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

CknSalad said:
As much as I am not a fan of Linus, I am glad that he is giving good exposure to mini-itx especially the custom mini-itx cases like Dan A4, Sentry, and Ncase M1.
Click to expand...
That's the main reason I subscribed to the channel. They have given SFF cases the promotion they deserve.
 
Here my setup:
Strix z270i
2x8gb 3200 cl16 rgb ram (g.skill)
1x 960evo
Silverstone sfx-l 500W
i7 7700k delided & relidded.
Noctua lh9i stock.
My max temps hover around 87 C under load (cinebench)
And my cinebench score is around 915-1028 which is kinda funny because it is overclocked onto 4.8Ghz vcore 1.28...
Anyone experiencing those kinda flexible results with the i7?
 
pionir said:
Here my setup:
Strix z270i
2x8gb 3200 cl16 rgb ram (g.skill)
1x 960evo
Silverstone sfx-l 500W
i7 7700k delided & relidded.
Noctua lh9i stock.
My max temps hover around 87 C under load (cinebench)
And my cinebench score is around 915-1028 which is kinda funny because it is overclocked onto 4.8Ghz vcore 1.28...
Anyone experiencing those kinda flexible results with the i7?
Click to expand...

some cores must be shutting down because of the overclock. run a stress test on prime 95 with these settings https://i.imgur.com/UhzFIPz.jpg and check on aida64 in the stress test monitoring window (without running the aida64 stress test) if you're keeping 100% activity . try running at 4.4ghz and you should get a higher score if it really was some cores shutting down during the cinebench benchmark.
 
mcheddadi said:
some cores must be shutting down because of the overclock. run a stress test on prime 95 with these settings https://i.imgur.com/UhzFIPz.jpg and check on aida64 in the stress test monitoring window (without running the aida64 stress test) if you're keeping 100% activity . try running at 4.4ghz and you should get a higher score if it really was some cores shutting down during the cinebench benchmark.
Click to expand...
Aida 64 reports 100% constant usage on 4.8 under prime95. Once i clocked it down to 4.4Ghz the new cinebench result is 876, which is worse than stock... i don`t know but after my delid it seems that all specs went down except the put in overclocks. For what the clock is, the benchmarks are not as good as stock... at least temps are nice :D
 
i went to stock settings (whole bios at standard even the clock) and now i am getting between 880 and 961.. this cpu drives me nuts
 
Looks like the most optimal overclock for i7 7700K or i7 6700K builds for this case is around 4.3 - 4.4 GHz. Very curious if users can at least comfortably hit 3.7 GHz for Ryzen 5 or 7 and perhaps 3.8 GHz when more AM4 mini-itx boards come out.
 
CknSalad said:
Looks like the most optimal overclock for i7 7700K or i7 6700K builds for this case is around 4.3 - 4.4 GHz. Very curious if users can at least comfortably hit 3.7 GHz for Ryzen 5 or 7 and perhaps 3.8 GHz when more AM4 mini-itx boards come out.
Click to expand...
Im far more interested in V2 having the AIO cpu mount option, as the temps would e far better with a delidded 7700k.
 
dondan said:
1. Rubber feets instead of metal feets for better grip with the same look
2. 92mm case fan mount point if you not use the hdd bay ( will also support Asetek 92mm AIO)
3. A cover plate for the GPU mounting hole
4. Better centered USB front port
5. Stamp-in motherboard stand-off instead of screwed ones
Click to expand...

For the V3 these are my wishes:
- It would be great to have a plastic plate anywhere, to attach a non protruding wifi antena inside the case, avoiding the faraday effect and the ugly antennas on the back. maybe above the gpu srcews in the back, you can design the antenna and provide it with the case, it would be great.
- A 3,5mm mini jack port in the front of the case for the headphones.
- An unified two slots hole in the back of the case, so the blower style gpus can exhaust freely without the middle plate.
- The case can be a lot lighter and incredibly efficient, if you design a custom power supply and a liquid cooling solution.
- A front led.
 
There is no faraday effect with this case. I cannot remember exactly but somewhere in the 279 pages here a user reported that he was using antenna inside the machine just fine.
 
Absolutely beautiful. I just wish it could have one 3.5" bay instead of or alongside the bunch of 2.5's. 6TB, 8TB or 10TB in one volume is nicer to have and cheaper than the same spread across two or three. Two 2.5's side by side have almost the exact same footprint as one 3.5, so I don't know why swappability/dual compatibility isn't more of a thing. The Sentry is the same deal too. I guess it's stick with the FTZ01 for now, with all its wasted space.
 
Just bought the FTW 3 in Green!!! now how do I sell this Founders Edition...Post on craigslist but no biters yet.
 
antilles said:
Absolutely beautiful. I just wish it could have one 3.5" bay instead of or alongside the bunch of 2.5's. 6TB, 8TB or 10TB in one volume is nicer to have and cheaper than the same spread across two or three. Two 2.5's side by side have almost the exact same footprint as one 3.5, so I don't know why swappability/dual compatibility isn't more of a thing. The Sentry is the same deal too. I guess it's stick with the FTZ01 for now, with all its wasted space.
Click to expand...
The drives aren't side by side in the A4, and that makes a difference. Also, volume counts, and a single 3.5" HDD takes up about the same space as anywhere from four to six 2.5" drives (though obviously cables become an issue with the latter).
 
I have a 5 GHz WI-FI. I can tell that with no antenna it works better than with an Asus antenna. And with my makeshift antenna (a 15mm piece of wire sticking out of the socket) it works even better. Download speeds from Steam are 11, 8 and 15 MB/sec respectively.
 
I got restless and ordered an LP53 so I decided to put it head to head against my C7 heatsink and use the winner

C7 vs LP53 Test

Specs:

Stock Intel 7700K used for this test. Undervolted to 1.12V and clocked at 4.3GHz
Asus Strix Z270i Gaming (bear in mind default settings on this board will overclock the 7700K to 4.5GHz on all cores)
Noctua a9x14 fan on both heatsinks
Noctua NT-H1 thermal paste
GTX 970 mini (waiting for my FTW3 to arrive)
Identical CPU and GPU fan curves (above 75 degrees, the CPU fan runs at 100% anyway)
LP53 is mounted with the heatpipe curve pointing down
C7 heatsink is mounted with heatpipes facing side to side

Test 1: Intel Extreme Utility Stress Test + Uniengine Heaven Loop, running concurrently for 1 hour, heatsink swapped, then test done again
Ambient conditions: Ambient 28-30 deg C. No airflow where the computer is situated. I literally had sweat on my brow the whole time sitting at my computer while running the test, so it was a pretty hard test

LP53 + Noctua A9x14: Averaging 91-92 degrees with occasional maximum spikes into 96-98 degrees

C7 + Noctua A9x14: Averaging 88-89 degrees with occasional maximum spikes into 93-94 degrees

Test 2: Observing temps on BF1

LP53 + Noctua A9x14: 85 degrees average (max 88 degrees) while gaming with maximum spikes into 90 degrees during map loading

C7 + Noctua A9x14: 81 degrees average (max 83 degrees) while gaming with maximum spikes into 87 degrees during map loading

My honest opinion:

The LP53 is pretty overrated. Sure, its made of copper, but thats the only good thing about it compared to the C7. It is absolutely dinky in size (similar to the NH-L9i size) and only has 2 heatpipes. Using it with the Noctua fan, a lot of available height for the CPU cooler is not used, and therefore wasted. Material type alone will not overcome the lack of fin surface area, especially since the size differences between the 2 heatsinks is so large. I also like the mounting solution of the C7 heatsink better (with a backplate under the motherboard). Plus it is also much easier (not to mention more elegant) to mount the noctua fan to the C7 heatsink compared to the LP53, because the fan is about the same diameter as the C7 heatsink, whereas it is larger than the LP53 heatsink.

I would take the LP53 over the NH-L9i, but with my setup, the C7 heatsink is clearly superior to the LP53 by 3-4 degrees on all loads
 
Last edited:
I've got currently Silverstone SX500-LG PSU on my A4 (7700K) and now considering to go with FTW3. Anyone smarter then me can share if my PSU is enough for the job or do I need to buy 600W PSU which seems to be recommendation according to EVGA.
 
Assuming you arent overclocking I think you should be fine. 95W TDP plus about 280W conservative watts from the GPU still puts you below 400W, with another 100W to spare. I think EVGA's recommendation is a little on the safe side and maybe assumes a whole bunch of fans and hard drives are being used as well, plus some added safety margin
 
Blackreplica said:
Assuming you arent overclocking I think you should be fine. 95W TDP plus about 280W conservative watts from the GPU still puts you below 400W, with another 100W to spare. I think EVGA's recommendation is a little on the safe side and maybe assumes a whole bunch of fans and hard drives are being used as well, plus some added safety margin
Click to expand...

Thanks for quick reply! Yeah no overclocking with this build.
 
I have finally got the last parts for my build and here is the result :)

wnvg1Ir.jpg

3b5ZExF.jpg


The list of current components:

Ryzen 1700
Biostar X370GTN
Cryorig C7 & Thermalright TY100
Sapphire RX570 8GB
Corsair CMK16GX4M2B3000C15
Corsair SF450
Samsung 960EVO 256GB
Crucial MX300 525GB

The RAM is currently running only on 2133 and I will probably change the RX570 with a Vega in the future. Next week I will receive a window kit and some led strips.
 
Last edited:
Just to add that with your comment

Blackreplica said:
Using it with the Noctua fan, a lot of available height for the CPU cooler is not used, and therefore wasted.
Click to expand...

With 4mm standoffs, this additional space does allow the Noctua a9 25mm thick fan to be used on the LP53, which is not possible on the C7 (which is slightly higher).
 
HoodedHusky said:
Just to add that with your comment



With 4mm standoffs, this additional space does allow the Noctua a9 25mm thick fan to be used on the LP53, which is not possible on the C7 (which is slightly higher).
Click to expand...
I just want to add that i have the noctua a9 25mm with my l9i and I cant for the life of me hear the terrible turbulence everyone else is reporting.
 
Blackreplica said:
I got restless and ordered an LP53 so I decided to put it head to head against my C7 heatsink and use the winner

C7 vs LP53 Test
C7 + Noctua A9x14: 81 degrees average (max 83 degrees) while gaming with maximum spikes into 87 degrees during map loading

My honest opinion:

The LP53 is pretty overrated. Sure, its made of copper, but thats the only good thing about it compared to the C7. It is absolutely dinky in size (similar to the NH-L9i size) and only has 2 heatpipes. Using it with the Noctua fan, a lot of available height for the CPU cooler is not used, and therefore wasted. Material type alone will not overcome the lack of fin surface area, especially since the size differences between the 2 heatsinks is so large. I also like the mounting solution of the C7 heatsink better (with a backplate under the motherboard). Plus it is also much easier (not to mention more elegant) to mount the noctua fan to the C7 heatsink compared to the LP53, because the fan is about the same diameter as the C7 heatsink, whereas it is larger than the LP53 heatsink.

I would take the LP53 over the NH-L9i, but with my setup, the C7 heatsink is clearly superior to the LP53 by 3-4 degrees on all loads
Click to expand...

I would say that the most cost-effective cpu cooling solution would be the C7 and Noctua fan. I remember reading a post that stated that for presumably the best cooling it would be the LP53 + Noctua fan + 4mm standoffs.
Plus, C7 comes with AM4 support now (just have to give them proof via a form). Looks like Meccabolix either did a really good job of delidding or won the silicon lottery.
 
Your pic is my reflection said:
Nope. There also is "+no VRM HS + no IO shield". I don't remember if there was a test of the C7 with such conditions, so it was not a fair comparison. And in equal conditions the C7 is ~3 degrees cooler though it have some turbulence noise of course.
Click to expand...
Isn't the Cryorig C7 not too bad noise wise with the slim Noctua fans.
 
I have to add that it really depends on what motherboard you've got. I have an Asus z270i where the airflow of an LP53 is almost utterly restricted by the VRM heatsink and the memory module. It actually performs surprisingly well considering that. If you have another motherboard than the LP53 might be the best choice.
 
Last edited:
Your pic is my reflection said:
I have to add that it really depends on what motherboard you've got. I have an Asus z270i where the airflow of an LP53 is almost utterly restricted by the VRM heatsink and the memory module. It actually performs surprisingly well considering that. If you have another motherboard than the LP53 might be the best choice.
Click to expand...

ah okay. Meccabolix has the Gigabyte Gaming 5 Z170 mini-itx board, so maybe that is why the LP53 performed better. It would be nice to hear from him directly though if he did extensive testing with the CryOrig C7 with the same current configuration as the LP53 (no I/O shield, 4mm standoffs).
 
pionir said:
yFBiXAQ.png
S90TVhK.jpg

Managed to squeeze 4.8Ghz with my noctua lh9i fan... are you guys getting similar results?
Click to expand...

Cinebench alone isnt nearly tough enough. You need to run a proper cpu stress test like intel extreme utility and then run your GPU 100% which dumps more heat into the chassis while the cpu is taxed. just my opinion
 
Blackreplica said:
the C7 heatsink is clearly superior to the LP53 by 3-4 degrees on all loads
Click to expand...

This makes sense because the C7 has more metal! They cheaped out on their fan though, so if you want the absolute best cooler, you might need to buy the best heatsink and the best fan separately! Good information!
 
You must log in or register to reply here.
Back
Top