“Fix it in post”

I hate those four words.

Those four words are the bane of my professional existence.  Those four words are the mating call of the lazy cameraman, the dog whistle of the unskilled sound recordist, the battle cry of the hopeless producer.  Those four words are what is wrong with post production.  Post production shouldn’t be about fixing footage that was screwed up on set.  Post production should be about telling stories with the excellently produced footage handed to the post crew by the production crew.  In some cases, those four words can even translate as “The guy who shot this really doesn’t know what he’s doing.”

I just did some work on a film.  A narrative.  It was shot on a Canon 5D Mark II in 1080p with audio recorded separately.   I won’t get much into the specifics of the project here, though I think it’s pretty cool and it’s being finished by a sound designer right now and I can’t wait to see what it looks like in the end.  Tragically, Plural Eyes had trouble synchronizing the audio in places and that made things a bit complicated when editing.

This however pales in comparison to what not having a script supervisor does.  I’ve seen it a lot in recent work of mine, both reality and narrative.  When you don’t have a script supervisor and your director and camera ops don’t notice the inconsistencies you wind up with some major continuity errors that make clips unusable.

About half of the shots on this recent project were rendered useless because of continuity errors.

When there are these sorts of errors, most of the time no one has noticed on set so no one utters those damnable four words until the footage has been screened and the line producer has told the director that if he or she intends to reshoot, they should consider paying for it with their own credit card.

The time it would take to go through a film and add or remove a Hawaiian shirt from an an actor and the expense of it make reshooting the cheaper alternative.  Not only that, but even with say, the budgets and money slung around ILM, you’ll still wind up with some pretty appalling looking fixes.  Oh, you say you want to know what these sorts of fixes look like even when you spend millions on them?  Have a look at this clip, 45 seconds in:

Star Wars Re-Edited

The scene you just saw was filmed when Jabba was just a humanoid like Han.  Short, fat, wearing furs and leather.  It added nothing to the film so the scene was cut from the film.  Once Jabba was around in the third film this scene had to be Ret Conned because Jabba could no longer be circled by a person without climbing all over him, or floating like an animated South Park character.

When Lucasfilm decided to add this scene to the “deluxe” versions of the film, the “mistake” had to be fixed in post.  You just could not reshoot it.  This was then a giant continuity error, something there really is no fix for and no way to prepare for when shooting.

When a continuity error is this bad, saying those four words is like a little child asking a parent to fix the pet that just died.  Some people think a lone editor can pull off the same results as ILM.  Sorry, I’m a magician, not a wizard.

I’m preparing to shoot a narrative short of my own, my first in a while, and it made me think long and hard about the importance of a continuity person.  In post production we can do a lot to overcome continuity errors.  We can flip negatives or resize the frame a bit to cut out a mistake but in the final analysis, the best thing is for the director to get it right on set.  This is why I’m so in favor of people learning to shoot on 16mm film.  Film teaches you not to mess around on set.  Film teaches you that it is expensive to get things wrong.  Film teaches you to get it right on set because when all you have is a razor blade, some tape and a wax pencil, there’s no fixing it in post.

Unless you’re George Lucas.


The Vicissitudes Of Media

I had a really unpleasant surprise this week when my computer, while in the midst of prepping media for an editing gig, decided to stop seeing one of two drives in a striped RAID configuration.  To fix this I had to unplug the hard drives in the RAID, plug them in with different cables (just in case) copy off the data to a safe, albeit slow hard drive and reformat the media drives, plug them in again with fresh cables, copy the media back and prep a second RAID just in case the first one decided to off itself again.  This has been going on now for nearly 3 days.

What a nightmare.  I have a new 2TB Western Digital Caviar Black HDD that I just formatted and put into Rosewill RX358 V2 enclosure in case I need the extra space again, only this time I did the smart thing and got a drive with some speed.  This one is 7200rpm whereas the other drive, which is still going after a day and a half (with about 8 hours left it says)  is as slow as slow can be and is infuriating me and making my clients uncomfortable.  I really hate USB 2.0 as well.  It’s too slow for me.  I want a USB 3 card but what’s the point when most devices still don’t use the format?  The nice thing about the Rosewill drives is they come with a PCI slot cover with a SATA cable attached so I could run it to the motherboard (I still have two SATA slots left) instead of running it to USB, but to install that I’d have to pull an existing PCI card because I’m using every slot.  I think there’s one slot left but the video card in my machine is so big that it takes up the slot.  What I feel like I need is another 3 PCI slots!  In the meantime, I am using both my RAID cards so I at least I can feel like I’m getting my money’s worth there.

But let me get back to the inciting incident.  What happened?  I was using Plural Eyes to sync up some media, I was ingesting new video into Avid Media Composer from a firewire Mac Drive and I was doing nothing else.  Suddenly one of the two drives in my striped RAID went down.  It still showed up on the RAID software, but was grayed out.  I suppose the Mac drive might have confused my computer or something in as much as I need to use third party software just to look at it, but in the end I couldn’t take the chance that it was a one time snafoo.  I didn’t have a spare RAID with the space for the media I was working with so I had no choice but to copy it to the slow external drive.  Had I known that freeing up the space on what I now have as a second RAID wouldn’t take more than a few hours, I’d have done that and copied everything over that way.  It would have been a lot faster, but given the amount of time it took to set up those drives and the amount of time it takes to copy nearly 1TB of data, I don’t know if it’d have made a meaningful difference in the time it took to set up.  Anyway, the damn USB drive is transferring data at 6.39 MB/sec and has 174 GB to do.  That’s 178176 MB.  That’s a lot of seconds.

Or about 8 hours.  Take your pick.

Anyway, I have regrouped the drives as a new RAID, reformatted them and checked for errors.  No errors.  So they’re getting copied to.  The data is safe.  I’m just angry that it took so long to fix something on a time sensitive project that was already taking forever because of having to sync up the sound to the picture.  Ugh.