Larry Jordan reveals Apple's FCPX Roadmap

Quote

Very interesting that they are bringing back the dual viewer. Huge move by Apple.

* SIGH *

I try to avoid these discussions...but they are so silly, I cannot help myself. Look at that statement. How ridiculous is that statement? That an NLE had to "BRING BACK" an edit staple, the Dual Viewer?? Who out there LOVES working with a single monitor? Like I have been saying since the release (and Walter put it so eloquently above in better words than I)...Apple is in damage control and nobody really wants to hear about it anymore. Nobody is waiting around to see "what's FCPX gonna do next - I wonder?!" The only people really pushing FCPX are the teachers / salesmen. I would love to see a poll of editors who switched to FCPX and love it.

I look at it from a worker bee POV...I have MANY projects to do / revise / redo / etc and I cannot wait for Apple to "put back" what should have been there at the start. Been waiting 2 years for FCP8 while Adobe and Avid have leapfrogged Apple in marketshare almost overnight. Symphony 6 is now a $999 upgrade from Express Pro!! SMOKE is $3500?!?! Holy CRAP! These systems were a gazillion $$$ not too long ago.

The bright side of Apple vomiting on itself RE: FCPX = The industry came down to our price point. I just want to say THANKS AGAIN to Apple for laying this gigantor egg and making the elite tools affordable for the worker bees. Cannot wait for FCPX.1

When life gives you dilemmas...make dilemmanade.

I don't think this is something Apple could have done earlier (my hunch).

Their "re-introduction" of features is tied very much to what may be a linear progression for the developers but very non linear for us.

We don't know for sure why they're taking this path but I'm going to guess, having been using the program since its inception. Dual Viewer wasn't possible until Multicam was done because that's where ganging (also coming back I'd say) and Dual Viewer were built. For whatever reason it made more sense for them to build that first and used multicam development to build out other features.

I see things like that happening in other areas as well. Once Roles is developed further along, we'll probably start to see more controls over what they display . . . which may well line up Roles as a horizontal stack like tracks.

The reason why we can't yet copy/paste individual filters is they still need to do work on the database management side. It's as if they need to take one data point in a cell and break it out into separate cells for that to work and that adds to the relationship complexity exponentially.

Broadcast monitoring couldn't happen until certain things between FCPX and the OS were developed or hooked since I believe Broadcast monitoring is specific to Lion going forward.

Again all speculation on my part but it just seems that they have path which is most logical for the developers but seems strangely sequenced and prioritized for the end user.

I wouldn't be surprised if Send to Motion comes back but it will be around the time or after Logic Pro X is ready because there may both need the similar things to happen in XML (my guess).

I need to double check this but I'm under the impression there may yet need to be changes in how ProRes relates to encoding on multi core multi threading based systems.


To put it another way . . .
The developers see the progression as ABCDEFG
We see is as ACDFBEG
and we wonder why B didn't happen sooner and why E came after B.

> they have path which is most logical for the developers but seems strangely sequenced and prioritized for the end user.

Which is more important: What's convenient for the software developer, or what's useful for the customer?

Users aren't obliged in any way to make room for excuses for the product. When you have angry customers, excuses are just what a company resorts to when it refuses to apologize. Apple needs to follow Toyota's suit and just admit its screwup. Even then, it doesn't even begin to mitigate the ill will of 2011, as Walter so eloquently put it.


www.derekmok.com

Actually what NEEDED (not convenient) by the developers. They can't leap frog what they can't leap frog. Give the speed of the updates (which I certainly consider fast especially from the development perspective) they're likely on focusing on the BEST sequence of events that allows them to role out updates as quickly as possible. For them it may well have been an ass backwards nightmare to build Dual Viewers only to have to change things to get multicam right.

They're already bumping into things as it is as they move with haste. The plugin issues are a good example. Sapphire Edge and Magic Bullet Looks were broken because they built in a workaround to an issue which, when Apple FIXED, broke the plugins. That's the kind of problems that happen when you're forced into an ass backwards approach to development. Fortunately, I suspect all parties anticipated it since both plugin developers had fixes out within a few days. It's also why there are probably a few more plugin developers waiting until Apple is done with this before they get commercial plugins out the door.

I think Apple is taking a development path to avoid the cobbled together underbelly they were left with with FCP. A feature won't be released until that part of the code base is done. Once done other related feature development begins.

It took a long time for Sony to finally release their XDCAM EX plugin for FCPX and I understand that was in part due to Apple as well. That MXF support will be happening may well be related to work being done both internal and APIs for codec and/or container support in AVFoundation. I also think that once they abandoned Snow Leopard compatibility they may be free to move forward on other things as well because of stuff not in SL that they have to support internally in FCPX.

derekmok Wrote:
-------------------------------------------------------
> > they have path which is most logical for the
> developers but seems strangely sequenced and
> prioritized for the end user.
>
> Which is more important: What's convenient for
> the software developer, or what's useful for the
> customer?
>
> Users aren't obliged in any way to make room for
> excuses for the product. When you have angry
> customers, excuses are just what a company resorts
> to when it refuses to apologize. Apple needs to
> follow Toyota's suit and just admit its screwup.
> Even then, it doesn't even begin to mitigate the
> ill will of 2011, as Walter so eloquently put it.

> We don't know for sure why they're taking this
> path but I'm going to guess, having been using the
> program since its inception. Dual Viewer wasn't
> possible until Multicam was done because that's
> where ganging (also coming back I'd say) and Dual
> Viewer were built. For whatever reason it made
> more sense for them to build that first and used
> multicam development to build out other features.
>

<EDITED BY MODERATOR>
Final Cut had dual screens from day one and didn't get multicam till YEARS later. Like a decade after Avid.
You have been embarrassing yourself badly with this "fanboy" behavior, trying to be some kind of freelance Apple
salesman in a community of people who know better.

Apple could have put dual screens in from the very beginning of FCX.
There are absolutely no excuses for this mess.

But there are explanations.

IMO this mess looks like an attempt by Apple to "own" the editing world with the same methods they are using to
try and cornet phones and tablets... obtaining intellectual copyright over every part of the editing process.

There is absolutely no reason for renaming everything with obtuse terms that don't directly describe their function...

Except that you can't copyright the term "bin", or sequence or tracks or.....

Given that they expected FCX to be a huge hit and further expand their market penetration, it makes sense that
they were looking to "own" the whole editing world as their "intellectual property". Same as all the phone and tablet companies
are doing suing each other over ownership of IP for every little inch they can get.

But it didn't work out too well for Apple.

I hope FCP becomes something useful for the general editing world down the road.
Right now it's for niche content.

But I thank Apple for giving the competition an opportunity.
Adobe, Avid, Autodesk.... They rose to the occasion.

After NAB 2012, honestly...
It feels like professional editing is back.

Now, let's not get personal. I'm not a big fan of FCP X, not until it turns the corner and blows my socks off, which is near impossible, as I compare FCP X to MobileMe- nice idea, but "why doesn't it do that?" Everyone is entitled to their opinion, and I do respect Craig's opinion and he brings up very good points to level the discussion, such as the database back end of FCP X, etc..

I liken it to MobileMe, which was a really cool idea, but the implementation was horrible at best- slow, inefficient as a piece of software, clunky and clicky, and way overpriced. It was so bad I cancelled my MobileMe subscription in the middle of the trial period. But the truth is, the NLE, online and color correction as well as the whole post landscape needed to change. Why? Because the times are changing and the processes, as well as the expectations have evolved over the years.

And yea, it leaves a lot to be seen.



www.strypesinpost.com

>There are absolutely no excuses for this mess.

I consider FCP X to be a beta software that is shipping. Craig probably sees it as a good beta software. I have my reservations, especially on the fact that it is such a tremendous resource hog (turning filmstrip off doesn't count, as the software was designed to run on filmstrip view), that roles has not been able to properly replace tracks, and that FCP X was designed from the POV of a software programmer rather than that of a film editor.



www.strypesinpost.com

When 10.0 first came out it was Alpha. IMHO to reach Beta it has to be "feature complete" (at least as a usable release). My opinion is that I like the direction Apple is going with it. There's still serious missing chunks. I think the audio features will be appearing in some relationship to Logic Pro X. I also think that's around the time the Send To functions will be implemented as well.

I've worked with developers before and when a company does a ground up rewrite the coders may make decisions which result in an odd sequence of a reintroduction of similar features.

Macromedia, not Apple, wrote the base for the original Final Cut Pro. Apple had some serious challenges as they developed it from what I understand. Media management was a mess for years. Trimming was never even close to as developed as Avid even up to the point FCP was EOLd.

Yes Apple wants to control everything and yes patents are a very real part of the issue. Apple dropped virtually all programs that came from out of house included FCP legacy.

If you read a bit abut the history of Macromedia KeyGrip the rights and agreements issues, Macromedia basically had no reasonable way out to sell. One will never know but I suspect part of the reason FCP was dumped so suddenly were that certain agreements may have been up for renewal.

From the history of Final Cut Pro

Before version 5 was released, Ubillos' group was hired by Macromedia to create KeyGrip, built from the ground up as a more professional video-editing program based on Apple QuickTime. Macromedia could not release the product without causing its partner Truevision some issues with Microsoft, as KeyGrip was, in part, based on technology from Microsoft licensed to Truevision and then in turn to Macromedia. The terms of the IP licensing deal stated that it was not to be used in conjunction with QuickTime. Thus, Macromedia was forced to keep the product off the market until a solution could be found. At the same time, the company decided to focus more on applications that would support the web, so they sought to find a buyer for their non-web applications, including KeyGrip; which, by 1998, was renamed Final Cut.

I can only guess but IP may have been why Apple decided to dump everything they purchased from outside.

I also think that Apple may have decided to tie everything very tightly together between AVFoundation (which was part of iOS first) and OSX going forward.

Quicktime itself is being EOLd. Maybe that's getting them out of some future issue as well. I'm not sure what'll happen on the Windows side at some point but currently WMP12 can play H.264 .mov. I'm not sure if this relates but I'm not sure it doesn't.

I do think Apple wants all their media products (whether consumer or professional) tied more closely to the OS. This might relate to Broadcast Monitoring and Lion. It may be that other things can't be released until Mountain Lion is. I certainly believe the odd order of the introduction of features is tied having to get some things done first for other things to be implemented. In that sense it's from the POV of programmers, not editors. Their priority is not ours and I think Apple allowing them to work that way because it'll be better in the long run.

In many respects I don't see FCPX that much differently than OSX. Cheetah was unusable. Everyone stayed in OS9. Puma had some interesting things but still wasn't as functional as OS9. It wasn't until Jaguar that people finally started to use OSX and it still was missing feature that OS9 had. By Panther things were usable even though people who were deep into OS9 still saw flaw. To get from Cheetah to Jaguar was March 2001 to August 2003 and Panther was October 2003. I think it was the Jaguar Macs were they defaulted to booting into OSX. That was an extremely long transition.

It's not a direct parallel but given where FCPX was starting from in code, AVFoundation, OSX ties and that all that, not just FCPX is a work in progress, one might understand the loooong transition to something that's mature. 18 months transition would make sense. FCPX is going to look more like a complete app by December of this year.

Unlike OS9 there was no transition period. Unlike any other Apple product or service EOL FCS was pulled from the market immediately. I don't think it was because they believed FCPX was a complete usable product.

Apple did know that FCPX would not fit the needs of "professional" market and big part of their installed user base. Quite a lot of "NDA involved people" told them it won't work for that market nor for themselves.
But Apple had to make a decision. Working on 3 Video NLE apps FCP/FCE, FCPX and iMovie would have taken a lot of resources. Rewriting FCP would have been a lot (a really lot) of work, while the "old coded" FCP had to be developed same time for updates, if they wouldn't have done imagine FCP 7.0.3 with no update for 2 or more years. What would have been the reaction in the public.
The bad thing Apple did was that they didn't really listen to the important beta testers and released a new NLE app at the wrong time while without notifying people that the old product line is not availble anymore.
And they took the wrong name and called it Final Cut Pro X, which made a lot of people believe it's an update -- but it' wasn't, it was a new application which has nothing to do with Final Cut Pro.
(A note: people should have looked a the naming of the FCPX prefs file: "com.apple.FinalCut.plist" -- no "Pro" in there).
So a little bit more efforts from Apple team would have maken the current 10.0.3 the 10.0.0 at a little later release time. Clarifying that this app was/is a different kind of NLE at an early state would have helped as well. Not taking FCPS out off the shelves would have been another point.

One thing I don't understand with all this discussions: A lot people say "FCPX is data base driven". What the hell they think FCP was? Or Avid or PPro or Edius or Media100 or Vegas or, or, or. FCPX is just based on another kind of data base.: SQL Lite 3. This kind is pretty established, always developing/changing -- so what? Apple uses it's own private Core Data inside SQL and therefore it's useless for a real lot of people who have to organize shows, movies etc.

People might discuss the UI, might discuss skimming, might discuss roles or keywords and so on. For some this works fine for others not.
I belong to the latter group.

And there are a real lot of serious bugs within FCPX -- still after a year.

Anyway, as said above: don't be personal with replies, but constructive. As always I recommend to set up an Apple developer account (free) and use the bug reporter to supply feature requests, minor and major bugs. This way you see how your comment is handled and when. BUT before sending make double checks and search the forums.

Some workflow tools for FCP [www.spherico.com]
TitleExchange -- juggle titles within FCS, FCPX and many other apps.
[www.spherico.com]

Quote
And they took the wrong name and called it Final Cut Pro X, which made a lot of people believe it's an update

Although the numbering sequence was better OS9 to OSX was not a simple update.
Probably a bit better analogy is Quicktime 7 to Quicktime 10. The latter is more AVFoundation that Quicktime.
I can see how marketing wanted Final Cut Pro but that in itself is the problem. One might argue that Randy should have also won his choice of the First Cut name over iMovie. I do think FCPX's import from First Cut (iMovie) is a useful feature in some cases. Having a client do their selects or rough in First Cut and importing into Final Cut is a lot better than entering selects numbers off a time code sheet.

Quote
A lot people say "FCPX is data base driven". What the hell they think FCP was? Or Avid or PPro or Edius or Media100 or Vegas or, or, or. FCPX is just based on another kind of data base.: SQL Lite 3. This kind is pretty established, always developing/changing -- so what?

Most NLEs are using it like a spread sheet, not a database. I'd bet if you actually were to break it out in database form you'd see a lot more complex relationships. Things like Roles and Smart Collections are examples but I think there's a lot going on under the hood that less obvious. There's a lot that's not yet fully implemented. Look at the full extended Inspector and you'll see a lot of the roadmap there. It does seem PPro is heading in that direction as well though.

Wow, things can get pretty heated around here.

"The plugin issues are a good example. Sapphire Edge and Magic Bullet Looks were broken because they built in a workaround to an issue which, when Apple FIXED, broke the plugins." These plug ins were fixed within a couple of days and work just fine now in 10.0.4

It seems to me that there is a lot of good in FCPX...yes, it was released way too early with too many bugs and missing features. Yes, it is still missing a ton of things and, yes, for the life of me I will never understand why they changed all the keyboard shortcuts that took me 11 years to learn, and, yes, we still need to be able to set in and out points in the clip browser that actually remain and in/outs in the capture window and, yes, fix the ram leak and plenty more. But what's done is done and time for people to stop whining about it. I use both 7 and X now for different types of projects but the whining gives me a headache.

Steve

steve-sharksdelight

craig seeman Wrote:
-------------------------------------------------------
> It does seem PPro is heading
> in that direction as well though.

I certainly hope not. As a developer, I love that Adobe is using XML for all of their project and settings files.

My software:
Pro Maintenance Tools - Tools to keep Final Cut Studio, Final Cut Pro X, Avid Media Composer and Adobe Premiere Pro running smoothly and fix problems when they arise
Pro Media Tools - Edit QuickTime chapters and metadata, detect gamma shifts, edit markers, watch renders and more
More tools...

I just upgraded to MC6. Had it for quite a few weeks, but I was dreading going over to Lion, although it doesn't seem so bad. Quirky, but Lion seems fine. Avid MC6, on the other hand feels really nice, and I dread using FCP 7 at work. Because it has no future and today there are better alternatives. I'll be upgrading to CS6 when it's released, but my eyeballs popped out at the part where sending a sequence to SpeedGrade causes everything to be rendered to DPX. It's a little like that joke where they'll release something this year, then next year they'll release an update that makes everything finally work.



www.strypesinpost.com

< "but my eyeballs popped out at the part where sending a sequence to SpeedGrade causes everything to be rendered to DPX" >

I missed that! So can you round trip it back to Premiere after you are done with it in SpeedGrade? I'm asking because I never worked with a DPX file before.

If you want dynamic link with Speedgrade and PP, you should make a feature request here : [www.adobe.com]

I know they are getting quite a few calls for this, and the more votes they get for something, the higher up the priority list it goes.

>So can you round trip it back to Premiere after you are done with it in SpeedGrade?

Yup, you can. PPro supports DPX.

[www.adobe.com]

However, you will certainly need a really fast drive array and loads of storage! Something that provides a consistent 150MB/s throughput for HD, like a thunderbolt RAID 5 with 5-7 hard disks.

I believe dynamic linking is in their roadmap, just that between the acquisition of SpeedGrade and NAB, they did not have time to integrate dynamic link between the two apps. I doubt it will be implemented as as part of an incremental update, because this would be huge, and would make SpeedGrade actually usable for many people. More likely the feature will be in CS6.5, if not 7, with further upgrade costs.



www.strypesinpost.com

Quote
Craig
Most NLEs are using it like a spread sheet, not a database. I'd bet if you actually were to break it out in database form you'd see a lot more complex relationships. Things like Roles and Smart Collections are examples but I think there's a lot going on under the hood that less obvious. There's a lot that's not yet fully implemented. Look at the full extended Inspector and you'll see a lot of the roadmap there. It does seem PPro is heading in that direction as well though.

Craig -- No offense, but you mix UI display, displayed options and underlying source code.
Things like Roles, Keywords, Smart Collections (or however you call it) had been there all the time -- though not visible for the standard FCP user, and the FCP interface in this case was not really flexible.

Now the interface is more flexible and a lot of people say "Wow, cool" -- I think it's very cool for a certain group of people. But there are others who say "it doesn't meet my needs".
I can understand both groups and as an open minded consultant, beta tester, part time programmer (with sometimes crazy ideas) I love to test new stuff and talk to others about that.

The most important thing is that software developers (small or big) have to be open minded as well. It doesn't matter what kind of data base is used, but the data base needs to have "published hooks". With FCPX-Motion there is an invisible start - even though Motion doesn't have really published(documented) hooks a user can do a real lot with this combination within a project and also can share the experience with others easily (most active ones Alex4d and Simon Ubsdell and there are many others as well).
Unfortunately there is no real good hook from FCPX data base to the outside world, it's there with XML but really veeery basic. So I/You have to understand those people who say "we're lost with FCPX" -- the bigger problem for some of them is that they are lost with Avid, Premiere and others as well -- their data bases are also private.

It's a kind of interim state for many users. Even though there had been enough of problems with FCP the architecture was way more open compared to other applications. This made FCP successful.

I had a discussion with one of the so called "Lovers" of FCPX some days ago. He really likes/liked FCPX for it's speed of editing, it's ability to work with roles and keywords and so on. But when it came to the final cut the exchange/editing of subtitles made him a part of the group "we're lost with FCPX" since there was no way because of bugs and missing hooks in the FCPX database/XML.
He went back to FCP (with some kind of pain) after he tried PPro and figured out that there isn't any way.

My conclusion: FCPX needs a real lot of development to make it more open in the way FCP was. This will take time. A lot of people though say "we need that now" another group will say "we don't need that at all". So the discussion will always be controversial.

I don't think I'm confusing the database with the hooks. I know the hooks are poor. I hear it often from the third party developers. The fact that some feature internal to FCPX still aren't developed such as Copy/Paste selective settings and round tripping with Motion may well examples of the still "primitive" state of such things in X. My own speculation is the next "leap forward" will happen with Logic Pro X. I'm guessing/hoping that's when we'll see a return to round tripping and, with that, another round of XML improvements.