F5F Stay Refreshed Hardware Desktop Overvoltage detected on i5 9600 at 1.344v!

Overvoltage detected on i5 9600 at 1.344v!

Overvoltage detected on i5 9600 at 1.344v!

Pages (3): 1 2 3 Next
S
SenpaiPleasee
Member
65
11-17-2018, 05:31 AM
#1
I already asked on asus forums, but they don't asnwer there... I have great airflow, case big tower cooler master h500p. And even i didn't buy most beefier cooler, for quiet. I have be quiet dark rock slim, which has 62C on benchmark for i7 7700k 4.5ghz. My cpu i5 9600kf and ram g.skill 3200mhz cl16. I am using xpm profile. My vccio and vccsa is high. Vccio 1.328, vccsa 1.224. I read this is, because motherboard overvolt these to compensate for bad ram. My ram is supported in qvl on my asus z390-i gaming motherboard. I want to lower these, but didn't have time yet, because it can cause boot loop and i can't find CLRTC switch on my motherboard and didn't feel like breaking my bios atm... I also i am not sure, if lowering voltage can't damage cpu, or only cause instability. I read this about other things and than i ended up with corrupted ram, by only slightly touching basic timings... Corrupted in memtest and i didn't overclock it, nor changed voltages and it was new, so there is like 0.0... chance it was anything else and bsod, after i did that. So i don't trust anything online. I read that from multiple sources btw... I am crunching foldinghome and my temps are ridiculous, i have great airflow and this cooler is still supposed to be decent and is running on 100% on 70C+. Yet i have up to 76C on stock... While during gaming, even with constant 100% cpu load, i had 64-66C max. in summer... Also cpu voltage is crazy. Cpu is jumping to 1.344v max. Which is for 5ghz overclocks, even less. No way on stock, i should be having voltage this high... There is no manual for oc section. From other site i found: there should be CPU Core/Cache Voltage and if you change this, it should allow you to set cpu offset for voltage. I suppose if you enter value like 0.049. If your cpu were previously reaching 1.35, it will cap on 1.301v. I never used it on my old computer, it didn't have any of these options, so i don't know what they do. And it is hard to find these things from forums, which often contradict themselves. Right know i don't even have this option in my bios, even when i set ai tuner to manual and i would like to avoid that. Because i have no idea what it changes in bios and i don't know how to send all these values manually. My motherboard website: https://www.asus.com/us/Motherboards/ROG.../overview/ I need to get these voltages down, please help!
S
SenpaiPleasee
11-17-2018, 05:31 AM #1

I already asked on asus forums, but they don't asnwer there... I have great airflow, case big tower cooler master h500p. And even i didn't buy most beefier cooler, for quiet. I have be quiet dark rock slim, which has 62C on benchmark for i7 7700k 4.5ghz. My cpu i5 9600kf and ram g.skill 3200mhz cl16. I am using xpm profile. My vccio and vccsa is high. Vccio 1.328, vccsa 1.224. I read this is, because motherboard overvolt these to compensate for bad ram. My ram is supported in qvl on my asus z390-i gaming motherboard. I want to lower these, but didn't have time yet, because it can cause boot loop and i can't find CLRTC switch on my motherboard and didn't feel like breaking my bios atm... I also i am not sure, if lowering voltage can't damage cpu, or only cause instability. I read this about other things and than i ended up with corrupted ram, by only slightly touching basic timings... Corrupted in memtest and i didn't overclock it, nor changed voltages and it was new, so there is like 0.0... chance it was anything else and bsod, after i did that. So i don't trust anything online. I read that from multiple sources btw... I am crunching foldinghome and my temps are ridiculous, i have great airflow and this cooler is still supposed to be decent and is running on 100% on 70C+. Yet i have up to 76C on stock... While during gaming, even with constant 100% cpu load, i had 64-66C max. in summer... Also cpu voltage is crazy. Cpu is jumping to 1.344v max. Which is for 5ghz overclocks, even less. No way on stock, i should be having voltage this high... There is no manual for oc section. From other site i found: there should be CPU Core/Cache Voltage and if you change this, it should allow you to set cpu offset for voltage. I suppose if you enter value like 0.049. If your cpu were previously reaching 1.35, it will cap on 1.301v. I never used it on my old computer, it didn't have any of these options, so i don't know what they do. And it is hard to find these things from forums, which often contradict themselves. Right know i don't even have this option in my bios, even when i set ai tuner to manual and i would like to avoid that. Because i have no idea what it changes in bios and i don't know how to send all these values manually. My motherboard website: https://www.asus.com/us/Motherboards/ROG.../overview/ I need to get these voltages down, please help!

N
nicilbar
Junior Member
42
11-18-2018, 08:18 AM
#2
Asus is known for OV Vccia and system agent voltages. This isn't due to "bad ram" in general. It stems from the way the BIOS is programmed. Adjust these settings manually. They shouldn't go above 1.20v; begin at 1.10v and check stability.
N
nicilbar
11-18-2018, 08:18 AM #2

Asus is known for OV Vccia and system agent voltages. This isn't due to "bad ram" in general. It stems from the way the BIOS is programmed. Adjust these settings manually. They shouldn't go above 1.20v; begin at 1.10v and check stability.

A
Antez03
Member
174
11-19-2018, 06:38 AM
#3
Check the numbers, ASRock is known for strange overvoltage settings and buggy firmware. I won’t buy from ASUS again for other reasons. I heard from an admin at ASUS Support that this isn’t official support, but the details were mixed up. He mentioned CPU controllers, not RAM, and didn’t mention voltage specs. I still feel uncertain about the RAM amount or the CPU voltage. I’m considering lowering it slightly to test stability, but I want to be careful and avoid boot loops. I’m planning small tweaks before jumping to a higher voltage. I hope the CLRTC switch isn’t hidden under the GPU, which is frustrating after all the disassembly. I found an option for CPU Core/Cache Voltage that lets you set a negative offset, but I haven’t tried it yet. As per your advice, changing the VCCS should help. I set it to 1.29375 then 1.28750, but the BIOS still shows 1.312V at full speed in Windows. It’s unclear if that’s the right value or just an average. The adaptive BCLK is auto and enabled, which might be contributing. I’m thinking it could be causing the instability. I might disable it to see if it improves things. My adjustments aren’t working, and I’ve been testing in 0.625 increments. Even with adaptive BCLK off, it’s still at 1.312V in BIOS. I’m confused about the exact range—min, max, or average. My RAM is now undervolting to 1.344V and sometimes spikes to 1.36V. Gskill says 1.36V is fine by default, but I’m not sure if it’s safe to drop below 1.35V. The CPU core voltage is now at 1.2V, which seems too low. The BIOS lists a 0.0625 increment option, but it doesn’t seem to apply. There’s also a fluctuation to 1.232V during about 5% load, which isn’t ideal. My RAM is now at 1.34V and sometimes jumps to 1.36V. The cooler website claims it’s optimized for 120W DPD and high cooling, but the BIOS curve suggests fans should run at full speed. This is confusing. I’ll skip the stress test for now and see how these changes affect stability.
A
Antez03
11-19-2018, 06:38 AM #3

Check the numbers, ASRock is known for strange overvoltage settings and buggy firmware. I won’t buy from ASUS again for other reasons. I heard from an admin at ASUS Support that this isn’t official support, but the details were mixed up. He mentioned CPU controllers, not RAM, and didn’t mention voltage specs. I still feel uncertain about the RAM amount or the CPU voltage. I’m considering lowering it slightly to test stability, but I want to be careful and avoid boot loops. I’m planning small tweaks before jumping to a higher voltage. I hope the CLRTC switch isn’t hidden under the GPU, which is frustrating after all the disassembly. I found an option for CPU Core/Cache Voltage that lets you set a negative offset, but I haven’t tried it yet. As per your advice, changing the VCCS should help. I set it to 1.29375 then 1.28750, but the BIOS still shows 1.312V at full speed in Windows. It’s unclear if that’s the right value or just an average. The adaptive BCLK is auto and enabled, which might be contributing. I’m thinking it could be causing the instability. I might disable it to see if it improves things. My adjustments aren’t working, and I’ve been testing in 0.625 increments. Even with adaptive BCLK off, it’s still at 1.312V in BIOS. I’m confused about the exact range—min, max, or average. My RAM is now undervolting to 1.344V and sometimes spikes to 1.36V. Gskill says 1.36V is fine by default, but I’m not sure if it’s safe to drop below 1.35V. The CPU core voltage is now at 1.2V, which seems too low. The BIOS lists a 0.0625 increment option, but it doesn’t seem to apply. There’s also a fluctuation to 1.232V during about 5% load, which isn’t ideal. My RAM is now at 1.34V and sometimes jumps to 1.36V. The cooler website claims it’s optimized for 120W DPD and high cooling, but the BIOS curve suggests fans should run at full speed. This is confusing. I’ll skip the stress test for now and see how these changes affect stability.

J
johnitipek
Member
80
11-19-2018, 07:41 AM
#4
LLC adjusts voltage dynamically to handle v-droop, which is standard for all motherboards. When overclocking manually, disabling c-states helps. For heavy overclocks, higher voltages are used and cooling must support them. Most 14nm chips typically run around 4.7ghz at 1.30v and 5ghz at 1.35v, with minor variations per chip. A weak memory controller isn't always the problem; a solid B-Die setup minimizes issues. For Intel, memory performance is crucial. Lastly, keep Vcore under 1.2v and begin at 1.10v.
J
johnitipek
11-19-2018, 07:41 AM #4

LLC adjusts voltage dynamically to handle v-droop, which is standard for all motherboards. When overclocking manually, disabling c-states helps. For heavy overclocks, higher voltages are used and cooling must support them. Most 14nm chips typically run around 4.7ghz at 1.30v and 5ghz at 1.35v, with minor variations per chip. A weak memory controller isn't always the problem; a solid B-Die setup minimizes issues. For Intel, memory performance is crucial. Lastly, keep Vcore under 1.2v and begin at 1.10v.

F
72
11-24-2018, 05:00 PM
#5
Maybe lowering the LLC to a lower setting would help. Right now I'm not pushing the CPU too hard. The temperatures are off, which is good. I have solid cooling with a 180W TDP and strong airflow, unlike my i5 9600K where I see high temps. It seems fine for me to keep RAM below 1.35V. G.skill Trident Z should be B-Die, but I'm not sure. Setting it too low right away isn't comfortable—I've heard it can lead to boot loops. I plan to start at 1.1 since there are no issues so far.
F
Fluffy_BananaZ
11-24-2018, 05:00 PM #5

Maybe lowering the LLC to a lower setting would help. Right now I'm not pushing the CPU too hard. The temperatures are off, which is good. I have solid cooling with a 180W TDP and strong airflow, unlike my i5 9600K where I see high temps. It seems fine for me to keep RAM below 1.35V. G.skill Trident Z should be B-Die, but I'm not sure. Setting it too low right away isn't comfortable—I've heard it can lead to boot loops. I plan to start at 1.1 since there are no issues so far.

D
DareDevu
Member
56
11-24-2018, 11:11 PM
#6
I would set the memory to 1.360 volts. Turn off "Power Down Mode" unless you're connected to a 4000mhz and CL16 setup with your memory modules, in which case you won't need more than 1.10 volts for the Vcc period. The same applies to the system agent.
D
DareDevu
11-24-2018, 11:11 PM #6

I would set the memory to 1.360 volts. Turn off "Power Down Mode" unless you're connected to a 4000mhz and CL16 setup with your memory modules, in which case you won't need more than 1.10 volts for the Vcc period. The same applies to the system agent.

B
banshee45
Senior Member
726
11-25-2018, 07:28 AM
#7
I didn't dispute your points, just wanted minor adjustments and a preliminary check, since I heard it might lead to boot loops. Both VCCIO and VCCSA now sit at 1.120v, yet the vcore stays high at 1.288 max and current remains around 1.279 across all cores during stress testing. Previously, with VCCSA set to auto and VCCIO at 1.2 (roughly 1.212), the vcore was only about 1.244 max, again under stress. Clearly, it's lower than expected. Temperatures still spike to 76°C and have been consistently high. It did help a bit, but temperatures didn't drop significantly. I plan to reduce the LLC setting, as I'm not currently pushing the CPU hard. However, it seems the issue isn't resolved. This cooler appears to be of lower quality, though it should have been reliable. I don't recognize a "power down mode" in the ASUS forums, which should be listed under DRAM Timing Control. BIOS settings show VCCIO at 1.35v and VCCSA at 1.36v, but BIOS doesn't mention a power-down option. Ram voltage is set to 1.36v in BIOS, though it reads 1.35v. I'll keep it at 1.35v. EDIT: lowered LLC from 2 to 4, still 1.12 on both channels. At 5GHz, it reaches 72.2°C during load, but not all tasks are identical. Maximums were around 64-66°C in summer gaming sessions. Crunching Corona gave 76°C or higher under stress. Still, it feels excessive for the stock unit. Maybe adding a second fan would help—it spins at only 1,100 RPM. ASUS seems to be confusing things. In BIOS it says higher LLC reduces voltage droop but increases overshoot, so I set it to 4 from 2 and voltage jumped to 1.312-1.328. They don't provide clear installation instructions, and their guidance is outdated from a decade ago. Even at the lowest LLC setting (level 1), VCCIO and VCCSA stay at 1.12v minimum! I haven't tried negative offset yet. Also, I don't know if voltage fluctuations on RAM are acceptable. EDIT2: I experimented with negative offset, expecting a max of 1.288v on vcore. I adjusted it to -0.040 (0.005 increments). During stress testing, max reached 1.225v with a drop to 1.234v. Temperatures stayed high at 65-70°C. This improved the temperature by about 5°C, which is minimal. Still, it's not ideal. I'm concerned about potential CPU damage from overvoltage, especially since I had my PC in the heat during summer. I didn't find any voltage-related warnings in reviews—only that it might overvolt RAM, and I checked only DRAM, vcore, and RAM voltage, which seemed normal. Also, the -5°C drop came from tighter fan curves, not voltage changes.
B
banshee45
11-25-2018, 07:28 AM #7

I didn't dispute your points, just wanted minor adjustments and a preliminary check, since I heard it might lead to boot loops. Both VCCIO and VCCSA now sit at 1.120v, yet the vcore stays high at 1.288 max and current remains around 1.279 across all cores during stress testing. Previously, with VCCSA set to auto and VCCIO at 1.2 (roughly 1.212), the vcore was only about 1.244 max, again under stress. Clearly, it's lower than expected. Temperatures still spike to 76°C and have been consistently high. It did help a bit, but temperatures didn't drop significantly. I plan to reduce the LLC setting, as I'm not currently pushing the CPU hard. However, it seems the issue isn't resolved. This cooler appears to be of lower quality, though it should have been reliable. I don't recognize a "power down mode" in the ASUS forums, which should be listed under DRAM Timing Control. BIOS settings show VCCIO at 1.35v and VCCSA at 1.36v, but BIOS doesn't mention a power-down option. Ram voltage is set to 1.36v in BIOS, though it reads 1.35v. I'll keep it at 1.35v. EDIT: lowered LLC from 2 to 4, still 1.12 on both channels. At 5GHz, it reaches 72.2°C during load, but not all tasks are identical. Maximums were around 64-66°C in summer gaming sessions. Crunching Corona gave 76°C or higher under stress. Still, it feels excessive for the stock unit. Maybe adding a second fan would help—it spins at only 1,100 RPM. ASUS seems to be confusing things. In BIOS it says higher LLC reduces voltage droop but increases overshoot, so I set it to 4 from 2 and voltage jumped to 1.312-1.328. They don't provide clear installation instructions, and their guidance is outdated from a decade ago. Even at the lowest LLC setting (level 1), VCCIO and VCCSA stay at 1.12v minimum! I haven't tried negative offset yet. Also, I don't know if voltage fluctuations on RAM are acceptable. EDIT2: I experimented with negative offset, expecting a max of 1.288v on vcore. I adjusted it to -0.040 (0.005 increments). During stress testing, max reached 1.225v with a drop to 1.234v. Temperatures stayed high at 65-70°C. This improved the temperature by about 5°C, which is minimal. Still, it's not ideal. I'm concerned about potential CPU damage from overvoltage, especially since I had my PC in the heat during summer. I didn't find any voltage-related warnings in reviews—only that it might overvolt RAM, and I checked only DRAM, vcore, and RAM voltage, which seemed normal. Also, the -5°C drop came from tighter fan curves, not voltage changes.

M
Master_Pepe
Junior Member
15
11-25-2018, 07:11 PM
#8
I understand you weren't disputing, I was simply trying to be precise. The readings look good. 76c offers ample overclocking capacity. For the stock model, keep it at 100c (limit). Running at 85c or lower during steady tasks is ideal.
M
Master_Pepe
11-25-2018, 07:11 PM #8

I understand you weren't disputing, I was simply trying to be precise. The readings look good. 76c offers ample overclocking capacity. For the stock model, keep it at 100c (limit). Running at 85c or lower during steady tasks is ideal.

N
NatsuNegao
Junior Member
25
11-27-2018, 12:24 AM
#9
While 7700k on a 5GHz processor only reaches 72°C seems off. Also my setup has excellent airflow. I chose this model because of its strong cooling performance. According to the review at GamersNexus, it offers great thermal management and should handle 180W TDP. It might be tricky to judge accuracy, as it depends on the workloads. These sites likely know how to properly stress-test CPUs. If I decide to overclock, I’ll consider adding a second fan. For now it’s working well. Thanks for helping me lower those voltages that were too high.
N
NatsuNegao
11-27-2018, 12:24 AM #9

While 7700k on a 5GHz processor only reaches 72°C seems off. Also my setup has excellent airflow. I chose this model because of its strong cooling performance. According to the review at GamersNexus, it offers great thermal management and should handle 180W TDP. It might be tricky to judge accuracy, as it depends on the workloads. These sites likely know how to properly stress-test CPUs. If I decide to overclock, I’ll consider adding a second fan. For now it’s working well. Thanks for helping me lower those voltages that were too high.

M
MarioCovrigel
Member
195
11-27-2018, 02:17 AM
#10
It’s good to hear! The cooler seems ready for higher performance builds, especially with the 7700K’s potential for overclocking. I think it can handle a higher v-core and likely support 180W, with frequencies in the 1.375 to 1.385v range and around 5.1GHz.
M
MarioCovrigel
11-27-2018, 02:17 AM #10

It’s good to hear! The cooler seems ready for higher performance builds, especially with the 7700K’s potential for overclocking. I think it can handle a higher v-core and likely support 180W, with frequencies in the 1.375 to 1.385v range and around 5.1GHz.

Pages (3): 1 2 3 Next