F5F Stay Refreshed Hardware Desktop The SoC voltage rise enhances stability for the Ryzen 2000.

The SoC voltage rise enhances stability for the Ryzen 2000.

The SoC voltage rise enhances stability for the Ryzen 2000.

Pages (2): 1 2 Next
A
ATacticalCat_
Member
201
11-29-2018, 04:48 AM
#1
I've faced difficulties with my R7 2700 recently. It often crashes with a black screen while fans spin and nothing else works. The problem began after swapping from an As rock basic board to a B450 Tomahawk, but it continued (not just the board). Updating RAM helped most of the time. I'm considering running an OC to 4 GHz at 1.35–1.375V, as many users say it works fine there. The chip seems capable of reaching that speed, handling CB R20 multiple times before the screen appears again. Some people mention raising the SoC voltage fixed it for them. I’m wondering if boosting the voltage would assist with the CPU upgrade. My Zen+ IPC is slow, so until then I’d like to squeeze more performance from this chip. I’ve used it at 4 GHz for gaming for days without issues, but stability tests always trigger the black screen again. That suggests my CPU might actually support 4 GHz.
A
ATacticalCat_
11-29-2018, 04:48 AM #1

I've faced difficulties with my R7 2700 recently. It often crashes with a black screen while fans spin and nothing else works. The problem began after swapping from an As rock basic board to a B450 Tomahawk, but it continued (not just the board). Updating RAM helped most of the time. I'm considering running an OC to 4 GHz at 1.35–1.375V, as many users say it works fine there. The chip seems capable of reaching that speed, handling CB R20 multiple times before the screen appears again. Some people mention raising the SoC voltage fixed it for them. I’m wondering if boosting the voltage would assist with the CPU upgrade. My Zen+ IPC is slow, so until then I’d like to squeeze more performance from this chip. I’ve used it at 4 GHz for gaming for days without issues, but stability tests always trigger the black screen again. That suggests my CPU might actually support 4 GHz.

D
dbeast22
Junior Member
3
11-29-2018, 07:13 AM
#2
SOC voltage mainly enhances memory stability. A crash at R20 suggests the issue isn’t with the memory. The thermal performance of the CPU seems adequate. If instability persists, consider raising the SOC voltage to 1.15V—the typical optimal range—and observe any improvement, though it may not be significant. It’s unlikely to cause harm.
D
dbeast22
11-29-2018, 07:13 AM #2

SOC voltage mainly enhances memory stability. A crash at R20 suggests the issue isn’t with the memory. The thermal performance of the CPU seems adequate. If instability persists, consider raising the SOC voltage to 1.15V—the typical optimal range—and observe any improvement, though it may not be significant. It’s unlikely to cause harm.

L
Lillmust
Member
153
11-29-2018, 07:35 AM
#3
Temperatures don't seem to be the main concern. The 79.5 °C reading appears acceptable to me. It might be a hypothesis, but I believe certain CPUs have limited memory controllers, which could explain the need for SoC voltage adjustments to improve stability.
L
Lillmust
11-29-2018, 07:35 AM #3

Temperatures don't seem to be the main concern. The 79.5 °C reading appears acceptable to me. It might be a hypothesis, but I believe certain CPUs have limited memory controllers, which could explain the need for SoC voltage adjustments to improve stability.

M
mmillaa
Member
197
12-02-2018, 03:28 AM
#4
Consider raising the SoC voltage between 1.05 and 1.125. Avoid exceeding 1.15v to prevent issues. If that doesn’t work, it might mean your CPU has a problem and you should lower the clock speed for better stability.
M
mmillaa
12-02-2018, 03:28 AM #4

Consider raising the SoC voltage between 1.05 and 1.125. Avoid exceeding 1.15v to prevent issues. If that doesn’t work, it might mean your CPU has a problem and you should lower the clock speed for better stability.

G
Grifo24O
Member
181
12-03-2018, 09:44 AM
#5
I've tested five memory configurations with my 2700X and resolved the problem. It was simple—avoid SK Hynix kits. My Samsung B-die sets worked perfectly. The Zen+ model offers 3733MHz stability and a top frequency of 3966MHz, making it hard to separate the I/O if needed. Enabling XMP automatically raises the voltage to 1.08V; you can safely go up to 1.200V for XMP support. Beyond that, you're targeting high performance. If boosting voltage isn't an option, consider a 2x8GB B-die bundle and you'll likely fix the issue permanently. This advice applies across every Ryzen generation—though older chips like the 3000 and 5000 had better memory controllers. You might find discussions online about XMP limitations with AMD CPUs.
G
Grifo24O
12-03-2018, 09:44 AM #5

I've tested five memory configurations with my 2700X and resolved the problem. It was simple—avoid SK Hynix kits. My Samsung B-die sets worked perfectly. The Zen+ model offers 3733MHz stability and a top frequency of 3966MHz, making it hard to separate the I/O if needed. Enabling XMP automatically raises the voltage to 1.08V; you can safely go up to 1.200V for XMP support. Beyond that, you're targeting high performance. If boosting voltage isn't an option, consider a 2x8GB B-die bundle and you'll likely fix the issue permanently. This advice applies across every Ryzen generation—though older chips like the 3000 and 5000 had better memory controllers. You might find discussions online about XMP limitations with AMD CPUs.

X
XxKripxDeMoNxX
Senior Member
536
12-03-2018, 11:31 AM
#6
Yea, I figured it could still be a RAM issue. Eventhough my current config is fine for the most part. Daily usage on stock seems fine. I'm using custom timings and voltages and tested stabity with MEMTEST86 all night long. Using XMP or running JDEC profile don't seem to change anything regarding CPU OC Stability.
X
XxKripxDeMoNxX
12-03-2018, 11:31 AM #6

Yea, I figured it could still be a RAM issue. Eventhough my current config is fine for the most part. Daily usage on stock seems fine. I'm using custom timings and voltages and tested stabity with MEMTEST86 all night long. Using XMP or running JDEC profile don't seem to change anything regarding CPU OC Stability.

R
roborober567
Member
195
12-03-2018, 11:52 AM
#7
You might want to turn off the "power down mode" in the timing settings. This could reduce throttling during idle periods, which seems likely where the problems arise. If you're using it under heavier loads, it should work well.
R
roborober567
12-03-2018, 11:52 AM #7

You might want to turn off the "power down mode" in the timing settings. This could reduce throttling during idle periods, which seems likely where the problems arise. If you're using it under heavier loads, it should work well.

W
Ward12
Posting Freak
895
12-19-2018, 04:20 AM
#8
I applied a different RAM stability test. I used Testmem5 with anta777 extreme settings and OCCT supporting both SSE and AVX2.
W
Ward12
12-19-2018, 04:20 AM #8

I applied a different RAM stability test. I used Testmem5 with anta777 extreme settings and OCCT supporting both SSE and AVX2.

P
PedroO_
Senior Member
522
12-19-2018, 04:34 AM
#9
It occurs during stress testing with tools like Prime95 and Cinebench. With the older, low-cost RAM, this would happen even when the system was idle.
P
PedroO_
12-19-2018, 04:34 AM #9

It occurs during stress testing with tools like Prime95 and Cinebench. With the older, low-cost RAM, this would happen even when the system was idle.

D
demizio1
Member
178
12-19-2018, 03:36 PM
#10
It's helpful to remember RAM might be the problem. I didn't think a weak memory controller on Zen+ would cause issues. I'll check RAM stability using those applications.
D
demizio1
12-19-2018, 03:36 PM #10

It's helpful to remember RAM might be the problem. I didn't think a weak memory controller on Zen+ would cause issues. I'll check RAM stability using those applications.

Pages (2): 1 2 Next