-
Technique advice please on a project with 10,000 subtitles
On this video
https://vimeo.com/180135440
you can see a subtitle sequence showing the position and altitude.This data comes from a GPS (Oziexplorer app) which logs to a track log in textual format. Then I got a friend to write a little prog in C to convert this into SRT. This is then loaded into Pro 13 via the Vegasaur route, and it uses the standard Sony Text FX to generate the subtitles. At the time of loading into Pro, the text font etc is specified using a custom template in the FX config. The position of the text is specified (when all the titles are loaded) using the track motion feature (the subtitles are loaded onto a separate track, obviously).
All this works…
The problem is that I get a subtitle every 3 seconds and for a 7hr flight that is 8400 subtitles 🙂 Ideally I would like it every 1 second.
Vegas becomes totally unusable with this lot loaded on the timeline. It does still work, but
– any operation hangs the program afterwards for about a minute (while Vegas is reallocating 8400 memory blocks, etc)
– it is rather fragile, and about the only safe operation is to just render the job and exit Vegas
– the Vegasaur plug-in works only once; you have to exit Vegas after the subtitle loadSo I do a render of the two tracks (there is usually no audio), to an intermediate mp4, set to a high quality of 50mbits/sec, and import that for the next stage.
The input project is 1080P 50FPS, 25mbits/sec from the camera (Sony FDR-1000V). The output is 1080P also, with the final result, heavily edited to remove boring footage, being 25mbits/sec.
While doing the intermediate render I also do
– contrast up a little (sony fx)
– a little bit of colour correction (sony fx)
– NewblueFX lens correctionThe problem is that this intermediate render takes about 100hrs! My PC cannot be used for certain things during that time (as posted in the other thread; any video related ops corrupt the rendering).
Is there any way around this?
One way I can see is to render the subtitles to a single video file and drop that onto the timeline. But in Vegas that will still take best part of 100hrs. (Maybe a lot less because one isn’t doing any FX on it, and the output file could be low res, since it is just the line of text). Or one could render it externally with e.g. Handbrake https://forum.handbrake.fr/viewtopic.php?f=11&t=35164&sid=7f069f95c2e9522a042a5dbfea60800a using the enhanced SRT method (I could get my C program changed to do that) and that could be run on some old PC. And there are things like ffmpeg (?) which can convert text into a video track.
If I rendered only the final-cut version, the render would be maybe 5hrs only. The basic issue with this is that Vegas is not usable for any editing with the subtitles loaded… also I am not sure if I can crossfade multiple tracks (probably can).
Some contents or functionalities here are not available due to your cookie preferences!This happens because the functionality/content marked as “Vimeo framework” uses cookies that you choosed to keep disabled. In order to view this content or use this functionality, please enable cookies: click here to open your cookie preferences.