Sorry, I don't, because I never used the stock fan.
My system idles around 32 degrees and, in the time that I have owned it, the highest temp reached has been 60 degrees.
-
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.
Ryzen 3000 Air cooler options, convex and undersized plates of popular choices not suited to die....
- Thread starter MegaPinches
- Start date
Sorry, I don't, because I never used the stock fan.
My system idles around 32 degrees and, in the time that I have owned it, the highest temp reached has been 60 degrees.
- Joined
- 9 Dec 2015
- Posts
- 807
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 33,188
Also worth pointing out not only does it turn out the esports whatever it's called cover the cores, but baseplates are primarily convex because you have a point of contact and heavy pressure in the four corners once it's attached. The idea is it gets bent slightly and provides better pressure in the middle as the pressure flattens the base. If the base starts flat then the pressure causes the flat base to become concave, which means less good contact over most of the surface area of the chip.
It's been a while but this started being done with some large cores on waterblocks. With waterblock 'wars' going on I think one company started making them convex such that they bent a little and flatted when under pressure giving better temps on the whatever large chips it was at the time. Since then many/most waterblocks and heatsinks are designed to be slightly convex.
Also the heatspreader, it's nearly worthless, it's there for protection not heatspreading itself. 99.9% of the heat will go straight from the chips into the heatsink, almost no extra heat will be transferred sideways across the heatspreader.
I'd bet of someone wanted to test by taking a cooler and using a laser or router to cut away a layer of the heatsink in contact with any of the heatspreaded but that above the dies themselves (ie comparing a full heatspreader contact vs contact only over the dies) you'd see negligible temperature differences.
It's been a while but this started being done with some large cores on waterblocks. With waterblock 'wars' going on I think one company started making them convex such that they bent a little and flatted when under pressure giving better temps on the whatever large chips it was at the time. Since then many/most waterblocks and heatsinks are designed to be slightly convex.
Also the heatspreader, it's nearly worthless, it's there for protection not heatspreading itself. 99.9% of the heat will go straight from the chips into the heatsink, almost no extra heat will be transferred sideways across the heatspreader.
I'd bet of someone wanted to test by taking a cooler and using a laser or router to cut away a layer of the heatsink in contact with any of the heatspreaded but that above the dies themselves (ie comparing a full heatspreader contact vs contact only over the dies) you'd see negligible temperature differences.
Last edited:
It's better than it sounds actually. My motherboard doesn't have a BIOS yet with the latest Agesa version. So the CPU is running at ~1.41 volts all the time. It used to boost up to over 4GHz while doing the slightest little thing on the desktop. But, with the chipset drivers released yesterday that has been somewhat sorted. While gaming it runs at about 57 degrees for the most part.
The best thing is that it's virtually silent. I must actually try running the fan at full speed to see what the noise level is like.
I ran fans at full speed to check the same and its surprisingly quiet with the DRP4.
Thanks for the info, saves me the time
It's a nice cooler. Easy to install, quiet and decent cooling. My only real complaint is the Aesthetics. They could have done something to make the top of the heatsink look a little better.
Soldato
I have the Dark Rock 4 on my 3700X. Played about two hours of cities skylines last night (freshly assembled with TG cryonaut) and not once did the CPU go above 50c. Very impressed, especially as my fans speeds were so low. And to think I was tempted to upgrade to the Pro model... seems I didn't need to
(One thing to note, I haven't got any boosts switched on)
Time to decrease amount of common sense inhibiting drinking...
CPU would get crushed by the pressure before base of the heatsink would bend!
There's convex shape in bases of some heatsinks, because Intel couldn't be arsed to do consistently flat heatspreaders and many of them are concave.
http://www.silentpcreview.com/article1366-page1.html
There's even the question about possibility of bending of CPU package over time with LGA socket's lack of firm support.
Can I ask what temps you're getting with it?
Soldato
This was my original cooler when I got my 3700x and I agree that it's not ideal.
This is what I have installed now, temperature under full load is 60C.
Also, it's huge - practically covering my ITX board.
- Joined
- 9 Dec 2015
- Posts
- 807
This was my original cooler when I got my 3700x and I agree that it's not ideal.
This is what I have installed now, temperature under full load is 60C.
Also, it's huge - practically covering my ITX board.
My reckoning is that there has gotta be a point when they are too convex for Ryzen, H7 must be. Good to know some of the others I mentioned are performing well though
Haven't used one yet, but the Thermalright True Spirit Power is also good. 6x8mm heatpipes. It's tall though and hard to find in some countries. If it is available, then it should cost less than Noctua and Be Quiet as well. I'm frankly surprised people don't talk about Thermalright enough.
It might have contact issues with the new ryzens though as with most high end coolers the base is slightly convex(damn you Intel). Also every cpu cooler review before this point has done on (usually) on an Intel build so it's hard to be sure how much all the old coolers actually cool the cpus.
It might have contact issues with the new ryzens though as with most high end coolers the base is slightly convex(damn you Intel). Also every cpu cooler review before this point has done on (usually) on an Intel build so it's hard to be sure how much all the old coolers actually cool the cpus.
Associate
I have a thermalright ultra extreme 120 black edition which I've had for a good few years, with a noctua fan. Managed to get an AM4 compatible bracket for it to use on the the 3700x. The plate wasn't that even so ended up sanding it down a bit to level it off.
Hasn't made a huge drop it temps - cpuz stress test all cores and it sits around 71c, compared to 75c or so on stock cooler - but is a heck of a lot quieter. Core speeds seem to be slightly higher too, +25/+50mhz during all core test.
Given how all this auto boosting works, it could also mean it's boosting higher more often but don't know how to measure that.
Good review and comparison here between Be quiet's dark Rock and pro 4
https://www.guru3d.com/articles_pages/be_quiet_dark_rock_pro_4_review,7.html
Appears under load 3c difference, but not sure if he used a 2nd fan on the ordinary Rock or not, does not feel its worth the extra 25-30 quid which is almost half the cost of the ordinary one.
In gurus review he also had an issue with clearance of his high ram modules, his 1st slow was sacrificed for the greater good.
https://www.guru3d.com/articles_pages/be_quiet_dark_rock_pro_4_review,7.html
Appears under load 3c difference, but not sure if he used a 2nd fan on the ordinary Rock or not, does not feel its worth the extra 25-30 quid which is almost half the cost of the ordinary one.
In gurus review he also had an issue with clearance of his high ram modules, his 1st slow was sacrificed for the greater good.
Cooler Master Master Air MA10M TUF Gaming Edition. Slightly nosier than Noctua but has had good reports for cooling which is up there with the Noctua and Be Quiet from memory. It also has addressable RGB and looks overall a lot better than beige. Can be had for just over £50.
https://www.coolermaster.com/catalog/coolers/cpu-air-coolers/masterair-ma410m-tuf-gaming-edition/
https://www.coolermaster.com/catalog/coolers/cpu-air-coolers/masterair-ma410m-tuf-gaming-edition/
Tested on an Intel chip, we really do need to see more tests on Ryzen 3000 chips to know if and how the non centralised layout of the chiplets affects cooling capabilities of coolers. If there is no difference then it's fine accepting results from tests like these on Intel chips but otherwise they don't have any relevance. Does anyone know if this kind of testing has been done yet by someone with access to a decent variety of coolers?
Quite true we need more better testing I googled and read many sites but no ones really done any full on testing with high end coolers and Ryzen 3000 chips yet let alone explored the die chips issue.