Thursday, November 10, 2005

Learning Curves - Steep they are - But worth it!

An AudioBox (any flavor - 1616HD, AB1616 or the latest and greatest - AB64) is a matrix mixer that is completely under software control and responds to Midi Show Control commands (MSC), the full spec is located on Richmond's site as in here. This is for V1.0 of the spec.

A better description of what an AudioBox is from www.kiss-box.com : It's a "
Digital Audio Playback, Signal Processing, Sound Placement and Spatialization fully utilized with the AudioBox virtual matrix and hard disk playback/recording system". Please note, the AB64 allows for recording as well as all of the preceding.

The venue I usually work in does not have nor need any live sound reinforcement system (we seat 92 and the stage is considered a "black box") and so our designers have limited experience in using a common mixing system.

That said, the most difficult part in programming the AB1616 or ANY AudioBox is to understand that some programming has to be done to get ANY output at all!

Let me back up a moment.

Any RSD AudioBox supports a combination of live inputs and disk playback channels linked to a similar quanity of output channels. The AB1616 I have supports 16 live or disk based (or some combination that can be changed on a cue to cue basis) "playback" channels (best to think of them as sources of audio that may or may not be a stereo pair, and, that can change on a cue to cue basis too) that can be routed (sent) to 16 output channels.

On top of that, you can adjust EQ, delay, fade up/down/pan levels all on the fly - depending on the needs of the show, under program control (using ABEdit and MSC) - or by using a CM Labs MotorMix, in conjunction with the RDS AB, you can make similar adjustments LIVE!

Sidebar: Ever adjusted all your playback levels perfectly in tech, only to find with an audience in the space, your levels are too low? With SFX, there is no way to adjust ALL or ANY of your output levels on-the-fly. With a RSD AB and a CM Labs MotorMix, you can assign submasters to various inputs & outputs and adjust them independently or in groups. For those of you familar with lighting boards, the RSD AB subs work just like the subs on a lighting board - they raise or lower all assigned channels in proportion.

Back to what is an AB and it's learning curve.

Understand this [and it's mind boggling, and stressful if you're doing it on a deadline] - these are what you can program (and in some cases HAVE to program) to get your expected output:

  • Submaster assignments (there are 32 subs that can be assigned under program control)
  • Submaster zero point [the AB is a unity gain box, so it doesn't boost the signal but can cut it, you can setup your subs so that they let more of a hot signal in or provide cut too. It's all in how you adjust the gain of the box]
  • Input assignment: Live feed or disk playback
  • Input level, EQ, delay
  • Crosspoint: level, input & output
  • Output level, EQ, and delay
  • Fade up or down speed and stop point on all Inputs and Outputs.
  • BTW, the EQ and delay can be adjusted in real-time.
ABEdit is the free programming and playback software provided for the Windows platform.

Free Windows and Macintosh programming software is available on Richmond Sound Design's site. I'll talk about the ABEdit learning curve in a later part. BTW, this all has a happy ending - at least for the ZCT's recent show: The Haunting of Hill House.

Note: It's also good to have all your effects gathered up, approved by the powers that be BEFORE you begin programming the box.

Because if you don't, it's pure hell. Making a stressful time even more so.

How many AudioBoxes are there?

Previously, I've said there are 3 kinds of AudioBoxes (all are available from RSD but only the AB64 is available as new):

  • First Generation: a 1616HD which sports 8 live inputs, 8 disk playback inputs and 16 output channels, both MIDI and SCSI inputs and uses SCSI from a PC or Macintosh to load wav files to the internal SCSI disk storage. MSC is the control software command language.
  • Second Generation: an AB1616 which sports 16 live inputs, 16 disk playback inputs (which only 16 can be active at any one time) and 16 output channels, both MIDI and SCSI inputs and uses SCSI from a PC or Macintosh to load wav files to the internal SCSI disk storage. MSC is the control software command language.
  • Third Generation: an AB64 which is the most flexible AudioBox to date. Instead of using a SCSI interface to load wav files, it is a box that is network aware and uses TCP/IP to transfer audio files and provide control. It also has dual MIDI inputs, outputs, and thru connectors. MSC is the control software command language. It also uses IDE drives and can support 2 drives in a mirror 1 configuration. Nifty, eh?

There are also hardened AudioBoxes available that are more suited to the theme park and entertainment industry. They are custom built for the client and contain, among other changes, solid state disks in place of SCSI or IDE drives and the customer's connector choice for I/O and power.

Major technical specs for all the "stock" AudioBoxes can be had on Richmond Sound Design site as well as Harmonic Function's site.

Back to points that can make your life difficult if you forget about them or OK and easier if you remember:

ALL AudioBoxes use audio files that are at the 48K sample rate. It's the AES standard - see RSD FAQ. No matter how they started out, 48K is how they wind up stored in the AB.

Furthermore, in processing by the ABEdit software, the file is changed to 48K sample rate and split into 2 mono files, a right and a left.

If a file is already at the sample rate of 48K, the ABEdit software only splits it into a right and left file, assuming it started out life as a stereo file.

I use Sound Forge to edit my audio files, and to rip CD tracks and to process these tracks further. Sound Forge also supports the batch conversion of files to another sample rate such as 48K - you can do that to save a step in the ABEdit software. I didn't on this trip because I wasn't very organized - or at least not organized like an AudioBox show needs to be.

Files are transferred from the computer to the AudioBox using SCSI (if the AudioBox in question is a 1616HD or AB1616) and via a network connection (TCP/IP) if an AB64.

Note: It's a good idea to develop a protocol/procedure that normalizes your files BEFORE you transfer them and resample them in the AB. In Sound Forge, I usually normalize music files to -16db - there's a preset for that and there is a preset for speech (voiceovers) too. Sometimes, I drop the level even more.

Here be the reason - yes, the ABs have analog inputs and outputs but they process audio in the digital domain. Too hot of an input signal WILL result in distortion - it's a crackle that coincides with the peaks of the audio signal. Reduction of the channel input level (in the AB) may alter it but won't get rid of it.....

You want a procedure that nomalizes the signal before it's transferred to the AB. You also want to consider whether or not you're going to adjust EQ in the AudioBox - some EQ will increase the level of the signal resulting in digital distortion - it may be a trial and error process. Most likely will be.....

You can adjust the level of the file that gets transferred by the ABEdit software at the time you resample and split to right and left channels - BUT - this is global for all files transferred in that session. If you normalize the file in your editor, that's your best choice. Then take the default level in ABEdit when you transfer the files.

Here's the gotcha (one of many):

The AudioBox is a unity gain device - you can cut levels, but not boost them. Depending on how you set up your overall gain structure, you may be able to increase the I/O levels (because you've been cutting the level all along) via the submasters. But you have to take into account what is the loudest signal you need in your venue and make sure your gain structure allows it. Trial and error again...
------------------------------------------------------------------------------------------------

Back to basic setup and what works and what doesn't quite....

Your computer needs to be a PC with a SCSI card or interface if you're running the ABEdit software.

SCSI cards and OS that I have tried are - W2K pro with a SIIG AP-20. I suspect a Kouwell KW-910U will work too as they look identical right down to the contents of the manual. These are PCI cards that you have to open up the computer to install. It's very easy and safe.

I had success with a Belkin F5U115-UNV USB to SCSI adapter under WinXP home; however when I moved it to the W2K machine, it crashed the box. Twice. 1st time I knew what had happened, second time I didn't. Time to install a "real" SCSI interface card. No crashes to date with the SIIG AP-20 on the W2K machine.

Audio (wav) file transfers from the PC to the AudioBox are much faster using a SCSI "real" interface card compared to the USB card.

I'm in the process of evaluating a RATOC firewire to SCSI interface for use by the AB1616 and I'll let you know how that turns out. So far, on a Dell running WinXP Home, it doesn't look promising......I haven't tried it on my sound machine running W2K yet.

Some Working procedures:

[at least, this is what I used during my initial voyage]

Basic routing on the AudioBox I used went like this:
So my content approvers could hear the effects via the amps and speakers they would hear the final effects -

Sound computer has an Echo Audio Gina24 2 in 8 out sound card in it which I routed Audio Out 1[Left] & 2[Right] (set up as the default output in Sound Forge) to the AudioBox inputs 1[Left] & 2[Right]

I then patched AB Input 1 to AB Outputs 1, 3, 5, 7, 9, 11, 13, 15
and then patched AB Input 2 to AB Outputs 2, 4, 6, 8, 10, 12, 14, 16

I created a cue that I could recall at will to recreate this routing w/o interferring with the real cues of the show.

Now, I also created a cue that said patch AB Inputs 1, 3, 5, 7, 9, 11, 13, 15 to AB Output 15 and another one that patched AB Inputs 2, 4, 6, 8, 10, 12, 14, 16 to AB Output 16.

This cue I left in my setup or turn-on cues - this let me plug in a headphone amp into outputs 15, 16 so that I could monitor the AudioBox on headphones.....letting me program the box w/o having to program it and run it "out loud" in the venue (or at home). I could also listen to Sound Forge this way too, helping the editing process.

ZCT only has 8 channels of amps so dedicating AB Outputs 15, 16 to a headphone amp is no big deal....

But I'm getting ahead of myself. This programming stuff.

Prep and approval of audio files BEFORE you program the AudioBox!!!!!!!!

You HAVE to have the idea that all sound content is appoved by the powers that be BEFORE you start to program the AudioBox.

Edit all your content, for level, looping, time, whatever and save it to disk with a name that makes sense - perhaps it names the cue page or what the cue is or SOMETHING that tells you when you're looking to place this file in the AudioBox what it is.

Also, make sure you still have a copy of the original file, and the newly edited version as well as all the ones that the director didn't like (because, at some point, he or she may ask you to put in the one they didn't like 3 days ago but do now.....)

When you are satisfied with the sound content for a particular cue and you can get the powers to OK the content, THEN transfer the file to the AudioBox.

If you don't do this, then you'll be caught in a loop where you have to constantly re-edit and re-transfer the sound content over and over again. Under the stress of Tech or the Direcor or TD hovering over your shoulder will make you wish you still used razor blades to edit audio.

Maybe you're more organized than I am - but let me tell you, it isn't fun to have to re-edit a file and re-transfer it because the loop isn't perfect (or, more likely, they didn't need it looped but do now....because something with the set design no longer works and they need fill)....

OK, you've got your content, you have it approved, they love it and you have it transferred to the AudioBox.

Now what? You're not done by any means but tomorrow is another day.

Next time out - some more basic programming and working procedures. I have to collect my resources on this so that it's more concise than what I've written so far.

I'm hoping to go into what the ABEdit manual is not as well as get permission to use some of Richard Ingraham's content from his web site (or link to it anyway). I already have permission to link to RSD's site - they have some very good papers on-site that I think actors, sound designers, directors, and other tech personnel can benefit.

Thanks

0 Comments:

Post a Comment

<< Home