Slashdot Mirror
Building The Ultimate Video Editing Suite
PlainBlack writes "Once upon a time, I was the Chief Engineer at a small TV station, but got out of that line of work about the time that people were talking about replacing video tapes with hard drives. Now I'm looking to build myself a professional grade editing suite using only open source tools so that I can dump as much money as possible into the hardware. My question to Slashdot is, what are the best open source tools for such a suite? I'll need both video and audio editing; a bank of wipes, fades, and other effects; a great paint program; and a titler (text overlays)."
Perhaps Kino and/or Cinelerra have some of the features you need ?n d http://www.robfisher.net/video/kino.html might show you what they can do.
The tutorials at http://www.robfisher.net/video/cinelerra1.html
a
Of course, it is open source.
Linux DV HOWTO on Kino plug-ins
Can anyone mention a program like VirtualDub for linux? I've looked at all the existing software packages, and they were either very featureful and non-intuitive, or quite simple but lacking in basic features (like supporting various codes in avi). What I'm looking for is something reasonably featureful (editing avi and mpeg and conversion, etc) and as easy to use as VirtualDub. Does such software exist?
He who laughs last is stuck in a time dilation bubble.
I've spent a good deal of time searching and there isn't much out there that's open source. My company is currently doing data mining, but that's only to pay for starting and running a video/digital film production company. I found Premiere to work well, but I'm planning to switch to Mac and use some of the tools available there. There is almost nothing available in open source for DVD authoring. There is some simple software out there, but it's command line and, while I'm used to console programs and do a lot of programming, when I'd doing something like editing (or DVD authoring), I want an easy interface so I can focus on what I'm creating, not on what I have to do to make my tools help me. (I've noticed an open-source-denial system that works like this: Question: "Why isn't there open source software that does this?" Programmer's Answer: "Why would you want to do that?" -- instead of admitting there is a desire for a program, but that there is little desire to develop that program.)
I've looked into Main Actor (from mainconcept.com) and am considering using it. I've tried Cinelerra and found it frustrating to get up and running. Under KDE, there is KDEnliven, a video editor in an early stage of development (and, IMHO, the one with the most potential in the long run). There's also Jashaka (or Jakasha-- something like that), which I've heard has a good number of features, but is not well supported or backed for future development.
From my point of view, there has to be at least one solid video editor that works with different formats, allows easy out to DV, VHS, and to AVI and MPG files, as well as a full featured DVD authoring program that makes it easy to import different video format files and allows easy GUI editing of the menus and play sequences.
I've only been in the open source world for 2-3 years, and recently looked back to where things where when I started and where they are now. Video editing is still not a priority and not a task I'd expect to do with open source software. Judging from what I've seen in the past few years, though, I'm hoping it'll be there in another 3-4 years.
jahshaka.org
This looked like it could be quite good a little while ago, and they seem to be coming along nicely. May be worth a look
- Gef
Now I'm looking to build myself a professional grade editing suite using only open source tools so that I can dump as much money as possible into the hardware.
Me too... except I don't have any money. At all. I've been browsing through: http://www.linuxartist.org/ and trying different things in their Video - Animation section.
[signature]
http://ffmpeg.sourceforge.net/
look at the number of project using it listed there.
"professional grade" can mean a lot of different things, from realtime manipulation of uncompressed streams down to anything that can edit DV.
Once you reach a certain budget level, you will be able to do everything you define as 'professional', after that what you are paying for is either speed of operation or storage capacity.
Honestly, with the state ofthe market as it is, saving a thousand dollars on software might buy you 30 minutes a day of increased hardware speed, while not spending that will probably cost you more than 30 minutes a day of software usability losses.
I know this won't be a popular answer on slashdot, but if you are going to be spending a five digit sum on this, you might as well devote a few percent of your raid-array budget to buy some commerical software, either Premiere or Final Cut Pro depending which sode of the mac/pc divide you prefer, and if your budget is much lower, pick up an recent secondhand Mac and get iMovie for free.
A pizza of radius z and thickness a has a volume of pi z z a
This post has a LOT to say about problems with OSs. No community is perfect, and the OSS community certainly has its faults. This post points out a major problem with the OSS community, as does this topic. OSS revolves around programmers. There is nothing wrong with that, but it means that most of the strengths and weaknesses of programmers translate into the strengths and weaknesses of OSS. Unfortunately, one of those strengths is the urge to explore intellectually, but the reluctance to relfect on one's weaknesses outside of the intellectual arena.
/. pointing to this earlier, if you want to look it up), a story listed the reasons programmers were not liked and trusted by users and how developers often treated users with disrespect because the users didn't have the computer knowledge the developers did. This is part of the OSS denial reflex. If you are looking for an OSS program that does something, and there is none out there, right away, everyone calls you a freeloader and demands to know why you would want to do that anyway. Instead of saying, "No, there is no OSS solution for that yet," the response is usually to claim that there is no need to solve the problem anyway.
For example (and there was an article on
Personally, I think the driving force behind "finished" OSS (by finished I mean programs easily used by anyone who can use a computer) is becoming (and already is, to some point) large corporations that are backing OSS development, like Sun and OpenOffice. Abiword is good, KWrite is good. But until OpenOffice was released there was no word processor with the polish, ease of use, and power of professional word processing software. A lot of that comes from the fact that most OSS projects are not paying developers and programmers to write the GUIs and other work that programmers often shun because it's a pain.
The problem is that, rather than write the interfaces and adding the polish, many in the OSS community would rather attack the person who says, "This is not ready for prime time," than to step back and examine the situation and dare to ask themselves, "Is there a valid reason this person is saying this?"
I use OSS whenever possible, and I look forward to the day when I can use only OSS. I have a list of all the OSS programs I've used in starting my company and we've (me and employees) already started discussions on how to pay back those projects (would donations work better, or volunteering man hours while programmers are on my clock). We expect a major jump in income in the next year, and when that happens, we will be contributing to projects we have used, either by money or time.
It's not about getting it without paying. It's about trusting OSS and not trusting companies who have everything to gain by selling software that has flaws and charging for a new version with those flaws fixed.
Oh, and one last rant -- I've bumped into a number of purists who feel all software should be OSS, and often these people are the same ones who leap into a rant of denial when someone says, "I can't use that program. It lacks features and needs an interface." At that point, they start blaming the user for stupidity, instead of accepting that not everyone is a programmer and there are many intelligent people who are experts in their fields, but don't have time to write their own programs. You can't have it both ways. If you want people using OSS, then you have to make OSS easily accessible and usable by all users.
Stumbled across this one:
http://www.exploits.org/v4l/
A great many people think they are thinking when they are merely rearranging their prejudices. -- William James
[shameless_plug] Hoodlumz Productions [/shameless_plug]
"Sic Semper Tyrannosaurus Rex."
Has anyone used their new piece of equipment, the vt3? Their flyer / toaster combo certainly was impressive.
I'm guessing they're using the same proprietary fractal wavelet compression that provided such beautiful video back in the day.
--- It is not the things we do which we regret the most, but the things which we don't do.
I''m the editor in a small (3 people total) production company. I think you need to look at how much aggravation you're willing to put up with and how "Professional" you want to be. Are you okay with telling your clients "we can't do that" if they ask for some particualr effect or style the OS software can't do? Are you willing to spend the time and effort an OS solution is going to take?
I think you may be overestimating the need for massive hard drives at the expense of ease-of-use, hardware availability and cost. We use a single 120 gig hard drive for all our editing. We edit an hour-long show each week as well as commercials, some corporate videos and a twice-a-year dance show that runs at about 5 hours of footage without having too many space problems. A second 120 gig drive would be more than enough. The video is compressed when captured to about 1 gig every 5 minutes, but it looks so good the viewers won't notice it.
The two main expenses for us are the BetaSP deck (about $10,000) and the computer itself (about $5000 with capture card). Final Cut Pro is $1000, but you may be able to buy an older version for less. Version 2 is fine, version 3 adds some better titling software, and version 4 has lots of bells and whistles you may not need.
If you're shooting on MiniDV you can cut about $10000 off the cost right there. MiniDV decks are cheap and you can capture over firewire so you don't even need an expensive capture card. You can even use iMovie if all you need are basic transitions and titling, but I think you can only use one video track (plus titling) and two audio tracks.
For editing graphics we use The GIMP until we can afford photoshop, but all titling is done in FCP.
Basically if you're going to be making money at this the up-front costs are well worth it. Especially on something as complex as video-editing software I'm happy to pay for a solution that "just works" instead of having to worry about computer problems when I'm working.
Compare it to 3D modeling. If you're going to spend 40 hours a week doing paid work would you rather use Blender for free and accept the limitations it has, or pay for Maya or 3DS Max and get your work done?
When most of the cost is the hardware (camera, computer, VTR), it may be worth it to pay the $1000 for software that will do what you want with minimal fuss.
I used to be an Avid editor but didn't have the budget or clients to afford one of those on my own. Based on some feedback from friends I decided to go with a a Matrox RT2500 and later upgraded to a Matrox RT.X100. The RTX100 is fantastic. It's basically a PCI card with a breakout box that has stereo audio in/out, and component and Y/C in/out. There's also two firewire ports on the back. It uses Adobe Premiere for its editor and installs a plugin which lets Premiere use the RTX100 for realtime effects. Basically anything you find in an online suite you'll find here as a realtime effect. Titling, wipes, ADOs, keying, colour correction, etc.
The RTX100 also comes with DVD burning software called ReelDVD. I've only used it twice so all I can tell you is that it works and has lots of features, none of which I've yet to really take advantage of.
I pretty much use that on a dedicated machine with Premiere 6.0, Photoshop, After Effects, and Sound Forge. I also use some open-source tools such as VirtualDub and DubMan. I haven't upgraded to Premiere Pro yet as the Matrox drivers are still in beta.
My only suggestion is that if you do get a RTX100, then buy one of the recommended systems to use it in. The Matrox forums are full of people who complain that the RXT100 doesn't work right or at all yet admit they don't have a compatible system. Especially watch out for via chipsets as the RTX100 won't work on those at all.
Prevent email address forgery. Publish SPF records for y
You can get them to want to create an attractive interface, but they don't have the deep design skills fostered by the commercial groups.
Because of this, it would be very difficult to get something as good and complicated as Final Cut Pro in an open source project.
I would personally recommend a balanced approach.
- Buy the best PowerMac G5 you can afford.
- Get and use Final Cut Pro
- Install X-Windows and you can use whatever open source products appear
Then you have the best of the open and proprietary worlds, and you can decide on a case by case basis which one you prefer. And your basic platform is about 50/50 open source.I think it's an unbeatable compromise, and it's what I run personally.
Even CmdrTaco has a PowerBook. What more can I say?
Hope that helps.
D
Adobe continues to make Mac versions of most of its apps. The buzz at the time was that Premiere was dropped simply because it simply couldn't even come close to competing with Final Cut Pro.
http://www.nmt.edu/~kscott/video/
Nathan's blog
"Especially watch out for via chipsets as the RTX100 won't work on those at all."
Which is why I went with this. It's not a Matrox and it only runs under windows, but it does get the job done.
There is almost nothing available in open source for DVD authoring.
Wait fifteen or so years for the DVD patents to expire, and you'll see more action on that front.
If you are looking for an OSS program that does something, and there is none out there, right away, everyone calls you a freeloader and demands to know why you would want to do that anyway. Instead of saying, "No, there is no OSS solution for that yet," the response is usually to claim that there is no need to solve the problem anyway.
Actually, the thinking here is that a novice user may have taken a wrong turn in solving a problem and is trying to ask for a specific solution that doesn't exist yet, and a response that you may perceive as meaning "Why would you want to solve that problem?" may in fact mean "Here's a substitute solution to what we perceive as your more general problem."
Example: "Where can I get Microsoft Office for free?" / "No, you can't do that, but try OpenOffice.org instead." Or: "How do I run Adobe Photoshop on Linux?" / "No, you can't do that directly. What do you need that GIMP lacks?"
I know it's not open source but it's free: http://www.avid.com/freeDV/index.asp It's limited in it's features but it's the industry standard interface(80% of Hollywood and TV is edited on Avid). I'd steer clear of Linux for editing. There are some professional grade editing suites on varying *nix platforms but they certainly are not open source or free.
Someone's already mentioned VirtualDub, but I'd like to point out AviSynth too. Although VirtualDub provides a simple GUI and can be good for video filtering and for re-encoding, its capabilities are limited to linear editing.
AviSynth, on the other hand, is very powerful script-based non-linear editing system. AviSynth operates as a frame-server--other applications (such as VirtualDub) load AviSynth scripts, and when they request frame N, AviSynth generates that frame and feeds it to the calling application. Most applications don't even need to support AviSynth explicitly; as long as they use Windows' native AVI handlers, AviSynth should work with them just fine.
It doesn't provide a GUI, although there are a number of third-party graphical front-ends to ease the script-writing process. It's harder to learn for non-programmers, but as a script-based system, in many ways it's more powerful, flexible, and precise than traditional methods.
An AviSynth script often looks something like this:
There are quite a number of third-party plug-ins. So far, though, AviSynth has been used primarily as a tool for processing TV captures and DVD rips, and the vast majority of its filters are different types of denoisers. There are few "effects"-based filters, but hopefully this will change as more and more people use it as an editing system. However, there is support for basic fades, pans, and wipes.
More information can be found at the avisynth.org site and on the doom9 AviSynth forums.
Dude, spend your money on the hardware, and then warez the software. You can buy it later when you have more money.
What do you mean by "professional grade editing suite?" My local CBS station uses DV for editing local commercials and news, you can see the difference in quality between their DV productions and work done on full uncompressed SD edit suites. They say they're saving up their money for HD equipment, but it's obvious they're just using it because they're the most amateurish of all the local network stations and they can't afford anything better.
There are no professional-quality open source editing systems. Period. All my video editor buddies (and I) are using Macs with Final Cut Pro, Boris Red, Shake, Peak or Logic, etc. I know one big shop that used Combustion on an SGI Onyx/IRIX box but they switched to Macs when the G5 came out.
Hardware is another issue. If you're just doing amateur work, DV will be fine, you just need a cheapo DV converter like the Canopus ADVC-1000. If you're a pro, you'll need an SD card like a Kona-SD, a realtime effects card like the Matrox RTMac series, and a hardware RAID to deal with high-bandwidth data streams with reasonable speed. One vendor I know claims that 4 IDE drives in a RAID will be sufficient for SD, but I'm more inclined to go with an 8 drive IDE RAID, a 4-drive SCSI RAID with 15kRPM drives, or a SATA RAID, all of which would beat the pants off any IDE RAID, and even do a decent job with HD.
Face it, Apple owns the pro editing market. There are some competitors like Avid and Combustion, those packages even run on PeeCees, sometimes even on IRIX. But those packages are in serious decline. Apple is rumored to be buying out the Autodesk video line (including Combustion, Cleaner, etc.) Why would you want to waste time cobbling together open source junk? Is your time worth nothing? The additional cost of professional hardware and software will pale in significance next to your tape decks and other non-computer equipment, you are trying to save money where that will do the most harm to the project.
I use OSS whenever possible, and I look forward to the day when I can use only OSS. I have a list of all the OSS programs I've used in starting my company and we've (me and employees) already started discussions on how to pay back those projects (would donations work better, or volunteering man hours while programmers are on my clock). We expect a major jump in income in the next year, and when that happens, we will be contributing to projects we have used, either by money or time.
I'd say probably a mix of both... to those projects that pretty well fit your needs exactly already, send money; to those that need some work, send patches.
Need a Catering Connection
I've asked about WYSIWYG HTML editors. The response is
Mozilla Editor.
(almost) always, "Why would you want that?" or "I always hand code my HTML. I don't like WYSIWYG editors."
The problem here is that many popular WYSIWYG HTML editors tend to encourage web page authors to use deprecated or semi-deprecated presentational markup (<b>, <i>, <table>, <font>, etc) rather than the structural markup that W3C recommends (<strong>, <em>, <div>, CSS, etc). Does there exist a mature WYSIWYG editor for XHTML Strict, even on the Windows platform?
I have been doing video processing mostly (things like making svcd/dvd copies of peoples wedding videos and 8mm movies and such), and only a little bit of online editing (quite a bit of offline editing, with online previewing with the 'customer' tho)
10 years ago, I did this with a SGI indy and a hardware mjpeg compressor, nowadays I use a pc with opensource software..
Given that I do little online editing, this is very workable for me. The big deal in my case is not user-friendly high quality editing but squeezing the highest quality out of each bit of compressed video that fits on the desired medium.
Doing this with OSS works well if you are prepared to put in the efford to get all the tools (you'll need at least avidemux, mplayer and the mjpegtools) and with the limited support of capture hardware, and when you need resolutions beyond approx super-vhs quality, you will want external, firewire connected hardware.
The advantages:
You have more control, given you have the technical knowhow to change the source code and recompile the tools. You will also need to have scripting abilities to integrate the different tools and make your own offline editing suite with them.
The disadvantages:
No good online editing due to lack of editing software with a decent user interface.
High cost timewise.
If however you do go this way, and you do have the ability to make it work, you will end up being able to provide exactly what your customers want, free from limitations of the original developers of the software you use.
With regards to DV, the biggest problem is not the compression, and then recompression after editing in the format required for the end medium, but the fact that in most cases people lack the space to use uncompressed intermediate files for editing.
This can be overcome by creating editlists using a 'preview' suite, and then using an offline suite which can do all the editing and post processing in memory, and as such you get the space advantage of DV editing without the quality loss that usually occurs with it.
Anyway, if you are in for quick result, get yourself a Mac and get the best of both worlds.
it's about time someone actually said the truth about open source.
Actually invest some time into user friendleness in gui, online help, tech manuals, installation ease, and all the other things decently written software has.