Here is the image:

• https://medium.com/@ageitgey/try-deep-learning-in-python-now-with-a-fully-pre-configured-vm-1d97d4c3e9b

It’s a a fully configured deep learning VM. It’s about 7.7GB.

I am running on:

• an i7 3770 quad core at 3.4 Ghz (not the i7 3770K which has no VT-d virtualization extensions – this one does.)

• 16 GB of DDR 3 ram that is clocked at 1600 mhz. I benchmarked it at 19000 MB/sec in Novabench.

• The GPU in the host is 1060 GPU with 6 GB of RAM – but the image file I linked above explains “Due to licensing and installation complications, there’s no GPU acceleration / CUDA support provided. So you don’t need an Nvidia GPU to try this out, but it also won’t take advantage of a GPU if you have one.”

• windows 10.

• I have an SSD. It’s not the fastest but the write speed is 214 MB/seconds and the read speed is 253 MB/seconds in Novabench.

• I turned the swap off entirely on the host system and have plenty of RAM left.

Without anything else running on the host computer, which is a pretty fresh install, this VM doesn’t feel native at all when entering full screen, regardless if I give it 2 GB (default), 4 GB, or even 8 GB of memory. I turned on the virtualization extensions in my bios and in the vmware settings too! I went from giving 1 CPU to 2 CPU’s to the guest. None of this helps.

why deson’t it feel “native”? I mean things like moving the mouse around, opening a new firefox windows, etc. It’s “obviously” a VM.

Is it not supposed to feel native or buttery smooth? I’m just at a loss. Additionally, I did an update from within it and it updated Ubuntu files at 500 KB/second! My connection is several megabytes per second.

I feel like I must be missing something….but what??