Apparently, the team over at NVIDIA decided to spend some of my money on a graphics card upgrade. I’m now saving up to buy the NVIDIA Geforce RTX 3090 BFGPU. This card looked intense in the demo based on the sheer size of it and the design aesthetic. I do really like the Founders Editions of the NVIDIA graphics cards. My GTX 1060 is still running right now after doing years of work. I supposed the next piece of the puzzle with this pending purchase is planning enough machine learning on my local machine to make use of the 10,496 NVIDIA CUDA cores this card is rocking. You could buy 3 of the Geforce RTX 3070 cards for that price, but housing all 3 of them in the case would be a lot more work. A nice alternative might be running two of the Geforce RTX 3070 cards which would net you double 5,888 Cuda cores or 11,776 total Cuda cores which is obviously more than the 10,496 cores the RTX 3090 card contains. 

https://www.nvidia.com/en-us/geforce/graphics-cards/30-series/rtx-3090/