Everything’s running smooth on my i7 10th Gen +2060 SUPER setup. There are some growing pains, of course, like some of the tooltips not appearing or some slowdown when switching between options, but all in all it’s a good (re)start.
The only thing I don’t really like is the new trimming tool. It’s a bit hard to use, especially when used on movie-length footage (I’m talking 1hr+, 2hr+!)
The carried-over ability to input timecodes for trimming is nice, but I’d much rather use the built-in 2.6.4 way of clickable buttons that place start/end markers on the video timeline for trims. Those were especially handy for projects that required me to use multiple small pieces of footage from one long video file.
Ida, this issue was written on the EA and beta , but not here on the 3.0.0 ! maybe best would be to add it to the list for the ones who didn’t tested the beta or EA before ?
Can you please make sure you are using the same encoder and settings? Also, please double-check your Preferences and set your memory usage accordingly.
For the time being, you will want to preview each model in order to download them for offline use. We will be adding a model download tool that will make this a much faster and more seamless exp.
Can you please share your logs for the first issue?
All of the same models are in this app, however, the UI is different here. Which model are you looking for?
Yes, the enhance model is used to Upscale. The lock is letting you know that the enhance model is being applied and that you cannot upscale without an enhance model chosen, however, you can change the enhance model selection by clicking on the drop down menu.
If this is only occurring with a specific video, can you also upload the video file? You can securely submit your files(s) to our Dropbox using the link below. Please be sure to send me a note to let me know you sent something.
Will there be any plan to support TensorRT for Frame Interpolatin models?
I’m using RTX 3090 and now processing with Proteus got much faster since TensorRT being supported.
On the othe hand, Chronos(including fast)/Apollo felt extremely slow.
It doesn’t seem like these models are utilizing my CPU and GPU as much as Proteus does.
But quality-wise these models are way better than 2.6.4 models as far as I’ve tested. I hope TensorRT being implemented into these models too.
Yes. I notice that it still behaves like this when switching to replay a preview that has already been rendered. You would imagine that it would just be playing the (preview) output at that point.
The other odd thing I noticed is that it doesn’t get out of synch when doing a Export.
I also noticed that it happens when changing from one preview to another. The left image is often from the previous preview, while the right side is the output from the currently selected one.
Some awesome things that I just found: Apollo model is really fast compared to Chronos Slomo.
Secondly, the missing x5 speed can be entered into the model with the command line and it seems to work correctly.
