Hi,

 

After having failed to setup passthrough with a GTX 970 and a GTX 1060, I've been finally able to have it working using a GTX 1660 Ti !

 

I was expecting some performance loss, but I don't know if the numbers I get are good or if there is something to improve.

 

The average 1660 Ti 3d result of passmark tests are around 11500 points. I reach around 9500 points, almost 20% less. Is this the range of performance loss that we get with virtualized gaming ? I'm using an i5-8400 (passing 5 of 6 cores to the VM) with 8 GB of ram.

 

Thanks !

Edited January 17, 20206 yr by delivereath

Have you isolated the cores assigned to the VM? (i.e. nothing else can use them)

What PCI-X slot is the card in? (i.e. verify that it's running in 16x mode)

 

I'm running a Win10 VM with a GTX 970 and performance has suffered in 3D games due to the PCI-X slot in use (16x but wired 4x).

Hi,

 

I'm running RTX 2060 Super and in 3DMark Time Spy the GPU performance tests is no difference to bare metal (default settings). Altough there is a CPU test as well and it's there where is a loss, and it depends on you VM configuration (how many cores you're passing and etc.) . 

 

Games normally got more performance downgrade because they're more depend on CPU & memory subsystems, but pure graphic benchmarks are mostly showing a bare metal performance.