This IS the Avid L2: Re: [Avid-L2] Re: DnX 23.98 Project to 59.94 file

Thursday, January 21, 2016

Re: [Avid-L2] Re: DnX 23.98 Project to 59.94 file

QT Change does the correct job in adding DF time code back to an AME frame rate converted file. I start my 23.976 at 59:00:00. This results in a coincidence point of 1:00:00:00 and 1;00;00;00 when viewing the 30 DF time code in Avid. Then once baked out of AME I change the time code to DF starting at 00;58;59;28 and again 1;00;00;00 lines up with program start.

I understand as you have pointed out before that 2:3 and 3:2 visually look the same but in the above workflow the 1;00;00;00 is a B Frame with 3 fields and not an A frame. I haven't had any rejections for this but a purist might raise a QC issue. To me the fact that it has a consistent cadence makes future pulldown removal a relative breeze. I get all kinds of complaints from the streaming side of our deliverables about our 2:3 pulldown field tapes with 29.97P graphic animations for lower 3rds and the artifacts that creates for them to convert the broadcast master to something streamable.

I agree that NDF time code makes much more sense for pulldown removal and tracking A frames in the final sequence but just like a Sony deck will frame rate convert 23.976 to 29.94i with either NDF or DF time code I'd like that flexibility so my program masters time code reflect actual clock time. If it's an element to be used as a source then NDF is the way to go to help track things down the line but program masters traditionally are DF for accurate clock purposes and most of my delivery requirements are for DF time code when delivering 59.94 masters.

I can't say I want to start delivering 59.94P to avoid the cadence issues but I get what you are striving for. I don't like 2:2:2:4 especially when it inevitably becomes a source down the line that of course will get slo mo'd etc...

---In avid-l2@yahoogroups.com, <cutandcover@...> wrote :

John,

The thing you say about 3:2 and 2:3 -- they are the same! 2:3 and 3:2 both look like 2:3:2:3:2:3:2:3:2 and so on. Why argue the difference? Are you concerned that the pattern isn't starting on two duplicate fields? I don't know any broadcaster or file delivery that doesn't accept it, unless they're only accepting inferior pulldown pattern delivery (2:2:2:4).

And the code thing: the reason why AME creates non-drop code from 23.98 is because that's the only way to do it and have the resulting code make any sense. If you want drop code, then you can't rely on a software conversion alone to manage that, because what you put in on any given second will come out on some other second. Non-drop means every second in both 23.98 code and 29.97 code always matches on the zero frame. This is the only truly acceptable way to CONVERT time code. If your delivery requires drop-frame code in the file, you better be doing it inside an NLE because that is the ONLY way to line up correctly in drop frame.

And as far as those places only willing to accept progressive - I'm on a campaign to make them accept 59.94p. That way I don't have to give them inferior and ugly looking 2:2:2:4 pulldown, and we can all move to 60 frame delivery some day.

On Thu, Jan 21, 2016 at 3:03 PM, bigfish@... [Avid-L2] <Avid-L2@yahoogroups.com> wrote:

Adobe Media Encoder will do it but it adds 3:2 pulldown not 2:3. Also it creates files with non drop code. I haven't found a way to change the time code to drop in AME but I use QT Change to make it drop frame after the conversion.

Given the streaming dominates broadcast quality better make sure the 2:3 or 3:2 pulldown is acceptable. Some cable networks no longer allow 2:3 pulldown or any interlace artifacts, of which 2:3 pulldown is deemed to be an interlace artifact. Bottom line is that when it goes to Hulu etc.... the split frames cause issue. Given they air old series TV like Rockford Files etc... I don't know why they can stream those but not modern day 2:3. I'd bet it's that much of the modern day is random cadence so it's harder to remove pull down. It has been pointed out by Mr. Oliver that much of the old series TV that was onlined on tape also had random cadence so again I don't know why they can't process modern random cadence like they seem to do for antique random cadence.

It's all just a moving target. My current series we finally jumped into 23.976 but then they want the tape backups to be 59.94 while the primary files are 23.976. I guess they intend to add pulldown or 2:2:2:4 to the files for broadcast but want the true 23.976 file for streaming concerns. Now we can't seem to figure out what time code they want on the slates. We had thought we'd be delivering everything at 59.94 with pulldown so everything is zeroed and slated in DF TC but now with the files being delivered at 23.976 the slates still read in DF TC. My gut says DF for actual running time is the way to go because I assume they will take my 23.976 file and create their own 59.94 file for broadcast purposes. We are still waiting for clarification on this.

If you want to stay in Avid you can do the mixdown in 23.976 and open a sequence in 59.94 project using the mixdown then adjust the motion adapter to your delivery requirement of 2:3 pulldown or 2:2:2:4 cadence. Wouldn't it be nice to have universal mastering capabilities with files?