Am 07.09.2016 um 23:15 schrieb Ryan Gleeson via DCPomatic: Wow, tough, but very clear and complete. - Carsten

Am 08.09.2016 um 15:44 schrieb Manuel AC via DCPomatic: I think that wasn't RIP, but frame HMAC? At least that was a recent thing between DCP-o-matic and the Barco ICMP, which was solved from both sides. - Carsten

On Thu, 8 Sep 2016 09:44:17 -0400 Manuel AC via DCPomatic <dcpomatic(a)carlh.net> wrote: [snip] Does anybody know what this "ASDCP test" is? Is there a tool included with asdcplib that tests DCPs in some way? Thanks, Carl

On Sat, 10 Sep 2016, Wolfgang Woehl wrote: That is what is puzzling me. Yes, as far as I can see DoM's MXFs contain a RIP for 3 partitions (tested with mxfdump from mxflib). I think we need to see the DCP in question to see if there is some corruption or problem with it. Best regards, Carl

Hi all, If needed, I can open a folder on my server for this kind of use: FTP access for DoM feedback till 200GB of space and 1 GB of bandwidth. Let me know if it is useful. Lilian Le 07/09/2016 23:54, Carl Hetherington via DCPomatic a écrit :

I know some people will object - but I always said it is a bad idea to create/distribute SMPTE DCPs NOW. SMPTE DCP is not yet an industry standard for distribution. In the last years, I have seen not a single SMPTE DCP in commercial cinema. The only ones I have seen were those I created myself with DOM for testing. Whenever I make 'real' DCPs, I always and still use Interop. Yes, SMPTE will probably become an industry standard within the next two or three years. The problem we are facing here now is that QC procedures in various forms are complaining about issues/features that do not cause playback failures in any equipment. Because equipment implementations are forgiving, but the QC is not. This is bad if it leads to content being rejected at festivals for reasons that do not cause problems on existing equipment. - Carsten

Am 09.09.2016 um 01:20 schrieb Carl Hetherington via DCPomatic: I agree that QC tools have to be strict in evaluating files. The question is, what does the QC tools operator make of certain complaints? 'Warning'? 'Error'? If 'opaque' QC rules decide which tools are 'allowed' to master content for festivals, that would be a bad development. Deluxe is a mastering facility. If they decide on the QC rules, they boost their own business in ruling out alternative DCP creation solutions. Add to this the fact that full specs for DCP formatting is only available in the form of very costly SMPTE documents... I do agree, though, that ISDCF-WTF is a must for SMPTE wrapping. - Carsten

On 09/09/2016 00:20, Carl Hetherington via DCPomatic wrote: OK, interesting! I'm about to embark on making all the DCPs for 3 UK festivals. Have always used SMPTE in the past and found it rock solid, so nervous about switching, but it sounds like Interop is a better choice. Can anyone vouch for DCP-o-matic's stability and correctness when creating Interop DCPs on OS X (Yosemite)? I'm also wondering about updating to v2.x from trusty old v1.83.0. But I ran a workshop a few months ago in Birmingham (UK) teaching people how to use DCP-o-matic, and we had crashes all over the place with v2.x. This was on Macs - I'm trying to find out whether Yosemite or El Capitan. Sorry, I should have checked at the time and captured logs, but the workshop was mayhem with 20 people making DCPs simultaneously, and everyone wanting answers to lots of questions! Stability is really key for me, so don't want to make the switch to v2 unless it's as solid as v1. Has anyone got an experience running latest DCP-o-matic on Mac and can report how stable it is? OK, maybe I was a bit harsh about Deluxe before. But I just have an overall feeling that DCP infrastructure is a bit of a stitch-up. Take DCP servers for example - they're basically pretty standard Linux PCs, with upgraded graphics cards - with the small exception of custom decryption chips. You could build your own DCP server with commodity components and open source software for a fraction of the price of a Doremi, but then no distributor will give you KDMs because your box isn't "professional". So you have to fork out thousands. This pattern of big players monopolising a market and protecting their position through FUD and claims that they are the only "trusted professionals" is nothing new and no doubt it'll all collapse in time. But right now the losers are the scrappy little indie cinemas, pop-ups and community cinemas who are often the places showing the most interesting films, but are priced out of being able to screen first-run releases as they can't afford DCP. And it's those first run films which would make their business models actually work so they can continue doing interesting things. So the loser in the end is film culture in general. So it pisses me off! Jim

Right now using OSX El Capitan with the latest test version, but I think Carl recommends using the stable version ;) A crash or bug from time to time, but really stable and fast in general. I don't see why an older version would be better, new versions have the most bugs fixed, optimizations and features. A special note about the Batch converter, we have always seen quite a few problems but they are never clear enough to fill a bug report. Sadly we try to avoid using it until we can do more testing. Manuel AC On Fri, Sep 9, 2016 at 8:05 AM, Carl Hetherington via DCPomatic <dcpomatic(a)carlh.net> wrote: -- http://akadcp.com/ https://www.facebook.com/akadcp

Interop was never a real standard, it was a really smart move from DCI to get all the pieces of the chain interopeable while SMPTE finished writing the actual standard. That's my vision on the formal side of it. Interop works for what majors call a movie, that's Flat or Scope, 24 fps, 5.1, multireel, no subtitles. Anything else you are gambling in how the projection system will deal with it. I won't even start with the little Interop monsters like 25fps or SMPTE-XML subtitles. SMPTE DCPs are all over the place on commercial cinemas: HFR releases, some 3D notably from Disney, and hundreds of smaller advertisement and whatelse. Wide releases from majors in the thousands of copies still try to avoid the odd and the known problems some really outdated theaters may have. This kind of release have already hundreds of problems to deal with that are not related to the chosen standard. When majors will move to SMPTE I'm looking for theaters to go and get some automation improvements. Personally, I'm waiting for projection format chosen based on CPL metadata, and TKR. Software server updates with SMPTE support have been available for years. I don't feel sorry anymore for a theater that doesn't maintain their equipement, and software updates are a really cheap part of keeping your machines working and ready. I have never released an Interop DCP I have never had a problem with any DoM created SMPTE DCP (but I don't do that many wide distributed DCPs to be honest) I have yet to know a theater that can only play Interop I hope the default standard can be predefined in the preferences, that will save many people the extra click. I hope this change doesn't mean SMPTE features will be forgotten in the future of DoM. We do festivals and subtitles and slideshows and many other things that are not a movie for a major. I firmly believe that SMPTE gives DoM more and better options to make its magic happen. Manuel AC On Fri, Sep 9, 2016 at 9:54 AM, Carsten Kurz via DCPomatic <dcpomatic(a)carlh.net> wrote: -- http://akadcp.com/ https://www.facebook.com/akadcp

The industry spent half a decade to shake most bugs out of the Interop tree. It was a successful process because people were using Interop. If you - not you but a DCP author, software house, system manufacturer - want to be able to offer your clients a contemporary feature set then you want to promote and push SMPTE/ISO DCP. E.g. Carl fixing the RIP and audio label issues is exactly what needs to go on - it happens because people are using SMPTE/ISO. Shaking bugs out of trees … no progress without it :) Wolfgang

Hi Jim, Thanks for the note on this. I have fixed my local copy of DoM to write a channel format to SMPTE audio MXFs, which should remove this Deluxe fail. Does anyone on the list send DCPs with any regualarity to Deluxe? Do you think we can get them a test from a new version of DCP-o-matic to see if it passes their validation? Best regards, Carl On Thu, 8 Sep 2016, Jim Dummett via DCPomatic wrote: