Other than the coolness factor do you really need to run more than 13 GPU's off of the same motherboard? I know there is economics of scale, but these motherboards cost more as well, offsetting the savings gained. Plus you need extra PSUs, more memory and a bigger SSD. So adding all this up the cost savings really isn't a lot more than say 2 smaller rigs.
Also, a point that has been argued before, if the rig experiences issues, then you have 19, 21, or however many GPUs you manage to get running down at once versus with smaller rigs you would have maybe 8 GPUs down while the others would remain running. While these videos are nice to watch for the geek factor, it seems to be pointless after a certain limit.
6 rigs are suggestable to use in the mining rig right! One rig can give 6 card pci riser I think. How it is possible that we can run a 12 GPU cards in the one processors with the windows os.
Since it has the riser limit of one PCI slot with the 6 cables are only gives 6 cards options. Please explain me clearly.
The newer motherboards support more than 6 PCIe slots, and indeed they already have board configurations with 8, 12, 13, and higher PCIe slots which is what this post is about. Also, Windows and AMD/Nvidia drivers have recently been modified to allow this many GPUs to run, where in the past there were software limitations.
My point is while it is now possible to run this many GPUs, at which point does it still make sense to do so. The motherboards typically cost more and from what I read you also need to increase the physical memory and SSD sizes to accommodate the increased memory/swap file requirements of supporting so many GPUs. It is possible you may even need to use a more robust processor that the typical Celeron often used and you definitely need extra PSUs to power everything. So when you add up all these costs as well as the extra headaches of trying to get it all working, is it really better to run a 19 GPU rig versus 2 or 3 smaller rigs?