-
Be STILL my BEATING HEART.
-
David Lawrence
“Imagine being able to use a laptop in the field to log and tag footage instantly, do quick assemblies and rough cuts, etc. Then export and finish in a more powerful, flexible, editorial system. Something like that would rock.”The Foundry’s Storm already has that covered and more.
twitter.com/illyalaney
nextLAB Mobile
SpeedGrade DI
Resolve
da Vinci 8:8:8 Renaissance
Color -
Check out Nuendo. A lot more bang or the buck. And when I was doing testing a year or 2 ago I found it read FCP OMFs better than Logic. Go figure. And use whatever vid and aud hardware you want. One of my favorite apps ever as you can map the keyboard however you want and the editing just makes so much sense you’ll become annoyed with other apps.
Robert Brown
Editor/VFX/Colorist – FCP, Smoke, Quantel Pablo, After Effects, 3DS MAX, Premiere Prohttps://vimeo.com/user3987510/videos
-
-
Gary, what’s with the aggressive tone? We disagree, and you’re describing me and my arguments as “full of it,” “leaving outs,” “dumb,” and “magic?” If we’re going to disagree, let’s do so with civility.
[Gary Huff] “FCPX does not, at least to my knowledge, use any sort of artificial intelligence engine to establish this context. The “context” of which you speak is a simple run of IF…THEN…ELSE statements in the code (more or less) that governs the behavior. Whether it does it at the moment that you place clips on the timeline, or whether it does it later with clips that are arranged in time by another NLE, the end result is the same. Undesired linkage could easily be fixed after import.”
You seem to be suggesting that the FCPX editorial paradigm is not fundamentally different from the FCP editorial paradigm, so translation is straightforward. I disagree here.
I think it’s roughly analogous to raster drawing versus vector drawing. Can you make the same image with both methods? Yes. Can you rasterize a vector drawing? Yes. Can you vectorize a raster drawing? Not nearly as well — the translation may look right, but it’s not going to be easy to modify because the contextual information about how the strokes were drawn doesn’t exist in the raster drawing. You’ll spend a lot of time cleaning up a vectorized drawing that came from a raster drawing.
I don’t think FCPX (or Illustrator) has artificial intelligence. I do think it has additional features and data structures that FCP7 (and Photoshop) do not, and that the way you manipulate edits (and curves) use those features and data structures. The raster-vector translator in my example above had to guess at what the vector app could have actually tracked in the first place.
I agree with you that there’s no reason why FCPX shouldn’t be able to take a legacy project in and place the clips correctly in time. Jeremy, Mark, and David and others have all outlined ways this could be done. Of course I think that “undesired linkage” could easily be fixed after import. I also think that FCPX already has all the tools you’d need to hand-magnetize a timeline.
I’m essentially offering an apologist argument to rationalize why Apple hasn’t included — and have stated that they won’t include — a much-needed legacy import feature: a translated timeline wouldn’t feel like a “proper” FCPX timeline without a lot of manual intervention.
I think I’m with you on this — I don’t need Apple to protect me from myself. I’d gladly work over a non-native timeline to get the edit in. So would everyone else on this thread, I’m sure. I’d rather hand-magnetize a timeline than eye-match it. Add it to the list of feedback Apple doesn’t seem to be considering.
[Gary Huff] “It’s curious to me because you can’t seemingly claim that all of this will be messed up when FCPX will import iMovie projects, so that must be an admission that iMovie has a lot in common with FCPX under the hood if that sort of importer was provided on Day 1. Yet there are those who claim that it only bares a superficial resemblance with the UI, so which is it?”
I don’t know. I haven’t used iMovie. Since FCPX can import iMovie projects and not FCP projects, I’d assume that iMovie’s architecture is far closer to FCPX than FCP7.
[Gary Huff] “Yet another out. No one is saying that an importer will do a 100% perfect job time in and time out. That would be dumb. However, I will say that an importer will be able to import a lot of projects that will play correction from time 0:00:00 to the end. There will probably need to be some connections made, but that’s it.”
I’m not looking for outs — I just see some shades of gray where you see black and white.
I agree that an importer could easily make an FCPX timeline that would play identically to the FCP timeline it was imported from. That just involves placing clips correctly in time. It just wouldn’t be ready to edit, as if the piece had been originally edited in FCPX with an FCPX methodology.
[Gary Huff] “Are you really unable to grasp that most of what FCPX magnetic timeline does it automatic linkage? You can still do the exact same thing in any current NLE, you just have to do it manually.”
There is nothing like a clip connection with an arbitrary linkage point in FCP7, Avid, or Premiere Pro, and because of this, you cannot replicate the full behavior of the magnetic timeline in anything but FCPX. You can come close in many cases, but not all; FCPX’s features and structures make the new timeline’s logic fully internally consistent.
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] “I agree that an importer could easily make an FCPX timeline that would play identically to the FCP timeline it was imported from. That just involves placing clips correctly in time. It just wouldn’t be ready to edit, as if the piece had been originally edited in FCPX with an FCPX methodology.“
Hi
I guess this may be significant – in legacy FCP the ‘project’ gets built in the project’s internal XML (or code} whilst the media ‘assets’ sit unchanged on the hard drives.
In FCP X the ‘project’ (=timeline) is built cumulatively into a database that is part and parcel of the AV Foundation asset database(s) – which is a developing process as the timeline is built in the editing process.So the ‘database’ construction going on in FCP X seems to be a relational order of magnitude more complex than in legacy-FCP – and part of that complexity may relate to the interaction with AV Foundation data being created in the editing process that is additional to the sum of the individual media assets in the project. So which bit of legacy-FCP project data goes where in FCP X’s data structures may be problematic?
[Jeremy Garchow in his article] “Just recently, an Applescript was released by foolcolor.net called foolcut (https://www.foolcolor.net/foolcut). It takes your FCPX timeline and transfers it over to Adobe After Effects… This is done with scripting and no API..”
At this point in time legacy-QuickTime is inherently compatible with AV Foundation – foolcut’s Applescript FCP X database-querying powers show that – but will that always be the case?
[Jeremy Garchow in his article] “My biggest gripe with FCPX is the lack of native digital media format support. It is time to realize that the world is not a Quicktime world, and hasn’t been for some time, and that’s OK! …FCPX needs to embrace the non-Quicktime wrapped formats (MXF, R3D, MTS, etc) and allow for direct access to these audio and video essences, instead of wrapping them in yet another proprietary format.”
Is legacy-QT compatibility another ‘transition process’ – like OS 9 Classic was, or Rosetta? Something that will be withdrawn at a later date? As Jeremy points out there is some sort of kludge currently going on…
Apple maybe don’t want to allow something now that they know they will break further down the road?
-
[Walter Soyka]Gary, what’s with the aggressive tone? We disagree, and you’re describing me and my arguments as “full of it,” “leaving outs,” “dumb,” and “magic?” If we’re going to disagree, let’s do so with civility.
First of all, “full of it” and “dumb” wasn’t even in reference to you, so thanks for pulling in two of those to attempt to bolster your “poor me” defense. And please don’t resort to being so incredibly sensitive that “leaving outs” and the “magic” crack are as offensive to you as “glee club” was to others. And, BTW, yes I’m aggressive, but I try to be fair and if you make a good point, I’ll concede it. But so far, most of your points seem to demonstrate an ignorance of software development.
I think it’s roughly analogous to raster drawing versus vector drawing.
It’s not even roughly, more like not at all. Vector-based drawing takes place at the moment of media creation. FCPX is not there in the camera when I’m shooting video, it comes in later when the process of dealing with video is the arrangement of it, and it works its timeline by automatically linking what it thinks (by IF…THEN…ELSE rules on a fundamental level) are proper linkages.
And it sometimes doesn’t even get them right when editing within it.
I’m essentially offering an apologist argument to rationalize why Apple hasn’t included — and have stated that they won’t include — a much-needed legacy import feature: a translated timeline wouldn’t feel like a “proper” FCPX timeline without a lot of manual intervention.
That’s fine and all. I would say that, from my point of view, Apple didn’t include one because FCPX is not the successor to FCP7. Instead, it is the “pro” version of iMovie. A newly formed Apple product, not an upgraded version of someone else’s code. I think to argue otherwise is really to bury your head in the sand (IMHO).
I don’t know. I haven’t used iMovie. Since FCPX can import iMovie projects and not FCP projects, I’d assume that iMovie’s architecture is far closer to FCPX than FCP7.
And this wasn’t so much referencing anything that you have particular said, but others who had tried to make the argument that iMovie and FCPX have nothing in common except a superficial resemblance in UI.
There is nothing like a clip connection with an arbitrary linkage point in FCP7, Avid, or Premiere Pro, and because of this, you cannot replicate the full behavior of the magnetic timeline in anything but FCPX. You can come close in many cases, but not all; FCPX’s features and structures make the new timeline’s logic fully internally consistent.I never said “arbitrary”, I said you could duplicate it by linking manually, which I can. I still don’t understand why you cannot see that FCPX’s timeline is just like the others, except with more automation of linking clips. That’s all it does. And sometimes it messes up as well.
-
[Gary Huff] “First of all, “full of it” and “dumb” wasn’t even in reference to you, so thanks for pulling in two of those to attempt to bolster your “poor me” defense.”
Well, since I was the one who had the original discussion with you about how hard it was to get stuff into FCPX in the first place, I assumed I was included in your original statement, “How about all those people that said that FCPX’s timeline was so “revolutionary” that it couldn’t translate older projects or out to other formats and so on. Guess they were simply full of it, eh?”
I pulled “that would be dumb” directly from one of your subsequent responses to me.
And “poor me” is not a defense. I’m happy to debate you on the merits, but I expect you to extend me the same courtesy.
[Gary Huff] “And, BTW, yes I’m aggressive, but I try to be fair and if you make a good point, I’ll concede it. But so far, most of your points seem to demonstrate an ignorance of software development.”
I’m asserting that the data in an FCPX timeline is a superset of the data in an FCP timeline.
How does that suggest an ignorance of software development?
[Gary Huff] “Vector-based drawing takes place at the moment of media creation. FCPX is not there in the camera when I’m shooting video, it comes in later when the process of dealing with video is the arrangement of it, and it works its timeline by automatically linking what it thinks (by IF…THEN…ELSE rules on a fundamental level) are proper linkages.”
I think that editorial is a creative process in and of itself. FCPX does not automatically link things willy-nilly with a series of if-then-else statements. It links them according to the decisions the editor makes, and the specific FCPX tools he or she uses, while editing: cut it into the primary storyline? Pull it out of the storyline? Connect it? Change the connection point? Position it? Compound it? Storyline it?
All these things affect the context of the clip and its relationships to other clips — and the context is both totally plastic and relevant to subsequent operations.
I picked the vector curve example specifically because FCPX doesn’t just represent the composition and output of the edit (as the FCP timeline does, and as a raster image does) — it also represents some information about how the edit was actually constructed, via storyline designation and clip connections, then it uses that information to guide the mechanics of subsequent editorial.
[Gary Huff] “I never said “arbitrary”, I said you could duplicate it by linking manually, which I can. I still don’t understand why you cannot see that FCPX’s timeline is just like the others, except with more automation of linking clips. That’s all it does. And sometimes it messes up as well.”
FCPX’s magnetic timeline may not always do what you intend, but it will always do what you tell it. It won’t mess up. It follows its rules consistently, and garbage in will make garbage out.
Do this in FCP7: Imagine a sound effect with with some build up leading to a crash. You want the crash to hit on a specific frame, shortly after an edit point, but the build-up begins before the edit. You lay the SFX in and adjust the clip connection point so the audio and visual hits match. Now, with FCPX, if you insert a new clip at the edit point right before the hit, the hit stays linked where it is supposed to, and the build-up will still correctly proceed the hit over the new clip.
FCPX manages the context of how the two clips relate — not just their absolute points in time — and that’s an interesting new tool for editorial, and a complication for meaningful import of legacy timelines.
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]I pulled “that would be dumb” directly from one of your subsequent responses to me.
Here’s the original context:
I thought I was pretty clear that the “dumb” would be anyone claiming that a FCP7 importer would be able to do a perfect job all the time. My point being would that it would be dumb to say that, which is why no one is. If you still insist that I called you dumb, then I’m simply flummoxed.
I’m asserting that the data in an FCPX timeline is a superset of the data in an FCP timeline.
How does that suggest an ignorance of software development?
Because you couch it in terms that FCPX is developing “context” for connecting your clips instead of just following a pre-determined IF…THEN…ELSE path that can be applied at, really, any time instead of merely at the initial moment of arrangement. You seem unable to understand how this would be a accomplished from a strictly programming point of view, which makes your perception of how an importer would work very limited.
[I]t also represents some information about how the edit was actually constructed, via storyline designation and clip connections, then it uses that information to guide the mechanics of subsequent editorial.See? Again with couching a simple process in language that implies FCPX is something of being “aware” of what is going on. It connects clips based on its proximity to others and where on the timeline it is placed. Why can you not conceive of being able to time-shift this process?
I think that editorial is a creative process in and of itself. FCPX does not automatically link things willy-nilly with a series of if-then-else statements. It links them according to the decisions the editor makes,Again with the language that suggests that FCPX is aware. It is not. All of the features you listed are ways of doing the same thing as before, just in different ways. Perhaps FCPX makes some mistakes on an import, but how hard is it to fix, really? Once you have a timeline that plays back normally, why does it seem to you to be such a huge deal to re-arrange things?
and the specific FCPX tools he or she uses, while editing: cut it into the primary storyline?
The primary storyline is easily analogous to “V1”, so I fail to see the issue. Whatever is on top is seen instead of what is in the primary storyline. So regardless of whether you made V3 the “primary” it is still layered.
pull it out of the storyline? Connect it? Change the connection point? Position it? Compound it? Storyline it?Again, you are simply putting too much stock into this. You seem to think it is something new rather than a re-writing of what we basically already know. All the video you do in FCP7 is layered. All the video you do in FCPX is layered. How it gets layered is beside the point. In FCP7, a video on V3 is going to be shown instead of the video on V1. In FCPX, the video in the primary storyline is not going to be shown if there is video placed on a layer that is above it, regardless of what terminology Apple uses to describe that upper layer.
This whole thread started because someone can take the FCPX timeline and duplicate it in After Effects. Why are you still unable to grasp this concept when the evidence is right there in front of you?
-
[Gary Huff] “My point being would that it would be dumb to say that, which is why no one is. If you still insist that I called you dumb, then I’m simply flummoxed.”
I’ll concede that one. I misread it the first time, thinking you were calling my entire argument dumb, and I apologize to you.
I stand by the rest of my remarks.
[Gary Huff] “Again with the language that suggests that FCPX is aware. It is not. All of the features you listed are ways of doing the same thing as before, just in different ways. Perhaps FCPX makes some mistakes on an import, but how hard is it to fix, really? Once you have a timeline that plays back normally, why does it seem to you to be such a huge deal to re-arrange things? … Why can you not conceive of being able to time-shift this process?”
We’re talking past each other, then.
As I have said over and over, I think it would be very easily doable to just get the clips into FCPX at the right points in time, then let the editor re-arrange these.
I think it’s a very hard problem to ask software to meaningfully and correctly translate from a timeline which lacks the storyline and connected clips to one that does.
For whatever reason, Apple has opted not to develop this feature.
FCPX doesn’t “know” or “understand” anything. It is not itself “aware” of anything. However, while FCP only stores information about a clip’s absolute point in the timeline, and does not store information about the relationship between clips, FCPX does store information about the relative position of clips and their relationships.
This information is captured at the time the edit is made, and adjusted when edits are changed. This information is an attempt to capture the editor’s intent, and is then later used for the series of if-then-elses that govern the magnetic timeline’s behavior.
I think the FCPX timeline is only superficially similar to the FCP7 timeline. I guess you think they’re very similar. I honestly can’t see how. It’s not magic, but I think there’s a very deep conceptual difference between the FCP7 open timeline and the FCPX magnetic timeline.
I think it could be time-shifted if a human is reviewing the decisions. I don’t think it can be easily time-shifted and done by a computer, which will have no context for the clip relationships from a multi-track EDL (which is what FCP7’s XML essentially is).
Anyone who expects software to automatically deduce the relationships between the clips in their edit and make the magnetic timeline work as expected is expecting a magic bullet. That’s all I’m saying!
[Gary Huff] “The primary storyline is easily analogous to “V1”, so I fail to see the issue. Whatever is on top is seen instead of what is in the primary storyline. So regardless of whether you made V3 the “primary” it is still layered.”
I agree that for straight playout, it doesn’t matter. For later manual connection, it doesn’t matter.
But for a serious attempt at translation, V1 as primary may not always be a good assumption. There could be checkerboarded primary edits on V2. The audio could be the primary.
[Gary Huff] “Again, you are simply putting too much stock into this. You seem to think it is something new rather than a re-writing of what we basically already know. All the video you do in FCP7 is layered. All the video you do in FCPX is layered. How it gets layered is beside the point.”
No — how it gets layered actually governs the behavior of the magnetic timeline as the edit continues.
[Gary Huff] “This whole thread started because someone can take the FCPX timeline and duplicate it in After Effects. Why are you still unable to grasp this concept when the evidence is right there in front of you?”
I’m arguing that FCPX’s timeline is a superset of FCP’s timeline. It contains everything FCP does and more. It’s easy to go out. 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.
It’s harder to go in, because those other apps don’t collect all the data that FCPX does.
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 -
I guess we’re at a standstill then until a FCP7/XML project importer comes out.
Reply to this Discussion! Login or Sign Up