4 non-SLI graphics cards (Quadro 4000) one PC

Sep 21, 2018
4
0
10
I have 4 of these rather old beasties and just want to play with a demo of a lot of monitors, doing some rendering, writing AI code, etc. It's for the classroom. Is there any conflict installing all 4? What features get turned off?
 

Rogue Leader

It's a trap!
Moderator


No features will be turned off however depending on the CPU/Board you use some of the cards will use the motherboard's PCIe lanes so they will be slower.
 

Rogue Leader

It's a trap!
Moderator


4 x16 size slots.

Can you give use a budget range? that would really help.



Not necessarily, since he is not running SLI the cards will fully function on 4 or even 1 pcie lane.
 

RobCrezz

Expert
Ambassador


Sure, but ideally you want x8 lanes.

If we are being pedantic then they dont need x16 slots like you suggest and you could put the card in a x1 slot.
 

Rogue Leader

It's a trap!
Moderator


Hardly being pedantic, he is doing a demo with some rendering and AI, not gaming. He doesn't need 8 pcie lanes, Mining which is a similar load runs GPUs on x1 slots. Also he can't always use an X1 or X4 slot as it needs to have an open back, not all board's slots have this.
 

Rogue Leader

It's a trap!
Moderator
Because it's an ai workload, even if he was rendering in 3d its for demo purposes. What you are suggesting is to have 8 lanes per gpu is going to need an x99 or x299 board and a 40 lane cpu, which is not cheap for either thing. I didn't say it had anything to do with mining, however I was using it as a comparison because of what he said he was doing. Mining, rendering, etc takes in a load and processes it and transfers it out. I doesn't need the speed of that live multi lane link to the cpu.

Again it's 4 cards NOT in sli for a demo of various capabilities. Don't get me wrong, there is a performance difference, but for the use case it's not a big deal.
 
Sep 21, 2018
4
0
10
When in doubt, at Microsoft, we would always say the answer is E) All of the above. This machine needs to be "every use" sitting in a lab of a scientific college and also used remotely in the classroom by professors to demonstrate models or setup of models or programming.

A "surevey" of personal super computing may include apps that almost never touch the GPU. OTOH, the GPU may get blistering hot animating Toy Story in real time for which, I think, it needs all lanes. Why older hardware? It has more versatility to teach parallel processing and we just happen to have them on hand. It puts some limits (no SLI, not massively powerful) yet incredibly diverse usage.

It is amazing to me how deep the group is willing to go to answer questions. I will be much more specific when I post. About to post another question that I can find no answers online and the manufacturer refuses to answer as they only directly support 2 specific processors. lol. Hack my own BIOS if I have to.
In answer to a question above, there is no budget and yet the budget is unlimited. These may go in an enclosure that hooks to the PCIe bus directly or via Thunderbolt to a blade system. It may be the same and hook up to a simple ROG GR8 II which is a pretty amazing machine for the $. They may be added to the botherboard of a Dell T7920 desktop with dual scalable processors or a simple 2-way E5-26xx v1/2/3/4 motherboard. The consensus of need is to mainly teach CUDA and do rendering. If we can do more with powerful CPU's, AWESOME.

In the mean time, give this Doc a quick overview?
1) Mining can be done on 1 lane why? Because it runs in a loop and only reports back when it has data?
2) What else runs well on 1x?
3) What applications are the most taxing on BOTH the CPU and GPU?
4) What 3d applications are not able to utilize a graphics card much at all?
5) What applications mainly rely on the maximum number of GPU cores?
6) How much does actively running a display slow down a GPU (should we consider PCioE given the use case stated above and have a portable thin client with multiple monitors that gets wheeled from lecture hall to lecture hall, demos, auditorium, dorm rooms, etc. as scheduled?)
 

Rogue Leader

It's a trap!
Moderator
1) yeah basically, Mining receives a load to process, processes it and then outputs it - in the simplest of terms. Remember the lanes are Input and output but not part of the processing. Games prefer lanes because the cpu and GPU are always communicating and feeding off eachother, bound to eachother to a point.
2) basically anything but a GPU. 1 PCIe lane is more than enough for most devices, except the most extreme of storage (Optane and NVMe) which can utilize up to 4 lanes.
3) Games really. Most anything else focuses on one or the other. Games are the only thing that have stuff like Game AI and tracking etc all reliant on the CPU, while the GPU is handling the graphics behind all that.
4) ALL 3D applications want a good graphics card. I don't know of anything thats reliant on the CPU only. Admittedly its NOT my area of expertise, but theres a reason these super expensive cards are out there.
5) 3d rendering, video rendering, data mining, CFD. Its a long list. take a look here: https://www.nvidia.com/en-us/data-center/gpu-accelerated-applications/catalog/
6) if the display is just showing 2D applications you can connect as many as it will allow its not going to slow it down. What will slow things down is displays showing 4k video or even more so displays rendering 3D. I'm not sure what you are asking there, what are you expecting this machine to do in terms of displaying?