aeduG

Oh I see. Thank you for the research. I will see how much I can invest, but it looks like I'm pretty much stuck with the cards.

Ok. I still don't understand why I would have to manually adjust the fan curve for 100% fan speed. Shouldn't the GPU be smart enough to do that when it's running 90 degrees hot?

Yeah thank you! I have two front in-take fans, one exhaust at the back, and one on top right above the cpu cooler. The case does not have holes for a side in-take unfortunately. Maybe I can mount a fan directly on the side of the GPUs?

Ok I get it. So this is why I'm getting these temperatures? The cards are not meant to be stacked like this?

Yes but but people recommend the 1080ti for deep learning. So that's why we bought it

It is the GTX 1080ti Turbo 11G from Asus. There is maybe 2-5mm of space between the cards:

Hello, I have built a quad GPU workstation. If I give all my GPU's 100% load, they heat up to 90 degrees but the fan speed is only at 80%. Is this bad? I plan to do heavy workloads, almost 24/7. And why are the GPU fans not going 100%? (see picture). I have space at the bottom and top of my case to add two more fans. Will this help? Thanks for your advice!

@Mr alex, @TVwazhere, @cynexit Thank you all for the recommendations. This helped a lot and I know what to buy now.

Ok I see. Maybe this will do just fine: Noctua NH-U14S TR4-SP3

I looked into threadripper and I can consider it but what cooling solution do you suggest? Water or air?

I think this is the max I can fit into the case/board/budget

ok I will have a look at ryzen

Hello, I would like to assemble a system for deep learning, but I am not an expert when it comes to selecting the optimal parts. As you can see below I wish to use 4 GPUs. I want to avoid possible bottlenecks. I hope to find someone in this forum who has experience with these kind of builds. Here is my tentative selection of parts: CPU: Intel Core i9-7900X X-Series (LGA 2066, 3.3GHz, Unlocked) SSD: Samsung 960 PRO (512GB, M.2 2280) POWER: Corsair HX1200 HDD: WD Black (6000GB, 3.5", Desktop, Workstation) COOLING: NZXT Kraken x62 inkl. AM4 CASE: Corsair Obsidian 750D BOARD: MSI X299 XPOWER GAMING AC (LGA 2066, Intel X299, E-ATX) RAM: 2x CORSAIR Vengeance LPX 4 x 16 GB (total 128 GB) GPU: 4x ASUS TURBO-GTX1080TI-11G Alternative AMD Build: Board: ASUS ROG ZENITH EXTREME (TR4, AMD X399, E-ATX) CPU: AMD Threadripper 1920X (TR4, 3.50GHz, Unlocked) COOLING: Noctua NH-U14S TR4-SP3 (16.50cm)

Thank you all for the replies. Yes, the workstation we build will be used for deep learning, not for games.

Hello, I would like to fit four Asus ROG Strix GTX 1080 Ti on this motherboard: https://www.asus.com/us/Motherboards/X99-E-10G-WS/ I read that the graphics card occupies 2.5 slots, but I could not find out how many slots each PCI-Express on the board uses. If it is 2, not all cards will fit, if it is 3, then they all fit in. Is that correct? Thank you for the help. P.S. my first time posting in this forum.