Ryzen 5 1600x Stock Voltages seem Very High?

Zarathustra[H]

Extremely [H]
Joined
Oct 29, 2000
Messages
39,038
Hey all,

Last year I posted here about the temps on my stepsons build seeming oddly high. I never quite got to the bottom of it, but the system worked, and since his application is light, and doesn't benefit much if at all from overclocking, I let it slide.

Fast forward almost a year later, and after a drive failure in his rig, I'm setting everything back up again. I decided to do some stability testing just to make sure, and I notice something odd.

upload_2018-10-21_19-38-22.png



Just sitting idle on the desktop, voltages are alternating between 1.287 and 1.550. This seems insane to me, and these are with the default settings on the MSI B350 Tomahawk motherboard.

1.) Am I just not familiar with the Ryzen chips? Do they run at higher voltages than I am used to?

2.) Could HWMonitor be reading incorrectly here?

3.) Or, are the default settings on the MSI B350 Tomahawk just really bad?

Would it make sense to go in and manually lower the voltages?

Appreciate any feedback/suggestions here.
 
I watch Ryzen master and the voltage is unlocked (range the cpu runs at) with turbo clocks come need of more voltage.

If it shows 1.4v for 2200g which means in default mode the cpu has that much of range of cap if temps are good as to boost longer clocks.
 
Last edited:
yes its a little high, due to the voltage being a bit too high on auto. bios up to date? I see you did last year but there have been several updates since then. start there as it may have been corrected in an update. if not id try setting it manually, see how it acts.
 
try hwinfo64 and see if you're seeing the same numbers, i've personally never used hwmonitor so i don't know if it works properly. my 1600 has a 1.050v VID and hovers between 1.16-1.29v(unboosted 3.4Ghz) with a max reported voltage of 1.31v(3.8Ghz boost voltage). i'd suspect the 1600x will be slightly higher than that. the 1.55v thing sounds similar to the issue people were having with the initial release bios for the zen+ chips which was fixed pretty quickly as far as i know. with my 1600 i had to set the LLC to 3 on my taichi otherwise my voltages were all over the place. are you using the ryzen or windows power profile?
 
yes its a little high, due to the voltage being a bit too high on auto. bios up to date? I see you did last year but there have been several updates since then. start there as it may have been corrected in an update. if not id try setting it manually, see how it acts.

Yeah, I have the latest BIOS on it from October of this year. That was the first thing I checked. Seems like MSI just screwed this one up to me.

try hwinfo64 and see if you're seeing the same numbers, i've personally never used hwmonitor so i don't know if it works properly. my 1600 has a 1.050v VID and hovers between 1.16-1.29v(unboosted 3.4Ghz) with a max reported voltage of 1.31v(3.8Ghz boost voltage). i'd suspect the 1600x will be slightly higher than that.

HWmonitor has always been my goto, but I've never used it on Ryzen before. Maybe that could be part of the problem. I want to remember that I double checked the values using Ryzen Master just to be sure though. At least I think I did. I'll have to take another look.

the 1.55v thing sounds similar to the issue people were having with the initial release bios for the zen+ chips which was fixed pretty quickly as far as i know. with my 1600 i had to set the LLC to 3 on my taichi otherwise my voltages were all over the place.

Hmm. I'll have to look for the load line calibration. I hadn't considered that could be the culprit. Those are usually set pretty conservatively stock.

I did go in and manually set the voltage to 1.35v Stock it looks like it was set to 1.424V

upload_2018-10-22_13-1-31.png


Its much cooler at 1.35V and still stable under load. Even this seems a little on the high end voltage wise though for stock clocks. Right?

To make things more complicated I also think I had some problems with the Corsair H110i GTX cooler I used in this build. At one point when it was really hot, and throttling like mad during a p95 test, I opened the case and started feeling the tubes to see if it was actually removing the heat to the radiator, and as soon as I did so, the temperatures suddenly dropped, and the auto fan slowed down the fans. At first I thought p95 had crashed, but it turns out it was still running at full speed, and it was no longer throttling.

I'm guessing there might have been a bubble caught in the line, impeding flow, and my touching it, loosened it? Either that or the cooler is suspect. I'm going to have to monitor it and potentially replace it.


are you using the ryzen or windows power profile?

Huh? I don't know what you are talking about here. Mind explaining?

Appreciate all the help.
 
I think he is referencing the "AMD Ryzen™ Balanced" windows power profile vs one of the stock windows power profiles.

Ah, I never realized there was a Ryzen power profile. Where does it come from? Is it when you install the AMD chipset drivers?
 
Just because I am curious. If you were to try to replicate proper stock configuration, what voltage would you set? Is 1.35v about right, or should it be lower?
 
yup and try moving the aio to the top of the case.

That was where I first tried to install it, but it would not fit.

While ai live the build quality of this case (it's truly spectacular, feels very solid) they did have a bit of a fail during the design. There simply is not enough space on either side of the fan slots to install a dual 140mm radiator in the top.

If you try to use the rear two fan slots the part of the radiator that extends beyond the fans hits the back of the case. If you try to use the front two slots, that same part interferes with the front.

I guess I could drilly my own mounting holes in the middle and center the dual slot radiator over the middle fan slot, but that might look really odd.
 
Back
Top