Ryzen 3rd Gen vs Temperature | PCSPECIALIST

Ryzen 3rd Gen vs Temperature

NoddyPirate

VALUED CONTRIBUTOR
Just for what it's worth as it comes up every now and again - a quick demo of the difference CPU temperature makes to boosts with Ryzen CPU's.

This is my 5600X (overclocked so the frequencies here won't match stock figures) using fan curves to force a high temperature initially and then a quick temperature reduction. The video is sped up a little so no one gets too bored.

Starting at 79 degrees and 4.544 GHz - speed improves by 82 MHz to 4.626 GHz with a 10 degree reduction in package temp.


On this chip the stock AMD cooler it seems will, generally speaking, keep it in the low 80's at full load. A decent air cooler will keep it in the low to mid 60's and an AIO in the 50's. That could potentially be a 250+ MHz difference in boost clocks from one end to the other - which is no small or insignificant change. Sustained boost periods will potentially be better also - they're just harder to quantify.

While the numbers will be different with different setups and different 5000 series CPU's, the effect will nonetheless be the same. Worth keeping in mind I think when choosing a cooler - and in particular when there is little price difference between cooling options.
 

NoddyPirate

VALUED CONTRIBUTOR
Post Script - I wish I had read this thread before I selected my own cooling option. :cry:

(Is that irony? I'm not sure anymore - ever since Alanis Morissette ruined irony for everyone.... :unsure: )
 

Bigfoot

VALUED CONTRIBUTOR
You
Post Script - I wish I had read this thread before I selected my own cooling option. :cry:

(Is that irony? I'm not sure anymore - ever since Alanis Morissette ruined irony for everyone.... :unsure: )
You could always upgrade your cooler yourself. I would have thought it would be a challenge you would relish.
 

NoddyPirate

VALUED CONTRIBUTOR
You

You could always upgrade your cooler yourself. I would have thought it would be a challenge you would relish.
Didn't take long for me to try something else for fun! :)

There has been something bothering me for quite a while now. One of my fan headers was empty and unused. This situation could not be allowed to continue. So I bought another £6 Arctic PWM fan for the fun - this one with an RPM range as close to my CPU Fan as I could get. :)

The experiment was to convert my Push Cpu Cooler into a Push-Pull with two fans instead of one. All in the attempt to lower my temps by any amount and see what extra MHz I could squeeze out of my PBO and Undervolt for the same power draw. It did not go well. :cry:

My CPU fan is in the CPU_FAN header and the new one went into the CPU_OPT - which simply mirrors whatever signals go to CPU_FAN through shared control - it is therefore impossible to control the CPU_OPT independently or directly. I knew having two different fans would be a problem and result in two slightly different speeds, but I was hoping it would work in my favour. Instead, the oppsite new Pull fan always ran at slightly lower RPM than the original Push one. In fact at minimum RPM on the original the new fan stopped altogether.

So I created a perfect obstacle for the air exiting the cooler and the result was 0.5 degrees warmer than previously. Hooray. :mad: Still, it passed 15 minutes of my day and I also learned a new skill - how to install a fan when you can't find any cooler fan clips to fit. Thank the Lord for Zipties! :D

IMG_5988.jpg

IMG_5986.jpg
IMG_5987.jpg
 

NoddyPirate

VALUED CONTRIBUTOR
If only someone had said to get a better cooler 😇
😀 It wouldn’t be funny if it wasn’t true!

But think of all the fun I wouldn’t have had in that case? And think how crazy my Overclock attempts would have become too!
 

NoddyPirate

VALUED CONTRIBUTOR
Might as well just rig up a sprinkler system...that's sprinkles LN2 🥶
I had cryotherapy the other week on a little skin lesion. I asked the Dermatologist if I could borrow her little bottle of LN2 if I promised to have it back the following week.

To say she looked confused is an understatement. I tried to tell her I wanted it for a Spyder on my computer but that seemed to make things worse......
 

FerrariVie

MOST VALUED CONTRIBUTOR
Just for what it's worth as it comes up every now and again - a quick demo of the difference CPU temperature makes to boosts with Ryzen CPU's.

This is my 5600X (overclocked so the frequencies here won't match stock figures) using fan curves to force a high temperature initially and then a quick temperature reduction. The video is sped up a little so no one gets too bored.

Starting at 79 degrees and 4.544 GHz - speed improves by 82 MHz to 4.626 GHz with a 10 degree reduction in package temp.


On this chip the stock AMD cooler it seems will, generally speaking, keep it in the low 80's at full load. A decent air cooler will keep it in the low to mid 60's and an AIO in the 50's. That could potentially be a 250+ MHz difference in boost clocks from one end to the other - which is no small or insignificant change. Sustained boost periods will potentially be better also - they're just harder to quantify.

While the numbers will be different with different setups and different 5000 series CPU's, the effect will nonetheless be the same. Worth keeping in mind I think when choosing a cooler - and in particular when there is little price difference between cooling options.
My brother just got a 5600X and he's on the stock cooler for now... CPU is going to the high-80s with brief spikes to the 90s when gaming, but room temperature should be around 26-29. He'll try undervolting, but it looks like only a good cooler will solve his issue.
 

NoddyPirate

VALUED CONTRIBUTOR
My brother just got a 5600X and he's on the stock cooler for now... CPU is going to the high-80s with brief spikes to the 90s when gaming, but room temperature should be around 26-29. He'll try undervolting, but it looks like only a good cooler will solve his issue.
That’s certainly in line with much of what I’ve read sadly.

I think you’re right about the benefit of a new cooler...
 
Top