DAN A4-SFX: The smallest gaming case in the world

G_pea_eS said:
Awesome results! I plan on deliding whatever processor I end up with. Did you use Conductonaut between the heatspreader and c7 as well? Also, please post what overclocking results you get with the delid and c7...

Are you going to be moving to a different cooler?
Click to expand...
I used Gelid GC-Extreme for the thermal paste. I read that liquid metal will bond to aluminum and I'm not interested in taking the risk.
I don't plan to overclock, I just want a quieter fan and lower temperature with PWM.
I'm waiting for Dondan's perfect air cooler. Meanwhile I don't plan to micro-optimise my cooling any further because it's a waste of time and money imo.
 
XstreamHard said:
As suggested before it would be easier to replace the 7mm mainboard standoffs with 5mm an just not use an IO shield. That would gain you more than enough room for the AXP100RH without any grinding.
Click to expand...
That's a good idea, I will consider that thanks!
 
promaty said:
More on CPU cooling in Dan A4-SFX

Unfortunately Cryorig C7 was a bit over-hyped. Although it does cool well at full speed, it is a clear loser in terms of performance / noise + you get an annoying high-frequency turbulence due to the fan spinning close to the side panel.

My recommendation for the best performance / noise:
1) Silverstone AR05 (36.7mm) - a clear winner
2) Noctua L9i (37mm) - if you want more compatibility, as it doesn't have the wide heatpipe of AR05 (95mm x 95mm vs AR05's 104mm x 92mm)
3) Cryorig C7 (47mm) - if you want some room for overclocking, as it cools better at full speed, but it is also much louder
4) Something custom if you have the time and energy to do it yourself

Some other coolers compatible with the case, which don't seem to offer any advantage:
Alpenfohn Silvretta (45mm) - inferior to C7, L9i and AR05
Xigmatek Praeton LD963 (45mm) - inferior to L9i and AR05
Phanteks PH-TC90LS (45mm) - inferior to L9i and AR05
Raijintek Zelos (44mm) - inferior to L9i and AR05
Cooltek ITX30 (30mm) - inferior to AR05, about the same as L9i, but wider (100mm x 94mm vs L9i's 95mm x 95mm)
Intel Boxed Cooler - ewww

Sources:

C7 turbulence demonstration by Meccabolix:


L9i vs C7 performance in Linus video:


Screenshot of capped noise, which shows L9i having superior performance / noise:
View attachment 16960

Note that C7 does cool better if you uncap the speed:
View attachment 16961

Note the video above includes AR06, not AR05. It also shows Phanteks PH-TC90LS inferior to L9i.

AR05 review shows it is superior to L9i. You can also see Cooltek ITX30, which is inferior to AR05 and about the same as L9i:
http://www.nikktech.com/main/articl...ow-profile-cpu-cooler-review?showall=&start=5

AR05 better performance:
View attachment 16959

AR05 lower noise:
View attachment 16958

This is also consistent with Dan's cooling test before, AR05 is better than L9i (without customization):
https://hardforum.com/threads/dan-a...-in-the-world.1799326/page-55#post-1041652691

Alpenfohn Silvretta inferior to C7:
http://www.tech-review.de/include.php?path=content/articles.php&contentid=16637&page=6

Xigmatek Praeton LD963 inferior to L9i:
https://www.hardwarebbq.com/intel-6...raeton-apache-iii-and-intel-stock-cooler-2/8/

Raijintek Zelos inferior to L9i:
https://elchapuzasinformatico.com/2014/02/review-raijintek-zelos/5/

I currently own C7, but I will replace it with AR05 because I cannot tolerate the turbulence (not as bad if you remove the side panel, but common).

Cheers ;)
Click to expand...


I'm hoping the LP53 with the 14mm noctua fan beats all of these so I don't have to go super custom. Ram height and other limitations on the nexus have me turned off. I'll be waiting patiently for my shipment while my L9i serves as a stop gap.
 
I totally forgot about the AR05. Was this dismissed earlier in the thread for some other reason? CPU cooler reviews are weird, for instance this review http://techicize.com/silverstone-ar05-review/ shows the AR05 throttling on a haswell i5 K chip, at over 30 degress hotter than the AXP200, which is converseley in a dead heat with the AR05 in the nikktech review linked above (on an ivy bridge i7.)

I'm just going to stop worrying about this now though and hope my LP53 does the job...;)
 
illram said:
I'm just going to stop worrying about this now though and hope my LP53 does the job...;)
Click to expand...
Me too! I've been obsessing over this issue and I don't even have all the components yet! I too ordered the LP53 and I have high hopes for it, all the reviews online are stellar and people are like "why did this thing never take off?!".
 
onmybikedrunk said:
Me too! I've been obsessing over this issue and I don't even have all the components yet! I too ordered the LP53 and I have high hopes for it, all the reviews online are stellar and people are like "why did this thing never take off?!".
Click to expand...

Probably due to no one really needing an $80 cooler that's only 40mm tall. We're creating this market (us in SFF). We're the ones pushing what's possible in these small form factors. By creating cases with these hard heatsink limitations, we're giving manufacturers a target to hit. The DanCase and the Sentry are popular, and hopefully these heatsink manufacturers see them as the trendsetters they are.
 
Is ever
cowsgomoo2 said:
I used Gelid GC-Extreme for the thermal paste. I read that liquid metal will bond to aluminum and I'm not interested in taking the risk.
I don't plan to overclock, I just want a quieter fan and lower temperature with PWM.
I'm waiting for Dondan's perfect air cooler. Meanwhile I don't plan to micro-optimise my cooling any further because it's a waste of time and money imo.
Click to expand...

Makes sense. However, according to Cryorig the C7's base is copper. I don't know how much of a difference it will make, but seeing as you already dropped 20 degrees it's probably not worth it...
 
Sorry for spamming but anyone with a silver case who is interested in selling please PM me thanks.

Can't wait for some real Ryzen news, hopefully their 65W cooler is low profile enough for this case, it would be funny if it was good, quiet and fitted!
 
Well, I think a lot of the reviews that you can find online on semi professional blogs are wrong. There are some in which a certain cooler performs superior than its competitiors and then there are others where it is the other way around. That should never ever happen. Furthermore the level of noise is always measured in db and not in sone which doesn't help at all if you want to judge a fan. Last but not least, in some reviews the coolers aren't even tested with a reference fan which makes it almost impossible to compare the coolers.

Whenever I take a look at the Silverstone Argon AR05 and compare it with the Noctua NH-L9i there are some things that I notice, which I think are very important.
- both coolers have almost the exact same dimensions and design, 90*90*37mm and two heatpipes. Therefore these coolers should perform more or less the same
- the fan speed of both fans is identical. This means, that the design of the fan is the only difference between them. Since Noctua is a brand known for their high quality fans, I doubt that the SilverStone fan can be much better than the the Noctua fan.
- So, what is left that could have influence in the cooling performence? The material quality, right? We all know, that copper transports heat better than alluminium, but in return it is far heavier and more expensive than alluminium. When I look at the data sheet of both coolers I see that the Noctua is more than twice as heavy as the Silverstone. I assume it has more copper build in, which should make it better.

So, why the hell should the Silverstone perform better than the Noctua?
 
Now everyone will buy Silverstone same as they did with C7 :)

I'll just say this ... L9i can keep 6700K stock from overheating while running torture tests - I ran the x264 encoding test with 16 threads for 6h - can be found on OCN and easily reproduced.

I will do the same test for L9i with 25mm Noctua fan and L53 with slim Noctua fan from L9i.
 
For custom CPU cooler combinations (LP53, etc.): keep in mind that there needs to be some space between the fan and the side panel, otherwise you will get the same turbulence as with Cryorig C7. I would say that there should be at least ~1cm of free space, which is in the range of Silverstone AR05 and Noctua L9i.
 
Shaav said:
Well, I think a lot of the reviews that you can find online on semi professional blogs are wrong. There are some in which a certain cooler performs superior than its competitiors and then there are others where it is the other way around. That should never ever happen. Furthermore the level of noise is always measured in db and not in sone which doesn't help at all if you want to judge a fan. Last but not least, in some reviews the coolers aren't even tested with a reference fan which makes it almost impossible to compare the coolers.

Whenever I take a look at the Silverstone Argon AR05 and compare it with the Noctua NH-L9i there are some things that I notice, which I think are very important.
- both coolers have almost the exact same dimensions and design, 90*90*37mm and two heatpipes. Therefore these coolers should perform more or less the same
- the fan speed of both fans is identical. This means, that the design of the fan is the only difference between them. Since Noctua is a brand known for their high quality fans, I doubt that the SilverStone fan can be much better than the the Noctua fan.
- So, what is left that could have influence in the cooling performence? The material quality, right? We all know, that copper transports heat better than alluminium, but in return it is far heavier and more expensive than alluminium. When I look at the data sheet of both coolers I see that the Noctua is more than twice as heavy as the Silverstone. I assume it has more copper build in, which should make it better.

So, why the hell should the Silverstone perform better than the Noctua?
Click to expand...

Here is a photo comparison of AR05 and L9i:

AR05:
ar05.PNG


L9i:
L9i.PNG



1) How are these designs "almost the same"?
2) AR05 is 104mm x 92mm, L9i is 95mm x 95mm
3) Where do you see two heatpipes of L9i ?

I think it is quite logical that AR05 performs better thanks to the heatpipes which transfer heat to the sides of the heatsink, and the heat is distributed more evenly throughout the cooler.

EDIT: sorry stupid me, L9i does have heatpipes as well, they are better visible in this photo:
promaty said:
Sorry, you're right, here is a picture where L9i heatpipes are visible:
View attachment 17019

So I guess the advantage of AR05 is that it is touching more fins by entering from the outside? In addition, AR05 pipes have more direct contact with the CPU. Not an expert of heat transfer, but I will trust the reviews :)
Click to expand...
 
Last edited:
sldr said:
I got a i7 7700K on a Noctua NH-L9i in my Dan A4 in my A4-SFX.

What tests should I run to get you guys some information?
Click to expand...

Just stress test the hell out of the stock 7700K with prime95 and aida and let us know how the noise levels and temps go (and if theres any throttling). If you had a video of the sound at idle/load too, you would win the internet

If you wanna score brownie points, run a GPU benchmark concurrently and see how the CPU temps fare as well
 
promaty said:
Here is a photo comparison of AR05 and L9i, I think you are confusing these with something else:

AR05:
View attachment 17009

L9i:
View attachment 17010


1) How are these designs "almost the same"?
2) AR05 is 104mm x 92mm, L9i is 95mm x 95mm
3) Where do you see two heatpipes of L9i ?

I think it is quite logical that AR05 performs better thanks to the heatpipes which transfer heat to the sides of the heatsink, and the heat is distributed more evenly throughout the cooler.
Click to expand...

Don't you see the heatpipes on the L9i?? They are more integrated into the fins, just making them harder to see. The AR05 does have the heatpipes in direct contact to the CPU IHS though. By the looks of the images the fins are a bit more tightly put on the Noctua...so I would guess that Noctua might have a higher fin area than the AR05...but that's just a guess from my side.
 
temp test..
i5 6600 linx 0.7 ambient:21℃
perfomance c7=lp53+a9x14 >l9i+ty 100=l9i+a9 pwm(2℃ high) >l9i (4℃ high)
 

Attachments

  • IMG_7581.JPG
    IMG_7581.JPG
    143.6 KB · Views: 246
  • IMG_7582.JPG
    IMG_7582.JPG
    141.5 KB · Views: 216
  • IMG_7588.JPG
    IMG_7588.JPG
    247.4 KB · Views: 229
one more..

modding is not necessary.

lp53 with l9x65 bundle fan..
Just fit!
 

Attachments

  • IMG_7583.JPG
    IMG_7583.JPG
    166 KB · Views: 264
  • IMG_7584.JPG
    IMG_7584.JPG
    172.1 KB · Views: 217
  • IMG_7585.JPG
    IMG_7585.JPG
    192.5 KB · Views: 240
  • IMG_7586.JPG
    IMG_7586.JPG
    140.9 KB · Views: 236
Last edited:
pellen said:
Don't you see the heatpipes on the L9i?? They are more integrated into the fins, just making them harder to see. The AR05 does have the heatpipes in direct contact to the CPU IHS though. By the looks of the images the fins are a bit more tightly put on the Noctua...so I would guess that Noctua might have a higher fin area than the AR05...but that's just a guess from my side.
Click to expand...

Sorry, you're right, here is a picture where L9i heatpipes are visible:
1.PNG


So I guess the advantage of AR05 is that it is touching more fins by entering from the outside? In addition, AR05 pipes have more direct contact with the CPU. Not an expert of heat transfer, but I will trust the reviews :)
 
Last edited:
Just got My ar05, but first, these are the temps for the Cryorig C7:

Idle:
c7 idle.JPG

Stress test with Prime 95:
c7 prime 95.JPG

Not bad, but the noise obviously is annoying at full fan speed.



Now for the SilverStone AR05, here is a video:



Idle speed: 1124 RPM

Full speed: 2320 RPM

Very very quiet at full speed, which is great! But these are the temps with a 6700k @ 4.4Ghz:

Idle:
ar05 idle.JPG

Stress test with Prime 95:
ar05 stress.JPG

Ouch, within the first minute it hit 100°C, I wouldn't recommend the SilverStone AR05 cooler if you are planning to overclock, but if you aren't looking to overclock its very very quiet at 100% fan speed (86°C max @ stock speeds).

This weekend i will try the lp53 with the a9x14 and post a video and some temps, i'm also waiting on a nexus to come sometime soon, i will do the same for that if i can fit it in.
 
Already thought that the silverstones perfomance isnt much better then the Noctuas. Doesnt make any sense that he would perform better.
 
Would a few people mind posting their GPU temps? I have a 1080 Founders Edition and am a little worried it will be too hot in there. Am I better off going for an EVGA ACX 3.0 / Gigabyte G1?
 
MrLahn said:
I've just received the Cryorig XT140 to put ontop of my C7, I'll test when I get home tonight and post my results. C7 fan is 15mm, XT140 is 13mm, so hopefully that'll make it quiet down a bit.
Click to expand...

I am really interested in this solution, if you don't mind please let me know how you would attach the fan to the cooler.
 
I am running the 6700k stock speed with the stock c7 and the asus 1070 strix gpu.
I ran some 2h For Honor while streaming last night and CPU temps did not exceed 65*.GPU stayed below 68*. I will try running som BF1 tonight with highest settings and post later.

On another note regarding the c7. I am running the Asus z170i pro gaming motherboard and i had to modify the mounting bracket since it was resting on a small row och 'chips' on the backside of the motherboard. The brackets seems to be a bit too big. So if you are running the same motherboard as me I would suggest you go with some of the other coolers than the C7.
 
Meccabolix said:
This weekend i will try the lp53 with the a9x14 and post a video and some temps, i'm also waiting on a nexus to come sometime soon, i will do the same for that if i can fit it in
Click to expand...

I'm really waiting for that :) if with videos then it will be great.
 
Meccabolix said:
Just got My ar05, but first, these are the temps for the Cryorig C7:

Idle:
View attachment 17020

Stress test with Prime 95:
View attachment 17021

Not bad, but the noise obviously is annoying at full fan speed.



Now for the SilverStone AR05, here is a video:



Idle speed: 1124 RPM

Full speed: 2320 RPM

Very very quiet at full speed, which is great! But these are the temps with a 6700k @ 4.4Ghz:

Idle:
View attachment 17022

Stress test with Prime 95:
View attachment 17023

Ouch, within the first minute it hit 100°C, I wouldn't recommend the SilverStone AR05 cooler if you are planning to overclock, but if you aren't looking to overclock its very very quiet at 100% fan speed (86°C max @ stock speeds).

This weekend i will try the lp53 with the a9x14 and post a video and some temps, i'm also waiting on a nexus to come sometime soon, i will do the same for that if i can fit it in.
Click to expand...


Exactly as I predicted. ;)

AR05 / L9i is ONLY for silent casual gaming. Get C7 or something custom if you want to overclock.
 
Delid 7700k stock clocks, stock C7, running on test bench (no Dan case yet) max temp playing Overwatch for 2 hours @165hz per core 80 | 79 | 79 | 79. While streaming the same game 30fps 720p to twitch, (still 165 fps locally) max temp 89 | 89 | 89 | 90. Inside the case it should get worse. I'll be able to post the same results from the L9i for more delid stats since my LP53 has such a long ship date.

I used Coollaboratory Liquid Pro on the die applied very meticulously. Noctua NT H1 thermal paste to mount the C7.

Voltage on 7700k reads 1.216 on CPU-Z when the CPU is stressed.
 
Last edited:
Also for people using AR05 / L9i I would recommend to go for a non-K i5 or i7 (i5-7600 for example). This is 65 W vs 91 W so you should have no problems whatsoever and run any game you want for the next 4 years with a decent GPU.
 
just go a reply back for some Samsung DDR4 VLP NON-ECC and UNREGISTERED memory. so basically should work with strix z270i motherboard with the nexus cooler and a big 120mm fan if nothing else on the motherboard gets in the way

DDR4 U-DIMM VLP 8GB 2133MHz 1Gx8 Samsung Low-Profile M4UR-8GS1VC0G-B
£69.00
DDR4 U-DIMM VLP 8GB 2133MHz 512Mx8 Samsung Low-Profile M4UR-8GSSWC0G-E
£57.40

Lead Times are 10 -14 days

Delivery is £47.50 via Fed-Ex. (canada from uk)
Prices are valid for 30 day. All pricing are British Pound Sterling
We accept all major credit cards (2% Surcharge) except American Express.

I asked a second quote for some 16GB sticks

UPDATE :

16GB 2133MHz 1Gx8 Samsung 2 2 M4UR-AGS1WC0G-B £96.20
16GB 2133MHz 1Gx8 Samsung 2 2 M4U0-AGS1WCRG £96.20
 
Last edited:
  • Like
Reactions: EdZ
like this
@arun21:
You can find temperatures regarding the GPU here: https://do cs.google.com/spreadsheets/d/1aZfTysWc4dtJC33qkLCjLkhPnzVLtbCtjpxoZeT92ZA/edit#gid=0
click on "Unigine Heaven", or "Gaming"

@Meccabolix
you call this "quiet"? I call it "annoying" which is the equivalent to the NH-L9i @ 2500rpm and above.
 
You must log in or register to reply here.
Back
Top