Arri D-21 vs. Red

[M Most]

That said, it sure seemed easier to get nice skin tones from the D-21 footy than the Red footy in the film out test.

I have dealt with both (I think Blair may be referring to me as one of his sources) and can confirm what Blair is saying. The LogCFilm matrices available from Arri are specifically tuned to yield images that work properly in Cineon log format when viewed through a typical print preview LUT, both in terms of their absolute values and the colorimetry they provide. In other words, while Red is basically targeted towards electronic display, the D21 when processed using LogCFilm matrices is targeted towards film output, and the colors and values are appropriate for that format. In my opinion, this is one of the primary reasons that Red has underperformed in "standardized" film targeted DI tests, and Arri has excelled in those same tests. This is not to say that one can do things the other can't, it is, however, to say that each manufacturer has made choices as to where to direct their efforts regarding intended output targets. Perhaps (and hopefully) the upcoming RedFilm transforms will help to address this. Even more hopefully, the Academy IIF proposal will be universally supported and basically eliminate the problem.

 

[M Most]

I was referring to HDCAM SR in my post, which I know has a better color space, but is still compressed. And it's yesterday's workflow. I fedex a tape to the network; they put it on a sat feed to the locals, who tape it for broadcast. I should just post a file they can all access, in a better codec.

And that file you post will almost certainly be far more compressed than HDCam SR.

I'm not saying that tape is the end all and be all, or that file based deliveries won't be a good thing (they will), but the advantages of a tape delivery - a physical storage element that is not likely to be lost and cannot be "accidentally deleted," as well as very wide compatibility, and good security, among other things - are quite real at the moment, especially at the network level. And quality, especially when using SR (about the closest you can get to being effectively uncompressed) becomes a non-issue.

 

[Blair S. Paulsen]

Mike is indeed one of the two known colorists I was referencing.

Based on the ASC/PGA CAS and the BSC tests I think its safe to say that stipulating a film out and DPX log format creates meaningful but inherently skewed impressions. Over the next decade the percentage of film prints shown in theatrical settings vs DCI spec digital projection is likely to shift considerably. The Academy IIF standard seems like an excellent way to take many of the variables out of the equation, and to whatever degree it is possible, future proof your artistic intent.

Cheers - #19

 

[James Brundige]

FWIW the level of compression in an SR master is so light compared to the MPEG-2 crunching the distributor is going to do I'm not sure it matters much. That said, a file based delivery option is overdue. My assumption is that possession of physical media in the form of an SR tape is considered a solid job security move by those who create deliverable specs. I would wager that over the next few years the nets will become more sanguine about multiple copies of their assets living on servers in an SRish codec as an alternative to the security blanket of a tape they can hold in their hand, but I'm not holding my breath...

Cheers - #19

Its a bummer that our masters are getting better and better, and the multiplexed distribution is getting worse and worse. Though, from what I read here, theater projection is pretty low spec too. Time for 5 K projection!