You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is your feature request related to a problem? Please describe.
Not a problem, but a nifty feature for us with many cpu cores (8 or more)
Describe the solution you'd like
batch upscale videos, it interpolates say 2 at a time. RTX upscale solution only uses 50% gpu usage. Don't know if that's just max output of the tensor cores or the RTX cores, which equals 50% cuda load though.
but simultaneously encode the upscaled frames per cpu thread all at once once all of them has been upscaled individually.
In handbrake, i convert movies to AV1 for tiny 1GB sized outputs, and it's done by opening one handbrake instance per movie, 7 at a time, and set affinity for each cpu thread. So i have say 6 cores spare, so i set each handbrake for each their cpu thread to give the best quality possible per movie, all running at once.
Describe alternatives you've considered
A frame assembler/photoshop/premiere pro, but same principle. 7 instanes, each given 1 thread each, and export the upscaled png sequence in each, and combinding all at the same time.
Additional context
Add any other context or screenshots about the feature request here.
The text was updated successfully, but these errors were encountered:
Is your feature request related to a problem? Please describe.
Not a problem, but a nifty feature for us with many cpu cores (8 or more)
Describe the solution you'd like
batch upscale videos, it interpolates say 2 at a time. RTX upscale solution only uses 50% gpu usage. Don't know if that's just max output of the tensor cores or the RTX cores, which equals 50% cuda load though.
but simultaneously encode the upscaled frames per cpu thread all at once once all of them has been upscaled individually.
In handbrake, i convert movies to AV1 for tiny 1GB sized outputs, and it's done by opening one handbrake instance per movie, 7 at a time, and set affinity for each cpu thread. So i have say 6 cores spare, so i set each handbrake for each their cpu thread to give the best quality possible per movie, all running at once.
Describe alternatives you've considered
A frame assembler/photoshop/premiere pro, but same principle. 7 instanes, each given 1 thread each, and export the upscaled png sequence in each, and combinding all at the same time.
Additional context
Add any other context or screenshots about the feature request here.
The text was updated successfully, but these errors were encountered: