G
ETTING STARTED
T
HE BUS
F
ILE OPERATIONS
E
DITING
V
IEW CONTROLS
E
FFECTS PROCESSORS
S
ETTINGS
W
INDOW SETTINGS
R
ECORDING
T
HE CONSOLE
T
RACK GROUPS
I
NTERFACE TRICKS
C
ONFIGURATION
S
PLIT16 AND JOIN16
B
UGS AND SOLUTIONS
To record your first sound go to Settings and select Add Track.
Do this again to add a second track if you desire stereo sound.
Hit the record button in the Playback window.
Type in a filename to record and select a sound format. WAV works better on intel platforms.
Use a mixer program like Mix2000 to select a sound source. Play a sound source through the input and adjust the sliders to get the level.
Hit the record button in the Record window to begin and pause recording.
Hit Save and Exit to save the recording. The recording is now saved on the hard drive under the filename you specified.
Turn the arrows to the left of each track green by pressing them.
To play back, hit the play button.
RECORD BUTTON - each track has a circular, record button on the left which can be either off/grey or on/red. To turn the record button on, turn the button red. Every track with the record on will be affected by edit, effect, or record operations.
MUTE/PLAYBACK BUTTON - each track has a triangular button on the left. When this is green, the track is bussed to the console and processed, allowing it to be heard on playback or saved when played to disk. When it is off, the track data isn't processed so to speed up processing, turn the playback for unused tracks off.
There are many ways to save and load a file, each with it's own pitfalls.
OPEN - This opens a file of any of the supported formats. The selections PCMLOHI and PCHHILO only apply when you intend to open a PCM file. Use them to set the byte order.
SAVE AS FILE - This saves the file as an edit list or a BCAST proprietary file. These formats can't be transported between platforms.
Edit list - this format requires all the original audio files to remain on disk. It only contains the edit list, console settings, and automation, but is the fastest to save. The official extension for this format is .ed but you can use whatever you want.
BCAST proprietary - this format is for saving audio data for every track with the console settings and automation but no mixdown. This allows the original audio files to be deleted without applying the console and automation. The official extension for this format is .bc but you can use whatever you want.
PLAY AS FILE - This saves the file as a conventional audio file. These formats can be transported between platforms. Be aware that Broadcast 2.0 considers the end of the longest track the end of the song, whether its audible or not. If the longest track isn't audible, you'll get long periods of silence at the end of the output.
WAV - mixes down the project to a WAV file with the number of output channels selected. This doesn't save the original data, so any track with its PLAYBACK BUTTON turned off won't be mixed in.
PCMHILO, PCMLOHI - mixes down the project to a raw PCM file with the number of channels selected. This doesn't save values for sampling rate, sample size, or the number of output channels. The byte order can either by LOHI or HILO depending on the selection. Use PCMHILO for most exporting.
Direct CD to Hard Drive transfer - The interface for direct CD to hard drive transfer is straight forward. Just specify a filename to copy to, specify the start and ending positions on the CD and do it. To specify the end of the CD, put the last track + 1, no minutes and no seconds.
This operation is known to work on 12x ATAPI drives. There are a few bugs in the device driver which have been worked around:
1 - If you've been using a CD player like Workman, quit the CD player, eject the CD, reload the CD, and then run the transfer. You may need to run the transfer twice after quitting Workman to get it to work.
2 - If the resulting sound file is choppy, there are lines in the .bcastrc file which are designed to work around this bug. Change the value of NFRAMES. The value of FRAMESIZE shouldn't need to be changed. For a 12x CD-ROM an NFRAMES value of 2 and a FRAMESIZE value of 2352 works best.
3 - Most of the time, the driver is just plain flaky. You need to record several takes and splice them together.
Broadcast uses a transparent edit list for editing. The user has no way of directly seeing the edit list, but the edit list can be modified by shifting edit handles. You need to turn on the edit handles feature to do this. Whether using edit handles or conventional editing, editing is accomplished by first turning on the record button for every track to edit.
Editing is performed the same way it is on a word processor. The original files are untouched. Edits can be adjusted or deleted if Show Edits is on. Editing for automation is also provided. You can cut, paste, clear, or insert segments of tracks or segments of automation. By changing which tracks have record permission, you can cut and paste between tracks. Editing the automation affects all the automation parameters, even the ones not showing.
SHIFT LABELS -You can determine whether editing tracks shifts the labels. Turn this off to prevent the labels from shifting with the audio data.
Broadcast introduces a new concept to hot keys: not having to press the Ctrl key. Just press the hot key for commands in lowercase. Press Shift for commands in uppercase.
PASTE SILENCE - This inserts silence of the length of the selection in front of the start point of the selection. Unlike regular PASTE, this doesn't delete the data in the selection. Use this for offsetting tracks.
TOGGLE LABEL - Insert a label at the cursor position or if a region is selected, insert labels at the start and end points. If a label is selected, it deletes it.
Broadcast was designed to be controlled by the keyboard although all the view functions are in the view menu. Expanding the track expands the vertical space allocated to each track and its automation display without zooming the wave itself vertically. Expanding Y expands the wave itself vertically, allowing waves to overlap. Fit either fits the selection or the entire wave to the screen.
The view menu stays open after you make a selection. To cancel the menu, click outside of it. Zooming is better accomplished with the Interface Tricks.
Effect operations alter every track with the record on, at the same time. Effect operations create an output file containing the affected region. BCAST is the preferred file format for effect files because of it's speed. WAV and PCM formats are available for directly saving the processed effect to a exportable sound file, with no console processing.
MUTE - replaces the selected region with silence.
CROSSFADE - performs a crossfade. The selected region is faded in and a region immediately in front of it, of equal length, is faded out. The two regions are then superimposed.
NORMALIZE - the track data is amplified so that the highest peak is the maximum level possible. Especially useful for avoiding the noise floor of cheap soundcards. The peak can be set above 100 in which case Broadcast clips off the peaks.
BALANCE - mathematically shifts every sample up or down so that the average sample value in a track is 0.
Frequently, the zero sample value of a sound source and the zero sample value of a soundcard don't match. This reduces the dynamic range in the recording. Mixing down many copies of the same track further offsets the zero value and reduces dynamic range. Balancing should be done before normalizing, overdubbing, or reverbing a badly balanced track.
REVERB - create a concert hall reverb effect. The reverb effect has many parameters for tweeking the effect. Although the default values simulate a general concert hall. Value sets can be saved and loaded.
Expect a long wait for high quality reverb. Reverb amplifies any noise or clipping. This effect mixes the data from every recordable track to create a realistic reverb meaning a mono track can be duplicated to two tracks and reverbed into stereo.
OUTPUT CHANNELS - set the number of output channels. Setting more output channels than the single sound card is capable of gives unknown results.
PLAY BUFFER - set the size of the buffer used for playback. The effect of this depends on your soundcard.On good soundcards, or rather, soundcards which OSS provides decent support for, the larger this is, the less breaks in playback but the slower the response to console adjustments. 256 gives smoothest playback. 64 gives best console response without breaks.
On cheap soundcards, the effect of this value is unpredictable. Sometimes a lower value gives smoother playback and a larger value causes breaks. On the Soundblaster, 80 and 256 are smoother than 128.
Be aware that playback buffer in the world of Linux is not the same as DMA buffer. The DMA buffer is fixed. The playback buffer is the amount of data Broadcast throws into the DMA buffer at a time.
In addition to all this, the playback buffer is also the playback buffer used during full duplex record. The playback buffer can't be set so large that the record loop doesn't have enough time to read from the record buffer.
AUTOMATION - you must turn this on in order for automation to take effect.
LOOP PLAYBACK - sets the boundries for looping playback. If a region is selected, the boundries of the loop are the selection otherwise the entire span of the wave is the loop. Once the loop is set, playback can be started from anywhere in the loop boundries. To unset the loop, select LOOP PLAYBACK again.
ADD TRACK GROUP - add a track group as seen below.
Window Settings
SHOW CONSOLE/TRACK GROUPS - Show the console modules or console groups
SHOW TITLES - Display the names of files comprising each edit
SHOW EDITS - Show handles marking the boundaries between different edits. Grabbing an edit handle works just like a cuepoint label. Move the edit around to adjust it. Edit handles in the same file can be deleted with the Edit/Clear function but edit handles between different files can't be deleted.
Only edits in tracks with record permission on are affected by adjustments.
HIGH QUALITY REFRESH - Display the waveform using high quality. This slows down the display.
The recording function creates a new file to record, records into the file, and pastes the file into tracks with the record permission on.
The number of channels to be recorded is determined by the number of tracks with record permission on. The first input channel is assigned to the first recordable track. The second input channel is assigned to the second recordable track, etc. Designating more tracks to record on than the number of input channels supported by hardware gives unknown results.
Full duplex playback is enabled when the playback triangle is lit. If playback is to be used during the take, it should be turned on at time 0 to ensure synchronization. If it is turned on after zero, the playback won't synchronize to the record position. Most soundcards support full duplex recording, but the obsolete drivers used in Linux don't. Harass 4front@ix.netcom.com to get full duplex recording implemented in Linux if you think you deserve it.
Each module on the console is laid out as follows.
EQ1, EQ2, EQ3 - three EQ processors per module. The left pot is level, the middle pot is center frequency, and the right pot is bandwidth.
EQ modules have two modes: EQ and BANDPASS. This setting is determined by the two radio buttons on each EQ processor.
In BANDPASS mode,
only the frequency range selected with the EQ passes through.
In EQ mode, the
original signal passes through and the frequency range selected is added
to it.
PAN - one slider per output channel. The first channel is the top slider. This interface allows more than two output channels.
FADE - adjust level for the track.
The signal path in the console is AUTOMATION -> FADE -> INVERSE -> BP -> EQ -> PAN -> output. This allows digital clipping to be avoided during EQ by turning down the fader.
After passing the track through the console, Broadcast passes it through any TRACK GROUPS it is a member of. Track groups do the same thing a console module does except they can do them to multiple tracks.
Create a track group under the SETTINGS menu. Members of a group can be edited by the group list button on the lower right.
FADE, PAN, POLARITY, and MUTE can be automated by turning on AUTOMATION in the SETTINGS menu.
To display the automation program, select items to view from the WINDOW menu. Automation is accomplished by creating automation points. Grab part of the automation line and drag it to create an automation point. Drag an automation point beyond a point to its left or right to delete it. Hold SHIFT down to automatically set the automation.
When a track is muted with automation, the data isn't processed during playback so processing time is reduced. This can be useful for projects where some tracks are used some of the time and other tracks are used other times, but you don't have enough processing power to play all of the tracks simultaneously.
For regions where the PAN or FADE is to remain constant, decrease processing time by setting the begin and end automation points equal. FADE sections of value oo are treated as muted to reduce processing.
Automation is processed before the console. Cut and paste automation points from the EDIT menu.
Special keys exist for interfacing with the MAIN SCREEN.
p -
start and pause playback on the main screen.
s -
to stop playback on the main screen.
r - record in
the main screen.
Cursor and Page keys -
allow you to zoom and move around the waveform as indicated in the VIEW
menu. This is the preferred method for zooming.
On the RECORD SCREEN other keys may be useful.
p -
start and pause recording.
s -
to save and exit.
l - add label
or toggle one if recording is paused.
^c - start over.
Cursor keys -
allow you to adjust the levels.
On the CONSOLE these keys have effects.
Cursor keys - allow you to adjust all the fades at the same time and scroll up or down.
Broadcast stores relevant configuration information in the .bcastrc file of your home directory. This file shouldn't be edited to avoid crashes, but deleting it should reset the settings.
For undo Broadcast uses the .bcundo file to store information.
For editing Broadcast uses the .bcclip, .bcautoclip files.
This distribution of of Broadcast comes with split16 and join16, a lossless compression scheme for 16 bit audio files of 1 or 2 tracks. The latest split16 and join16 REQUIRE Bzip2 to be installed on your system.
Bzip2 is the bleeding edge of compression technology, but split16 further enhances it to compress sound as much as 60%. Just run split16 to compress a file and join16 to decompress it. To compress BCAST files, regardless of the number of tracks, specify 1 channel for the compressor. This only works with 1 or 2 tracks.
Other solutions for Windows have copied this scheme, but they don't have bzip2 technology, they cost money, and they've been around shorter. The compression ratio depends on the complexity of the waveform.
X Error of failed request: BadMatch (invalid parameter attributes) Major opcode of failed request: 70 (X_PolyFillRectangle) Serial number of failed request: 76 Current serial number in output stream: 263