare time - please test VCore with your voltmeter different LLC levels. I as many others here would really appreciate it!
 

·
Meddling user
Joined
·
7,423 Posts
UEFI 0601 settings as attached txt, but manually set VBOOT/VDIMM 1.35V and SOC 0.900V

View attachment 0601_Base_Profile_setting.txt

Screenie 1 captured at 22:11:31
Screenie 2 captured at 22:11:53
Screenie 3 captured at 22:12:07


ProbeIt points:-

VCORE: 1.425-8V
SOC: 0.893 (steady)
RAM: 1.354/5V
1.8V PLL: 1.802/4V
1.05V SB: 1.060/2V

Note: Even when sat in UEFI a load is created on CPU.

OS let idle for ~15min then capture, Balanced stock OS power plan.


ProbeIt points:-

VCORE: 0.596-0.916V (as expected for stock CPU)
SOC: 0.892/6V
RAM: 1.354V (steady)
1.8V PLL: 1.803/5V
1.05V SB: 1.052/3V

Now running HCI v6.0, 3200MHz The Stilt Safe, but TRC 44 TRFC 256. The SOC of 0.900 for 3200MHz is quite low compared with 5x Ryzen gen 1 tested on C6H. I could get some to OS, but no way HCI active. Currently using F4-3200C14D-16GTZ, what I used the most on C6H.


ProbeIt points:-

VCORE: 1.308/20V (as expected for stock CPU)
SOC: 0.888/9V
RAM: 1.353/4V
1.8V PLL: 1.805/7V
1.05V SB: 1.057/9V

ProbeIt points to me seem much improved than C6H for sure. Basis for this is readings of idle vs load. For example RAM I used to see a LL effect just like on VCORE, SOC, IIRC only PLL/SB didn't have it.
 

·
Registered
Joined
·
8 Posts
Hey guys I am new to the forums and new to this motherboard.

I am currently having issues raising my BCLK over 101.4, anything over 101.4 will cause an unstable overclock (my computer freezing in Windows after running prime95/Cinebench).
I uploaded pictures of my current BIOS settings.
https://imgur.com/a/0Nkx0aD

Things I have tried:
-If I lower the performance enhancer level to default it will be stable but I want to boost for single threaded applications.
-I set the offset voltage as high as +0.075 and it doesn't even allow a stable 102 let alone 101.8 BCLK. Thus changing the voltage really had no effect in going past a stable 101.4 BCLK
-I have tried 102 BCLK with stock memory settings (and all other settings stock) with just performance enhancer level 3, boost enabled, and offset voltage +0.075. The result was the same as with all of my memory settings and other settings.

Also I have an image of HWiNFO during a stress test (this is with 101.6 BCLK, +0.025 offset voltage, performance level3, and enabled precision boost):
https://i.imgur.com/bD1O3AS.png
This stress test later proved to be unstable when it froze not too long after.

I have a noctua DH-15 inside my Meshify C for cooling. I have 2 140mm intake fans, 2 fans on top (120mm and 140mm) and one 120mm fan in the back.

My current theory is that my motherboard needs the extra 4pin CPU power cable (currently I only have the 8pin CPU power cable), but it was my understanding I only really need that cable for LN2 overclocking?
 

·
Registered
Joined
·
1,367 Posts
Hey guys I am new to the forums and new to this motherboard.

I am currently having issues raising my BCLK over 101.4, anything over 101.4 will cause an unstable overclock (my computer freezing in Windows after running prime95/Cinebench).
I uploaded pictures of my current BIOS settings.
https://imgur.com/a/0Nkx0aD

Things I have tried:
-If I lower the performance enhancer level to default it will be stable but I want to boost for single threaded applications.
-I set the offset voltage as high as +0.075 and it doesn't even allow a stable 102 let alone 101.8 BCLK. Thus changing the voltage really had no effect in going past a stable 101.4 BCLK
-I have tried 102 BCLK with stock memory settings (and all other settings stock) with just performance enhancer level 3, boost enabled, and offset voltage +0.075. The result was the same as with all of my memory settings and other settings.

Also I have an image of HWiNFO during a stress test (this is with 101.6 BCLK, +0.025 offset voltage, performance level3, and enabled precision boost):
https://i.imgur.com/bD1O3AS.png
This stress test later proved to be unstable when it froze not too long after.

I have a noctua DH-15 inside my Meshify C for cooling. I have 2 140mm intake fans, 2 fans on top (120mm and 140mm) and one 120mm fan in the back.

My current theory is that my motherboard needs the extra 4pin CPU power cable (currently I only have the 8pin CPU power cable), but it was my understanding I only really need that cable for LN2 overclocking?
Same goes for me, but I have tried it with Manual OC. 102 BCLK takes ages for booting windows and then eventually when loaded 10+min, Its totally laggy and freezes then. 100-101 works fine. Never tried 101.6 :)
Also tried 103+ etc, with no success booting at all.

And by the way: I have both power cables plugged in so no need to try. Doesn't make sense anyways.. :)
 

·
Registered
Joined
·
93 Posts
Hey guys I am new to the forums and new to this motherboard.

I am currently having issues raising my BCLK over 101.4, anything over 101.4 will cause an unstable overclock (my computer freezing in Windows after running prime95/Cinebench).
I uploaded pictures of my current BIOS settings.
https://imgur.com/a/0Nkx0aD

Things I have tried:
-If I lower the performance enhancer level to default it will be stable but I want to boost for single threaded applications.
-I set the offset voltage as high as +0.075 and it doesn't even allow a stable 102 let alone 101.8 BCLK. Thus changing the voltage really had no effect in going past a stable 101.4 BCLK
-I have tried 102 BCLK with stock memory settings (and all other settings stock) with just performance enhancer level 3, boost enabled, and offset voltage +0.075. The result was the same as with all of my memory settings and other settings.

Also I have an image of HWiNFO during a stress test (this is with 101.6 BCLK, +0.025 offset voltage, performance level3, and enabled precision boost):
https://i.imgur.com/bD1O3AS.png
This stress test later proved to be unstable when it froze not too long after.

I have a noctua DH-15 inside my Meshify C for cooling. I have 2 140mm intake fans, 2 fans on top (120mm and 140mm) and one 120mm fan in the back.

My current theory is that my motherboard needs the extra 4pin CPU power cable (currently I only have the 8pin CPU power cable), but it was my understanding I only really need that cable for LN2 overclocking?
Is your memory stable at 3500 MHz ? Have you tested it at that speed ?
Why are you running your ram with 1.48v ?
 

·
Meddling user
Joined
·
7,423 Posts

·
Registered
Joined
·
8 Posts
Is your memory stable at 3500 MHz ? Have you tested it at that speed ?
Why are you running your ram with 1.48v ?
My memory is stable (30min of memtest and LinX) with anything below 3550MHz 14-14-14-28 timings

Is that bad I have 1.48v?
I was following the Ryzen DRAM calculator.
https://i.imgur.com/FFnyhYz.png


My memory isn't the problem though, I can't get past 101.4 BCLK with 2133 MHz with default timings.



Okay that's good to know. Hopefully, someone knows why I can't go past 101.4 BCLK
 

·
Meddling user
Joined
·
7,423 Posts
Well calling it a night. Impressed how easily 3200MHz was achieved for HCI and the lowness of SOC. I just picked a random figure of 900mV to use, may see how low I can go to pass HCI before moving on to P95, RB, etc. CB15 got a nice boost from MEMCLK/IFCLK increase :) , CPU-Z seemed unaffected.


Just in case others want Raja posted before in C6H thread a C7H wallpaper, link.

Okay that's good to know. Hopefully, someone knows why I can't go past 101.4 BCLK
I used to find at x BCLK on C6H my WiFi dropped, perhaps something you have "connected" is causing issue.

Personally I don't tinker with BCLK. Perhaps by changing eCLK to async may help, but it has a latency hit, check PDF in OP.
 

·
Registered
Joined
·
8 Posts
Well calling it a night. Impressed how easily 3200MHz was achieved for HCI and the lowness of SOC. I just picked a random figure of 900mV to use, may see how low I can go to pass HCI before moving on to P95, RB, etc. CB15 got a nice boost from MEMCLK/IFCLK increase :) , CPU-Z seemed unaffected.


Just in case others want Raja posted before in C6H thread a C7H wallpaper, link.

I used to find at x BCLK on C6H my WiFi dropped, perhaps something you have "connected" is causing issue.

Personally I don't tinker with BCLK. Perhaps by changing eCLK to async may help, but it has a latency hit, check PDF in OP.
I will test this and come back to you. It don't plan to use async but using this feature should rule out some possibilities.

What do you mean by something I have "connected"? Like my my GPU or RAM? I have a EVGA GTX 1080 FTW, besides that I only have a slow Crucial SSD and a 3TB mechanical HDD. Also I have the non-wifi version of the motherboard.

EDIT:

I used async and I still couldn't run 102Mhz BCLK :(
https://i.imgur.com/IS1xcV6.jpg
 

·
Registered
Joined
·
43 Posts
I'm have been able to stablize my cpu on 4.3Ghz all cores, with 1.394v showing in HW Info beta version as the vid. With IBT @ very high, and temps around 75c. I doubt I'd do anything in my 24/7 stuff that will be that CPU intensive. I believe I'm i bios 0401. Are there voltage discrepancies in that bios as well? Apart from the cpu volts, all other voltages are on auto. What could be ways to lower the cpu voltage, like PLL voltage?

I can boot on all auto voltages at 4.4Ghz, and run Cinebench once or twice but the autovoltage are close to 1.46v at that point, which I guess is a little too high for Ryzen 2?
 

·
Registered
Joined
·
530 Posts
I'm have been able to stablize my cpu on 4.3Ghz all cores, with 1.394v showing in HW Info beta version as the vid. With IBT @ very high, and temps around 75c. I doubt I'd do anything in my 24/7 stuff that will be that CPU intensive. I believe I'm i bios 0401. Are there voltage discrepancies in that bios as well? Apart from the cpu volts, all other voltages are on auto. What could be ways to lower the cpu voltage, like PLL voltage?



I can boot on all auto voltages at 4.4Ghz, and run Cinebench once or twice but the autovoltage are close to 1.46v at that point, which I guess is a little too high for Ryzen 2?


Are you looking at SVI2 Voltage on hwinfo? I doubt it’s hitting 4.4 at 1.46 Vcore


Sent from my iPhone using Tapatalk Pro
 

·
Registered
Joined
·
43 Posts
Are you looking at SVI2 Voltage on hwinfo? I doubt it’s hitting 4.4 at 1.46 Vcore


Sent from my iPhone using Tapatalk Pro
You're right its 1.49v.


This is with ram at the default speed, 2133mhz or something.
 

·
Registered
Joined
·
530 Posts
You're right its 1.49v.






This is with ram at the default speed, 2133mhz or something.


1.494 is way to high you don’t want more then 1.42 spikes imo if you’re using PE


Sent from my iPhone using Tapatalk Pro
 

·
Registered
Joined
·
43 Posts
I agree, but PE states were disabled when I was trying this, and voltage was on Auto. I think on 0207 and 0401 they aren't present. I see them in 0601.
 

·
Registered
Joined
·
380 Posts
Hi majestynl,

could you redo your AIDA Latency Benchmark with the latest AIDA version 5.97? 🙂
For me it showed a loss of 2ns in latency. Would be interesting if you see the same and it would be better for comparison 🙂

Thanks!
And here it is, as promised! No difference for me...sorry
That is even better 🙂
You bet the 60ns barrier 😉
 

·
Registered
Joined
·
530 Posts
I think I’ve gotten a stable OC at 4.1 Ghz and Offset voltage at +0.13125 for a total 1.344 Vcore LLC3 under heavy load my Vcore is 1.319 seems like the best I can do atm even with my h115I and noctua fans since I can’t download Corsair link and adjust it I spike up to 78C


Sent from my iPhone using Tapatalk Pro
 

·
Meddling user
Joined
·
7,423 Posts
Good morning, Vietnam!!! :D

Nice sweet post even with room at 19C, IIRC some gen 1 CPUs had cold boot issues. Quick little 30min circuit on RB with tweaked RAM as done last night. Stock 2700X is impressing me a bit more than yesterday, ~3.95GHz in RB, below is 29 of 30min.


What was more impressive was how it held so well MHz in Y-Cruncher. Captured each test, but VST as attention got diverted by breakfast :p . Comparing VID and SVI2 seems like no VDROOP :headscrat , using LLC [Auto] for CPU/SOC, so would think is AMD stock.


@elmor

My bug list for UEFI 0601:-

  • 1.05V SB [Auto] is 1.115V on ProbeIt, stock settings. Using 1V manually gets me ~1.06V on ProbeIt when in UEFI and ~1.05V in OS at idle/load.
  • Bluetooth disabled in UEFI is still detected in OS and functional.
 

·
Overclocker
Joined
·
6,943 Posts
Hi guys,

I just did a fresh install of Windows 10 pro build 1803 but when i go to the Asus CH7 hero Wifi page to download drivers, i saw that there are some missing...

I see no USB drivers, now my USB isn't working properly and i can't find the driver anywhere...
 

·
Premium Member
Joined
·
888 Posts
Discussion Starter #820
PSA C7H Peeps

SB voltage on [Auto] IMO seems crazy excessive on UEFI 0601.

My build was uneventful up til posting and using UEFI.

I flashbacked UEFI 0601 prior to 1st powerup, removed power from board, then did clear CMOS. I then powered up, went to do screen shot of Main page and CPU page, rig rebooted :eek:. I tried again and it rebooted :eek:. I stopped taking screen shots via F12.

Checked UEFI and found SB voltage readout at 1.094V, applied a DMM to ProbeIt point and it read back as 1.115V :( . A manual setting of 1V yielded in 1.064/5V, same as what my C6H would be when set at 1.05V or [Auto] in UEFI and when measured at ProbeIt point.

SOC was read lower in UEFI than Ryzen Gen 1, using DMM on ProbeIt it was ~0.800V. Near enough 0.075V-0.125V lower than several Ryzen Gen 1 CPUs I used on C6H. So SOC [Auto] was as it should be.

RAM [Auto] was 1.202V on ProbeIt point, all good.

PLL [Auto] I never measured, but 1.8V manually set was read back via DMM on ProbeIt point as 1.8V steady as it was on my C6H.
My SB was identical to yours! Elmor or Stilt did mention something about voltage the other day when I reported my PLL voltage as low - seemed to think it was most likely a readout error. Maybe one of them can clear this up for us. Do your other voltages report as you would expect?

I have my DMM - will have a test in a while and report back.

There's a small issue on the first batches of boards where the droop from the VRM output to the SB input is a relatively large (~50mV). The read point for both software and ProbeIt is at the VRM output, so it will read higher (before droop). Ie the VRM is outputting ~50mV higher than what the SB is actually getting. Not pretty but it shouldn't be a cause for concern. If you set it to ~1.0V in BIOS to get a 1.05V readout, technically you're undervolting the rail.
 
801 - 820 of 12401 Posts
Top