For my part, I don't care too much if Kdenlive has the sum of all the features of other software packages considered professional and mimicks each and every one's workflow model.

Is a BetaCam less professional than a 35mm Arriflex? Is a KEM less professional than a Moviola? Is Blender less professional than Maya? Depends.

The lack of a standard format for exchange between media editing software (the second point you make in your last post) is the real issue here, and the one that makes Kdenlive (and e.g. Blender, when it comes to EDL.s) feel less "pro" than their proprietary competition. They may not be open source, but when it comes to interoperability with other applications (be it of the same vendor or not) they certainly seem more open than their FLOSS counterparts.

So, what is long overdue is some sort of OpenEDL or really open open media framework, an HTML of the world wide FLOSS media editing application web. This needs diplomatic skills more than coding wizardry and would be a perfect task for enthusiastic non-coders. I don't know if I have enough spare time to handle this, but I would definitely take it under consideration. And I'm no coding wizard, that's for sure!

ArtInvent (and others, of course), why don't we team up and take this into our hands? First step would be to look for other efforts that go in the same direction and join forces with them.

Hi, my name is Gabriel, I'm a certified instructor for Adobe's Premiere Pro, After Effects, and Autodesk's Maya, 3ds Max and Combustion.
I like Linux, KDE and free software in general and really hope one day it gets to pro status, or at least good enough for prosumer level work.
I'm keeping an eye on Kdenlive for video editing, and like the cleaness of it's interface specially. Haven't tried it yet, so I can't comment, but I think is right headed.

One suggestion I'd give, if I may, by seeing the screenshots would be to optimize the space used by the different elements of the interface with the purpose of maximizing work areas (timeline, monitors, effect controls) over resourse access areas (project panel, effects/transitions selection, undo list, upper buttons bar, etc.).
When I edit I'm constantly looking for ways to have taller timelines and bigger monitor areas in every program I've used. If you manage to provide just that, you'll have gained many editors hearts right away! :)

Kudos to the developers. Keep the good work!!!

AlanJames1987 ....... Are you still around?

Linux badly needs a good video editor and in my opinion it needs to be coded from scratch in C or C++ and not be dependent on external software packages like ffmpeg, MLT and the type of Linux OS you are using such as Ubuntu, Fedora, Mandriva, etc. Look at all the Kdenlive packages available for the different Linux OS's, should be one like GIMP, FireFox, Open Office.

This of course would be a major job and could be handle the same way GIMP, FireFox, Open Office are developed. And according to the December 2009 issue of HDVideoPro magazine Final Cut Pro was developed by Apple as a "loss leader" to sell macs. So perhaps Mark Shuttleworth - Ubuntu founder - wouldn't mind donating some money to devolop a Linux video editor

I recommend you download ArtistX 0.7 - an Ubuntu based OS. It has tons of preloaded graphics and audio software to try out and a few video editors such as:

Kdenlive (easiest to use)
VideoCut 0.20
StopMotion
Open Movie Editor 0.0.2008
Cinelerra

There really is no way not to be dependant on something like ffmpeg. Ffmpeg probably represents several hundred, if not thousands, of man-years of coding. Reimplementing that in Kdenlive would be madness.

Any video editor for Linux, that is "selfcontained", will not be able to handle more than a few of the most basic/simple encodings.

The "split" in kdelinve is basically sound. Mlt on top of ffmpeg is good, and it works.

Not to say that everything is perfect - because it is not.

One could consider delivering Kdenlive with MLT and FFMPEG as "total package". However, this would really mean delivering it in source only. No distro wants that, so...

Thanks for the reply madsdyd.

RE: "Any video editor for Linux, that is "selfcontained", will not be able to handle more than a few of the most basic/simple encodings."

I would like to use DCRAW in this example. As you know DCRAW is a C program that will convert most camera RAW files into TIFF, now apply this to a Linux video editor. I don't understand why a C programmer can't come up with a similar program to convert video/audio formats to other formats, and then implement it into a self-contained video editor. More info on DCRAW: http://www.cybercom.net/~dcoffin/dcraw/

Dave Coffin - the developer of DCRAW - also offers consulting services and I'm sure would be able to provided lots of help on such a project, for a fee of course. This is were the likes of Mark Shuttleworth, RedHat, SUSE, etc. come in to help with the finances of such a major project. RedHat (RHT) has a market cap of $5 billion and I'm sure Shuttleworth could raise some money.

Apple developed FCP as a "loss leader" for Macs, why can't the Linux community do the same thing for a Linux video editor?

I think it is silly to compare DCRAW with 8000 lines of code against kdenlive and its supporting libraries (mlt, ffmpeg, sdl, libx264, and many more) with easily over 1 million lines of code.

I don't think you have any idea how hard it is to write quality software that is stable in all environments and that is fast enough to do the job. Video codecs in particular require many optimizations that few people are skilled enough to do well. Just as most doctors can't do brain surgery, most software developers can't do speed optimizations in assembly language.

The Linux community as a whole is not a for-profit corporation and it makes no sense for you to expect it to behave like one. If you want Red Hat, Novell, etc to finance a professional NLE project for you, you should be contacting them instead of posting about it here. I predict they will laugh and then have a guard escort you out of the building after you propose that all necessary components need to be written from scratch. :)

If you really really want to make a company make a (/another) videoeditor for linux, why not get some votes together for one of these Ubuntu Brainstorm 'video editor wanted!' ideas?

http://brainstorm.ubuntu.com/random_ideas/?keywords=kdenlive

(Sorry if I've stepped in late here, but...)

As far as music production goes, MuseScore (http://www.musescore.org/) has been my finale replacement for quite some time now, and is quickly approaching 1.0. All that's left to do is incorporate an easy way to export individual parts and it's set to compete with finale.

Apart from Audacity (which is great, but the Linux audio framework has gone on without it) Ardour is a powerful multitrack editor with a Track-and-Bus approach. http://ardour.org/

Regarding functionality in a NLE: I really would like to see as little functionality as possible, regarding importing and exporting. IMHO a NLE should only handle raw audio/video streams and do that perfectly. Conversion from/to formats should be done by tools that are specialised in that, like ffmpeg and mencoder. A I guess a "professional" would know from formats and codecs and how to convert these >:-)

Two big problems with Kdenlive is the missing codecs, such has MPEG-4, H.264 and Flash. And all the different Kdenlive packages for each linux OS.

I'm on Mint 8 and when I downloaded Kdenlive 0.7.5 it was missing all the codecs listed above.

(Warning: This is going to be very ranty.)

Just two words: software patents. Those are hard to stop. If we succeed on the national level (parliaments), the lobbyists just try it one level up - the EU, the WIPO or any other uncontrollable lawmaking agency they see fit to call into life ad-hoc. Don't blame Kdenlive, don't blame your distro.

Two solutions though:
1) stop using tech altogether. Whatever gadget you buy, from whichever vendor, you are most likely - if only indirectly - shoving your money up the *** of some pro-patent lobbying farts
2) lobby your distro to implement a simple means that lets you install the patent encumbered components if you so chose

I only want to emphasize that this thread is extremely important to me. I'm writing a variety of articles for beginners editing video in Linux, I'm a newbie myself to the topic, and the comments and suggestions made here have given me some needed background on the subject from the professional POV. It would be a grave error to ignore or discourage the contributions from the pros. I don't get the impression that anyone here is saying "Do it like it's done in [your favorite Win/Mac NLE here]", but they are pointing out that certain features in commercial apps are features that are worth having in Kdenlive.

The Ardour DAW isn't trying to copy Pro Tools, but you can be sure that its devs pay a *lot* of attention to what's going on in the world of commercial DAWs.

Edit: I must add that I do enjoy using Kdenlive, it's a wonderful program.

Conversion always means quality loss, and most important TIME means MONEY for professionals, which HATES to convert or even CAPTURE. If you are in a news workflow you need somekind of solid or DTE disk to avoid capture, so as you can see conversion is totally aout of the question.

Let`s say you have 80 hours of video (the ussual for an average documentary with many interviews/inserts), how longs would you think it could take to make that conversion into something KDEnlive can edit... well, your customer will run away to some Mac user, profesional means to take less time to do more and having a competitive solution make you even more compettitive don`t you think?

What you propose is good for editing a family video, and having all the time in the world, and the just upload the video to youtube.

Kdenlive needs desperately Sequencing/nesting IMHO (to avoid intermediate renders)

One basic functionality for 35mm / 16mm film editing is EDL list. While capturing the dv footage provided by telecine lab it should be able to capture footage as well as timecode data. This data should be preserved while cutting the clips too. And in the end kdenlive should be able to generate those cut list / EDL list for a negative cutter to cut the film manually. Here in India we follow this workflow with Avid. As of I know there is no one software which is capable of doing this, which is very essential and basic.

More info about EDL:

http://en.wikipedia.org/wiki/Edit_decision_list

I'm paid to edit video and I work on a range of things from promotional stuff through productions of live events and other bits and bobs. I produce 2-4 videos a week. I work almost exclusively with final cut however I've been watching the progress of kdenlive and also openshot with great interest.

The progress is very encouraging and I'm really impressed with what I can now do with kdenlive.

For the first time I think I could now actually use it for work but what I'm hoping is that kdenlive will get to the stage where I WANT to use it for work.

There is so much positive stuff I could say about it already.
I like the render options and separate rendering process.

Fairly essential deal makers/breakers...
* Drag to insert a clip between two clips
* Layer modes like dodge, burn, difference etc.
* clip opacity doesn't work for me and I need it.
* hi pass, low pass audio filters.
* stereo audio tracks should be linked mono tracks. With keyframing for pan, vol.
* Sound level meter with peak indicator
* fully parametric eq - ideally with built in analyser
* timecode overlay effect

Things that would be great...
* syncing up camera angles and easy jumping between them. Like the multicam interfaces in other pro NLE packages.
* fluid speed scaling. I set key frames in the video and link them to keyframes in the sound track. The program alters the speed of the video track to line up those keyframes. Ie my video can beat to the music more easily.
* Clip properties like position, cropping, distorts, opacity. I'd like to be able to interact with the preview window as if it was a drawing canvas with keyframes too.
* I don't want to see clip names on the timeline unless I look at the properties - visual clutter.
* I'd like to have left click drag selection to select multiple items on the timeline.
* I'd like an interface for importing clips from sdhc or hd based cameras. I'd want it to be able to handle fixing the aspect ratio of mod/moi files ie. turn them into proper mpeg2 files with a completed header.
* A media/clip library with simple tagging and searching - backend on a media server? A separate but integrated app.
* Would be great to set up a client/server rendering model so that the rendering can actually happen on a separate pc. A separated but integrated app. eg add batch conversions, I can use it without a project just to convert a file? etc.

Just my 2p worth.

Well done on a great app so far.

John