here are samples
The trial version warns about the logo only, it doesn’t mention any opaque background.
Logically, it would be pointless to release a trial version which affects the whole of the original picture in this way. That wouldn’t allow to really try the software.
I still do think that is an issue.
Can an owner of a licenced version check and confirm this?
Need the footage you used to do that.
The trial version tested on my clips does not affect the video like yours.
FYI:
I just tested also on a very old weird videocd capture that uses a custom matrix BVOP
No problem either.
I’ve (after days) got it to download one of the Chronos Fast models (chf-v3-fp16-1152x704-ox.tz)
and it’s previewed parts of a clip, but when you move to a different section of a clip and press preview again it says “downloading mod…” again.
Surely it shouldn’t be downloading the model again when it already has it (it must have it as it previewed earlier parts of the clip. I’ve not changed any setting since previewing the previous section so shouldn’t it be using the same model file (chf-v3-fp16-1152x704-ox.tz) that it already downloaded instead of wasting time and bandwidth by downloading again?
TaskManager shows it using 13.1 Mbps of the network (so it is downloading), still taking ages though. I did try setting input/output rules in Windows Firewall to allow things but that didn’t seem to make any difference so I removed the rules. With no Firewall rules it is downloading (when I see no reason for it to be downloading it again) but taking quite a while (but now not as long as before before it previews - before it just took ages then gave an error message. It still does for some models).
So can it please be changed to not re-download model files if it already has them?
The images were compared with and without license in CLI, and the watermark does not affect the tint.
I think it’s another bug that makes the colors look wrong.
The tendency of the bug is similar to the BT601 and BT709 conversion errors that have been occurring for some time.
It has been reported in the bug thread and will eventually be fixed.
Original
No Auth
Colorspace Mismatch(BT601,BT709 issue simulate) Red ribbon is yellowing.
I’m maybe wrong, but it seems the change in brightness (not color) is due to the logo of the trial version which is in transparancy on the video result, certainly the transparancy is not at zero %, if it’s that, this is not a bug or an issue, it’s normal.
It’s not good to say all these confusing things.
This is a poll to help see what might be the cause of the model download issues with this version of Topaz Video AI.
0 voters
0 voters
0 voters
0 voters
here: clip
everyone is welcome to try. settings I used:
Proteus enhancement, no stabilization, no motion deblur, H265 Main10 (Intel) encoder, mp4 container
I found that when converting a DVD 25fps to 4k 60fps (Chronos Fast + Proteus), with the Replace Duplicate Frames function turned on, the final video has small jumps, jerks, and distortions that are unpleasant to the eye. Without this function, the video seems to be smoother…
This is a matter of course. The function of replacing duplicate frames is to insert more frames, which of course is more prone to problems.
I notice thats not the clip that you posted originally.
It seems to me that the ratio: size-quality-ease of processing, it is best to put H265 Main10 (NVIDIA) MKV on the input. Is it so?
You need to compare that in a video player. I use Media Player Classic.
Note that Video AI will always display the color wrong both left and right picture.
That is why you don’t notice the issue.
it is a different clip but it is affected in the same way.
the other clip I prefer not to share for personal reasons.
Have you tried other non-intel encoders?
I am new to the program. Is there a detailed description of the functions of different AI models somewhere? I try the error method, and I have to constantly rewrite the video. Efficiency is extremely low. For two months of work, I have only a couple of clips ready. In other cases, errors constantly appear in different places, due to the incorrect use of models.
[image]
yes, I have the issue here in MPC-BE, need to do some test, but i think it’s maybe related to the Mjpeg format you have → intel Encoder in H265 Main10, i’ll test other combinaison, hold on , and will re encode your original video too.
You could try decreasing the sensitivity value (for the duplicate frame detection) to see if that helps (in case it’s skipping frames that aren’t really duplicate ones).
eg. the default value seems to be 10, and that might think some frames which are very similar but not identical are the same. It probably depends how much compression (or maybe picture noise) is in the frame. If there’s compression artefacts that differ between frames but the frames are really the same then setting sensitivity too low won’t be able to detect that they’re the same.