Yes. All records should be listed in one command-line argument, following the -r argument on the command line; separate the record names by `+' characters. For example, the command
wave -r 237+237n &starts a WAVE process group, opening a separate WAVE signal window for each of record 237 and record 237n. Each signal window corresponds to a separate WAVE process; thus you may browse through and edit each record independently. The signal window assigned to the last record in the group (in this case, 237n) is the master signal window; it contains a
It is also possible to synchronize by clicking on a point of interest in the
master signal window. To do this, press and hold
while clicking
with the middle mouse button. The master signal window will be unaffected,
but the other windows are redrawn so that they begin at the time corresponding
to the point of interest. This feature is particularly useful if you have
a derived record such as in the heart rate signal analysis example (see
section 3.2) open in the master signal window, and
the raw signals in another window; you can then click on a feature in the
derived signal and instantly see the corresponding raw signals.
Remote-control applications such as wavescript and wave-remote can also start or control a WAVE process group. All signal windows in the group move synchronously in response to commands from wavescript or wave-remote. You may change the record that is open in any signal window, either via the Load window or using a remote-control application. You may quit from any signal window without affecting the others.
Note that there are a few important limitations:
Taken together, the first two of these points imply that if you decide after opening a record that you would like to add another record to the group, it will be necessary to exit and start again, naming all desired records in the group on the WAVE command line when you do so.
If you wish to move through two records sampled at the same sampling frequency in lockstep (for example, to view digitally filtered signals and the original unfiltered signals side-by-side), another approach is to create a single header file that names the signal files for both records. It is not necessary for all signals to be in the same signal file, or even on the same disk drive.
George B. Moody (george@mit.edu)