Question Is 1.4v safe for 24/7 use

Toggle sidebar Toggle sidebar
Sort by date Sort by votes
Eximo

Eximo

Titan
Ambassador
May 27, 2010
25,839
4,186
139,340
What CPU? Memory controller is in the CPU, so the greatest risk is it eventually failing.

1.35v is Intel's rated maximum, not sure what AMD's is. Some references to 1.5 volts as operable, but not for long term. 1.4 might be okay with AMD.

If you upgrade regularly, shouldn't be too big a deal. If you want to keep this CPU for a long time, I would take it down a notch and try to get it around 1.35 volts.
 
Upvote 0 Downvote
Eximo

Eximo

Titan
Ambassador
May 27, 2010
25,839
4,186
139,340
No one can really answer that. The CPU will either die or it won't. It may get unstable before outright failure and you can back off on the overclock then.

The way you should look at it is if replacing the CPU or memory is an unexpected cost you can afford. If not, you shouldn't tempt fate. If so, go for it. These CPUs haven't been around long enough for anyone to have real data. High end overclockers have broken them certainly, but that only tells you were the outright failure points are.

For example, I am blasting 1.416 volts through my CPU to get 5Ghz, if it melts, I can easily afford a replacement. Something I thought of before delidding it and watercooling.
 
Upvote 0 Downvote
A

Andrewbandrew05

Reputable
Jun 30, 2019
243
17
4,615
I haven't oced my cpu, I've just oced my ram, and I'm not sure if too much voltage running through them will kill my cpu or my ram or what-not or if it's only temps that will do that. Also, I was running prime-95 and unplugged my monitor to prevent screen burn in. When I came back after a while and plugged it back on nothing showed on the screen. Did my pc crash or what happened?
 
Upvote 0 Downvote
Eximo

Eximo

Titan
Ambassador
May 27, 2010
25,839
4,186
139,340
Voltage will certainly kill the memory controller or the memory. All circuits (PN junctions) eventually break down due to electron migration. More voltage makes that process happen faster. The whole principle of operation is to have one material with an excess of electrons and one with a lack of electrons separated by a barrier (junction) voltage and current force those electrons to jump from one side to the other. And you get your amplification or switching effect as desired.

DisplayPort is quite finnicky if that is what you were using, but, yes it could have crashed. Check the event logs.
 
Upvote 0 Downvote
E

EndEffeKt_24

Commendable
Mar 27, 2019
659
157
1,340
I run my memory on 1.44v and I consider 1.45 a barrier that I dont want to break. Micron rates their chips for up to 1.5 as far as I recall.

Not on the topic, but you should really check if your overclock is really beneficial with the timings beeing that lose. Does your infinity fabric even clock high enough to be in sync mode?
Maybe stepping down to like 3600 and tighten the timings would give you better latency and lower voltages.
 
  • Like
Reactions: drivinfast247 and TJ Hooker
Upvote 0 Downvote
T

TJ Hooker

Titan
Ambassador
Apr 15, 2014
11,474
1,623
72,390
Running at that speed means your infinity fabric is only running at half the speed of your memory clock. Compared to ~3700 MHz and lower where it runs at the same speed as memory clock. Having IF at half the speed of memory clock has a negative effect on latency. You're better off running at around 3600-3800, whatever the highest speed you can get and still have IF running 1:1 with memory clock. How far you can OC the IF will vary from one CPU to the next, same as any other form of overclocking.
 
Upvote 0 Downvote
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom