Maybe there's two related features here:

1. use of reels to reduce DCP size when there is long still-image (or
   black) content

2. use of reels to reduce DCP size when there is repeated content

I was thinking more of 0000002, which could probably done silently (or almost
silently). I agree that the same solution for 0000001 would probably be a bad
idea.

I would still opt to have an option to actively enable it, as another reel segmentation option. Maybe even 'split by content' could be a side switch for this? I think users should have the ultimate choice
to control the reel structure. A bit hypothetical, but what if users want a strict 4GB reel segmentation due to storage media, but there is repeat content that is larger than 4GB?

The problem is, some exotic cases may need/want reel re-use AND absolute reel size limit? Well, maybe too exotic... But then it would be good if both would only work exclusively, so the users notices he
can only use either option.

• Carsten

The more I think about it, the more I think that this should be selectable under reel options. That would still allow users to force a single reel, to segment strictly by content, or size. Then, as a fourth option, allow to reuse repeated content. Also because, it will be a special case, only a few users will want to use that feature, and then I think it needs to go there in the GUI, as that is the only place to arrange reel structure. I think there will be no collisions with the other options in real life.

Should we reflect this in timeline view as well, I guess yes? Ah no, that only shows the formal reel sequence, not the actual asset references.

But, would that be an interesting option to show content/reel/asset references in timeline view? Would certainly be interesting for loaded DCPs to dig into their structure?

• Carsten

If I may add something to this, it would be creating a "block" to repeat (or a number of them) that would do the job with more efficiency:
a=ABCD
b=EFGH
Final DCP using a, a, b, b, a, a, b, b or another customizable choice, like n times a.
Even better, n+1/2 times a, creating an nx(ABCD)+AB (if AB has half the a duration). That would mean utilizing the "duration" feature on a reel of the CPL.

In a similar use, using a DCP or an essence of a DCP (audio or video) repeatedly in the CPL, as a "block" or as one of more blocks would do magic.

The CPL structure can be a very powerful tool for getting more content with less data usage.
The benefit would be not only on the data volume department, but also on the time demanded for the creation of some kind of DCPs.
Slideshow DCPs, test patterns, videos used as background on different occasions would not only be smaller, they would also be faster rendered into DCPs., needing less rendering and less hash calculation time.

Just to add my two-pence worth:

I can see the value of this feature, but I for one would like to be able to disable it.

I tend to follow the KISS principle. One CPL, one reel, no funny business, regardless of the content - to leave as little room as possible for running into a bug on some weird old projector which hasn't had a firmware update in a decade.

Of course there needs to be an option to activate this deliberately.

Trying to condense my opinions from my forum posts on this topic:

We either need to accept a considerable change/addition to the GUI to arrange this manually (e.g. an asset/reel based timeline editor), or, we could find a 'simple' way that reuses assets automatically after we enable that feature. I vote for the latter, as it probably means less work, and, as it is a feature most users won't use, it would be less irritating than offering another timeline editor.

If it should assemble CPLs reusing assets automatically, it has to be quite smart, though. But I think it's possible by either flagging content pieces, or by implementing/reusing another reel segmentation mode (e.g. 'single reel', 'split by content', 'split by size', 'reuse/reference').

• Carsten

Thinking about this some more, I think your idea of the separate reel mode might be the way to go. If you select that, DoM looks at what you've got and tries to repeat content wherever it can (prioritising video, I guess, then audio and subs).

A fully automated procedure, though, would call for criteria that could make a difference.
An example: One makes a DCP with the following pictures or videos:
ABCD ABCE ABCD ABCE (spaces are just for the sake of distinction)
What would constitute a block/reel?
ABC?
Two separate reels of ABCD and ABCE?
ABCD ABCE?
Any of the above would be greatly beneficial, of course, but having a saying on what constitutes a repetitive reel/block, may have more applications and be more efficient.
Or not.
Again, the opportunity for one to import three times the same video file in the content list to make a DCP and it resulting in one third the size of what it would otherwise be is superb!
Low capacity servers, network bandwidth, storage media, encoding and hashing duration would only benefit.