So I'd like to dip my toes into the world of machine learning. I have most of STH's Starter ML build already as my home server. Except for the GPU, and also, it's in a 2U rackmount chassis. Specifically, the Supermicro SC825TQ-536LPB; the "LPB" part of the model name stands for "low-profile bracket", i.e. it only accepts half-height PCI cards. It looks like GTX1060s don't come in half-height, so won't fit in my chassis naturally.
Just wondering if anyone has any thoughts on the cheapest way to get a GTX 1060 (or better) up and running with minimal cost. I'd rather not buy a new chassis. Or maybe drop down to a GTX 1050 for now, which apparently do come in half-height form factors?
I also haven't given any thought to cooling. I can't remember the last time I had a dedicated graphics card. I've done only headless servers and integrated graphics systems for years now!
Just wondering if anyone has any thoughts on the cheapest way to get a GTX 1060 (or better) up and running with minimal cost. I'd rather not buy a new chassis. Or maybe drop down to a GTX 1050 for now, which apparently do come in half-height form factors?
I also haven't given any thought to cooling. I can't remember the last time I had a dedicated graphics card. I've done only headless servers and integrated graphics systems for years now!