Yesterday I was shown a demo of a great sound library program called Soundminer. It’s very exciting: 58 metadata fields, SQL search functions, Rewire support for multichannel monitoring, Quicktime support for auditioning against picture, all kinds of fun stuff. But I get ahead of myself…
A sound company’s or sound supervisor’s bread and butter is the sound effects library. Attitude, communication, and all those customer service things are very important, but without a sound effect library there can be no post-production sound. A critical factor in the sound effects library is how easy is it to get to the necessary sounds?
When I started working in this industry eight years ago, most Supervising Sound Editors would use a process similar to this: Take the continuity list that shows the division of the film into reels and scenes. For each scene create a list of spots. Search the sound effects library for appropriate sounds for each spot. Listen to the CDs and DATs and make lists of the chosen sounds for each spot. Print the lists. Have the assistant load the sounds into the computer from the source CDs or DATs. Have the assistant copy the sounds to the editors’ drives and make copies of the lists for the editors.
For example, scene 20 is a car chase through New York with gunfire between the two cars. So spots might be Good Guy Car, Bad Guy Car, Good Guy Guns, Bad Guy Guns, NYC Traffic Day. It might get even more specific since the most dynamic sounds for the scene would probably be the cars. So there might be spots for Engine Revs, Tire Squeals, and Brake Slams.
Supervisors would use a database program like Filemaker Pro, Leonardo, or even something on a ancient Alpha Micro computer to search for sounds. Type in “Engine Rev” and see what goodies pop up. Or maybe there was that one recording of a really beefy police car who’s engine sounds might be perfect, so the supervisor types in “Crown Vic Engine”. The list of the matching sounds would be displayed, and then the process of pulling the DATs or CDs from the shelf, finding the right track number and listening to the sounds would begin. Between all this searching, finding, listening, and the all the loading, copying, and photocopying that the assistants would do, the process could take weeks.
Thankfully now we have large quantities of cheap disk space, high speed networks, and some really great sound library software. Now most supervisors have made a point to get their entire sound effects library loaded onto hard drives and ready for instant access. These might be Firewire drives that sit in their office or they might be network storage hanging off a server. With the new sound library software, supervisors can type in keywords to search for in the database and then click a button next to the description to immediately start auditioning the sounds. Choices can be made and put into a “pull list” or “bin” depending on the terminology used. And then with another click of the button, the chosen sound effects can be downloaded on to a cutting hard drive from or server or set of master drives, and imported into an open Pro Tools session.
As you can probably imagine, this saves HUGE amounts of time.
Three years ago this May, Cameron, the Supervisor I work with, and I set up a system like this for ourselves. We used a program called MTools from Gallery and it changed everything. Suddenly the process that used to take weeks could be done in several days. And we loved it for a long time. MTools is basically a series a utilities applications that interfaced with a Filemaker Pro database. The database would list all of the information about the sounds, and by clicking a button, the path to the sound would be handed off to a program called Dcode which would audition the sound from the server. Dcode is the same program that would copy of the files from the server when the pull list was sent to Pro Tools.
Unfortunately MTools has also been a bit buggy. It’s a problem with a lot of sound software. We’re definitely a niche market and it doesn’t attract the top developers. The Filemaker part of the equation was pretty rock solid. Some people have complained about the speed of Filemaker search hundreds of thousands of sounds but I never found it to be too bad. The problem typically stemmed from the strange behavior of Dcode.
Sometimes it would refuse to run. As soon as it started, it would immediately quit. The solution was to throw out the preferences file and copy “fresh” Dcode from the install disk or the server. Then if you auditioned your first sound, Dcode wouldn’t quit if you hit Command-Q. There is this button with a down arrow that is supposed to rebuild waveform overviews, but by clicking this button, Dcode will quit after you’ve auditioned your first sound. It also has a problem remember the path to the save directory. You run the program, select the directory you want all your sounds transfer into, spend an hour selecting the best sounds, click the transfer button, and it throws up a bunch of error messages. You realize that there’s no longer a path listed in the destination field. There was literally I time two weeks ago when I sat on the phone with Cameron for two hours while we tried to get Dcode to copy the selected sounds to a local hard drive.
We have decided that we need to find a better solution. That is where the demo of Soundminer comes in. It does all the basic functions I’ve talked about flawlessly. And there’s a lot of extra functionality that we never had with MTools. You can customize the layout, font size, color, etc. as easily as you can with a program like iTunes. (You can make customizations in Filemaker Pro and in general it’s a pretty easy database program to learn, but doing something simple like change the font from 12 point to 18 point because the editor has bad eyes is a bit involved.) You can set in and out points on the sounds your auditioning and it will only transfer those portions to your editing session. It has a function where it will find matching sounds based on the characteristics of the sound itself, not just the keywords you typed in. And about a half-billion other things that are great.
I suspect that I will spend most of next week in the office switching us over to a new Soundminer-based library.