-
New blog post from Philip Hodgetts. Worth the read.
-
Not including preproduction. All of the things you mention can happen while the script is being developed, even if it’s in flux. We do that all the time. Script is nowhere near done, but casting, location and other productions tasks are being worked on.
And if they tried this in the film days, what does that say for us now?
If fcpx can help complete tasks faster, then we will still have jobs, or the money holders will go find places that will get it done. I wish I could tell all of our clients we need more time. The best thing we can try to do is educate them for the next time, but it still doesn’t always work that way.
-
[Jeremy Garchow] “Just to be clear. Again. This is not about shot selection or “knowing” the footage. This is about the computers making certain parts of your job easier/faster.”
And that’s the part I agree with.
[Jeremy Garchow] “Does it rally take that much time for a cursory organization? Yes, you have to watch the footage, and tag it, but if I know these 17 clips are “Interview 01″ to get me started, what else do I need on day1?”
No, it doesn’t take much time for cursory organization.
You don’t get much value out of cursory organization, either.
[Walter Soyka] “The difference between searching FCPX and searching a good XLS log or DAM is minimal;”
[Jeremy Garchow] “I disagree. It’s in fcpx, immediately accessible for edit, instead of sifting through another application only to send over part of what you want, not exactly what you want. Assts Managers are great, but for quick hit editorial, not really.”
Fair, but again, I was trying to compare working with metadata across apps to not having metadata at all. Of course it’s faster to not have to change contexts. I’m not arguing against that!
FCPX adding metadata is a big deal, and it’s significant. I’m not trying to downplay that.
And what are we really talking about here? Aren’t we talking about automating editorial to save time?
Whether it’s a computer making a rule-based edit based on encoded metadata or a person making an intuitive edit based on their memory, or anything in between, the foundation is the same: what’s in the footage? That’s where metadata helps.
I’m saying that it takes time — significant time — to create valuable and meaningful metadata, which can then actually save time later when you’re actually cutting. I’d consider both of these tasks to be part of editorial today. Do you disagree?
[Jeremy Garchow] “And what about metadata in production? Panasonic has a wifi camera that lets you log while shooting, basically giving a rudimentary preorganized shoot to editors. That cuts prep time down even more. The monkey work that the computer does is more time for me to address the real time duties. Like Herb said, you can’t shoot and edit a film sketch in a day, but could you a tapeless sketch?”
My argument here against a technological solution to the time crunch is that it’s not permanent. You won’t get to keep the time you’ve saved. The production will take it back.
If enhanced metadata can cut a day of editorial off a week-long project, you don’t get to keep that day to make your project better. As soon as you only need until Thursday to deliver what you used to deliver on Friday, Thursday becomes your new deadline.
I am really not arguing against metadata. It’s a good thing. We need more of it, and we need more tools like FCPX that can use it.
I’m arguing that metadata is not a sustainable solution to the “not enough time” problem. I’m arguing that there is no technological solution to that problem, because (Soyka’s Law) “expectations rise at the same rate as capabilities.”
Walter Soyka
Principal & Designer at Keen Live
Motion Graphics, Widescreen Events, Presentation Design, and Consulting
RenderBreak Blog – What I’m thinking when my workstation’s thinking
Creative Cow Forum Host: Live & Stage Events -
[Walter Soyka] “And what are we really talking about here? Aren’t we talking about automating editorial to save time?”
Yes. And Philip is also saying that FCPX works faster (because it does).
[Walter Soyka] “I’m saying that it takes time — significant time — to create valuable and meaningful metadata, which can then actually save time later when you’re actually cutting. I’d consider both of these tasks to be part of editorial today. Do you disagree?”
Somewhat. I think the “significant time” is where this argument lies.
Ok, let’s break this down to a tv show.
The tv show is usually shot to a script. The script is usually marked with scene, and then there’s takes. No question that on set some of this stuff gets changed/reordered/rewritten.
Now, let’s just say that we used a camera that was able to wirelessly metalog on set. Scene, take, reel, and some other general information.
Now, let’s say Fcpx was more capable, or had greater third party support.
You feed fcpx or an application a final script with scenes, fcpx then lays out the entire script by scene (which of course the order could be user defined), and each scene now has an audition clip of all the takes per scene sorted by rating.
By no means is this a complete episode, but think of the amount of time that was saved right there. I am not advocating that this is the best way to go, but I bet it will be the way it skews, and that is part of this whole argument otherwise we’d still be editing reel to reel.
This will happen especially for younger folks for which computers and tagging are quite literally a way of life. They communicate by metadata, even if they don’t know what that word means.
Without an editor watching one single frame, the show is extremely and roughly put together as written. It obviously won’t stay that way, as things start to get fleshed out it might all change, but days and days of organization/prep was saved right there. It’s not that crazy. Editors will still have a very important role, no question, but asking for more time won’t be in our vocabulary for much longer.
-
[Walter Soyka] “(Soyka’s Law) “expectations rise at the same rate as capabilities.”
Well said! “Soyka’s Law” is a keeper.
_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl -
[Jeremy Garchow] “Without an editor watching one single frame, the show is extremely and roughly put together as written. It obviously won’t stay that way, as things start to get fleshed out it might all change, but days and days of organization/prep was saved right there. It’s not that crazy. Editors will still have a very important role, no question, but asking for more time won’t be in our vocabulary for much longer.”
Interesting argument. If I’m understanding correctly, it sounds like you’re proposing metadata as a way to potentially automate the manual work an assistant editor would do. Following Walter’s argument, the assistant’s task would shift to entering the metadata. Is this the basic idea?
_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl -
[Jeremy Garchow] “Now, let’s just say that we used a camera that was able to wirelessly metalog on set. Scene, take, reel, and some other general information. “
Can FCPX access the scene and take metadata that you can create using a KiPro? Legacy can’t and it’s somewhat frustrating to know the info is there and it can’t be used. Right now Legacy can only access reel#, timecode and clip name, which can be composed of the scene and take metadata, but you can’t sort or search independent of clip name.
Herb Sevush
Zebra Productions
—————————
nothin’ attached to nothin’
“Deciding the spine is the process of editing” F. Bieberkopf -
[David Lawrence] ” Is this the basic idea?”
Assistant editors, editors, whatever, so yes, more or less. Along with FCPX just being faster, in this example audition clips and the magnetic timeline come in to play.
Again, I’m not advocating this is the way to go. I’m not sure if it will turn out a better end product, but it will certainly speed things up.
Looking on the bright side of this, if your deadline is Friday, and you’re done shooting the Friday before, instead of having three prep days and four edit days, you might have six edit days.
Jeremy
-
All well and good, Jeremy — but there are too many assumptions in there for us to reasonably argue. Avid and Premiere Pro also already have tools for script-based editorial. That’s not my line of work, so I can’t really comment on how useful or not they are.
I agree with Philip that “faster” and “more automated” are good design goals. Who wants to work slower and harder?
I agree with Philip on the main point of his article: there’s a lot of real merit in the FCPX approach. I disagree with Philip on one point, too: I think it’s too early to call FCPX an inevitable winner right now. (I also think it’s too early to call it an inevitable loser. I’m currently more intrigued by Oliver Peter’s observation that different market segments are adopting different NLEs.)
I have a lot of respect for Philip and his work, and I’m trying to offer some thoughtful critical commentary on the ideas he’s sharing. I agree with him (and with you) that using metadata will speed up editorial. I also agree that as more editors adopt metadata-driven workflows that confer speed advantages, other editors must follow suit to remain competitive.
I’m trying to point out that actually creating that metadata is still largely a manual process, and that even if it’s informal for most editors today, it currently accounts for a lot of editorial time. If someone else creates the metadata instead of the editor, you are saving editorial time, but not total project time. The more an editor relies on metadata for editorial, the more important the quality of that metadata is. Prep still needs to go on the schedule somewhere.
I think that’s a reasonable objection to make. If Philip can meet it (which he certainly may be able to do), I think he’ll have a much stronger and more persuasive argument for others who think his claim might be a bit strong.
I’m further arguing that making the editorial process faster does not solve the time crunch problem. Will enhanced metadata tools let editors work faster? Yes! Will enhanced metadata tools give them extra time to focus on aspects of their jobs they must neglect now, due to time constraints? No! They’ll just be expected to deliver comparable results in less time than they have today. The ability to edit faster will just make post clients expect work sooner.
Where I lose Philip entirely is when he postulates that people are completing their projects four times faster — today — with FCPX. He’s saying that the industry is pre-destined to adopt FCPX because it’s — worst-case — twice as fast. Today.
He’s not qualifying it at all. He’s not saying who these people are or what sort of work these people are doing. He’s not suggesting there is any other path to this sort of dramatic improvement. He’s implying that anyone not using FCPX (unless they can’t use FCPX because they need “support for layered Photoshop files” or “selective copy and paste of attributes”) is wasting 50% to 75% of their time and their clients’ post budgets.
I’m challenging that. Shane is challenging that. I’m asking who is now taking a day and a half with FCPX to finish a week’s worth of work, because that idea is a little hard to believe. I’m asking if the scope of work under consideration is the same, or if some of that improvement comes by moving work that is still necessary outside of the area of consideration.
If it’s true, and if FCPX lets today’s editor do the same work four times faster, then great! Proving it would change a lot of people’s minds about the product.
Walter Soyka
Principal & Designer at Keen Live
Motion Graphics, Widescreen Events, Presentation Design, and Consulting
RenderBreak Blog – What I’m thinking when my workstation’s thinking
Creative Cow Forum Host: Live & Stage Events -
[Jeremy Garchow] “Ok, let’s break this down to a tv show.”
I think in this whole break down you are making some incorrect assumptions about workflow.
1. A film is never cut according to “best” takes decided on set. When I deliver a “first assembly” it is almost never made up completely of the “circle” takes. I have made my own judgement calls about what I think is best and that’s not always what was picked on set or location. Sometimes I’ll use pieces of all takes to build a scene, because an actor might have been better at the beginning of a “bad” take than in the supposed “best” take.
2. If an assistant has to load info anyway, then from the editor’s POV, both systems are equal. It may be faster for the assistant, but it’s irrelevant for the editor. Unfortunately there’s no way to know, because collaborative editing – as is possible in Media Composer, Lightworks or FCP 7 – is all but impossible in FCP X.
3. Not all scene/take data is always entered. If you are going from the written script supervisor’s notes, then that is already sorted out for you. Simply pick your favorites or subclip then – either way, one NLE is about as fast as the other.
4. If you are working on a film show, a lot of the scene/take info was already organized by the colorist. If you get an ALE or an XML, then no need to re-enter it. Many digital shows still go through a telecine process so the workflow is the same.
5. If the editor wants to use Avid’s ScriptSync, then FCP 7 or FCP X are completely inadequate. Although ScriptSync requires some assistant prep work before editing can begin, it has one distinct advantage over all other NLEs. You can click on a line of dialogue in a bin (script text is displayed) and all coverage can be reviewed for that line of dialogue, even though it occurs within the body of each media clip.
The reason to view all the footage is not to generate a bunch of metadata, it’s to make creative editorial decisions based on camera work and acting performances. One NLE isn’t going to make this any faster than another. It’s just whether you like one mode of organization or another.
– Oliver
Oliver Peters Post Production Services, LLC
Orlando, FL
http://www.oliverpeters.com -
[Herb Sevush] “Can FCPX access the scene and take metadata that you can create using a KiPro?”
Right now, no. But it will from Panasonic P2 material (as FCPX has an XML to go off of).
I expect this will change, but who knows.
Reply to this Discussion! Login or Sign Up