[David Lawrence]

[Bill Davis] “If you’re wedded to the idea of local storage of full-rez HD master files, I agree. But I got totally schooled last year on a series of HD uploads for network spots. Via modern H264 encoding I was able to achieve more than a 10 to 1 data reduction with no visible degradation of the signal for HD broadcast uploads.”

Bill,

H264 is an outstanding delivery codec. I use it for client deliverables all the time and have learned to get encodes that are virtually indistinguishable from the ProRes masters. One of the tools I use is the x246 codec. If you’re not using x264, stop reading this post, download the codec and make some compressor settings. Not only is it faster than regular h264, the quality is amazing.

[Bill Davis] “What’s wrong with ingesting field tapes via laptop into proxy files that can be uploaded to the cloud for use by distributed teams – those should be good enough to do most of the edit process directly to and from the cloud. Then when you get back to the base with your field cards, files, HDs or whatever, you could just plug them in and the background rendering starts (just like now) automatically replacing your comp files with full rez versions.”

Nothing. If you’re talking about a networked, proxy-based workflow, as @Tim Wilson points out, that’s already here. I was responding to this:

[Bill Davis] “What if the box no longer even matters. What if everything does eventually end up stored in the cloud?

What if “editing” is something you do merely by manipulating the metadata via the cloud?

Shoot wherever. Upload everything. “

At some point, you’ll need to re-link your proxies to your masters for finishing. Where will the masters live? On the cloud? I don’t think so. For high quality footage, this scenario is not gonna happen for a very very very long time, if ever. Local storage is simply faster, cheaper, more reliable and more accessible. And if anything, masters are only getting bigger. Wait till 2K and 4K gets more affordable.

On the other hand, bandwidth trends – especially in the US – are bleak. Take a look at this 2009 study on US data caps. Forget about media production, if you care about the viability of the cloud as a media consumption platform, you might want to keep net neutrality and data caps on your radar and let your elected officials know how you feel. This is a huge and defining battle and things are not getting better.

I can imagine field scenarios where FCPX – even in its current state – would be a fantastic organizing and assembly tool. But the benefits will come from being able to work quickly on a laptop in the field. The need for the cloud scenario seems like an edge case that for the most part is better solved in other ways.

Actually, one of the best distributed production collaboration tools I’ve ever used is unique to both Apple and FCP7 — iChat Theater Preview. I use it with clients and they are blown away at seeing their changes in real time. My LA partners and I collaborate with it on a regular basis. It’s relatively light on bandwidth and makes a lot of sense for field collaboration. Shoot in the field, edit in the hotel room, get feedback on the cut via iChat preview, iterate, rinse and repeat. iChat preview is probably one of the things I’ll miss the most when I finally leave FCP7. Now that Apple seems obsessed with full-screen layouts, I wonder when we’ll see something like this again.

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Chris Harlan] “Add to that the major Net Neutrality war over “last-mile” that will dominate the next decade. “

Outstanding point, Chris. Net Neutrality is a huge issue that isn’t going away. Nor will the political influence of the telecom and cable industry. The US lags far behind other countries in bandwidth and isn’t even in the top ten for downstream speed, let alone upstream speed. The idea that the cloud will be viable asset storage for billable productions anytime soon is laughable.

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Craig Seeman] “Keep in mind though ingesting to the cloud can be major time consuming given file size vs upload speed.
Once uploaded though, the ability to do long distance collaboration would be of value to some.”

Can be? LOL, My partners and I FedEx bare hard drives back and forth between SF and LA on a regular basis to share the terabytes of footage. We both have the fastest upstream bandwidth possible in our areas and it would take weeks to ingest to the cloud. Add to that the fact that every major US ISP is capping data and this cloud scenario becomes a total pipe dream for any kind of serious work. Sorry, not gonna happen anytime soon.

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Jeremy Garchow] “On a quick turnaround project if someone walks in and says, I need this quote from Gary, I can go right there, type “Gary” and start skimming, instead of finding/opening the interview bin and look around, double click the shot into the viewer, make sure it’s the right one, if not, go back to the bin, double click the next shot, etc. FCPX is fast.”

I think the thing that’s getting lost in this sub-thread is the difference between tagging and logging. They’re very different processes. Tagging is fast and general, but logging takes time, especially for interviews. You can speed up an interview, but only so much. You just have to put in the time and listen.

I like the organizational tools in FCPX. I think they’re one of the best things about it. I think the skimmer’s nice though the UI is sloppy (try making a range then clicking inside it, then try clicking outside it, oops!). At least that can be fixed.

So that said, for b-roll tagging/rough organization, these new tools are great. For logging though, I find the skimmer doesn’t add much. Still need to J K L thru the material. In your example above, if you had ten “Gary” interview clips, the skimmer won’t help you find a quote unless you know the material. Once we have this, maybe it will be less of an issue. But until then, tagging only gets you part way there. Someone still has to take the time to listen.

Unless you’re going to just start throwing things on the timeline and start cutting in place. I’ve been on high-pressure projects where that was definitely the case. No time to organize, just start cutting now. Not my favorite way to work but sometimes there’s no choice. In this case, tagging goes out the window and I really need to use the timeline as a scratchpad – but that’s another conversation 😉

I do like being able to favorite different soundbites in the same clip, though. That’s a win.

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Walter Soyka] “FCPX must ultimately place clips at specific points in time, which is the way every other application on the planet manages time-based media. FCPX’s extra clip relationship data will be lost in the process.”

I think the point that I (and maybe @Gary) am not getting is what is it about this extra relationship data that is in any way meaningful about the editor’s intentions? Especially since all relationships must be expressed in reference to the primary storyline.

For example, if it were possible to connect clips to secondary storylines, or even other clips, then I’d be able to create meaningful relationships that actually reflect my grouping intentions. But given the constraint of everything having to connect to the primary storyline, whether it makes sense or not, I just don’t see any meaningful value.

Jeremy’s recreation of Kevin Federline’s timeline is a good example of this. Even though the timeline looks the same on the surface, the relationships have nothing to do with the editor’s intentions.

If we assume that:

V1 = primary
V2 = connected at In point IF no transitions
V2 = secondary connected at In point IF with transitions

rinse and repeat for both video and audio…

Doesn’t that get us 90 to 100% there for a useful percentage of cases? What would we really be losing?

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Jeremy Garchow] “I talk about it in the article, with screengrabs! :0) It is preliminary “bucket” organize. ie throw related clips in a “bin”.”

Ah yes. I Just saw the thumbnails at first and missed the keyword list. Yes, that’s something I like in FCPX a lot. In fact it’s even easier to favorite regions than you say, just hit F, no need for the control key!

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Jeremy Garchow] “I can organize over 1000 clips in 15 minutes or less.”

Can you be a bit more specific about what you mean by “organize” in this example?

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Walter Soyka] “If you guess at the primary storyline and connected clips, but you guess wrong about the connection points, or guess wrong about what’s primary and what’s connected, the rippling will be wrong.

FCPX works because it encourages you to establish these relationships in context at the time you make the edit. I don’t think an imported timeline missing these contextual cues would feel like an FCPX timeline which relies on them.”

OK, makes sense.

But the more I think about it, I’m not sure it would be that big a deal. FCPX forces all connections to be in relationship to the primary. I find that to be one of its biggest constraints and weaknesses. What if the relationships expressed by track groupings have nothing to do with a notion of “primary”? As long as vertical and horizontal relationships in time are maintained, grouping wouldn’t matter and could left for the editor to decide after import.

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Walter Soyka] “If you make mistakes, the magnetic timeline features still won’t work right, but like many here, I’d certainly rather have my edit in with limited magnetic functionality than not have it at all.”

Walter — just curious, what specific features are you thinking wouldn’t work right?

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl

[David Lawrence]

[Mark Morache] “Being of the “there must be a pony in here somewhere” school of philosophy, I’m expecting that the long delay from Apple for an update to FCX is because of the extra work it’s taking them to put the tracks back in the timeline. “

Oh how I wish you’re right about that. Not holding my breath tho…

[Mark Morache] “Seriously, they could have created their non-conflict timeline with tracks. They could have kept a version of connected clips by locking clips together, in tracks. They could have given us 64bit with backround rendering, crazy powerful metadata, even a version of the magnetic clips, and still given us our tracks. “

Word. The UI is an abstraction of the guts under the hood. The trackless magnetic timeline is a design choice, not an engineering choice. Apple could have absolutely kept tracks if they wanted to. In fact there’s evidence in the program itself that at some point tracks were there. I would love to have been a fly on the wall when they decided to dump this for their “magical” new paradigm.

_______________________
David Lawrence
art~media~design~research
propaganda.com
publicmattersgroup.com
facebook.com/dlawrence
twitter.com/dhl