I've decided to set my voltage to about 1.47v to test and see if it'll be enough power to "drive" a phenom 2 x4 [email protected]. I know there's both vdroop and vdrop, and one of them is setting my voltage to around 1.42v. The real issue is when I run prime95, it'll drop EVEN FARTHER DOWN. This is beyond frustrating, because I'll have to account for two drops and sort of guess how high up I'll need to go. So that already lacking 1.42v will be 1.39, and then my system will crash shortly afterwards. How can I curb that second drop? I know the vdrop/vdroop is impossible to work around, but surely there's something I can do about that second drop at full load? I don't want to have to crank my shit up to 1.55v and risk burning it out...
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
Navigation
Install the app
How to install the app on iOS
Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
More options
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
Issue with voltage...
- Thread starter zero940
- Start date
BillParrish
Supreme [H]ardness
- Joined
- Aug 25, 2006
- Messages
- 7,519
Short story, slightly more Vdrop and I am guessing more Vdroop at high load currents compared to idle or light load.
Voltage drop across a component (cap, inductor, etc.) = Current x internal component resistance.
Going from idle or light load to full load will make the current jump from a few amps to over 50amps and the voltage loss as you have noticed jumps as well. The voltage regulation circuit will compensate for most, but not all of that.
I am also thinking that like Intel the voltage regulation circuity/controller has a "programmed in" amount of Vdroop that increases as the load current increases because it is large changes in load current that cause the voltage regulation circuit to overshoot the target voltage while it compensates. If like Intel this is a constant linear decrease but the line has a different slope than actual voltage so as the cpu load increases so does the voltage offset or "droop".
I could not find exact info in AMD docs online, http://support.amd.com/us/psearch/P...Tech+Doc+Processor&ostype=&keywords=&items=20
I think you have to order it.
• Refer to the AMD Family 10h Processor Electrical Data Sheet, order# 40014, for electrical details of
AMD Family 10h processors.
Short of measuing the actual voltage going to the cpu with a voltmeter under full load I dont know how you could compensate and if you even should. The trouble is that if you set the bios voltage high(er) to make up for the full load conditions you might be in a voltage danger zone at idle or light load. But I have not messed with an AMD since Thunderbird and Barton and without the tech doc on the loadline offset I am shooting in the dark here.
Voltage drop across a component (cap, inductor, etc.) = Current x internal component resistance.
Going from idle or light load to full load will make the current jump from a few amps to over 50amps and the voltage loss as you have noticed jumps as well. The voltage regulation circuit will compensate for most, but not all of that.
I am also thinking that like Intel the voltage regulation circuity/controller has a "programmed in" amount of Vdroop that increases as the load current increases because it is large changes in load current that cause the voltage regulation circuit to overshoot the target voltage while it compensates. If like Intel this is a constant linear decrease but the line has a different slope than actual voltage so as the cpu load increases so does the voltage offset or "droop".
I could not find exact info in AMD docs online, http://support.amd.com/us/psearch/P...Tech+Doc+Processor&ostype=&keywords=&items=20
I think you have to order it.
• Refer to the AMD Family 10h Processor Electrical Data Sheet, order# 40014, for electrical details of
AMD Family 10h processors.
Short of measuing the actual voltage going to the cpu with a voltmeter under full load I dont know how you could compensate and if you even should. The trouble is that if you set the bios voltage high(er) to make up for the full load conditions you might be in a voltage danger zone at idle or light load. But I have not messed with an AMD since Thunderbird and Barton and without the tech doc on the loadline offset I am shooting in the dark here.