radical GPU cooling

Generic forum to discuss Octane Render, post ideas and suggest improvements.
Forum rules
Please add your OS and Hardware Configuration in your signature, it makes it easier for us to help you analyze problems. Example: Win 7 64 | Geforce GTX680 | i7 3770 | 16GB
Post Reply
ristoraven
Licensed Customer
Posts: 390
Joined: Wed Jan 08, 2014 5:47 am

Hi,

I think I just invented a new consept for radical GPU cooling. I have been a hobbyist airbrush painter (painting stuff even I don't want to put on wall) for a few years and now I was looking the compressor, like the monkey looked a bone in Space Odyssey 2001, then it hit me, heureka. I will use some duct tape, aim the air hose directly to the GPU fan and see how it goes. I haven't tried this yet. What could go wrong? Sure this makes some noise, but, airflow could be cranked up to 6 bars, so this should cool that thing quite radically.

Patent not pending. :)
User avatar
glimpse
Licensed Customer
Posts: 3740
Joined: Wed Jan 26, 2011 2:17 pm
Contact:

I guess You're going to see slightly lower temps & slightly higher clock speeds..but I wuoldn't expect too much.. air is air & it would be far from watercooling + with quite some extra noise =)

Had a simillar idea, to make a shrowd out of thick paper & add server grade fan (120-140mm) spining 2-3kRMP just to see how that goes, but when I heard those fans I changed my plans =DDD..
ristoraven
Licensed Customer
Posts: 390
Joined: Wed Jan 08, 2014 5:47 am

:D:D

Nothing can beat inventing. :D

Well, I need to do something, probably buy whole new PC. Around august, so this compressor trick could work in the mean time. I try this today, if this works so that temps stays somewhere below 70 C, I'm good. Now it's constantly at 85 C limit under load.. I have looked up also a small silent airbrush compressor at around 70 USD, so not bad, if this duct tape concept works.
ristoraven
Licensed Customer
Posts: 390
Joined: Wed Jan 08, 2014 5:47 am

Okay, did a quick test :)

Temps dropped down to 70 C with airbrush pointing to the GPU fan. Then as a control test, I left the computer case open, pointing it to a open door (a bit chilly day today), temps went to 74 C and stayed there. So the airbrush gave 4 C benefit. Conclusion: I think I'll just leave the case open, pointing it to a open door.. :ugeek:
User avatar
glimpse
Licensed Customer
Posts: 3740
Joined: Wed Jan 26, 2011 2:17 pm
Contact:

ristoraven wrote:Okay, did a quick test :)

Temps dropped down to 70 C with airbrush pointing to the GPU fan. Then as a control test, I left the computer case open, pointing it to a open door (a bit chilly day today), temps went to 74 C and stayed there. So the airbrush gave 4 C benefit. Conclusion: I think I'll just leave the case open, pointing it to a open door.. :ugeek:
=DDD You can have a huge fan to blow to Your PC too..- that might cut few degrees more =) or simply get much faster spinning fans on Your case & that will get the same effect =)
ristoraven
Licensed Customer
Posts: 390
Joined: Wed Jan 08, 2014 5:47 am

:D

Thanks, but nah, although I have one excellent fan for that, I think I settle on this setup for the time being :D

Or maybe when it gets hot out side, then that fan and the fridge would work..

Okay I stop here. :D
User avatar
smicha
Licensed Customer
Posts: 3151
Joined: Wed Sep 21, 2011 4:13 pm
Location: Warsaw, Poland

Ristoraven,
Original idea. If this works for you... why not.

2x Titan X (freshly tested) with two slot distance from each other and a very well ventilated case go at 62-70C at 60% RPM at a very mild noise (not comparable to watercooling but it's not that bad). Cranking up Titan X fans to 100% (jet noise) reduces temps to 48C. Sick result for air, but the noise drives me crazy.
3090, Titan, Quadro, Xeon Scalable Supermicro, 768GB RAM; Sketchup Pro, Classical Architecture.
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
User avatar
glimpse
Licensed Customer
Posts: 3740
Joined: Wed Jan 26, 2011 2:17 pm
Contact:

smicha wrote:Ristoraven,
Original idea. If this works for you... why not.

2x Titan X (freshly tested) with two slot distance from each other and a very well ventilated case go at 62-70C at 60% RPM at a very mild noise (not comparable to watercooling but it's not that bad). Cranking up Titan X fans to 100% (jet noise) reduces temps to 48C. Sick result for air, but the noise drives me crazy.
have You tried to OC with those fans on 100%? =) curious how far that would go, what one could get from such setup =)
User avatar
smicha
Licensed Customer
Posts: 3151
Joined: Wed Sep 21, 2011 4:13 pm
Location: Warsaw, Poland

glimpse wrote:
smicha wrote:Ristoraven,
Original idea. If this works for you... why not.

2x Titan X (freshly tested) with two slot distance from each other and a very well ventilated case go at 62-70C at 60% RPM at a very mild noise (not comparable to watercooling but it's not that bad). Cranking up Titan X fans to 100% (jet noise) reduces temps to 48C. Sick result for air, but the noise drives me crazy.
have You tried to OC with those fans on 100%? =) curious how far that would go, what one could get from such setup =)
1520 MHz on core (no OC memory 3300 - locked in MSI AB, but in precision X 4000 would be the target) + mild OC in voltage to 1.21V. I didn't want to push them higher on air. On water I'd get 150 score easily with crazy low temps. From what I read on overclock - OC for gaming not always gives great results and it's hard to exceed 1500 mhz. Octane makes use about 85% of Titan X power, that is probably why it OCs so easily.
3090, Titan, Quadro, Xeon Scalable Supermicro, 768GB RAM; Sketchup Pro, Classical Architecture.
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
Post Reply

Return to “General Discussion”