Michael Phillips
1
COW Coin
Forum Replies Created
-
Some of my thoughts on the first example got typo’d out, now that re-read it. I meant to say getting a job in an all “Avid shop” a certificate can at least help filter out those with some level of “familiarity” versus those who don’t. Not that someone who isn’t certified isn’t qualified – it’s one of those things that gives a head start to the conversation.
You can certainly go though a book like Media Composer 6:
https://www.amazon.com/Media-Composer-Editing-Essentials-Learning/dp/1133727980/ref=sr_1_1?s=books&ie=UTF8&qid=1364937889&sr=1-1&keywords=media+composer+6Which is based on entry level certification on your own. The benefit of any class situation is the ability to ask the questions usually not covered in a book which is usually “how” but also when and why?
Michael
-
If you’re looking to get a job in an Avid job – either permanent or via freelance, the certification gives some idea of the base level of training achieved. So it’s a good starting point. If you’re your own one-man shop with your own clientele, it’s really up to you whether you need certification or not, or just learn more on Media Composer functionality via the myriad of tutorial videos available.
Michael
-
The DNG RAW format has advantages for both day and night shooting as it give you more dynamic range, including highlights. Are you shooting under controlled lighting situation, or is it more run-n-gun style with available light only?
DNxHD or ProRes will get you about 12.5 hours of dailies per 1TB storage
DNG will get you 2.43 hours of dailies per 1TB storage
Roughly a factor of x5. DNG or uncompressed RAW shooting is not for the faint of heart when it comes to storage and archive. Not to mention the time it takes to copy, backup, etc.
I would do some testing with RAW shooting in the day and night situations you plan on shooting in to compare. I would highly recommend that you shoot the BMC Film LOG if you go with either the ProRes or DNxHD format as your camera original to give you that extra “extra” in color correction.
But you have it right – if you were to shoot DNG, the process is to open in Resolve (even the free Lite version) and create HD dailies (1920 x 1080) as ProRes or DNxHD. Be sure you settle on REEL ID when making dailies – if using Avid, then just check “use filename” for REEL ID. It may be the same for FCP or FCPx but it gets a little wonky on longer filenames if EDLs are very introduced into the workflow for conform.
Edit with the low data rate proxy, then you can conform in Resolve, to the original DNG files for color correction. The free Lite version will also be fine as you only want to master to HD – the $999 version is required if you wanted to finish and output at higher than HD resolution.
Michael
-
How do you want to finish? At 2.5K, HD? both, etc.?
There is no reason not to consider a transcode to a more friendly editorial format. The goal of editing is to tell the story and do so with the best performance possible so the editor can concentrate on the flow – I don’t understand why there is such a need to edit “native” when either the system, the workflow, and such don’t really require it.
Transcode to either a ProRes or DNxHD codec depending on system. There is a ow data rate for either one (36Mb/s) that is great. Once done, you can go back to Resolve for the final conform and color – pretty much as most studio theatrical films are done. Once back in Resolve, you have access to the RAW parameters and a kick-ass color correction.
Don’t forget audio and audio workflow in all of this – audio brings so much to the story and viewer experience – even more so than what resolution or camera was used, or whether it was RAW or not. Get the best audio you can on set.
Michael
-
4 hours is roughly 310GB of DNxHD so you should be good there.
Michael
-
Aside from the AMA issues for now, I think you might want to rethink your whole 23.976 for PAL 25fps delivery. Note that every “studio” type film or program that originates here as 23.976 or 24.000 gets pulled up to 25fps as part of the conversion process – it is a frame for frame conversion essentially resulting in a program that for all intent and purposes looks like it was shot 25p to start with. Subtle differences is that motion is now 4.1% faster that the original program.
Now, you should also know that most if not all DVD players in Europe can play back a 23.976 disk – the player and/or television can do the conversion on the fly. This is most likely some combination of pulldown insertion and interlace to progressive conversion depending on quality of player.
If you have the opportunity and capability to shoot 25p, and the PAL DVD is your only deliverable, then it is something to consider.
Now – to the project type – if shooting 23.976 I would offline, online and master at 23.976 especially in an offline (DNxHD 36) and relink back to camera originals – you’ll have to open your files in a 23.976 anyway, transcode there, then move that to a PAL project, yadda yadda yadda – too many hoops with potential for errors. Take the final 23.976 master and deal with it from there for your PAL as a single pass transcode/conversion.
I’ll be posting a blog one of these days/weeks about how to do a file based universal mastering within Media Composer (no need for Symphony) with audio and pitch correction for audio. Just have to find the time. This allows for frame for frame rate conversion resulting in a much higher quality conversion than Media Composer’s mix & match. It takes a few steps, but worth it, quality wise if you have the time, and not the budget. 😉
Michael
-
You might want to look into something like this:
https://www.vidsharx.com/for a two seat low cost scenario. I have not used it myself, but have heard from a few folks it worked well in their two system configuration.
Michael
-
Forgot about the security issue with the burn-in, so the consolidate won’t work… then my suggestion of just their sequence and the watermarked QT might be something to try. The manually created EDL can work with a V1 track (and to hell with audio edits that are not straight cuts) will be painful depending on complexity of timeline. I am involved in a DI now where a single reel has 1100 picture events!
Michael
-
Ah…. the trailer workflow. It would be great if Media Composer incorporated an EDL Trace feature like in the old linear days, and I believe Lightworks could do it – basically trace back from a “master” what was the record side timecode back through the same sequence to its original sources. Although layers may need to some manual intervention but you would conform to the sequence spans first, then have the ability to match frame back to original sources.
This is not an easy task in your current situation as all metadata has been flattened in the Quicktime creation. It might make more sense to consolidate the sequence and hand that over than making QT, but they probably want to make small files to send over an FTP. A feature I had designed for Media Composer several years ago was a flattened file like a QT being merged into a sequence so it was now using that picture and sound for the cutdown.
And this might be the way you get the same result, but you have to manage the timeline more carefully. With the QT reels, ask that they give you a bin with the timelines in them. In your system, edit the QT track as the top track where cut points all align. Now when editing, be sure to keep all tracks as part of the edit. Send back timeline minus your QT tracks and edits…
In order to relink with just the flattened QT files, you will need not only the source timecode, but the reel ID to be logged which I assume you are getting from the burn-in. Sounds like a laborious process prone with errors. I would go the consolidated route first. 20 minutes of DNxHD 36 is only 5.5 GB…
Michael
-
And none can wonder whether it is a first calculated step in going private?
Michael