Question DeepCool AK400 4-heatpipe cooler, 220W TDP, $34.99 @ Amazon. Opinions?

VirtualLarry

No Lifer
Aug 25, 2001
56,229
9,990
126


I ordered two from Newegg, after watching the videos for installation on Amazon (AM4).
 

UsandThem

Elite Member
May 4, 2000
16,068
7,380
146
Look to be good budget coolers, a little bit better than the Cooler Master 212 EVO at the same price.

Of course, there are better beefier options when going up another $15-$20 like the Mugen 5, Big Shurikan 3, or Be Quiet Shadow Rock 3.
 

Motostu

Senior member
Oct 5, 2020
496
527
106
No opinion on this model, but I have a DeepCool GAMMAXX GTE V2 on my 5600 (180W TDP I think), and it seems to be doing well. Nice and quiet, and keeps temps reasonable.

The CPU interface on the cooler wasn't perfectly flat, and being the old overclocker I am I lapped it. Not sure if necessary with good thermal paste, but makes me feel better.

Thanks for posting, and update us when you've had a chance to use it. I may throw in a faster CPU at some point and might be looking for something that can dissipate more heat.
 
Last edited:

VirtualLarry

No Lifer
Aug 25, 2001
56,229
9,990
126
Sorry, was in a hurry. Forgot to mention, planning on using them both each with a Ryzen 9 5900X. Do you think that the 220W TDP is accurate, or that they will be enough? Load is ging to be DC work (if I can finally get my main desktop stable again) and maybe some mining. Mostly PrimeGrid, which I haven't been able to do, since I think either some of my mobo's VRMs are friend, or my CM AIO leaked. It doesn't make any noise any more. Hence I want to go with a Semi decent air cooler this time.

Case for each will be these:

Hopefully mesh front and three intake fans will keep the air moving and cool.
 

Tech Junky

Diamond Member
Jan 27, 2022
3,365
1,118
106
When looking at coolers I tend to look at the ones with at least 6 pipes. These provide more surface area for the heat to dissipate. I also add a second fan for push/pull but also redundancy if one fails.

A $50 or $100+ cooler do the same thing. Now design can make a difference for instance can you rotate 90 degrees and ain it towards the top for exhaust and pull GPU heat away at the same time? Do you have a fan directly over the vrm sink? Can you work around the cooler with your hands w/o removing it to get something in or out? Do you want 2 or 3 fans? RAM accessibility sometimes makes a difference to some people if you're using all 4 slots.
 
  • Like
Reactions: john3850

gorobei

Diamond Member
Jan 7, 2007
3,654
980
136
i bought this a few weeks ago. they had a $5 discount code. so around $35.
71YEiWsyLlS._SL1500_.jpg



6 pipes and includes 2 fans. for ~$35, i think it is a much better option than any 4 pipe 1fan cooler.

i think i found it thru a hardwarecanucks video. the amazon page has a bunch of versions with color and rgb options with pricing varying widely.
 

BoomerD

No Lifer
Feb 26, 2006
62,681
11,025
136
i bought this a few weeks ago. they had a $5 discount code. so around $35.
71YEiWsyLlS._SL1500_.jpg



6 pipes and includes 2 fans. for ~$35, i think it is a much better option than any 4 pipe 1fan cooler.

i think i found it thru a hardwarecanucks video. the amazon page has a bunch of versions with color and rgb options with pricing varying widely.
Only a 1 year warranty...that would concern me.
 

gorobei

Diamond Member
Jan 7, 2007
3,654
980
136
Only a 1 year warranty...that would concern me.
thermalright is one of the old school brand names. while they have done some entry level stuff recently, their higher end models have been reliable as anyone else. i suspect the 1year has more to do with the rgb models with included rgb fans. well made heatpipes usually last far longer than the first pc they go into, but rgb is just asking for defects over time.
 

Tech Junky

Diamond Member
Jan 27, 2022
3,365
1,118
106
Only a 1 year warranty...that would concern me.
The only thing that needs a warranty though is the fans. The metal either works or doesn't. Fans that come with most air towers suck anyway. I tend to not use stock fans on anything and have a box of them sitting around for a rainy day.
 
  • Like
Reactions: aigomorla

Justinus

Diamond Member
Oct 10, 2005
3,167
1,509
136
I got a deepcool AK620 to use on my new 5900x and discovered it allowed the (completely stock) 5900x to hit 95C peak temps in a ycruncher stress test. When I removed it, it was clear the base of the cooler was too convex to seat really well on the AM4 heatspreader. I might yet lap it but in the meantime I switched to a U12A I had spare.

I imagine the extra convex base helps it seat better on Intel, especially LGA1700 with the bending memes.

I would think the base shape is probably similar on the AK400.
 
  • Wow
Reactions: VirtualLarry

VirtualLarry

No Lifer
Aug 25, 2001
56,229
9,990
126
I got a deepcool AK620 to use on my new 5900x and discovered it allowed the (completely stock) 5900x to hit 95C peak temps in a ycruncher stress test. When I removed it, it was clear the base of the cooler was too convex to seat really well on the AM4 heatspreader. I might yet lap it but in the meantime I switched to a U12A I had spare.

I imagine the extra convex base helps it seat better on Intel, especially LGA1700 with the bending memes.

I would think the base shape is probably similar on the AK400.
Read up on this more, because, IIRC, the nominal stock temps (for the 5000 series in particular) are SUPPOSED to hit 95C and stay there (Turbo thermal limit).

Where you would see if that cooler was effective, would be overall benchmark performance @ 95C compared to a lesser cooler.

It may not have to do with base concavity at all.

As an enthusiast, that explanation doesn't sit well with me either, but I seem to recall Richard Hallock mentioning it.
 

Justinus

Diamond Member
Oct 10, 2005
3,167
1,509
136
Read up on this more, because, IIRC, the nominal stock temps (for the 5000 series in particular) are SUPPOSED to hit 95C and stay there (Turbo thermal limit).

Where you would see if that cooler was effective, would be overall benchmark performance @ 95C compared to a lesser cooler.

It may not have to do with base concavity at all.

As an enthusiast, that explanation doesn't sit well with me either, but I seem to recall Richard Hallock mentioning it.

The U12A I put on is objectively worse at cooling than the AK620 according to reviews, but under the same conditions it cut the peak temp down to just below 90C. It leaves a flat imprint in the thermal paste with no obvious gaps like the AK620 had. I don't have any other explanation than convex coldplate. The largest spot where it clearly didn't mate to the IHS well is over one of the CCDs.

Basically observed behavior is higher average temps and lower peak temps on the U12A, lower average temps and higher peak temps on the AK620.
 
Last edited:

gorobei

Diamond Member
Jan 7, 2007
3,654
980
136
The U12A I put on is objectively worse at cooling than the AK620 according to reviews, but under the same conditions it cut the peak temp down to just below 90C. It leaves a flat imprint in the thermal paste with no obvious gaps like the AK620 had. I don't have any other explanation than convex Coldplay. The largest spot where it clearly didn't mate to the IHS well is over one of the CCDs.

Basically observed behavior is higher average temps and lower peak temps on the U12A, lower average temps and higher peak temps on the AK620.
i've found that generally trustable sites like gamersnexus et all, can have varying results when it comes to cooling. some raved about the vetroo v5 but GN slammed it. some liked the U12A while others had it mediocre on their testing. i suspect it comes down to varying testing methodology and review samples QA since some get early access. you would probably have to test 5 retail samples of any given model to make any assessment with confidence.

hopefully LTT labs will do tests with larger samples than 1 review unit.



found the review with the thermalright peerless assassin.
 

aigomorla

CPU, Cases&Cooling Mod PC Gaming Mod Elite Member
Super Moderator
Sep 28, 2005
20,839
3,174
126
Only a 1 year warranty...that would concern me.

Please tell me this is sarcasm?
Because a heat pipe cooler either works forever until you damage it so the gas inside escapes, or it doesn't work from the getgo, because the heat pipes are empty.

And if your talking about fans, i can say without a doubt, only Noctua honors warrenty on fans dying, but the fan needs to be not clogged with dust, and has to be a genuine motor fail.

All the other vendors will basically laugh at you if you try to RMA a fan, and it will be on the mood of the Customer Support department, to replace it or not.

i've found that generally trustable sites like gamersnexus et all, can have varying results when it comes to cooling.

Its really difficult to do an accurate test, because during every test to have a fair comparison, you need to hold environment constant.
That means you need a temperature controlled room, and you need the same test bed for all the test.
You modify one of those, the accuracy on the comparison will be skewed.

GN tries hard to be as accurate by doing so, and so far i feel they are the closest in accuracy.
 
  • Like
Reactions: VirtualLarry

Tech Junky

Diamond Member
Jan 27, 2022
3,365
1,118
106
And if your talking about fans, i can say without a doubt, only Noctua honors warrenty on fans dying,
I was playing around with orientation of in vs out while the system was running with Arctic's and nicked one of the blades on something which caused a microfracture around the hub of one of the fans. I couldn't see it but, I could hear it as it had just enough wobble to make noise in the airflow. I e-mailed them for a replacement and one showed up a week later in the mail no questions asked. They provided good service and that's why I recommend them besides their good performance in my systems.
 

gorobei

Diamond Member
Jan 7, 2007
3,654
980
136
Its really difficult to do an accurate test, because during every test to have a fair comparison, you need to hold environment constant.
That means you need a temperature controlled room, and you need the same test bed for all the test.
You modify one of those, the accuracy on the comparison will be skewed.

GN tries hard to be as accurate by doing so, and so far i feel they are the closest in accuracy.
i'm not trying to impugn GN in general, they do a serious attempt to come up with a valid testing method, but steve has mentioned in videos that they cant afford to do multiple samples(due to time needed and money outlay), just multiple runs on the same unit.

GN and Linus' newly christened Lab32 have both spent the money for thermal and fan testing chambers, but you really need to do multiple samples. if GN did a similar offer to the 3090 power issue deepdive where you send them a unit for testing and they ship it back to the owner, they could get a significant sample size. basically through some coordination, consumers/fans of GN or LTT would buy a cooler from amazon/egg/etc and have it shipped to the labs for testing and then the lab would ship it back to the consumer. the cost of shipping could be baked into the revenue of the video/review if their labs become the goto source for figuring out good/bad cooler/psu/etc.
 

BonzaiDuck

Lifer
Jun 30, 2004
15,699
1,448
126
I want someone to deliberately search for and find a comparison review or a few comparison reviews (you can crosswalk them with the transitivity axiom) which identify a heatpipe cooler that performs better than the ThermalRight Le Grand Macho.

I've grown so comfortable with this cooler by findings I made five years ago that I no longer search. And yes, it's a big cooler in need of a sufficiently-wide midtower case. But I was only interested in performance.

I want someone to identify something better on performance grounds alone, confined exclusively to the range of all heatpipe coolers.
 

dlerious

Golden Member
Mar 4, 2004
1,772
719
136
i've found that generally trustable sites like gamersnexus et all, can have varying results when it comes to cooling. some raved about the vetroo v5 but GN slammed it. some liked the U12A while others had it mediocre on their testing. i suspect it comes down to varying testing methodology and review samples QA since some get early access. you would probably have to test 5 retail samples of any given model to make any assessment with confidence.

hopefully LTT labs will do tests with larger samples than 1 review unit.



found the review with the thermalright peerless assassin.
Updated video.
 

gorobei

Diamond Member
Jan 7, 2007
3,654
980
136
Updated video.
yeah the deepcool ak620 is probably better on fans and support, but given that i cant find it listed for less than $65 i dont know that it is $30 better than the thermalright.

i didnt look that hard but it looks like the base on the peerless assassin is offset from the tower and the ak620 is centered. might be relevant if you have motherboard clearance issues.

nice that HC bought a 2nd sample to verify the performance.