examine content fails

[carl]

There is some MXF support so may be worth a try.

[Carsten]

Hmm. Weird issues. Must be a bit disappointing to run into all this with your first DCP-o-matic steps. I usually recommend not to go the single image/split audio route. Yes, I understand it may be attractive if you come from something more complex than rec.709, h.264, etc,. However, people run into all sorts of problems, very often not related to DCP-o-matic.

Most DOM users do not have to fight so hard to get their first DCP done. In most cases, it's just a matter of a few clicks and some patience.

Now, may not sound appealing to you, but why not create a DCP from a common interleaved rec.709 format, so you have something to preview, then try again for the sake of the larger color gamut.

- Carsten

[dermot shane]

hy not create a DCP from a common interleaved rec.709 format, so you have something to preview, then try again for the sake of the larger color gamut.

if i cannot make a high quality DCP from either the DNxHR444 or Jpeg2000 sequence then i think the best option is to pay to get a professional on th case!

fortuantly it only takes around 45 min to make a DCP, so the "try, render, cry" cycle is not too bad....

the audio wraped in a Qt is 6 frames longer 2pop-to-2pop, maybe it is not respecting the 24frame export value, and instead refrenceing the master timline's 23.98 ?

will try MXF next

[dermot shane]

i re-rexported the 5.1's as a Qt, and as MXF-op-atom x6 tracks

for reasons i do not understand the length of the Qt now matches the length of the Jpeg2000 sequence, so i'm running with that, see how the next screening goes tomorrow

huge thanks!

d

[Carsten]

if i cannot make a high quality DCP from either the DNxHR444 or Jpeg2000 sequence then i think the best option is to pay to get a professional on th case!

I am not talking about your capabilities, but the issues you are experiencing beyond your control - e.g. bad/questionable Resolve audio export, Mac NTFS file count restrictions, etc.

DNxHR is still fairly new and it seems FFMPEG support for it is experimental unfortunately.

Remains the question wether the XYZ J2Ks from resolve use a DCI compliant profile - if not, the DCI server/decoder may crash or do other weird things. Yes, there is an EasyDCP backend for Resolve, but does the normal J2K export actually use DCI profiles?


- Carsten

[dermot shane]

Remains the question wether the XYZ J2Ks from resolve use a DCI compliant profile - if not, the DCI server/decoder may crash or do other weird things. Yes, there is an EasyDCP backend for Resolve, but does the normal J2K export actually use DCI profiles?

it seems to, there was no issues with the DCP playback, and the color was accurate and represented the full colors in P3, only the audio was problematic, and that is hopefully sorted - but with out me understanding why it went off the rails in the first place <sigh>

did send a note to the Resolve forum, no reply as of yet, i guess no one has seen this, or they use the plug-in

[dermot shane]

all went well, played back in a cinema, looked great, sounded great... awesome software, awesome support from Carl and Carsten

my workflow;
graded in Resolve, ACEScc, DCi2kscope
set ODT to P3-DCi
set frame rate to 24
set in and out points to the top and tail 2pop's
exported picture to Jpeg2000 seq, debayer and sizeing set to best, caches bypassed, took aournd 7hours - about the same time that it takes to export a DPX seq, the choke point is the de-bayer settings and renders of noise reduction and OFX plug-in's while exporting.
exported 5.1 audio a single 6ch Qt/24bit, took around 20 min
in DCP-0-Matic set a project to DCi2k scope
imported audio Qt
imported jepg2000 file seq
trimed 1fr off top and tail of both elements to get rid of the 2pop's
created DCP, took around 45 min
copied DCP to a disk and sent it to a cinema for a check screening

pretty simple process with great resualts

i will still send bonded deliverables off to facilities, but will create festival/QC screeners in the future with DCP-0-Matic

[carl]

Hi Dermot,
Great! Thanks for reporting back. It's always good to hear of effective workflows.
All the best, Carl

[Carsten]

Great - Dermot, could you supply a short snippet of DNxHR444 to Carl for evaluation?

(or try the latest Beta on your DNxHR file?)

- Carsten

[dermot shane]

i tired the beta from last week without any joy. but here's a short clip from a film, rendered into DNxHR444 to play with
https://we.tl/olnpbWEHJb
There's two versions, graded in ACESlog, one with the 709 ODT, one with the P3-DCi ODT
thanks for giveing it a go!