Lossless codec, preview window, disable audio (IW3)

[RudiRastelli]

Hi Mr Nagadomi !

Would be nice if we could choose a lossless codec like FFV1 to avoid to many re-encodings.
I prefer to use 'handbrake' to make my final encodings.

Another cool feature would be to have a preview window to watch the effects when we change some settings.
It must not play the video... just move a slider to select the time and it only shows an anaglyph picture for example.
(to say the truth i would prefer an interlaced output because if have a passive 3D monitor)

and last... a checkbox to select if output should have audio or not.
I use 'mkvtoolnix' to add the audio from the original video to the 3D video.

Regards
Rudi

[nagadomi]

Loss less codec is already supported. Video Format=avi, Video Codec=utvideo

For preview, it would be difficult to see it without actually seeing it in VR.
See
https://github.com/nagadomi/nunif/tree/master/iw3#recommended-workflow-for-slow-processor-or-very-large-video
for output options for preview. (In GUI, you can specify Max FPS=0.5)
Also, dev branch supports Stereo Format=Cross Eyed you can quick check without anaglyph glasses.

For audio, ideally all audio channels and subtitles should be exported. However, it has not yet been supported.
audio is basically copied without re-encoding, there is very little processing time to adding audio. If it is not needed, simply delete it.

[RudiRastelli]

Yep... i know IW3-desktop and it's fantastic for people having a powerful rig and a VR headset.
Even if you owe all this and ure willing to keep putting the VR glasses on and off for fine-tuning, it will be almost impossible to find the perfect setting because our eyes are able to correct certain inaccuracies. This is not too bad at first, but if your eyes have to do this for hours, it will stress them and your brain as well.

Take the screenshot above... it looks perfect in 3D because our eyes are compensating that the main characters are not perfectly in focus. For me this means that in a 3D view I can't decide if something is not quite right.
In contrast, in an interlaced view you can see it immediately, because your eyes can't fool you.

I can only recommend watching it yourself with “Bino 3D Player” on your 2D display. Throw in an already rendered 3D video, pause it at a scene with good depth and inspect it.

As an addition to my post above:
I wish IW3 would have the following workflow...

• Select the input video and output folder
• Select whatever frame to preview (maybe the frame at start time setting)
• Find my setting for that frame
• Hit START to let do IW3 it's magic

[nagadomi]

iw3 does not support keyframing or transitions for settings, so there is nothing you can do with a detailed preview like the one above.

[RudiRastelli]

For audio, ideally all audio channels and subtitles should be exported. However, it has not yet been supported. audio is basically copied without re-encoding, there is very little processing time to adding audio. If it is not needed, simply delete it.

Will using "--vf an" completely remove audio from the output ?

My usual workflow is to use MKVtoolnix to add audio-tracks, subtitles and chapters from the original video after 3D conversion is done. Every time i have to deselect the audio track from the IW3 output... it's a bit annoying.

[nagadomi]

No, audio is added by the program and the vf option only applies to the video stream, so the vf option cannot control audio.
Even if an option to remove audio is added, you’ll still need to manually add audio and subtitle tracks, so I don’t think it makes much of a difference.

[RudiRastelli]

The vf option cannot control audio.

And again i've learned something new... thx

Have you ever thought about integrating audio-/subtitle-/chapter-processing into IW3?
It would be great to specify which tracks are copied into the 3D video.