This is technically a feature request which I should lodge elsewhere (and will do), but I wanted to run a suggestion up the flag pole and see if anyone else saluted.
I think it would be very useful if the program was able to split an existing DCP into 'reels' of a certain length (approximate). I think the 'cut' point may have to be specified manually or performed by the program if it can detect cuts between scenes. There may or may not have to be overlaps as with conventional film reels.
The reason this may be useful is to do with subtitling. I am basing this suggestion on feedback I've had, and before I lodge as a formal feature request, I need to understand better why this is being suggested to me. By posting here first, maybe there is someone who can short cut this process?
Splitting a DCP of a feature into reels
-
- Posts: 2804
- Joined: Tue Apr 15, 2014 9:11 pm
- Location: Germany
Re: Splitting a DCP of a feature into reels
The reason is simply that most cinema projectors need subtitles to be split in reels as well, because they lack sufficient memory to store the whole subtitle track in one piece. It is not an absolute must, but if you try to include cinecanvas or SMPTE subtitles within a single-reel full length feature, you will most probably experience lost subtitles in a lot of locations.
Technically, projector-rendered subtitles, as they were originally conceived, are still not a solid feature, at least not in DLP systems. If you browse the net, you will find a lot of issues with them disappearing or not working at all. This function is built into the projector heads not by the individual cinema projector manufacturers, but by Texas Instruments, originally under the name 'Cine Canvas'. However, one can say that they did not implement it the most robust way. If you prepare subtitles, you have to tune them in different ways to increase the probability for them to appear correctly (e.g. FontCompressor).
If you want to be absolutely sure about when, where and how they appear on screen, you need to 'burn' them into the video file - which currently is the only method supported by DCP-o-matic. That means of course, if you also want to distribute a non-subtitled version as well, you need to create two complete DCPs with audio and video track files. Obviously, then it's not necessary to split it into reels.
- Carsten
Technically, projector-rendered subtitles, as they were originally conceived, are still not a solid feature, at least not in DLP systems. If you browse the net, you will find a lot of issues with them disappearing or not working at all. This function is built into the projector heads not by the individual cinema projector manufacturers, but by Texas Instruments, originally under the name 'Cine Canvas'. However, one can say that they did not implement it the most robust way. If you prepare subtitles, you have to tune them in different ways to increase the probability for them to appear correctly (e.g. FontCompressor).
If you want to be absolutely sure about when, where and how they appear on screen, you need to 'burn' them into the video file - which currently is the only method supported by DCP-o-matic. That means of course, if you also want to distribute a non-subtitled version as well, you need to create two complete DCPs with audio and video track files. Obviously, then it's not necessary to split it into reels.
- Carsten
-
- Site Admin
- Posts: 2548
- Joined: Thu Nov 14, 2013 2:53 pm
Re: Splitting a DCP of a feature into reels
Yes, it's on the to-do list. As far as I am aware DCP reels are spliced together with frame and sample accuracy, so you can split them anywhere you like (except, I think, when a subtitle is on-screen).I think it would be very useful if the program was able to split an existing DCP into 'reels' of a certain length (approximate). I think the 'cut' point may have to be specified manually or performed by the program if it can detect cuts between scenes. There may or may not have to be overlaps as with conventional film reels.
That's interesting, I didn't know that, thanks Carsten!The reason is simply that most cinema projectors need subtitles to be split in reels as well, because they lack sufficient memory to store the whole subtitle track in one piece. It is not an absolute must, but if you try to include cinecanvas or SMPTE subtitles within a single-reel full length feature, you will most probably experience lost subtitles in a lot of locations.
Best
Carl
-
- Posts: 2804
- Joined: Tue Apr 15, 2014 9:11 pm
- Location: Germany
Re: Splitting a DCP of a feature into reels
http://www.dlp.com/downloads/pdf_dlp_ci ... _Rev_C.pdf
http://www.dlp.com/downloads/pdf_dlp_ci ... _rev_a.pdf
There is also a reoccurring issue with IMB systems supporting 2k and 4k content in 2k projectors when showing 4k content with subtitles. The scaling reference may be wrong, so that titles get positioned based on 4k, but are finally shown in 2k, thus the titles are not visible.
Then 3D titles are coming up as well now...
It's still quite a mess, and so, prerendered subtitles are not such a bad idea after all.
- Carsten
http://www.dlp.com/downloads/pdf_dlp_ci ... _rev_a.pdf
There is also a reoccurring issue with IMB systems supporting 2k and 4k content in 2k projectors when showing 4k content with subtitles. The scaling reference may be wrong, so that titles get positioned based on 4k, but are finally shown in 2k, thus the titles are not visible.
Then 3D titles are coming up as well now...
It's still quite a mess, and so, prerendered subtitles are not such a bad idea after all.
- Carsten
-
- Posts: 72
- Joined: Mon Jul 28, 2014 9:37 am
Re: Splitting a DCP of a feature into reels
Hi, any news of this. I want to recreate a reel of an existing DCP which has everything split into reels.
Ideally I would like to just recreate a MXF for one reel (which has had a flash frame that has been fixed). I have the TIFFs for the reel. Can I recreate XMF and pop in into DCP structure?
Alternatively I have all the assets for each reel and can recreate the whole thing, is this possible?
Ben
Ideally I would like to just recreate a MXF for one reel (which has had a flash frame that has been fixed). I have the TIFFs for the reel. Can I recreate XMF and pop in into DCP structure?
Alternatively I have all the assets for each reel and can recreate the whole thing, is this possible?
Ben
-
- Site Admin
- Posts: 2548
- Joined: Thu Nov 14, 2013 2:53 pm
Re: Splitting a DCP of a feature into reels
2.4.11 has basic support for creating DCPs in reels, but that doesn't quite get to what you want to do. Currently you can only make a VF referencing the whole DCP, not just parts of it. I'll think about how we could do that: maybe an option to split DCP content into constituent reels.
You should be able to recreate the whole DCP, though... just put the assets in to DCP-o-matic as normal. If you added the TIFFs for each reel in separate steps ("add folder" each time) you should then be able to make a DCP with reels by choosing "split by video content" for reel type in the DCP panel. This is all quite new though so you may need a following wind. Let me know how you get on if you try it!
You should be able to recreate the whole DCP, though... just put the assets in to DCP-o-matic as normal. If you added the TIFFs for each reel in separate steps ("add folder" each time) you should then be able to make a DCP with reels by choosing "split by video content" for reel type in the DCP panel. This is all quite new though so you may need a following wind. Let me know how you get on if you try it!