Re: CaWave

[mooney@oxygen.aps1.anl.gov (Tim Mooney)]

Ben-chin, Mark,

I've started using IDL/CaWave and have a few comments:

1) This is great!  It's much easier to make channel-access calls as the
EPICS model suggests in IDL than it is in 'C'.  In fact, it's actually
easier to do CA calls right than wrong in IDL.  I was able to get some
crude but useable scan/display software (my first IDL program) going in
three days--much more quickly and with much better performance than a
scan application I did (wrong) in 'C'.

2) CaWave.pro detects that it is working with IDL, rather than PVWave,
by the occurrence in a directory path of the string 'idl'.  Until a
more comprehensive mechanism is developed, that string comparison
should be made case insensitive.

3) I've noticed that caput()s don't actually take effect until
capendevent, or the next caget() of any PV; capendio does not flush the
output queue as I've come to expect from working with other
channel-access software, and an earlier call of the form 'capendio,
time=x' also does not cause a queue flush by time 'x' after the
caput().  (In case it matters, the PV in the caput() call was not
attached to a monitor.)  I don't think this behavior was intended, and
it disagrees with my reading of the online documentation (very helpful,
by the way).

Tim Mooney