New possibilities from Production 2.0?

Production 2.0

The delayed “Production 2.0” event… took place in Soho, London last night. There was nothing to get that excited about, certainly not much on the nature of digital production… I haven’t covered previously.

The organisers presented a workflow using a Panavision Genesis that basically allowed rushes to be viewed immediately after shooting on a laptop, projector, or even an iPod. Most of this is thanks to the Codex Digital Recorder, a disk-based uncompressed video recorder that can transcode on-the-fly to a variety of different formats. All good stuff.

Also present were the Hat Factory to provide on-set VFX and editing capability, though that was very much a case of –insert VFX facility here– rather than them presenting anything in the way of innovation. Also present were transmissions bods Sohonet, though their exact role in all of this was very unclear. I would guess that if it’s your aim to bounce data around the world, that’s where they can help. I certainly wouldn’t consider them an integral part of the system though.

It was interesting to actually see it all come together in the flesh as it were. There were no apparent hiccups anywhere along the line, it seemed to work fairly smoothly (although we were practically in laboratory conditions), and I have no doubt that Codex Digital can in fact deliver on what they’re offering (although I am still waiting for the promised email to say that the footage from the event is available to view online).

Also of interest was the discussion about the workflow for the Wachowski (siblings?) forthcoming film, “Speedracer”. They made use of up to 7 Codex Digital Recorders, and their workflow was to send data off to the four corners of the Earth after each shoot, where it was colour-corrected, composited, and edited overnight as needed, and then sent back. At the dailies session the next day, the results were auto-conformed, and the production was able to watch a segment of the finished film rather than individual takes in isolation. Absolutely incredible, but I can almost feel the pain and heartbreak that the overnight crew must have gone through to make it happen.

The entire event left me wondering what the actual, tangible benefit of all of this really is. The only conclusion offered by the seminar was that it allows things to happen faster. “And it’s easy”, but as the fellow sat next to me pointed out, “of course it’s easy if you’re the designer of the system”. Because, at the end of the day, is browsing through a set of folders and files to find a shot as easy as spooling through a tape for “most” people? It remains to be seen.

UPDATE: The edited highlights can be viewed online now… 

Posted: January 31st, 2008
Categories: Opinion
Tags: , , , , ,
Comment from Miguel - 2/18/2008 at 7:55 pm

Hi Jack

Thanks for the feedback. There are many points I could respond to, but maybe the obvious one to answer is “what was the point?”

I think the main point is that this isn’t nice ideas, this apparently seamless workflow from camera to post is just that – seamless. What you saw is production-proven, and not by the system designers, as we are no longer production people – but we have been, and we set-out to improve things.

For instance, on Speedracer, the way you write it is as if the pressure was put on the editors and artists by the workflow. Quite the opposite, the demand was already there from the Director and Producer, we were able to make their lives much easier by eradicating delays and uncertainty. Look at the quote on the current (updated) website from the senior AE, he certainly felt it was of benefit, as have directors, producers, DoPs and VFX supers.

I would welcome the chance to chat over some of the issues you feel are of concern, but one other area you raise here, which is worth addressing publicly is of organisation. Is it easier to spool through a tape or look through thousands of files? The answer is, and I guess this is central to your interest, it depends on how well you are organised!

If you don’t log onset, you will have to look through all your rushes in real time anyway, and need access to a tape deck in order to do so. At that point you will either make a log, or digitise as you go along – increasing the time required by having to shuttle, pre-roll and capture. All in all, a very lengthy process we are all familiar with. If you have logged onset then you have the option of manually inputting that log into your NLE and batch digitising – still a much longer than real-time process. With the workflow that was shown at Production 2.0 someone can log onset (even just to the simple level of marking circle takes) and the database can automatically create the required post format based on the logs. That can then be imported into the NLE in a faster than realtime process, at which point the editor and/or Director can view the shots, with all the metadata attached in a far more intuitive manner than shuttling up and down a tape.

Hope that helps…

Comment from Jack - 2/19/2008 at 1:13 pm

Hi Miguel,

Thanks for your comments. Certainly this workflow, when done right, is superior to a more traditional workflow, not to mention potentially faster. In the long-term this is definately the way to go, and if were up to me, it would be my first choice to use a workflow like this.

The main problem as I see it, and it’s certainly a short-term one, is that people are thrown into the deep end. Yes, it’s easier for people who know what they’re doing, in a well-organised environment, to work like this, but for everyone else, there may be a huge learning curves for all involved who have a limited knowledge of computer systems (and I include experienced editors in that category). Yes it beats the hell out of shuttling up and down a tape, but then that tape’s not likely to accidentally delete itself (or any of the other thousands of things that can happen)- things that people inexperienced in the digital arena may be confounded by when the first start out.

I’m absolutely not trying to say that we should stick with video because it’s easier, I’m saying that the technology will take a long time before people are as comfortable with it as with anything else.

Add your comment











*/ ?>
Comment from Miguel - 2/18/2008 at 7:55 pm

Hi Jack

Thanks for the feedback. There are many points I could respond to, but maybe the obvious one to answer is “what was the point?”

I think the main point is that this isn’t nice ideas, this apparently seamless workflow from camera to post is just that – seamless. What you saw is production-proven, and not by the system designers, as we are no longer production people – but we have been, and we set-out to improve things.

For instance, on Speedracer, the way you write it is as if the pressure was put on the editors and artists by the workflow. Quite the opposite, the demand was already there from the Director and Producer, we were able to make their lives much easier by eradicating delays and uncertainty. Look at the quote on the current (updated) website from the senior AE, he certainly felt it was of benefit, as have directors, producers, DoPs and VFX supers.

I would welcome the chance to chat over some of the issues you feel are of concern, but one other area you raise here, which is worth addressing publicly is of organisation. Is it easier to spool through a tape or look through thousands of files? The answer is, and I guess this is central to your interest, it depends on how well you are organised!

If you don’t log onset, you will have to look through all your rushes in real time anyway, and need access to a tape deck in order to do so. At that point you will either make a log, or digitise as you go along – increasing the time required by having to shuttle, pre-roll and capture. All in all, a very lengthy process we are all familiar with. If you have logged onset then you have the option of manually inputting that log into your NLE and batch digitising – still a much longer than real-time process. With the workflow that was shown at Production 2.0 someone can log onset (even just to the simple level of marking circle takes) and the database can automatically create the required post format based on the logs. That can then be imported into the NLE in a faster than realtime process, at which point the editor and/or Director can view the shots, with all the metadata attached in a far more intuitive manner than shuttling up and down a tape.

Hope that helps…

Comment from Jack - 2/19/2008 at 1:13 pm

Hi Miguel,

Thanks for your comments. Certainly this workflow, when done right, is superior to a more traditional workflow, not to mention potentially faster. In the long-term this is definately the way to go, and if were up to me, it would be my first choice to use a workflow like this.

The main problem as I see it, and it’s certainly a short-term one, is that people are thrown into the deep end. Yes, it’s easier for people who know what they’re doing, in a well-organised environment, to work like this, but for everyone else, there may be a huge learning curves for all involved who have a limited knowledge of computer systems (and I include experienced editors in that category). Yes it beats the hell out of shuttling up and down a tape, but then that tape’s not likely to accidentally delete itself (or any of the other thousands of things that can happen)- things that people inexperienced in the digital arena may be confounded by when the first start out.

I’m absolutely not trying to say that we should stick with video because it’s easier, I’m saying that the technology will take a long time before people are as comfortable with it as with anything else.

Leave a Reply

Your email address will not be published. Required fields are marked *