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

[mirgus]

So the Noctua has a better cooling performance then the Cryorig C7 ! Ambient temperature with bot CPUs is the same (25 degrees, messured with an Arduino + TMP36GZ).
The Noctua should be quieter with an 25mm fan and maybe a little bit cooler. Also same thermalpaste used !

Here are some noise recordings with my microphone. Sorry for the coil noise in the background, its from my power supply. 2460RPM is the max speed of the Noctua fan.















I uploaded the soundfiles, if someone wants to play it with their music player for direct comparison

 

[Petroj]

Interesting. My mind was set on the C7 but now I might rethink that and get a NH-L9 with a 25mm fan.

I wonder what Noctua would be able to accomplish if they made a version of NH-L9 with a one centimeter thicker heatsink. It would be interesting to compare the temps of such a version and the original

 

[mirgus]

Well i thought, that the cpu would be 5-8°C cooler with the C7, because its just bigger and has 2 more heatpipes.
I dont know whats the adventage of the Noctua here, i guess the heatpipes + the cooperbase has a better quality.
I would be interesting, how much the surface size differents is between the cooler. I will calculate that when i make the new photos
If you really want the quietest of those coolers, get the Noctua. Its quieter with the stock 15mm fan and u can replace it.

Although the Noctua packaging is freaking gorgeous


I am happy i can test the 2 coolers (i bought the C7 myself). There arent comparable tests at all and so i can test parameters and i know, that both are tested with them (fan rpms, ambient temperature etc).
I think i will buy the NH-L9i and sell the C7


I will post here a short version of my comparison test if its finished
Gonna mount the 25mm Noctua fan tomorrow and test it

 

[lscollo1234]

If anyone is looking to sell their kick starter case shoot me a PM. I'm willing to buy any color and ill pay a small markup if it comes to it.

 

[akhilv1]

Anyone know how long custom psu cables need to be for this case? I suppose 24-pin depends on the motherboard but how about the 6/8-pin and the 4/8-pin?

 

[tviceman]

i7 6700k power consumption results:

voltage 1.152-1.168v
Hyperthreading ON
Prime 95 - 143 watts

voltage - 1.152-1.168v
Hyperthreading OFF
Prime 95 - 120 watts

voltage - 1.296-1.312v (stock motherboard setting)
Hyperthreading ON
Prime 95 - >170 watts


I ran Prime 95 until it completed it's second pass and reached max temperatures.
Hyperthreading added more than 20 watts according to the Kill-a-watt meter.
Temps with Hyperthreading disabled were about 8-9 degrees lower.
Temps with Hyperthreading on and voltage at 1.168v hit a maximum of 88 C.
I manually aborted the torture test at the stock settings because temps hit 98 C quickly into the second pass.

More here: https://hardforum.com/threads/6700k-4-2ghz-stress-test-power-consumption-results.1918376/

 

[akhilv1]

i7 6700k power consumption results:

voltage 1.152-1.168v
Hyperthreading ON
Prime 95 - 143 watts

voltage - 1.152-1.168v
Hyperthreading OFF
Prime 95 - 120 watts

voltage - 1.296-1.312v (stock motherboard setting)
Hyperthreading ON
Prime 95 - >170 watts


I ran Prime 95 until it completed it's second pass and reached max temperatures.
Hyperthreading added more than 20 watts according to the Kill-a-watt meter.
Temps with Hyperthreading disabled were about 8-9 degrees lower.
Temps with Hyperthreading on and voltage at 1.168v hit a maximum of 88 C.
I manually aborted the torture test at the stock settings because temps hit 98 C quickly into the second pass.

More here: https://hardforum.com/threads/6700k-4-2ghz-stress-test-power-consumption-results.1918376/

I'm curious to see what kind of peak power usage is achieved coupled with something like a 1070 in a regular gaming benchmark. Would you be willing to run a quick game with the kill a watt?

 

[tviceman]

I'm curious to see what kind of peak power usage is achieved coupled with something like a 1070 in a regular gaming benchmark. Would you be willing to run a quick game with the kill a watt?

Unigine Heaven topped out at 206 watts with my video card at stock settings. I know there are better, more hungry games / benchmarks than that, but that's all I got right now. I'll try out an actual game in a few days and report back.

 

[akhilv1]

Unigine Heaven topped out at 206 watts with my video card at stock settings. I know there are better, more hungry games / benchmarks than that, but that's all I got right now. I'll try out an actual game in a few days and report back.

Thanks for testing it out.

 

[mirgus]

Well what can u do, if the cooler space is limited and you want lower temps?

Guess whos head will roll soon

Edit:

So guys, the Aida64 Stress CPU measurements are finished.



If u want the quietest setup and coolest setup, then the Noctua NH-L9i + an 92x92x25mm fan with a decent quality is the way to go. If thats to expensive, take the stock NH-l9i, still quieter then the C7 and a little bit cooler.

The ambience temperature when i measured with the Noctua 92x92x25mm fan were 1 degree higher then the other measurements.


And here the new soundfiles. The Noctua A9 25mm pwm is louder at 550rpm then at 850rpm. I double checked it.





Feel free if u have a suggestion

 

[stelsevo]

Oh crap. I already bought C7
Now I'm going to sell it and buy L9i + NF-B9 Redux
Mirgus, thanks!

 

[mirgus]

I will sell my C7 too
I will make some high res pictures on sunday and will leave a link here
At least i could use him for comparison

Can u see my linked pictures in the post above correctly or is my browser bugged ?

 

[stelsevo]

Can u see my linked pictures in the post above correctly or is my browser bugged ?

I see your pictures

 

[raildex1]

mirgus sell L9i - https://hardforum.com/threads/dan-a...-in-the-world.1799326/page-55#post-1041652691

Dan has already tested the cooler in the A4.

And also with the 25mm fan.

The C7 is better than the L9i.

T318 vs R31 is the current thing to test - dan already tested T318 with AXP-100.

What someone needs to do is make 20mm version of AXP100 fan or A14 and combine with the better of the T318 or R31. Then use Gelid Solutions GC-Extreme

Then delid the CPU And put some CLU on the die - although problem is that since this PC is designed for moving around this may pose some issues.

If you want quietness though then the Noctua L9i is still the best afaik.

 

[mirgus]

Edit:
This is referred to the Cryorig C7 test which daniel has done

Nope he tested the Cryorg C7 with another CPU. Many things arent good to test under such test conditions

First how was the ambient temperature with the different tests? Are they really the same? Did anything of the test setup change? This can make easy 5°C differents. Then the Cryorig C7 was tested with another CPU and not the same as with the other coolers. U just simply cant "calculate" the temperate difference between those different CPUs, as my test shows. Just because it says a CPU as a TDP of XX Watt, it doesnt mean it produces that much heat. Then u still have the influence of the IHS, how well is the thermalpaste applied (between die and IHS) and the CPU Temperature can variate imense on different CPUs with the same model. Every CPU is different.


So i know for sure, that the Noctua NH-L9i is better in any way (noise, temperature) then the C7 in my test conditions (25°C ambient temperature, all case fans off and sidepanel opened)

The T318 or R31 might be better, didnt test them. Because they are made for server platforms, they arent really interesting for the most users.


If the Dan Case arrives, i will try to retest it in tha A4-SFX case

 

[Ozymandias]

Then delid the CPU And put some CLU on the die - although problem is that since this PC is designed for moving around this may pose some issues.

No need to delid an Intel HEDT CPU, they have solder between the CPU and IHS.

 

[mirgus]

Another test of the Corsair SFX 450
https://www.computerbase.de/2016-11/sfx-netzteil-corsair-sf450-test/

Short version, the corsair is better in anyway then the silverstone ST45SF-G. Also the corsair has 7 years warranty ! Best choice atm

 

[Philfreeze]

Your forgetting the most importent fact when comparing copper to aluminium. Copper is 3,32 times as heavy as aluminium. In a portable case like this, it would make ZERO sense, to choose a heavier material when the diffrence in performance is this tiny.

That is true if you want to carry the case around. 200g Is quite a bit of a difference.
I was only talking about the cooling performance in this comparison, it just didn't fit the stuff I have learned in EE school.

I asked Dynatron some month ago about this, because they wrote on their page a TDP of 135W for the full copper T318 and 165W for the R31. They told me that they perform on the same level. It is an error in the datasheet.

Yeah I kind of expected something like that.

Btw. does somebody know why so many people think Aluminium is "more efficient at dissipating" as Ca11idus put it? I haven't found any comparable data to support this claim.
As I said, convection and radiation are the two big things you have to consider and these two factors (mostly convection though) say how 'efficient' the heat is conducted from the heatsink to the air.

 

[Ca11idus]

That is true if you want to carry the case around. 200g Is quite a bit of a difference.
I was only talking about the cooling performance in this comparison, it just didn't fit the stuff I have learned in EE school.




Yeah I kind of expected something like that.

Btw. does somebody know why so many people think Aluminium is "more efficient at dissipating" as Ca11idus put it? I haven't found any comparable data to support this claim.
As I said, convection and radiation are the two big things you have to consider and these two factors (mostly convection though) say how 'efficient' the heat is conducted from the heatsink to the air.

I don't know for a fact that the aluminum fins are better. I just know that they were tested and found to be better. The test environments were probably different. But why are the vast majority of the available heatsinks not all copper? It's not a price issue because people will spend $80+ for a heatsink if it performs well.

Outside of what you have learned in EE school have you tested the two to know for a fact that the data that you were taught translates from paper to real world efficiency? I like to see tests and facts.

Just like carbon fiber. In paper it looks amazing but if you use it as a building material for a boat you will have little to no indication when it is about to fail. It will just snap. Again, in paper it is an amazing material but it is won out by other materials for boating.

There should be plenty of data like the older thermalright full copper heatsink. Even msi's full copper heatsink golden edition gpu that came out a few years ago.

 

[Ca11idus]

Copper has got better thermal conductivity.

Aluminium - 200WmK" role="presentation" style="box-sizing: border-box; margin: 0px; padding: 0px; border: 0px; font-variant: inherit; font-stretch: inherit; line-height: normal; font-family: inherit; vertical-align: baseline; display: inline; word-spacing: normal; word-wrap: normal; white-space: nowrap; float: none; direction: ltr; max-width: none; max-height: none; min-width: 0px; min-height: 0px; position: relative;">200WmK200WmK
Copper - 400WmK" role="presentation" style="box-sizing: border-box; margin: 0px; padding: 0px; border: 0px; font-variant: inherit; font-stretch: inherit; line-height: normal; font-family: inherit; vertical-align: baseline; display: inline; word-spacing: normal; word-wrap: normal; white-space: nowrap; float: none; direction: ltr; max-width: none; max-height: none; min-width: 0px; min-height: 0px; position: relative;">400WmK400WmK
(from here, also here)

But thermal conductivity within the solid material is only a part of the story. The rest of the story depends on where one wants to dump the heat into.

Liquid coolant
Copper heatsink (one may also call it heat transfer block) will perform better than aluminium.

Air with forced convection
In other words, there's a fan blowing onto the heatsink. Copper heatsink will perform better than aluminium.

Air with natural convection
I've saved best for last. It also looks like it's the O.P.'s case too.

With natural convection air, the copper heatsink perform only marginally1 better (in °C/W) than aluminium. This is because the bottleneck isn't in the transfer withing metal. When you have air with natural convection, the bottleneck is in the transfer between metal and air, and it's the same for Al and Cu.

1 I might add that the marginal increase is often not worth the cost of Cu.

This curve demonstrates the non-linear relationship between heat transfer and material thermal conductivity. The curve is generic. It applies to any application having both conduction and convection components to the total heat transfer. [Radiation is typically small and is ignored in this calculation.] The shape of the curve is the same regardless of the application. The quantitative values on the axes are not shown because they depend on the power, part size and convective cooling conditions. They become fixed for any given application and set of conditions. It’s obvious from the shape of the curve that heat transfer depends on material thermal conductivity but there is also a point, a knee in the curve, where increasing thermal conductivity produces negligible improvement in the heat transfer.
(source, emphasis mine N.A.)

Copy and pasted http://electronics.stackexchange.com/questions/255731/copper-or-aluminum-heatsink

 

[Ca11idus]

Copper has a higher specific heat, so a block of copper would cool better than a block of aluminum, but aluminum has a better coefficient of thermal conductivity so it tends to radiate heat faster through fins.

This is why you generally see radiators and fins made out of aluminum, even high end heat sinks for computers tend to have copper heat pipes with aluminum fins.

In a case like the A4 where you are fighting for space and airflow (thiner fan for the t318 or the r31) it just works better.

 

[Ca11idus]

I want copper base on copper fins to be amazingly better than copper base on aluminum fins with my t318. But I'll have to wait for a comparison. Again, I'm not saying that copper isn't a better conductor. It probably isn't that much better to justify producing fully copper heatsinks.

 

[iodine]

So i know for sure, that the Noctua NH-L9i is better in any way (noise, temperature) then the C7 in my test conditions (25°C ambient temperature, all case fans off and sidepanel opened)

Thank you mirgus for doing such a big test and sharing it with all, but i do think that your test setup has a big impact on why yor test shows different results than for instance dondan's test in the A4.

In the chart it says you are using the maximus VII Impact motherboard. I think that this particular motherboard with the vertical powerdelivery daughterboard, and vertical sound card, will have a negative impact on airflow when using small cpu coolers as the C7 and the NH-L9i.
When installed with fins vertically, half of the air flow will hit the power-daughter-board, and the other half of the air will hit the soundcard and the graphichs card(when not installed with riser cable as in the A4).
if installed with fins horisontally, half the air will hit ram modules other half towards back-IO.

This will result in a lot of the hot air heading back in the vecinity of the fan, causing it to reuse some of the hot air. i think that the C7 with its special fan-shroud, designed to increase its air intake from the side of the fan, will be reusing more hot air with the VII impact compared to the NH-L9i which mirgus tested with fans with standard shrouds.

in short: i would like to see these coolers compared using another motherboard, and with the vga card mounted with a riser cable, or in the actual A4.

btw im a kickstarter backer waiting for 2 cases, and have 2 C7 ready which i will keep

 

[Pantera_666]

even high end heat sinks for computers tend to have copper heat pipes with aluminum fins.

You are absolutely right... But i found something interesting, check this:

For some reason they made it full copper on this "super high end" gpu...

 

[bAMtan2]

If u want the quietest setup and coolest setup, then the Noctua NH-L9i + an 92x92x25mm fan with a decent quality is the way to go.

I'm confused. Wouldn't that be too big to fit in DAN A4-SFX?

 

[mirgus]

Thank you mirgus for doing such a big test and sharing it with all, but i do think that your test setup has a big impact on why yor test shows different results than for instance dondan's test in the A4.

In the chart it says you are using the maximus VII Impact motherboard. I think that this particular motherboard with the vertical powerdelivery daughterboard, and vertical sound card, will have a negative impact on airflow when using small cpu coolers as the C7 and the NH-L9i.
When installed with fins vertically, half of the air flow will hit the power-daughter-board, and the other half of the air will hit the soundcard and the graphichs card(when not installed with riser cable as in the A4).
if installed with fins horisontally, half the air will hit ram modules other half towards back-IO.

This will result in a lot of the hot air heading back in the vecinity of the fan, causing it to reuse some of the hot air. i think that the C7 with its special fan-shroud, designed to increase its air intake from the side of the fan, will be reusing more hot air with the VII impact compared to the NH-L9i which mirgus tested with fans with standard shrouds.

in short: i would like to see these coolers compared using another motherboard, and with the vga card mounted with a riser cable, or in the actual A4.

btw im a kickstarter backer waiting for 2 cases, and have 2 C7 ready which i will keep

Thats why both cpu's where installed the same way, fines pointing to the back of the case


And dont believe every marketing bullshit like "air intakes". It doesnt effect the perfmonce of the fan at all
The C7 was higher then the ram and the extra pcb and it didnt perform better then the noctua on the same hight as the ram and pcb

I'm confused. Wouldn't that be too big to fit in DAN A4-SFX?

Nope it will fit

 

[bAMtan2]

Nope it will fit

I remember dondan saying 40mm is the limit, but I don't see this mentioned in the user manual or on the website, so I could be wrong.

take the stock NH-l9i, still quieter then the C7 and a little bit cooler.

There seems to be some disagreement about this One says C7 is cooler than L9i: http://www.funkykit.com/reviews/cooling/noctua-nh-l9i-compact-cpu-cooler-review/3/

Two say C7 is close to L9x65 (which is much bigger than L9i): http://aphnetworks.com/reviews/cryorig-c7/3 and http://www.eteknix.com/cryorig-c7-low-profile-cpu-cooler-review/5/

 

[mirgus]

Well i dont care about that tests. Why? Because i cant know for sure that the coolers had the same ambient temperature and if they used the same thermal paste. If they use the thermal paste which is included with the coolers, then there u already have different results.

Thats the reason why i tested them against each other

Again i dont want the Noctua to win, i bought the C7 by myself. Which tests you will trust is your thing, i personally know that my test conditions were absolut the same with both coolers.
Altough the Noctua have the option to mount a 25mm fan and he is much quieter with him and offers the same cooling perfomance then the C7 at much lower RPMs and noise

 

[Ca11idus]

You are absolutely right... But i found something interesting, check this:

For some reason they made it full copper on this "super high end" gpu...

For some reason indeed. A full copper heatsink gives the consumer the nice warm fuzzy feeling that they are getting more bang for the buck. These cards were priced outside of most people's budget because they were binned.

 

[Ozymandias]

To add more fuel to the copper v. aluminum debate, here's an article with scientific tests, by a thermal engineer. TLDR: with a basic heatsink design, copper and aluminum have very similar performance. However, the larger the heatsink or the further the source of the heat (CPU) is from the radiating surface (fins), copper's lower internal resistance wins out. With the R31 vs the T318, I wouldn't expect much difference at all but performance should technically favor the T318.

 

[swooby]

dondan, what is the current/latest realistic earliest that new orders will be 1) accepted and 2) shipped to the US?

Thanks!

 

[Ca11idus]

dondan, what is the current/latest realistic earliest that new orders will be 1) accepted and 2) shipped to the US?

Thanks!

Late January.

 

[dondan]

The container is on the way

 

[Nathan Spears]

The container is on the way

Congrats!!


Oh, and can I have one?!?! Sign me up!!!

 

[illram]

Mmmmm me likey.

 

[Hermit2001]

Copper has got better thermal conductivity.

...but aluminum has a better coefficient of thermal conductivity...

Lol, which side are you on?

 

[ranix]

Will this psu fit in the case : https://www.techpowerup.com/228288/enermax-unveils-the-revolution-sfx-power-supply-series ?

no fan under 60% load with 550 W, seems quite good on paper.

 

[Jspr]

Will this psu fit in the case : https://www.techpowerup.com/228288/enermax-unveils-the-revolution-sfx-power-supply-series ?

no fan under 60% load with 550 W, seems quite good on paper.

It must be a mistype, as their on site says 30%
http://www.enermax.com.tw/home.php?fn=eng/product_a1_1_1&lv0=1&lv1=101&no=337

"Smart semi-fanless control
Fanless mode before 30% of system load for..."

 

[swooby]

Late January.

For which? Ordering or shipping?
I took Dan's statement(s) about the container to mean that the current batch may only satisfy current orders and a new batch may be needed to fill new orders.
If new orders won't be taken until late-January, then when would those new orders ship to the purchasers?

Thanks!

 

[dondan]

The current batch is for Kickstarter and new orders. Take a look into the FAQ on my side. It will be available again early 2017 (January)