News Thermalright AXP90

It looks like Thermalright is working on a new heatsink in the sub 50mm height class:

利民新产品 - 电脑讨论 - Chiphell - 分享与交流用户体验

利民新产品,　　　＿＿＿＿＿　　 ｜早上好 ｜　　　￣￣∨￣￣　　　　∧＿∧　　　　 (´・ω⊂ヽ゛　　　 /　　_ノ⌒⌒ヽ、(￣⊂人　 ノシ⌒　　ノ⊂ニニニニニニニニニニ ...,电脑讨论,讨论区-技术与经验的讨论 ,Chiphell - 分享与交流用户体验

www.chiphell.com

 

[Hardly]

I give this cooler a thumbs up, it works as good as my Black Ridge without the need for VLP RAM or maybe cutting the z390 phantoms heatsink. The downside is that it too tall for the S4M w/ a noctua fan.

90C @ 4.6Ghz on a 9700k on noctua thermal paste

Cool, thanks for the testing!
Are you saying that you have to cut the VRM heatsink on the phantom, or that you don't have to?

 

[AURMEND]

Cool, thanks for the testing!
Are you saying that you have to cut the VRM heatsink on the phantom, or that you don't have to?

I already cut mine for the black ridge but it looks like it will fit but with worse for the VRM, RAM, and CPU because it will be blowing hot air right on it it.

 

[Nichronos]

I give this cooler a thumbs up, it works as good as my Black Ridge without the need for VLP RAM or maybe cutting the z390 phantoms heatsink. The downside is that it too tall for the S4M w/ a noctua fan.

90C @ 4.6Ghz on a 9700k on noctua thermal paste

I am also with 9700k on this mobo
Did you undervolt it? My is 4.6 @ 1,15v stable and never exceeds 77C* in Aida64 FPU stress with Noctua l9i + NF a9x25 fan.
On 4.9 @ 1,25v is where i go beyond 90C* and this is why i need a better cooler...

My system:
CASE: CustomMOD FX 4.7
CPU: Intel Core i7-9700K
MB: ASRock Z390 ITX/ac
RAM: ViperSteel 16GB 4400MHz
GPU: Gigabyte RTX2070 ITX
SSD: Samsung 1TB SM961
PSU: Delta 500AB Modular FLEX

 

[AURMEND]

I am also with 9700k on this mobo
Did you undervolt it? My is 4.6 @ 1,15v stable and never exceeds 77C* in Aida64 FPU stress with Noctua l9i + NF a9x25 fan.
On 4.9 @ 1,25v is where i go beyond 90C* and this is why i need a better cooler...

My system:
CASE: CustomMOD FX 4.7
CPU: Intel Core i7-9700K
MB: ASRock Z390 ITX/ac
RAM: ViperSteel 16GB 4400MHz
GPU: Gigabyte RTX2070 ITX
SSD: Samsung 1TB SM961
PSU: Delta 500AB Modular FLEX

I have a -100mv offset

 

[Nichronos]

I have a -100mv offset

You sure? 100mV via the BIOS would be impossible, only by some 3rd party Software like Throttlestop.
50mV on this motherboard equals 130mV on Asus Z390-i

EDIT: i am 1000% sure you fucked up and actually using a +100mV instead of negative, that is why you have 90+ on 4,6GHz

 

[Salulu]

Anybody been able to find any temp comparisons between the aluminium and copper versions?
Curious how much better the copper version actually is considering its nearly twice the price and weight (so extra shipping cost with Superbuy etc.)

 

[AURMEND]

You sure? 100mV via the BIOS would be impossible, only by some 3rd party Software like Throttlestop.
50mV on this motherboard equals 130mV on Asus Z390-i

EDIT: i am 1000% sure you fucked up and actually using a +100mV instead of negative, that is why you have 90+ on 4,6GHz

Wrong, just changed it to no voltage offset -100mV 4.5Ghz = 1.02V
0Mv 4.5Ghz = 1.14V

Second, I was comparing it to the first batch black ridge I had and it performed as good.

Right now I am running a NH-L9i w/ 120mm fan

unless im looking at the wrong voltage. CPU VCORE or VID?

 

[Vlad502]

Anybody been able to find any temp comparisons between the aluminium and copper versions?
Curious how much better the copper version actually is considering its nearly twice the price and weight (so extra shipping cost with Superbuy etc.)

My guess is 3°C - if AXP90 is 7°C better than L9i and AXP90FC is 4°C better than C7G, so AXP90FC vs AXP90 will be like C7G vs C7:

 

[Nichronos]

Wrong, just changed it to no voltage offset -100mV 4.5Ghz = 1.02V
0Mv 4.5Ghz = 1.14V

Second, I was comparing it to the first batch black ridge I had and it performed as good.

Right now I am running a NH-L9i w/ 120mm fan

unless im looking at the wrong voltage. CPU VCORE or VID?

I just put an -100mV and removed my fixed voltage to test it, and i can confirm i experienced the same insane temp levels as you, the fixed 1.15v is 1000% better!

 

[AURMEND]

I just put an -100mV andc removed my fixed voltage to test it, and i can confirm i experienced the same insane temp levels as you, the fixed 1.15v is 1000% better!

huh... could have been bad mounting also. well lets see how much my NH-l9i can push w/ 3000rpm 120*25mm Noctua

 

[Vlad502]

bilibili

 

[thelaughingman]

bilibili

just using the google translate app here but the 2nd cooler in the 50mm category is BR v2 with NF A9x14 right? so about same temps (+- 1C) for more noise (+3.5db).

 

[troybot]

just using the google translate app here but the 2nd cooler in the 50mm category is BR v2 with NF A9x14 right? so about same temps (+- 1C) for more noise (+3.5db).

I have one in a Dancase A4. It works similar to my BR (but I cut mine to fix high ram), but gives me unlimited height clearance without mods. Took a few weeks to deliver to germany, even during the crisis (but I ordered before it started)

Before:

After:

 

[thelaughingman]

I have one in a Dancase A4. It works similar to my BR (but I cut mine to fix high ram), but gives me unlimited height clearance without mods. Took a few weeks to deliver to germany, even during the crisis (but I ordered before it started)

Before:

After:

nice build man! I'm tempted to get this cooler but have concerns about the noise. I see you already cut open the side for unobstructed airflow but I only have the stock side panel. And from my own experience running the L9a with A9 PWM fan it was really loud. The air turbulence is just unbearable. Have you tested noise with the stock side panel?

 

[troybot]

nice build man! I'm tempted to get this cooler but have concerns about the noise. I see you already cut open the side for unobstructed airflow but I only have the stock side panel. And from my own experience running the L9a with A9 PWM fan it was really loud. The air turbulence is just unbearable. Have you tested noise with the stock side panel?

I never tried with the stock side panel for the reason you probably guessed, with my own panel I can have a larger slower moving fan for similar enough performance. I would be concerned about turbulence since it would be less than a mm from the (stock) side panel.

Regarding the L9a, I also looked into it, but 2 heatpipes had me a bit worried with my 9700k. Everything I read also said that for some reason the 14mm fan outperformed the standard a9, which I found interesting, but it would still struggle with my cpu. So it was either this or a c7g, but I couldnt find a c7 cu / g in stock.

 

[thelaughingman]

I never tried with the stock side panel for the reason you probably guessed, with my own panel I can have a larger slower moving fan for similar enough performance. I would be concerned about turbulence since it would be less than a mm from the (stock) side panel.

Regarding the L9a, I also looked into it, but 2 heatpipes had me a bit worried with my 9700k. Everything I read also said that for some reason the 14mm fan outperformed the standard a9, which I found interesting, but it would still struggle with my cpu. So it was either this or a c7g, but I couldnt find a c7 cu / g in stock.

right now I'm using the L9a with the A9x14 and ghetto fan duct so it's alright, inaudible at idle around 48-50C - I'm running the 3900X in Eco mode. The only annoyance is when the fan ramps up as I put on load - I have the fan curve quite aggressive at 70C. Temps drop right after but the oscillation is not pleasant. Thus I have a replica Black Ridge from Taobao coming this week so hopefully it will be a better experience overall

 

[Xoroxronis]

From my testing (DAN Case-3700X) the AXP-90 Copper with NF-A9x14 VS NH-L9a with fan duct is 3-4C cooler.
I have done many tests (88W PPT and 100W PPT, Cinebench, Blender, Prime, always the same frequencies, voltages, power draws from the wall, ambient temps etc) and always the results was 3-4 degrees cooler.

But i'm intersted if anyone tested with 3900X/3950X? Not Eco full 142W PPT?

 

[ignsvn]

Ever tested different fan configurations eg pull VS push?

 

[Xoroxronis]

No, with this case/cooler i haven't.

 

[Poblopuablo]

Does anyone have the Intel version of the axp 90 than can confirm the same mounting locations for the brackets? I want to ensure they are the same.