Prototype Project Orthrus - Smallest SLI mATX case (V2)
- Thread starter LukeD
- Start date
Finally managed to whip this up. Updates in the first post as well. Thermal images coming soon as well.
Benchmarks - Noise / Power Consumption / Temperature:
Test Configuration:
CPU - Intel 6800K (@3.4GHz)
RAM - Corsair Vengeance LPX 3200MHz 16GB (2x8GB)
MB - EVGA Micro 2 X99
GPU 1 - nVidia GTX 1080
GPU 2 - nVidia GTX 1080
CPU AIR Cooler - ID Cooling IS-VC45
GPU AIR Cooler - Founders Edition Blower
CPU AIO Cooler - Kelvin S24
CPU CL Cooler - Kelvin S24 (cpu block/pump combo)
GPU CL Waterblock 1 - EK-FC1080 GTX - Nickel
GPU CL Waterblock 2 - Koolance GPU 220
Thermal Paste - EK
Benchmarks - Noise / Power Consumption / Temperature:
Test Configuration:
CPU - Intel 6800K (@3.4GHz)
RAM - Corsair Vengeance LPX 3200MHz 16GB (2x8GB)
MB - EVGA Micro 2 X99
GPU 1 - nVidia GTX 1080
GPU 2 - nVidia GTX 1080
CPU AIR Cooler - ID Cooling IS-VC45
GPU AIR Cooler - Founders Edition Blower
CPU AIO Cooler - Kelvin S24
CPU CL Cooler - Kelvin S24 (cpu block/pump combo)
GPU CL Waterblock 1 - EK-FC1080 GTX - Nickel
GPU CL Waterblock 2 - Koolance GPU 220
Thermal Paste - EK
And here are the thermal images corresponding to tests 6, 7, 8 and 9 from the above table
Awesome job LukeD, looks amazing! I'm wondering if making the GPU panel have more perforation would make running open shroud coolers a better option than blowers. It looks like when running air throttling is occurring on the GPU, but you have additional headroom on the CPU side even with that 140w processor. So having the GPUs use open shroud coolers could reduce their temps enough, so they don't throttle. I'm also wondering if because of the isolation between the two compartments if open shroud coolers will have any significant impact on the CPU. I also wonder if using an AIO for CPU and using open shroud GPUs would be best for thermals as seen in Bitwit's testing of a more standard case, . As the front AIO won't raise air temps that much on the incoming air allowing the better cooling of open shroud GPUs to get better temps than blowers. The only negative side I could see to this would be case temps may get too high at this point as both the CPU and GPUs are dumping their heat into the case. Anyway great work as always and I would love to see the additional testing if you have the time.
If you were to go with an open air GPU cooler in this case, intake temp from a radiator at the front wouldn't be much of a problem.. simply because you'd want that as the exhaust to dispose of the GPU waste heat, that is from the otherwise lack of exhaust vents at the GPU compartment.
(From my outside perspective and lack of research)
Also, the Bitwit video is mostly useful in that component layout and setup, as just a longer backplate on the blower style GPU with front rad intake would seriously affect temps and adding another GPU would give an effect as well, and the length front to back etc.
(From my outside perspective and lack of research)
Also, the Bitwit video is mostly useful in that component layout and setup, as just a longer backplate on the blower style GPU with front rad intake would seriously affect temps and adding another GPU would give an effect as well, and the length front to back etc.
Yeah I agree that Bitwit's setup isn't the same, but what I think is important is looking at these different configurations to see how a certain cooling setup effects each component. Seeing as the GPUs are the component that is running the hottest, I would like to see if open shrouds can solve this problem without having to resort to water cooling for a very high end setup. What it will also help see is how good the case is at exhausting air as at that point all the heat is being dumped into the case from the shrouds and AIO. Overall I think it would be a good addition to the set of test to really see if what the case can handle and what are its current failures.
Lol. Yes, tests 3 and 4 i took out because i had incomplete data sets. It was some tests I was running after drilling holes in the case 
So i should have put that in the explanation that the vents weren't present on test 1 and 2.
Also please note that GPU throttling in CPU AIO + GPU AIR combination only occurred when the fans at the front of the case were off.
Otherwise there was no throttling experienced when the front fans were pushing air through the case.
I would love to do some open / fan cooler based GPU testing however I only have 1 card like that (EVGA 1080) that I put the EKWB on.
I could put the old cooler back on but i'd need another card like it to do proper SLI testing.
In Australia they are crazy expensive, over $1000 so I'm not too keen on spending that money right now.
Unless someone wants to send me their card in for testing
Also another thing to note was that, Test 9 worked really well, exhausting hot air out the front of the case, and the fans pushing that air out.
It actually worked really well for the PSU as well, because the reverse air flow kept the PSU really cool.
So i should have put that in the explanation that the vents weren't present on test 1 and 2.
Also please note that GPU throttling in CPU AIO + GPU AIR combination only occurred when the fans at the front of the case were off.
Otherwise there was no throttling experienced when the front fans were pushing air through the case.
I would love to do some open / fan cooler based GPU testing however I only have 1 card like that (EVGA 1080) that I put the EKWB on.
I could put the old cooler back on but i'd need another card like it to do proper SLI testing.
In Australia they are crazy expensive, over $1000 so I'm not too keen on spending that money right now.
Unless someone wants to send me their card in for testing
Also another thing to note was that, Test 9 worked really well, exhausting hot air out the front of the case, and the fans pushing that air out.
It actually worked really well for the PSU as well, because the reverse air flow kept the PSU really cool.
LukeD, you didn't convert the CPU Idle temps after the 1st test, but thank you .
.
Lol. Yes, tests 3 and 4 i took out because i had incomplete data sets. It was some tests I was running after drilling holes in the case
So i should have put that in the explanation that the vents weren't present on test 1 and 2.
Also please note that GPU throttling in CPU AIO + GPU AIR combination only occurred when the fans at the front of the case were off.
Otherwise there was no throttling experienced when the front fans were pushing air through the case.
Also another thing to note was that, Test 9 worked really well, exhausting hot air out the front of the case, and the fans pushing that air out.
It actually worked really well for the PSU as well, because the reverse air flow kept the PSU really cool.
I figured it had to do with that complication you encountered
I guess that depends on your definition of throttling, I can't see it really throttling at those temps after the no case fan tests, but GPU Boost 3 won't be as giving on the clocks, bringing them closer to the advertised boost clock which (surprise!) some stubborn people will consider thermal throttling
Woop woop, I wonder how many would actually want to use front exhaust even if it would be superior...
I would, and I'd put it over my monitor blowing down hot PCMR air over my face.. letting my hair wave in the winds it creates. (Perhaps a bit over dramatized... but still!)
In GPU-z there is a field there called PerfCap Reason.
When the temps get to 85c it will say "Thrm" - meaning its thermo throttling.
After test 9 i did a test where I allowed the GPU's 120% power and they could hit 1970 mhz on water
In the FLIR camera i could see the entire window shade was yellow/white because of all the hot air keeping it toasty
Shame its the middle of summer here in Australia, it could really be used to keep your feet warm
When the temps get to 85c it will say "Thrm" - meaning its thermo throttling.
After test 9 i did a test where I allowed the GPU's 120% power and they could hit 1970 mhz on water
In the FLIR camera i could see the entire window shade was yellow/white because of all the hot air keeping it toasty
Shame its the middle of summer here in Australia, it could really be used to keep your feet warm
Oh this next one is going to have a very special purpose
I don't want to spoil the surprise All will be revealed in good time
Oh this next one is going to have a very special purpose
All will be revealed in good time
Protocase seems to be pretty excited for Project Orthrus btw!
And noticed you've been retweeted by Fractal Design, I wonder how many you managed to inspire through that...
Also...Is it gonna be a show piece out on Computex? Know you prolly won't spoil it but I could imagine some forward thinking company wanting to showcase their hardware in there.. kinda
Similar threads
