Does amfeltec splitter/cluster need high end MB? Tweet

[jfharper]

So the splitter is 4x to 16x...I have to look what the cluster is.

Would the splitter/cluster work on a regular MB designed with lower number of pcie slots and a 28 lane cpu?

For example, my MB handles 2 pcie slots at 8x each where I have two 980 ti's installed.. If I removed, one gpu, left the other in the MB for the display, added the splitter/cluster with 4 cards, and plugged that into the other pcie mb slot, the splitter/cluster would be at 4x.

But what I don't understand, the splitter runs 4 at 4x, so is that 16 lanes or 4? I read there is a drop in performance for the cards on the splitter/cluster but people use them on high-end mb's.

Just wondering if running a splitter now would help, then when I expand later, getting high end mn, could the splitter be used then as well...since they are a $250-$450 investment.

[mr_seq]

I've been having the same question for a year now. Only stumbled upon this yesterday:

Scene compilation is done to 100% on the CPU - the GPUs are for rendering only. That means, PCIe speeds are not relevant here. Of course, the time until the first image appears includes the, scene upload time, actual render time for the first sample/pixel and the tone map time. Depending on the scene and resolution you can ignore upload (how much data needs to be transferred to the GPUs?), rendering (how fast does it render?) and tone mapping (how big is the image you are rendering?) or you cannot.

So I think it's indeed (compressed?) into 4 lanes, but it's only important to send and receive the scene. Drop in performance would be more noticeable when e.g. gaming.
A high end mobo looks unnecessary if you look at all the GPU mining vids out there..

I found this very useful: https://youtu.be/M5z53A5RW7I?t=1m12s
The Cubix expander is basically a more robust way, but comes at a price.
If you're looking at something less expensive and with only 3 GPU's: https://mineshop.eu/cables/pcie-adapter ... ots-detail

Can you keep us updated? Very interested in this since my MacPro 5,1 is maxed out and I'm building an pci-e extender myself.

[JoshuaReeves]

Well, personally, I can’t see any reason to put large numbers of GPUs on one system. It creates a much more costly single point of failure. MB, memory, cpu, PSU, or storage failure takes down 2 dozen GPUs? One GPU has hw issues and it takes down the other 23? No thanks. After hash rate, THE most important attribute of a mining rig is stability/uptime. A rig that large is highly unlikely to exhibit that vital quality. In my opinion, you’re trying to create a solution to a problem that doesn’t exist. The system you’re looking to build will become very complicated, as you try to configure multiple PSUs to power all the cards, etc. Also, even if you had a driver that could handle 24 GPUs, you’d have to have a multi-core/multiple thread CPU to handle the driver concurrency necessary to efficiently drive that many. I’m sure it’s an interesting problem to think about, but I don’t see it as the least bit practical or desirable in a real-world production mining operation. Of course, that’s just my take on it. I have a medium sized farm, and I wouldn’t touch a rig like that. I wouldn’t even build an 8 GPU rig. The only 8 GPU rig I’ve seen that I’d consider is the PandaMiner, since it’s a integrated design that was intended to run 8 GPUs reliably from the get-go.