DAN A4 V4.1 build with Ryzen 5600X

[nightshift]

This seems like an ideal setup. Best current 65tdp processor with a gpu that just sounds fitting to it and ideal for this case in size and power consumption. I'm also interested to see the regular 5600 from January.
I would very much like to see a comparison between the L9a + fan duct and the AXP90 Full Black. Is the "full black" also full copper? If these cpu's can't go higher than 80c at full load with a bare L9i/a, that sounds great. Actually now that I think about it, what I would like to see is a comparison video of these new gen Ryzens (especially the 5600x) tested on various low profile coolers starting with the L9a, then Blackridge, etc. It would be so easy to make for these reviewers and so relevant for the people yet nobody did it yet. All tests are focused on performance and not temperature, let alone low-profile point of view.
Nice Keychron K6 btw

 

[luke2122]

This seems like an ideal setup. Best current 65tdp processor with a gpu that just sounds fitting to it and ideal for this case in size and power consumption. I'm also interested to see the regular 5600 from January.
I would very much like to see a comparison between the L9a + fan duct and the AXP90 Full Black. Is the "full black" also full copper? If these cpu's can't go higher than 80c at full load with a bare L9i/a, that sounds great. Actually now that I think about it, what I would like to see is a comparison video of these new gen Ryzens (especially the 5600x) tested on various low profile coolers starting with the L9a, then Blackridge, etc. It would be so easy to make for these reviewers and so relevant for the people yet nobody did it yet. All tests are focused on performance and not temperature, let alone low-profile point of view.
Nice Keychron K6 btw

Thanks man, K6 is really great.
And yeah, Full back version is also copper, and compatible with both inter and and cpu as well as additional backplate. I think its performance would be similar to Blackridge but better than Noctua L9a in Dan A4. Also I do agree that waiting several weeks for 5600 is a good idea if you can wait.

 

[nice_big_cup_of_stfu]

After having read a lot of threads in this forum, I just completed my build besides the GPU ( waiting for the upcoming RTX3060 Ti ASUS DUAL MINI version, which I think is the best choice for me in this A4 case so far) so I temporally use my old MSI 1080 before 3060ti.
Overall, have to say I am very satisfied by good temperature and decent performance, in terms of the new AMD cpu, I locked my 5600X's all cores with 4.5G by 1.25V, and the idle temp is about 45-48 degrees and when I play 3A games at 1440P high setting, it goes up to 50-62 in most cases.

Wow, great temps!

Like @drmarkuse, I'm using a 5600X that gets up to 80+C hot with a L9a and fan duct. With PBO on, can get into the 90s.

And CPU-Z is showing my voltages at 1.15V or so and nowhere close to getting up to 4.5GHz.

 

[nice_big_cup_of_stfu]

Okay some additional testing:

First, I also an ASUS B550i with the latest BIOS. I also have the NH-L9a w/ Fan Duct. I noticed that the PPT is set to 76W in Ryzen Master which is apparently a bug in the ASUS implementation - the 5600X is supposed to have an 88W PPT. So I set it to that.

However, running Cinebench R20, I'm getting only ~3.9-4.0GHz all core boost with 88W PPT and everything else in BIOS basically set at stock. Temps are around 78-81C.

I do get ~4.65 GHz single core boost.

I set the AutoOC in BIOS, and get ~4.65-4.8 GHz in ST with temps at ~77C. For MT, I get ~4.1-4.2 GHz and temps are around 80-82C.

So ST performance is good, but MT is well below what other people are getting out there (4.4 GHz all-core boost seems to be what most people are getting) and I am definitely doing a lot worse than others when it comes to temps.

Now it could be that I simply lost the CPU lottery, but given the BIOS bug where PPT is set incorrectly, I'm curious if my CPU also isn't boosting as high in MT as it could be for the voltage.

 

[daddelbud]

Okay some additional testing:

First, I also an ASUS B550i with the latest BIOS. I also have the NH-L9a w/ Fan Duct. I noticed that the PPT is set to 76W in Ryzen Master which is apparently a bug in the ASUS implementation - the 5600X is supposed to have an 88W PPT. So I set it to that.

Sure it's a bug?
In most reviews they have noted the same limit, and I also see it on my Gigabyte B450 ITX board

 

[nice_big_cup_of_stfu]

Sure it's a bug?
In most reviews they have noted the same limit, and I also see it on my Gigabyte B450 ITX board

Supposedly AMD said it's supposed to be 88W but the AGESA set it to 76W

Anywho, looks like high temps are an issue for a lot of people:
Might try and set 1.2V/4.5G

 

[luke2122]

Wow, great temps!

Like @drmarkuse, I'm using a 5600X that gets up to 80+C hot with a L9a and fan duct. With PBO on, can get into the 90s.

And CPU-Z is showing my voltages at 1.15V or so and nowhere close to getting up to 4.5GHz.

That sounds weird, it seems the only variable is the cooler which should't cause a large difference at temps. what is worthy to mention is that I didn't even carefully note the temps after turning on PBO and just manually set my voltage to 1.25 and clock speed to 4.5GHz to test and use. Up to now the ASUS BIOS still stays at a rough version, so maybe you can wait for the next update to see if there is an improvement for that.

 

[nice_big_cup_of_stfu]

That sounds weird, it seems the only variable is the cooler which should't cause a large difference at temps. what is worthy to mention is that I didn't even carefully note the temps after turning on PBO and just manually set my voltage to 1.25 and clock speed to 4.5GHz to test and use. Up to now the ASUS BIOS still stays at a rough version, so maybe you can wait for the next update to see if there is an improvement for that.

Yeah there is the new update coming that lets PBO2 undervolt. So I might wait for that.

I'm curious if the l9a is indeed near its limit. The 5600X can draw same PPT as a 3700X and that's about the limit for the l9a.

What temps are you getting in synthetics at 4.5G/1.25V?

Notably I am getting high 70s gaming in games that use multiple cores. In SC2, for instance, 3 cores are going up to 4.6 GHZ. So the heat isn't too bad for what it is, but I am shocked how low people are getting here

 

[luke2122]

Yeah there is the new update coming that lets PBO2 undervolt. So I might wait for that.

I'm curious if the l9a is indeed near its limit. The 5600X can draw same PPT as a 3700X and that's about the limit for the l9a.

What temps are you getting in synthetics at 4.5G/1.25V?

Notably I am getting high 70s gaming in games that use multiple cores. In SC2, for instance, 3 cores are going up to 4.6 GHZ. So the heat isn't too bad for what it is, but I am shocked how low people are getting here

Probably regardless of other factor, this is indeed attribute to the coolers’ performance. My current idle temps can stabilised around 38-42 degree at 4.5G/1.275V with 1100 fans rpm and 10℃ room temps, the list as below shows different temps under diverse types of games I normally play, it can be a reference for you:
The Witcher 3 — 1440p high setting — 62-68℃
PES2021 — 2160p with ultra setting — 50-56℃
Civilisation 6 — 2160p with ultra setting — 52-56℃
NBA 2K21– 2160p with high setting— 55-62℃
Hades — 2160p with ultra setting — 45-50℃
pillar of Eternity — 2160p with ultra setting — 43-48℃

 

[tmuy3000]

Hey! I was going for an extremely similar setup.
Case: Dan A4v4.1
CPU: AMD Ryzen 5 5600X
GPU: ASUS Dual GeForce RTX 3070
Motherboard: ASUS ROG STRIX B550-I GAMING Motherboard
CPU Cooler: Alpenfohn Black Ridge Low Profile CPU Cooler - 120mm
RAM: CORSAIR - Vengeance LPX 16GB (2PK 8GB) 3.6GHz
SSD: Samsung - 970 EVO Plus 1TB
PSU: Crosshair SF 600W

The main difference is that I was able to obtain an ASUS Dual GeForce RTX 3070. After doing some research it seems that both ASUS Dual GeForce RTX 3070 and the 3060ti version are 2.7 slot. Wouldn't this mean that they cannot fit in the Dan A4 case?

 

[luke2122]

Hey! I was going for an extremely similar setup.
Case: Dan A4v4.1
CPU: AMD Ryzen 5 5600X
GPU: ASUS Dual GeForce RTX 3070
Motherboard: ASUS ROG STRIX B550-I GAMING Motherboard
CPU Cooler: Alpenfohn Black Ridge Low Profile CPU Cooler - 120mm
RAM: CORSAIR - Vengeance LPX 16GB (2PK 8GB) 3.6GHz
SSD: Samsung - 970 EVO Plus 1TB
PSU: Crosshair SF 600W

The main difference is that I was able to obtain an ASUS Dual GeForce RTX 3070. After doing some research it seems that both ASUS Dual GeForce RTX 3070 and the 3060ti version are 2.7 slot. Wouldn't this mean that they cannot fit in the Dan A4 case?

Yeah The limit is 2-slot, So far Asus hasn’t released two-slot 3060ti and 3070, so I end up with buying gigabyte eagle 3060ti OC which is perfectly matched in this case. But if you want to get 2-slot ASUS 3060ti you can wait for its mini dual version. But I don’t know the exact releasing date.

 

[nice_big_cup_of_stfu]

Figured out some temp issues - bought some authentic Thermal Grizzly Kryonaut (who knew that fakes existed...), re-seated my NH-L9a, and really applied tight pressure on the CPU and backplate. Updated to the latest BIOS as well, which allows for curve optimization in PBO2.

Moreover, I adjusted my fan curve w/ the fan duct - when it is very aggressive, with only mild undervolts on my 5600X, I peak in the high 60s in Cinebench R23 (all-core boost holds at around 4.2GHz) with power draw being my limit (76W stock). With 88W power draw, I peak at around 79°C at an all-core boost holding at 4.4 GHz. And for fun, I set it to PBO on with limits up to the motherboard... peak of 88°C with all-core boost holding at 4.45 GHz.

With that being said, I value silence, so I set a very quiet fan curve (40%) and I typically idle in the 50-60°C range and peak at 80 in gaming (3 cores running full-bore).

 

[luke2122]

Figured out some temp issues - bought some authentic Thermal Grizzly Kryonaut (who knew that fakes existed...), re-seated my NH-L9a, and really applied tight pressure on the CPU and backplate. Updated to the latest BIOS as well, which allows for curve optimization in PBO2.

Moreover, I adjusted my fan curve w/ the fan duct - when it is very aggressive, with only mild undervolts on my 5600X, I peak in the high 60s in Cinebench R23 (all-core boost holds at around 4.2GHz) with power draw being my limit (76W stock). With 88W power draw, I peak at around 79°C at an all-core boost holding at 4.4 GHz. And for fun, I set it to PBO on with limits up to the motherboard... peak of 88°C with all-core boost holding at 4.45 GHz.

With that being said, I value silence, so I set a very quiet fan curve (40%) and I typically idle in the 50-60°C range and peak at 80 in gaming (3 cores running full-bore).

Sounds better, it seems that you’ve already found a balance, silence is a significant merit if using Nh-l9a. When I use AXP90, the noisy of turbulence is a little bit annoying although it can offer better temp control. I haven’t considered this problem before my build.

 

[tmuy3000]

Yeah, that was completely my fault. I should have never assumed the ASUS Dual GeForce RTX 3070 was a dual-slot card. The set-up you have right now looks extremely clean. I guess I'll try and return my 3070 or trade it with someone. Back to trying to get a 3060ti or 3070! Ill post my build here when I finally complete it.

 

[nice_big_cup_of_stfu]

Sounds better, it seems that you’ve already found a balance, silence is a significant merit if using Nh-l9a. When I use AXP90, the noisy of turbulence is a little bit annoying although it can offer better temp control. I haven’t considered this problem before my build.

Yeah I ran into the same issue playing with the Black Ridge cooler - its larger area means the CPU seems to get hotter slower in synthetics (can sustain higher boost clocks initially) but the fan needs to be REALLY loud

Meanwhile, on the NH-L9a w/ Fan Duct, properly mounted of course, I now keep my fan curve at 50% at 65°C, 75% at 75°C, and 100% at 85°C - and it at 100% is quieter than the Black Ridge at 75%. So in practice, I'm sustaining similar clocks/thermals for the noise (and in actuality, the CPU never quite hits 85C on Tdie, so it is always somewhere below 100%).

 

[fongalv]

Yeah I ran into the same issue playing with the Black Ridge cooler - its larger area means the CPU seems to get hotter slower in synthetics (can sustain higher boost clocks initially) but the fan needs to be REALLY loud

Meanwhile, on the NH-L9a w/ Fan Duct, properly mounted of course, I now keep my fan curve at 50% at 65°C, 75% at 75°C, and 100% at 85°C - and it at 100% is quieter than the Black Ridge at 75%. So in practice, I'm sustaining similar clocks/thermals for the noise (and in actuality, the CPU never quite hits 85C on Tdie, so it is always somewhere below 100%).

Hi, I'm a little curious about your findings with regards to the coolers. When you mentioned that performance/noise level wise, NH-L9A>black ridge(& AXP-90 from luke2122), were you using the stock black ridge 92mm fan?

 

[londiste]

NH-L9a has a much better fan than Black Ridge's stock. If comparing with stock fans, Black Ridge is considerably louder at the same performance level. I have both NH-L9a-AM4 and Black Ridge but have not had a chance to compare them directly. I do have a C7 Cu that used to cool the same CPU as NH-L9a-AM4 and same CPU as Black Ridge, all at 65W power usage level.

C7 Cu is slightly better than NH-L9a-AM4 with the same fan (2500rpm Noctua's NF-A9x14 PWM) and Black Ridge got on par with normal 2200RPM NF-A9x14 PWM. All running at about 75% at full load.

After upgrading to 5600X which runs at 76W it took a little powerup to keep that in check and Black Ridge shines with the 120mm fan. NF-A9x14 ran the full 2200rpm at full load to keep this in check, NF-A12x15-PWM gets the same result at 1300rpm.

 

[daddelbud]

... NF-A12x15-PWM gets the same result at 1300rpm.

On top of the L9a heatsink?
I didnt have much success with that

 

[londiste]

On top of the L9a heatsink?
I didnt have much success with that

Black Ridge

 

[Hellow]

Blackridge fan is louder, but much better !