Little Frog in High Def

EDIT BAY POSTS

The thirty-first episode of THE EDIT BAY is now available for download.

This one is all about editing with a laptop. Pretty common now, but not so much back in 2003…on the set of a David Mamet film.

(Here’s the tweet from the director of Toy Story 3, editing on an airplane)

To play in your browser or download direct, click here.

To subscribe to this podcast in iTunes, CLICK HERE.

A little less than a year ago, I reviewed the new Avid Media Composer v5 software.  Buried in that otherwise good review…at the bottom of the review…I mentioned that even though AMA was cool and nice, you couldn’t export an AAF or OMF from a project that contained AMA material.  You’d get an error stating that the media needed to be Avid media in order to export an AAF.  This pretty much put the breaks on my wonderful plans to capture in FCP (to ProRes), import those files into Avid Media Composer via AMA…edit them…and then export the cut via AAF, import back into FCP via Automatic Duck Pro Import 2.o (zero quality loss, as it would reference the same media you captured with FCP), then take that into COLOR to color correct, and then finish and output with FCP.  There are a few reasons for wanting to do this, but I won’t go into that now…later.

Well…now you CAN do that.  And thanks to Twitter, and the great community of Twitter people I follow, I found out how.

Twitter person @jayfriesen (fellow Montanan) was complaining that he couldn’t export an AAF with footage he imported via AMA.  That this was a huge workflow hole that needed to be fixed.  He has a Twitter follower named @joshpetok who chimed in; “did you try using a linked AAF and turn on the ‘use AAF edit protocol’ checkbox?” Well, I have done that before, and it still didn’t work.  I thought I did anyway.  Well, that got people in that tweet bundle (group of people tagged in a tweet) to have this discussion between ourselves like “Really?  Serious?  it works?” and “Why didn’t the beta folks tell us about this” and “I am on the beta and I knew nothing about this.”  And so on.

Well, I tested it.  Guess what.  IT WORKS!

First thing I did was import some Canon DSLR footage (Canon T2i) into FCP, transcoding to ProRes 422 as I did so.  I then launched Avid MC 5.5 and used LINK TO AVID AMA FILES to load the footage into my bin.  I make a quick little sequence:

Nothing fancy, a few clips, a dissolve, layering. Then I exported an AAF:

I clicked on OPTIONS to get to the window where I need to adjust things like LINK TO MEDIA and USE AAF Protocols. So I used these exact settings:

That resulted in Avid churning out an AAF file without complaint. No fuss, no muss. OK now, bring this into FCP via IMPORT>AUTOMATIC DUCK PRO IMPORT 2.0…using these settings (that match the clip settings):

And a second later, the sequence and a folder with the clips appeared. I double clicked on the sequence…exact match:

NOW… it doesn’t end there.  Apparently this isn’t new.  This bug was secretly fixed in version 5.0.3.7.  Because @joshpetok said; “I’m on 5.0.3.7 at my current gig. When I get a min, I will test here.”  And follows up with; “Works in 5.0.3.7. Make sure AAF Edit Protocol checkbox is on.  Only issue: clips can’t be embedded. Link manually in resolve (sic).”  Meaning that you have to manually relink the media in the RESOLVE color corrector.  @joshpetak: “it’s basically a better version of an EDL with metadata.”

That is cool!  This means you can capture with FCP…AMA to Avid for editing…and now either AAF back to FCP to go to Color, or AAF to RESOLVE.  Now it is later (remember, earlier in the article I said later that I’d talk about why you’d want to do this…now is later).  Why would you want to do this?  Well, if you want an inexpensive capture station (FCP), but want to edit in Avid MC, because you are either more comfortable with it, or want to use the shared project workflow it offers…or like MC over FCP for whatever reasons you have.  Or a client supplies you with ProRes footage.  And then you want to export an AAF for audio mix (which you couldn’t do with AMA’d footage before), or you want to get this back to FCP, or some other application for final touches.

I have worked on a couple shows that were shot on film.  The shows I worked on were shot on Super 16mm, and it was, and still is, an expensive process.  The film stock isn’t cheap (although cheaper than some tape formats), but then you need to add into that the film processing and telecine to tape.  And we wouldn’t receive the footage for the day’s shoot until the following day, when we needed time to capture it.  So the editor might see the footage perhaps two days later, or in the afternoon of the next day, depending on the edit bay situation.

And, because it was an expensive process, the shooting ratio was small.  On average you might see between two and six takes, depending on how good the actors and crew were in getting the shot just right.  Once they got the shot, they might shoot a safety…but there’d be plenty of takes that might have flubs, or something bad happening, but they still print because part of the take was good.  If something went wrong, they’d shoot more.  But, regardless, the shooting ratio was pretty low. And because it was low, the amount of time needed to review the footage, and produce a rough cut, was relatively short.  A week for a rough cut was totally doable…on a 30 min show.  60 min shows have more time…two weeks, or 12 days.

Then came tape….and more recently, tapeless.  Now the shooting format was cheaper.  And because of this, directors are shooting more…A LOT more.  And shooting longer takes.  Sometimes getting two to seven takes in ONE “take.”  Meaning that they don’t stop and re-slate, they just say “RESET…let’s take it from the top” and don’t stop the tape and roll again.  That is fine, we can subclip or add markers/locators to separate them.  But what this is really starting to do is make the amount of footage that the editor has to look at and deal with, increase ten-fold.  Yet, and this is the clincher…we have the SAME AMOUNT OF TIME to sort through the footage and present a rough cut.

That’s right.  The shooting ratio jumps to ten times the amount we used to get…but the time allotted to cut this footage remains the same.

I have an editor friend who is dealing with this right now.  On EVEN STEVENS, we’d have perhaps 3 hours of footage for the 22:42 min show.  But he is on another Disney Channel show where he regularly sees six to 8 hours of footage…multiple takes buried in one roll.  This takes time to sort through.  Producers and networks want the best take used in the show.  Well, this requires that the editors actually watch all the footage…and then compare all the takes. Several times. We need to see the subtle differences that make one take better than the other.  Comparing 3-4 takes is a bit quicker than looking at 10-12 takes.  It takes time.  But the big problem is, we aren’t given any additional time to do this.  And on shows that are Union, you can’t work longer hours to do this.  Well, not on the books anyway.  Yet, we are expected to take the extra time to do this.

But that brings us to the overall issue of dealing with lots of footage, and the expectations producers/networks have in terms of our work schedule.  To many, what we do is mysterious enough.  But many people don’t seem to grasp that when an editor is given 80 hours of footage, we need time to look at everything.  If we work 8 hour days, we need two weeks (40 hours a week) to review the footage. Given a 50 hour week (10 hour days…which is more of the norm), that is a week and a half…8 days.  The problem is that producers want to see results after a week.  They want to see some sort of cut…a string out, a rough cut.  Something.  So we have to start building the cut the instant we watch the footage.  And what this has me doing is cutting something, finding a better take later, replacing what I edited, getting that to work.  Then, oh dear, the second half of yet a later cut was better than the one I have, but the one I have has a better first half.  But, great, the continuity is off, so I can’t make that work.  My cutting can end up being more haphazard.

The message here to producers… who I doubt read this blog so they won’t get the message… is “please give us time.”  If you want a quality project, please give us time to review the footage.  If you don’t have it in the budget to do that, please shoot less.  Well, this won’t work out well in the documentary world, because they shoot what they shoot.  But in the “reality” world where there are multiple takes (yes, there are), and in the narrative world… you need to note that the more you shoot, the longer it takes us to sort through the haystack to find the needle.  So please keep that in mind.

This is why the cut we turned in wasn’t as good as it could have been.  Or the one that we did deliver… that cut that was really really good…we worked 24, 48 hours straight to make it that way.  Or we worked multiple 16 hour days.  That seems to be the expectation lately.  That we put in the long long hours required to produce the cut that they expect us to produce in the short amount of time they give us.

This, my lovely wife, is why I was working late all week long, and got home long after you were asleep.  And why I am cranky when I get up at 6AM to help with the kids.  And sleep until noon on Sunday.

EDIT: Read the comments!  They are full of great discussion…

OK, it has been a while since I did a spontaneous blog post…dealing with an issue I am faced with at this moment.  But, it is one that has always been in the air, and has driven me to make the decisions I have made.

ARCHIVING TAPELESS MEDIA

I know that the current thought for rock solid archiving is LTO or DLT tape drive backups.  Because this is what banks use to archive all their data.  Yes, you have to buy a deck, and while it doesn’t cost as much as, say an HDCAM SR deck or D5 deck…it isn’t cheap.  And then you have multiple options…LTO3, LTO4, DLT.  And multiple types of software to backup/recover this data.  Not much of a standard there.  Although when I was at the DV Expo, there seemed to be a consortium of LTO drive makers that all were backing the new way of archiving, one that made things very simple, and made the drives show up like hard drives, and you could grab what you wanted. Still, to many places it is still cost prohibitive, and somewhat complex (if you don’t get newest) software.

The issue I’m facing today is that the first two seasons of a show had their camera masters archived to LTO4, using NETVAULT software.  They did this when they were housed at a post facility.  So archival and recovery was easy, as it was under the same roof as they were editing.  But then the post house closed, and auctioned off all of their equipment.  And the production didn’t buy the LTO deck.  So now they had all these tapes with their source footage, and no way to get anything off of them.  It’s like when I asked for an HD copy of a show I edited for my reel, and I was handed an HDCAM tape.  Well…nice, but, I don’t have an HDCAM deck.

So today the production told me that they needed to retreive some footage from a previous season to use in the current season.  But all they had were these tapes.  And the post house was gone.  It was now my job to find a way to get this footage off the drives.  I called around, and found a place that had the drives, had the software, and could do just what I needed.  And it was relatively cheap…but not when your budget is already spent, and low low low as it was.  There was a per/GB charge that wasn’t too bad, but then a bay rental fee.  And the total was steep given the budget, but what could we do?  We needed the footage.

This situation makes me feel much better about how I archive the current season of the show.  My current solution is to archive to hard drives.  And not one drive, but two per archive.  A manual “RAID 1″ if you will.  I backup all the cards to two hard drives…redundancy.  So I get 1TB Hitachi hard drives (around $60 each), and I have this SansDigital 4 drive eSATA dock ($99)…and I simply slap bare SATA drives in, backup, take them out, put them into Webietech drive cases ($7), and I’m done.  They are easy to access…and I don’t need my DOCK to access them.  Any single drive SATA docking system will work.  And they are cheap, so I can get a couple to keep around in storage if I wanted.  Then I can access any of the drives, any time I want.  Heck, I can even slap them onto a drive tray on my MacPro!

When I hand off drives to clients, I get drives with enclosures, so they can access them any time on any machine they want. Two drive redundancy if they want…they pay for the drives.

I know that LTO options are getting standardized, but still, when a client wants the source backups, and the get handed one of these…they now are in the situation described above.  With a tape in hand and no easy means to access the footage.  And from experience, clients want the easiest thing possible.

The thirtieth episode of THE EDIT BAY is now available for download.

This one is about killing babies. No, not ACTUAL babies. Segments of shows that you worked really hard on…poured your life into…don’t work. You have to kill them.

To play in your browser or download direct, click here.

To subscribe to this podcast in iTunes, CLICK HERE.

No, that isn’t some odd Frankenstein-vampire thing.  It is a term that is often used by documentary editors to explain how we edit certain interview clips.  Much like Frankenstein’s Monster was made from body parts of different people, a “Frankenbite” is a soundbite made up of statements from several sentences from an interview subject.  We might need half of one sentence, and the other half of another, or a small word to piece together a couple sentences, or to add to make what they say sound right (mean, make sense).

Now, to keep things ethical, which is one of the main rules of documentaries, you shouldn’t do this to make the subject say something they didn’t say.  Rather, we are trying to condense what they are saying, or make what they are saying clearer…make sense.  Not everyone is concise in how they say things, nor can some people explain things in a way that makes sense to the normal everyday viewer.

It’s a bit different than a “pull up.” A “pull up” is when an editor cuts out “uhm,” “uh,” “you know,” “like,” or any number of paused-in-thought-words, or stumbles, repeated words…or empty air when the subject is formulating what they what they are about to say.  We make “So I, uh…I ran across the str…the road to the store, uhm, where I, you know, ran into Harrison Ford, like…uh…buying beer” into “So I ran across the road to the store, where I ran into Harrison Ford buying beer.”  Basically just cleaning up the statement.

Frankenbyting is where we try to fix what people say.  Take the following statement: “The school I went too, back in 1988, I was in third grade at the time.  I was listening to our teacher talk about Andrew Jackson, the president who forced the Indians to march hundreds of miles from the east coast to Oklahoma…many of them died.  Mr.Braeburn said he was a hero for moving these dirty Indians away from civilized people, but I…I couldn’t stand for that, so I raised my hand…I am Indian…I confronted that…told him he was wrong for saying that.”  I might want to chop this up and make it more clear.  ”Back in 1998, (when) I was in third grade, our teacher, Mr Braeburn, (said) that President Andrew Jackson forced Indians to march hundreds of miles to Oklahoma, (and) man of them died.  He said that he was a hero for moving these dirty indians away from civilized people.  I am Indian…I couldn’t stand for that.  I confronted (him) (and) told him he was wrong.”

So what I did was make what the subject said clearer…but say the exact same thing.  Because their speech pattern was so broken up that it make it difficult to follow.  So I rearranged things to make it clearer.  But I didn’t have all the words I needed to make it right.  You’ll note that the words in parenthesis, (these things), those are words that aren’t in that sentence…bridging words that I need to find.  I will listen to other parts of the interview in order to find those words.  But not only do I need to find those words…they need to sound right.  Someone might say “and” differently, depending on what they are talking about…or “but.”  Or any other word.  So I have to find that word, and it needs to sound right, fit the sentence.  Have the right inflection.

Was this time consuming…you bet it was.  But now there is software out there that can help us.  Scott Simmons of The Edit Blog over at Pro Video Coalition mentioned this on Twitter.  He said that he used GET, from AV3 software, to search for the word “but” to help him build his Frankenbite.  GET, and PhraseFind on the Avid side, is designed to use waveform prediction (corrected by Phillip Hodgetts in the comments below) which uses pattern matching of audio waveforms to catalog narration, and allow us the editors to search for soundbites, or words, by simply searching for it.  Now that is what I call handy!

Now, this can have an evil side.  This can be, and has been, used to make people say things that they didn’t really say.  I would hope that it would go without saying that this is highly unethical, but many people do this.  So that the person would say what they need them to say to further the story they are working on.  Just plain wrong.

Now, the intention of this article wasn’t as marketting for GET or PhraseFind.  It stemmed from the Twitter post where Scott mentioned that he used GET to find a word, and then I commented how it’s great for Frankenbiting, and more than a few people hadn’t heard the term.  But that software is darn useful, and I like pointing out useful tools to editors any chance I get.

The twenty-ninth episode of THE EDIT BAY is now available for download.

This one is about ads with contests.

To play in your browser or download direct, click here.

To subscribe to this podcast in iTunes, CLICK HERE.

Simply brilliant…

(Credit goes to ToksterJokester over on digg.com)

Hey people…because I have been too busy lately to post any cool tips or tricks or workflow fun (due to an upside down schedule), I would like to point out a blog that IS doing this.  Scott Simmons over on the Editblog at PVC is posting lots of quick tips for Avid and FCP.  And he is asking for readers to submit tips as well.  This is a great way for editors to share the great and often hidden tricks and tools they use when editing.  If you have a tip to submit, click here.

Thanks Scott!

Today Avid Technology issued a press release outlining the changes and new features in their latest version of the Media Composer software, Avid Media Composer 5.5.  Among the new features is one that particularly excites me…and it has been very difficult to keep quiet about this (I’m on the Avid Beta, so I’ve been playing with this for a while).  Media Composer 5.5 now works with AJA hardware, specifically, the AJA IO Express.  What is major about this announcement, other than adding another third party partner to their list (they opened up MC5 to the Matrox MXO2 Mini last year) is that this isn’t just another monitoring solution…this is a capture solution as well.

Yes, you can capture and output using the AJA IO Express.  Capture to Avid MXF media.  Have deck control so that you can capture and output with accurate timecode. That’s a big deal!  Sure, it doesn’t offer the hardware acceleration that the DX line of hardware does, but it is a great economical way to get your footage into and out of your system.

Not long ago…last year, around NAB in April…Avid announced Media Composer 5, and I blogged about it.  One of the things that I mentioned that was HUGE was monitoring via non-Avid hardware, specifically the Matrox MXO2 mini.  That was a huge thing…now we could monitor what we were doing without the expensive Avid hardware.  Because that was a gripe of a lot of people, that yes, the Media Composer software is now inexpensive, but still they needed to shell out upwards of $8000 for a Mojo DX  just to see what they were doing on a monitor.  They didn’t need to capture, they just needed to monitor…as they shoot tapeless.

Still, this left more than a few people asking, why the MINI?  We want this to work with AJA, Black-Magic.  We want Avid to open up to the hardware we already have so that we can capture using the same card in Final Cut Pro as well as Avid.  The answer Avid had to that was…”baby steps.”  They were slowing working towards this…soon more and more hardware will be added. Sure enough, we have new hardware, and the ability to capture and output…for under $1000.  And it works on a tower and laptop (if the laptop has an Express34 slot).

What else did they announce?

PHRASE FIND.  This is much akin to GET from AV3 software.  This software will index all of the spoken words in a project, and allow you to search for something someone said.  For the times where you know that the interviewee talked about their life in Montana, but you don’t know where in the 2 hour interview they said this.  You could spend the two hours looking for it, or use Phrase Find to find the word MONTANA within seconds.

Phrase Find isn’t part of the Avid Media Composer package…it is what they call an “addon” product. But, it is one that will be well worth the money, with the amount of time it saves.

They have also added native support for HDCAM Lite, expanded support for Euphonix hardware (no surprise since they bought the company, and enhances their Smart Tool functionality…allowing us to directly manipulate transitions on the timeline with the mouse.

Avid is advancing their Media Composer software faster than I would have imagined.  It is very difficult to keep up with them now.  Every time I turn around, BAM, new Media Composer with great new features. BAM! support for more hardware.  Adding new functionality, and making changes to old functionality to make it better.  Listening to editor’s needs and addressing them at the speed of light.  Avid Media Composer is emerging as the leader in keeping up with current technologies…and losing the image of the old stagnant system that while rock solid, was very antiquated.

In the past I would find my self cursing under my breath…or rather loudly for people around me to hear…that Avid couldn’t do this or couldn’t do that….cursing that I wasn’t working on my lovely Final Cut Pro system.  But now more and more often I find myself on Final Cut Pro cursing and swearing and wanting to be using Media Composer.  Honestly, I want both systems at my beck and call.  And Avid is making it easier for me to do that.

The twenty-eigth episode of THE EDIT BAY is now available for download.

This one has me lost in translation.

To play in your browser or download direct, click here.

To subscribe to this podcast in iTunes, CLICK HERE.

No, you can’t do that now…not without third party plugins.  But I wanted to mention two situations of me needing to do this recently, and two of the third party options I used.

SITUATION #1:

I am working on a show business talk show that airs weekly on cable.  This show is shot in a small studio to three HPX-300 cameras, recorded onto P2 cards.  AVCIntra-100…1920×1080 at 23.98fps.  Now, we have a VERY fast turnaround time.  The show needs to be edited in three days.  But the biggest rush is getting the footage back from the set and ready to start cutting in the same day.  Because a producer comes in and pulls out all the selected bits he wants in the show…and we don’t want him to be here until 3:00AM!

Typically we backup the cards to small drives, then import the footage into FCP via Log and Transfer.  Well, the cards are all 64GB cards and each one has about 75-80 mins of footage total (4 cameras, multicamera shoot).  There are eight cards…four usually fill up, and the other four go about 1/4 full.  Offloading these takes a while.  We don’t have that Panasonic AJ-PCD35 PCIe card reader that makes this VERY fast.  Nope, we have to rely on my Powerbook G4 and ShotPut Pro.  It takes a while.  Offload the cards…then ingest into FCP.  Even coming in native AVCIntra took time.  And the first show did have the producer waiting forever.  So this wouldn’t do.

So I suggested an alternative.  Offload the cards immediately to G-Raid drives that we use for editing…then use CALIBRATED MXF Import to access the footage and allow editing to begin right away.  Then backup the cards again for archive.  This way we can start editing right away, after only 3-4 hours of offloading.  And we need this solution on multiple machines, as the main editor uses one machine, I, the online editor, use another, and the assistant yet another.  So we needed an economical solution we could all use.  I already had Calibrated on my MacPro, so it was an easy choice.

All we do is find the video files and drag them right into the FCP Browser.  They link to the audio right away and we can start multiclipping and cutting…well, after we do a bit of labeling of the bins as for what card they came from so we can group things properly.  It works rather well.  The only drawback is that the individual P2 files come in their original 4GB chunks.  So if you run the camera for a long time, you will have one shot broken into multiple clips.  Log and Transfer will stitch these together as one long clip.  That’s handy.  We had to work around this by Multiclipping the smaller 4GB clips, and then just stringing them out on the timeline.  It works fine.  And you do need to work in a ProRes timeline, as there is no setting for AVCIntra.  But that is fine, it renders quickly…well, quickishly…on my end.

SITUATION #2:

I am tasked to edit a promo for a product.  I am asked to do this on Friday.  It is due Monday.  The footage is all RAW P2 file backups on three 2TB USB drives.  I do have notes as for selects, but I have literally hundreds of hours of footage to look at.  There is no way that I have time to load all of that into FCP’s Log and Transfer interface and click through the footage.  Then choose my selects, and wait for them to import.  Nope…I need a faster solution.

For this I was on my laptop, that has MXF4Mac and P2 Flow.  MXF4Mac, like Calibrated and Raylight, offers you the ability to open the MXF files directly from the VIDEO folder.  And with the ability of the MacOS to allow you to preview footage without opening the application simply by clicking it and pressing the space bar is a big big bonus.  This allowed me to scan through all the MXF files by having that preview option open, and simply pressing the DOWN ARROW to select the next clip.  I was able to scan the footage VERY quickly.

When I found a shot I was looking for, I’d use MXF4Mac to send that shot to FCP…or P2 Flow to access and send the entire card to FCP (takes a matter of seconds).  After only a day, I was able to locate and import all the footage I needed, and work with the footage off of the USB drives.  I then cut the promo, output a QT of the rough cut for notes, addressed the notes, then exported the final.  All without converting one frame.

MXF4Mac is a bit more expensive than Calibrated, but it offers a few more features.  Like the ability to stitch the footage so that one take is not broken into several small chuncks, but actually imports as one chunk.  And you can view and edit metadata, and even MAP metadata so that you can then send a batch capture list to FCP, or just send the MXF native, and have that metadata appear in existing FCP columns.

Yes, I do primarily advocate converting your footage to QT when working with P2.  Why?  Because more often than not I am working in situations with 4-6 editors, and multiple seats of third party MXF readers can get expensive.  And organizing that amount of footage in that format is easier for assistants and editors to manage.  But, if you are a sole editor, and want access to all that metadata, and want to edit quickly, I do recommend these third party options.

The twenty-seventh episode of THE EDIT BAY is now available for download.

This one has someone else watching the cut, and giving me darn good notes.

To play in your browser or download direct, click here.

To subscribe to this podcast in iTunes, CLICK HERE.