• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

*** Official Ryzen Owners Thread ***

Permabanned
Joined
23 Apr 2014
Posts
23,553
Location
Hertfordshire
No. 1.4v is safe for 2nd gen Ryzen, not 1.45v. 1st gen Ryzen safe max was 1.425v.....................................1.45v is ok for a bit of benching but defo not for any sort a 24/7 running.........................unless you is running under Phase ? :D

Goes way over 1.45v when boosting at stock settings. Where did you get the 1.4v safe from?
 
Soldato
Joined
26 May 2014
Posts
2,952
Goes way over 1.45v when boosting at stock settings. Where did you get the 1.4v safe from?
Only for short bursts. AMD have said that 1.5V+ is fine at stock because the chip only spends very short periods at those voltages, only when it has thermal headroom and only when just a couple of cores are loaded. The safe all-core voltage for 24/7 use is much lower. I don't know about official sources for the exact cutoff, but Buildzoid has said that you want to stay under 1.4V when overclocking manually.
 
Permabanned
Joined
23 Apr 2014
Posts
23,553
Location
Hertfordshire
Only for short bursts. AMD have said that 1.5V+ is fine at stock because the chip only spends very short periods at those voltages, only when it has thermal headroom and only when just a couple of cores are loaded. The safe all-core voltage for 24/7 use is much lower. I don't know about official sources for the exact cutoff, but Buildzoid has said that you want to stay under 1.4V when overclocking manually.

If you are working on your PC all day its hitting those voltages all the time though. Just wondering where the "safe" info comes from is all. :)

Im running 4.2 @ 1.381 24/7 as I didn't like the voltages it uses when "normal" boosting, although I expect its fine.
 
Soldato
Joined
14 Dec 2003
Posts
5,683
I don't think I've ever had a CPU fail on me, however old or whatever volts I've put through it. I suspect someone running 1.45v daily wouldn't have any issues in the time that the CPU is still relevant and not outdated. Might shorten the lifespan but by the time it fails it'll be obsolete anyway!
 
Permabanned
Joined
23 Apr 2014
Posts
23,553
Location
Hertfordshire
I don't think I've ever had a CPU fail on me, however old or whatever volts I've put through it. I suspect someone running 1.45v daily wouldn't have any issues in the time that the CPU is still relevant and not outdated. Might shorten the lifespan but by the time it fails it'll be obsolete anyway!

Now I think on it, I have not had a CPU fail either. Good call, trying my 2700X with 1.6v later. :p
 
Soldato
Joined
26 May 2014
Posts
2,952
If you are working on your PC all day its hitting those voltages all the time though.
It's irrelevant how often it hits that voltage really - only for how long and how hot it gets in the process. There really aren't many common desktop workloads where it's going to hit those voltages often, and the CPU will automatically ramp down both voltage and clock speed once it's reached what it considers a safe limit (in terms of time or thermals). Spikes to 1.5V+ are usually over inside one update in a program like HWiNFO64 (2000ms by default), and only occur under very light loads when only using a core or two. Anything heavier and both clock speed and voltage come tumbling down. On the previous MSI board I had this 2700X in, voltage dropped all the way down to ~1.3V with all cores loaded, with a boost of ~3.95GHz, which is probably about right for bone stock. On the Crosshair VI I replaced it with, I'm getting ~4.15GHz all-core boost at ~1.375V under full load with the 'Performance Enhancer' level two setting and no other manual tweaking. The single-core boost on both boards was the same 4.35GHz limit with voltages bursting to 1.5V+ during those short periods. I've left HWiNFO64 open for hours at a time whilst just using my PC for desktop work and some gaming, and the averaged voltage over the whole period is always somewhere between 1V and 1.1V, so on average the CPU is seeing very low voltage levels in common desktop usage over a long period of time.
 
Permabanned
Joined
23 Apr 2014
Posts
23,553
Location
Hertfordshire
Seems to me it hits those higher voltages fairly often, just in average PC usage. Using PE level 4 at the moment, does 4225 all core and standard 4.35 boost. Though the all core voltage is over 1.4 the temperatures are totally fine. Not sure its even worth using boost vs all core clocks as its not used in games that I can see, PUBG which appears to be using one core at high loads and the rest intermittently but it just locks it out at 4225 all core, though I have not tried many other games.

Even heavy loads(handbrake) im "only" hitting 69 degrees so expect the voltages are fine. Not disagreeing as such, just difficult to know what is "safe" and what is "working as intended". :)
 
Associate
Joined
12 Mar 2017
Posts
1,115
Location
Ireland
Even if it's B-die, you won't always get rated speeds. My R7 1700's IMC can't do more than 2933Mhz reliably, I always end up getting memory errors if I try higher even though initially it seems stable, even after 10000% memtests I inevitably end up getting errors after a few days.
Most I can get stable with a B-die kit is 2933Mhz with tighter C14 timings, which is a shame since apparently I got a really good kit that can do 3200 C12, but the CPU IMC is just letting it down.
 
Soldato
Joined
1 Mar 2010
Posts
14,366
Location
5 degrees starboard
Even if it's B-die, you won't always get rated speeds. My R7 1700's IMC can't do more than 2933Mhz reliably, I always end up getting memory errors if I try higher even though initially it seems stable, even after 10000% memtests I inevitably end up getting errors after a few days.
Most I can get stable with a B-die kit is 2933Mhz with tighter C14 timings, which is a shame since apparently I got a really good kit that can do 3200 C12, but the CPU IMC is just letting it down.


Still doing 3333C14 on Gskill Ripjaws 3200C14. This is with 1700 at 3.6 all cores. Not too aggressive but on stock cooling and fully stable.
 
Associate
Joined
12 Mar 2017
Posts
1,115
Location
Ireland
It's not really when my old 4.8Ghz i7 2600K was better in single threaded workloads. But I got it for the extra cores, so it's doing its job thus far, just a shame that I can't squeeze more out of it.
 
Soldato
Joined
12 Mar 2006
Posts
22,990
Location
N.E England
Swapped out my 1600 recently and upgraded to a 1700x when they were on offer, the upgrade only cost me 50 quid after selling the 1600 :D

Two more cores and managed to squeeze 4Ghz out of it.
 
Soldato
Joined
6 Aug 2009
Posts
7,071
It's not really when my old 4.8Ghz i7 2600K was better in single threaded workloads. But I got it for the extra cores, so it's doing its job thus far, just a shame that I can't squeeze more out of it.

What motherboard do you have? Is the RAM single rank? It may not be your IMC. Looking at reviews there seems to be quite a few factors in play with Ryzen memory.
 
Permabanned
Joined
12 Sep 2013
Posts
9,221
Location
Knowhere
While the supplied 2700x cooler is half decent for me it keeps revving up and down like it's on the edge of the required temp level and has to keep speeding up to pull the temperature back down which can be annoying, that's with a Corsair C70 case. I just ordered a Be Quiet Dark Rock Pro 3 to replace it, I've wanted to get a proper high end air cooler for a while so now I have this one should last longer than me. :D
 
Soldato
Joined
26 May 2014
Posts
2,952
While the supplied 2700x cooler is half decent for me it keeps revving up and down like it's on the edge of the required temp level and has to keep speeding up to pull the temperature back down which can be annoying, that's with a Corsair C70 case. I just ordered a Be Quiet Dark Rock Pro 3 to replace it, I've wanted to get a proper high end air cooler for a while so now I have this one should last longer than me. :D
There should be a setting in your BIOS somewhere to control fan ramp up/down time. Increasing that a bit should stop the fan spinning up every time the temperature briefly spikes. Even my Noctua NH-U14S exhibits that behaviour if I don't increase ramp up time a bit.
 
Associate
Joined
23 Feb 2009
Posts
2,395
Location
Bournemouth
While the supplied 2700x cooler is half decent for me it keeps revving up and down like it's on the edge of the required temp level and has to keep speeding up to pull the temperature back down which can be annoying, that's with a Corsair C70 case. I just ordered a Be Quiet Dark Rock Pro 3 to replace it, I've wanted to get a proper high end air cooler for a while so now I have this one should last longer than me. :D

Yeah i also can hear the fan revving up and down all over the place ;).

I agree with you about the stock cooler that comes with the processor as half decent indeed.

I have a xigmatek pc case so you are not the only person to hear the fan going up and down like a yoo yoo.
 
Permabanned
Joined
12 Sep 2013
Posts
9,221
Location
Knowhere
There should be a setting in your BIOS somewhere to control fan ramp up/down time. Increasing that a bit should stop the fan spinning up every time the temperature briefly spikes. Even my Noctua NH-U14S exhibits that behaviour if I don't increase ramp up time a bit.

I'll remember to check that if needed in the future, Cheers

Yeah i also can hear the fan revving up and down all over the place ;).

I agree with you about the stock cooler that comes with the processor as half decent indeed.

I have a xigmatek pc case so you are not the only person to hear the fan going up and down like a yoo yoo.

You should have a go at changing the fan settings in the bios as lltfdaniel suggested.
 
Back
Top Bottom