[Paul Dickin]

[Craig Seeman] “Paul, that’s what I meant by common timeline integration.”
Hi
It all comes down to semantics 😉

[Craig Seeman] “Motion specific assets…”
I reckon those are gone. In future everything will be a shared asset.

[Craig Seeman] “I’m not sure what’s likely given that Apple hasn’t mentioned or shown…”

I reckon Apple have been very shrewd and focussed with their Sneak Peek.

They knew there would be a barrage of reaction (mostly anti) from the existing post production fraternity about the changes in the new software, and have carefully dropped enough hints for the most forward thinking brains (like Larry Jordan and Philip Hodgetts) to get some sort of damage limitation out there before the official FCP X announcement.

Which is what is happening all over the FCP-focussed parts of the web.

[Paul Dickin]

[Craig Seeman] “At the very least it will remain that way but it seems possible they’ll be integration at least with a common time line.”
Hi
I think that is very much a “faster horses” way of looking at it 😉
It seems that going forward it will be the core components of the OS (rather than separate applications) which will be handling everything – components like AVFoundation, CoreData, CoreAnimation etc etc.

All that is required of a new ‘application’ like FCP X or son-of-Motion etc will just to be a GUI interface to manipulate the basic media assets using the OS-provided toolset.
So everything can only be ‘totally integrated’, and as a ‘timeline’ will be a low level ‘composition’, defined by its own metadata, any GUI can work on it any way you want.
It seems to me? 😉

Obviously they have got to ‘make-this-all-happen’, which may take a while.
Possibly until 2013 according to some rumours….

[Paul Dickin]

[Martin Curtis] “…converting to ProRes really really fast”.”
Hi
ProRes in this context is a “faster horse” – which is not what FCP X is about…

All FCP X needs to do is convert the source media into a frame-by-frame display file that the computer graphics card can show you – at whatever scale you require.

Here’s what Apple have published about the process, using their QuickTime replacement called AVFoundation, which is an OS X/iOS component, independent of FCP:
Editing.
AV Foundation uses compositions to create new assets from existing pieces of media (typically, one or more video and audio tracks). To assemble audiovisual constructs from one or more source assets, you can insert assets into instances of AVMutableComposition…. offering insertion, removal, and scaling operations…

You use a mutable composition to add and remove tracks, and adjust their temporal orderings. You can also set the relative volumes and ramping of audio tracks; and set the opacity, and opacity ramps, of video tracks.

A composition is an assemblage of pieces of media held in memory. When you export a composition using an export session, it’s collapsed to a file.“
https://developer.apple.com/library/ios/#DOCUMENTATION/AudioVideo/Conceptual/AVFoundationPG/Articles/00_Introduction.html%23//apple_ref/doc/uid/TP40010188
(my emphasis).

So they’re saying everything is in memory until such time as you choose to Export it into whatever format you choose.

What the sneak peek showed us was a GUI representation that exactly corresponded to the description I’ve quoted. Each ‘clip’ or ‘combined clip’ we saw Randy working on would have been a graphical timeline instance of a complex ‘composition’, which is an AVFoundation file that has been created on-the-fly in memory to allow editing to proceed.

There seems to me to be no reason for it to have to exist ‘in memory’ in some legacy FCP 7 compressed codec during the intermediate editing stages.

It remains to be seen whether the Export process refers back to the original source media, to maintain the highest quality (which would take time I guess), or if the intermediate memory renders which have been created for editing are used.
Any straightforward ‘first generation’ transcoding will involve decompressing into memory, so maybe the FCP X editing intermediate files could be more-or-less ‘invisible’, just like a normal format-to-format re-encode?

Disclaimer: I’m not an expert in any of this, I’m just looking at what Apple showed us, and reading what they’ve published. To me it all makes excellent sense 🙂

[Paul Dickin]

[Chris Kenny] “You might have thousands of merged clips. I don’t think you can just dump all of them in a sequence and work from there. You have to be able to work with them the same way you can work with standalone clips, prior to editing them into a sequence.”

[Andy Mees] “…I’m suggesting that the internal data model of a sequence (ie AVComposition) would certainly seem to be able to be used to describe (and therefore to map) these otherwise ‘virtual’ clips…”
Hi Chris and Andy
Thanks for the clarification- I’m feeling for a non-technical understanding of all this. 😉

It seems to me that what we call ‘sequences’ in FCP 7 will in FCP X exist as OS X/iOS system-level AVMutableCompositions – which look to me in theory to be capable of becoming – recursively – as complex as necessary…

Everything contained within the ‘sequence’ will be either AVAssets (clips), AVCompositions (merged clips) or AVMutableCompositions (combined clips and other nested-type editing blocks). This would include multi-cam capability, which would just be another multi-video-track AVMutableComposition.

All this is going on at OS level, and the database to define all this myriad of AVAssets/Compositions/MutableCompositions has to be part of the OS’s AVFoundation methodology.

So FCP X’s role in this process is
a) to provide a GUI to manage the development of this data – creation of new assets/compositions etc etc
b) to provide tools to modify the underlying AVAsset data – colour correct it etc etc
and
c) maybe to be a database-management supertool to handle the extensively layered datasets that the editing process will require of the OS.

So the ‘project’ in future will not at all like it has been done in the past 🙂

[Paul Dickin]

Hi
I’ve had a look too. Interesting stuff…
I’ve already posted my comments:
https://forums.creativecow.net/readpost/335/1238
https://forums.creativecow.net/readpost/335/1242

[Paul Dickin]

[Bret Williams] “we’re discussing whether FCP X will be able to import a FCP 7 project… What are you discussing? “
Hi
I was speculating that rather than FCP X being lumbered with extra bloat to enable the ‘import’ of the XML (or native code or whatever) from an FCP 7 project it might be more elegant to put the burden of that export onto an ‘enhanced’ FCP 7 – given that it seems in all probability to require a major media asset repurposing.

And given the likelihood in future of huge additional FCP X sales to non-FCP 7 users. And the fact that under an App Store download distribution model it would be of benefit to minimise bloat.

The whole whether-it-will/whether-it-won’t tone of this forum seems scare-mongering and specious to me – the facts as we have them from the sneak peek and Apple’s Developer web site are enough to make it clear that Apple is continuing its Pro Apps commitment 😉

What on earth are consumers going to need Closed Captioning functionality for home movie creation on their iPads for? AV Frameworks enables full functionality for continued professional media creation. IMO… 🙂

Also I am speculating about some of what Philip Hodgetts might have to say in his forthcoming pre-announcement FCP X Webinar, where my reply is OnT I thought – rather than cross-posting all over the forum 😉

Edit: I realise I’m probably not the Paul that Brett’s comment was aimed at – but I was the last Paul to post….

[Paul Dickin]

Hi
I’ve spent a while going through Philip Hodgetts’ blog, and his linked blog/presentation by Chris Adamson about AV Frameworks (Apple’s QT replacement methodology), most of which is above my head 😉 but some sentences in Apple’s developer pages about AV Framework are written (more or less!) in clear English…

Quote:
“The primary class that the AV Foundation framework uses to represent media is AVAsset…
AVAsset is not tied to particular data format. ……such as a QuickTime movie file or an MP3 files (amongst other types).

Each asset contains a timed collection of tracks that are intended to be presented or processed together, each of a uniform media type, including but not limited to audio, video, text, closed captions, and subtitles… Assets may also have metadata.
In a complex composition, however, there may be multiple overlapping tracks of audio and video.

Editing.
AV Foundation uses compositions to create new assets from existing pieces of media (typically, one or more video and audio tracks).
To assemble audiovisual constructs from one or more source assets, you can insert assets into instances of AVMutableComposition…. offering insertion, removal, and scaling operations…

You use a mutable composition to add and remove tracks, and adjust their temporal orderings. You can also set the relative volumes and ramping of audio tracks; and set the opacity, and opacity ramps, of video tracks. A composition is an assemblage of pieces of media held in memory. When you export a composition using an export session, it’s collapsed to a file.”

That’s all (cut/pasted) from:
https://developer.apple.com/library/ios/#DOCUMENTATION/AudioVideo/Conceptual/AVFoundationPG/Articles/00_Introduction.html%23//apple_ref/doc/uid/TP40010188

Chris Adamson has a slide that summarises that:

AVAsset
A collection of time-based media data
Sound, video, text (closed captions, subtitles, etc.)
Each distinct media type is contained in a track
An asset represents the arrangement of the tracks. The tracks
represent the traits of the media’s presentation (volume, pan, affine
transforms, opacity, etc.).
Asset ≠ media. Track ≠ media. Media = media.
Also contains metadata (where common to all tracks)
https://iphonefall2010.crowdvine.com/system/talks/presentations/000/014/595/chris_adamson.pdf?1289428982

That description of ‘how the new way works’ seems to me to correspond pretty much with what we saw happening on the sneak peek timeline, the various clips on the timeline GUI representing what’s being described in my quote.

That means a FCP X ‘project’ is just a database document, and all the ‘editing’ is going on in the continuing creation of new complex AVAssets.

So to import a FCP 7 project is going to require a mammoth reconversion process to create the new AVAsset ‘compositions’ needed to work with the old FCP 7 project converted into a FCP X database.

I would guess this will be best handled by a new transitional point version of FCP 7 to handle the conversion procedures, to be distributed after FCP X is made available?

[Paul Dickin]

Hi
Software Update could deliver a ‘Pro App Update’ to install system level stuff as in the past?

[Paul Dickin]

[Chris Kenny] “…the Event Library is rooted on the storage device, with no reference to a specific project file…”
Hi
As is all the organised metadata.
Does everything – media management, clip timecode and metadata – all get handled at a global (total) level by the OS rather than the application?

(Which sort of is the current FC Server’s methodology, as I understand it).

[Paul Dickin]

“It takes a worried man to sing a worried song…” Woodie Guthrie et al.

[Craig Seeman] “…FCPX 3, two years from now if they’ve have everything worked out “
Hi
That’s about it, in business planning terms, I reckon.
Same as the old FCP – it wasn’t until v2 that it became usable, it wasn’t until v3 that it became fully featured, and it took the v4 rewrite to make it capable of becoming an ‘industry standard’.

So FCS 3/FCP 7 (or A. or A.) will have to do for the foreseeable future until FCP X 3 or 4.1 comes along 😉
In business planning terms Apple are likely to be secretive because there’s a lot still to do.
To some extent that relies on other worldwide technical standards being updated to be fit for the 21st century – like timecode (and, as mentioned elsewhere, EDLs etc…)

Seems reasonable to me to apply old tools to old standards, yet look forward to hugely improved solutions to doing things more efficiently.

“…I’m worried now, but I won’t be worried long.” 🙂