tbirdparis
Member
Hi all,
I just wanted to run my conversion/transcoding workflow by the experts (ie, you people), to see if I'm compromising the quality of my video or not.
Am using a GH13 in PAL land (so, 50p). Sometimes I'm shooting 1080, in which case I just use Voltaic (on Mac) with custom settings to convert to ProRes HQ, size and frame rate set to compressor native, no gamma correction and with the export as 24p (reverse telecine) option enabled. For the 1080 stuff, this works fine.
More often though, I shoot stuff at 720p so I have the option of either using the footage at normal speed or at 50fps played at 25fps, ie, real in-camera 50% slow mo. My workflow for this is that I convert all my MTS files with Voltaic first (to turn them into ProRes HQ), which gives me transcoded QT ProRes HQ clips at 50fps. So far so good. Then, I've set up custom settings in Apple Compressor to do two things. One is to convert the 50fps clips to 25 by dropping every second frame which gives me a more film-like motion that I like, while keeping the speed the same as the original. The other setting is to convert the clip to play all 50 frames but at 25fps, so I get clips that are (obviously) twice as long and with smooth, non-faked slow mo. It all seems to work fine.
My question is.. By converting first in Voltaic to ProRes HQ, and then doing a second step in Compressor, am I transcoding twice? Ie, is compressor re-transcoding the data a second time when it exports, which would mean I'm losing some quality and adding more blech to the data? Ideally, all I want compressor to do is take the video files that are _already_ in ProRes HQ format, and in one case just throw away every second frame, and in the other case, just tell the video file that it's now to be played at 25fps. In effect that's what it's doing, but I'm worried that it's not just doing that.. but that it's also re-transcoding the data a second time which (I presume) is messing with the quality. Is this likely to be the case?
If anyone has any comments to make on this workflow, I'd appreciate it. I know that FCPx (which I use) now allows you to directly ingest MTS files, which is all good and fine. But I like being able to prep my 720p stuff as I want it before going anywhere near FCP. I prefer having a nicely organized folder of QT ProRes HQ video files, with a 25fps (dropped frame normal speed) version and a 25fps (50p playing at 25 for real slow mo), so I can then choose what I want to take into FCP.
Basically it would be good to know if Compressor is intelligent enough to "know" not to drag a video file through transcoding calculations again when the file is _already_ in the destination format. If all that it does is keep the original data, and just change the frame rate as per my description above, then I guess there's no problem here. Any ideas?
Cheers!
I just wanted to run my conversion/transcoding workflow by the experts (ie, you people), to see if I'm compromising the quality of my video or not.
Am using a GH13 in PAL land (so, 50p). Sometimes I'm shooting 1080, in which case I just use Voltaic (on Mac) with custom settings to convert to ProRes HQ, size and frame rate set to compressor native, no gamma correction and with the export as 24p (reverse telecine) option enabled. For the 1080 stuff, this works fine.
More often though, I shoot stuff at 720p so I have the option of either using the footage at normal speed or at 50fps played at 25fps, ie, real in-camera 50% slow mo. My workflow for this is that I convert all my MTS files with Voltaic first (to turn them into ProRes HQ), which gives me transcoded QT ProRes HQ clips at 50fps. So far so good. Then, I've set up custom settings in Apple Compressor to do two things. One is to convert the 50fps clips to 25 by dropping every second frame which gives me a more film-like motion that I like, while keeping the speed the same as the original. The other setting is to convert the clip to play all 50 frames but at 25fps, so I get clips that are (obviously) twice as long and with smooth, non-faked slow mo. It all seems to work fine.
My question is.. By converting first in Voltaic to ProRes HQ, and then doing a second step in Compressor, am I transcoding twice? Ie, is compressor re-transcoding the data a second time when it exports, which would mean I'm losing some quality and adding more blech to the data? Ideally, all I want compressor to do is take the video files that are _already_ in ProRes HQ format, and in one case just throw away every second frame, and in the other case, just tell the video file that it's now to be played at 25fps. In effect that's what it's doing, but I'm worried that it's not just doing that.. but that it's also re-transcoding the data a second time which (I presume) is messing with the quality. Is this likely to be the case?
If anyone has any comments to make on this workflow, I'd appreciate it. I know that FCPx (which I use) now allows you to directly ingest MTS files, which is all good and fine. But I like being able to prep my 720p stuff as I want it before going anywhere near FCP. I prefer having a nicely organized folder of QT ProRes HQ video files, with a 25fps (dropped frame normal speed) version and a 25fps (50p playing at 25 for real slow mo), so I can then choose what I want to take into FCP.
Basically it would be good to know if Compressor is intelligent enough to "know" not to drag a video file through transcoding calculations again when the file is _already_ in the destination format. If all that it does is keep the original data, and just change the frame rate as per my description above, then I guess there's no problem here. Any ideas?
Cheers!