Michael Phillips
1
COW Coin
Forum Replies Created
-
Long cuts as well as long takes on original files longer than 10 minutes gets sluggish as well.
Michael
-
As Job summed up admirably, there are distinct differences in playback frame rates between 23.976 and 24.000. This is a .1% difference between the two. So a 1 hour program at 24.000 will be (1 hour x .1%) longer when played back at 23.976. Or faster and shorted if going from 23.976 to 24.000. This is true for picture playback as audio does not use timecode as a playback speed, but as a counting method where sample rate is what determines playback speed against whatever clock is used for a given frame rate.
So… given 24fps at 48000 samples per second, any audio file that is not 48000 samples will be slowed up or slowed down. So a file recorded at 48048 in a 48000 project will play back .1% slower. If you download Sound Device’s Wave Agent, you can play with sample rate to timecode and see what happens.
This same .1% is what is referred to as the “fractional” rate and is part of life in NTSC based countries (North America, Japan, etc.). What is referred to as 30fps or 60fps can be and most likely (30 x .1%) 29.97 or (60 x .1%) 59.94.
Many cameras allow you to record 59.94 or 60.000 for example so that only adds to the potential errors. For example, the Canon 5D when it first came out was true 30.000 and not 29.97 which caused sync issues when that was first used.
People tend to refer to 29.97 as 30 because humans are either efficient or lazy and it is much easier to say “thirty” than it is to say “twenty nine point nine seven”.
My latest blog offers a spreadsheet that does some of the fps to sample conversion as well as the number of +/- semitones needed to maintain pitch when changing speed on audio:
https://24p.com/wordpress/?p=195Michael
-
You content may be 23.976, but the NTSC 23.976 and HD 23.976 are not the same thing when it comes to timecode management. There is no 24fps timecode or format defined by SMPTE for standard definition formats. 24 and 23.976 are “fabricated”formats in the NTSC signal with forms of pullodwn; 2:3:2:3 and 2:3:3:2, etc. In 1992, Avid derived a method to which native 24fps editing could be done from NTSC and PAL source within SMPTE constraints. Tape machines and the like only ran at 30NDF, 30DF or at 25fps. So capture and layback was designed to work native 23.976/24 but within PAL and NTSC – thus the timecode counting issue.
When HD was defined and implemented, SMPTE spec includes a 24fps timecode. There is no 23.976 timecode, nor is there a 24fps DF timecode. Just as there are no 50fps and 60fps SMPTE defined timecodes.
In the world of file based, a lot changes, many parts of MC need to be updated to better serve all this, but it works if your are cognizant of what is happening. If your audio was recorded 24fps TC, then I would bring them into a 23.976p project to start.
Michael
-
Are you in an NTSC 23.976p project type?
Michael
-
The dongle is an additional $500, so software + dongle = $1,799. As noted already, if you let your $299 year update fee lapse after 12 months, you are buying another license. At that point you now have two Media Composer licenses:
1. Original version of MC with dongle stuck at that version. That one will work as long as OS and such are supported.
2. A second Media Composer licensed for the latest version at time of purchase. This is a software license. If you want to keep using a dongle, then you pay another $500. The old dongle will not work, nor will Avid update it.Basically you can do a spreadsheet to look at your costs over time. Purchasing a license and keeping up the yearly upgrade subscription (no dongle) is:
Year 1: $1299
Year 2: $299 (comes to $25/month)
Year 3: $299
etc.This is the only offering that leaves you with a version of MC should you ever decide to stop payments.
The full-on subscription model is exactly like Adobe’s meaning you stop paying, you have no editing solution. You can subscribe by the month or commit to a year which gives you a discount.
Michael
-
iNews and other tools as part of the new production solutions:
https://www.avid.com/US/Solutions/byNeed/news-production.html#overview -
They should…but import and AMA insist on adding .mov to the Source File name so you will have to deal with that in the relink dialog box. There is a setting for “ignore extension”.
AMA and import are just two different ways to get into MC – if there was a transcode happening, then AMA+transcode is at least 2x faster than import+transcode. But in your case, if indeed a fast import, then it will be just as fast as an AMA consolidate without the extra .new master clips hanging around.
If on OS X, you can always do a final sequence mixdown to your ProRes delivery format and export “Same As Source”regardless of what the timeline codec was to start with. But if it is all ProRes and rendered, then you should be able to export ProRes without the mixdown.
Michael
-
Good to know,thanks. I was just testing with whatever else I had on hand, but I would suspect that other solutions would have this type of support. I just wanted to contrast what Wraptor in AME CC as is, could not do.
Michael
-
There are some basic differences in UI nomenclature as well interaction. For example, getting into trim mode is different. I can’t remember now but it was and may still be. Biggest difference is how it connects to thee different broadcast tools as part of an collaborative workflow and the integration thereof.
Michael
-
Any reason why ProRes over native MXF DNxHD? I am going to assume the ProRes files are .mov? The only way you are going to get that to work is AMA link to the ProRes, import AAF and do a relink to selected and hope that the REEL ID is consistent between the AAF timeline and the new files.
Look at both TAPE and SOURCE FILE columns for the sequence (clips in sequence) and the AMA Linked clips. Are the the same? If not, how close?
Michael