From: Tim M. <or...@le...> - 2013-05-30 10:38:51
|
On 05/28/2013 11:00 AM, Tom Breton (Tehom) wrote: > I'm going to answer the biggest issue first: > >> In the not-so-minor annoyance department, I found the user-interface >> dialog absolutely bewildering. It seemed to have little to do with >> expression controllers, and only after a great deal of trial and error >> was I able to work out what each item was supposed to do. However, >> once I figured it out, I found the new features quite powerful. >> >> From a programmer's point of view, the one-size-fits-all approach to >> the controller interface may be elegant, but from a user's perspective >> it is not. For the sake of clarity each controller should have its >> own specialized interface dialog that clearly presents only those >> options appropriate to that controller. > OK, clear interface design is not my strength and I welcome help. > > But all the options are available to all the controllers. For instance, > you might apply vibrato to expression controllers; strings with > expression vibrato + modulation vibrato is sometimes useful. So > specializing on controllers vs pitchbends wouldn't help. One of my basic concerns is whether or not it is appropriate to make all options available to all controllers. Is vibrato really appropriate for Pan (10), Chorus (93), Reverb (91), or Sustain (64)? Perhaps the appearance of vibrato options in the dialog could be made conditional, based on the controller chosen. A problem with the term "Vibrato" is that it refers to the frequency modulation of an audio tone. "Tremolo," on the other hand, usually implies amplitude modulation. Perhaps replacing "Vibrato" with "Vibrato / Tremolo" would make it clear that this option is not limited to frequency. "Vibrato wavelength" seems a bit murky. I think it would be clearer if, instead, the number of vibrato cycles in the selection could be specified. Or, better still, the vibrato frequency in hertz. The number of half cycles of vibrato could be calculated from the available time interval, the vibrato frequency, and the current tempo. > Looking over the dialog, I see a number of places where the text assumes > pitchbends or is just unclear. Let me run these possible renamings by > you and everyone else: > > * "Pre-bend" -> "Before Ramp"/"Before Bend" > > * "Value (%):" -> "Start at value:"/"Start at value (%):" > > * "Duration (%):"" -> "Wait this long (%):" > * Tooltip: "How long to wait before ramping, as a percentage of the > total time" > > * "Bend Sequence" -> "Controller Sequence"/"Bend Sequence" > > * "Ramp duration (%):" -> "Ramp duration (%):"/"Bend duration (%):" > * Tooltip: "How long the bend or ramp lasts, as a percentage of > the remaining time" Yes, I think these changes would greatly clarify the dialog. > * "Preset" -> I don't know. I can see where a user might not guess > what "preset" means, but what's a better word or short phrase to > convey the meaning? > * Tooltip for preset: "Use this saved or built-in setting. You > can edit it. For saved (but not built-in) settings, that will > become the new saved setting" I have no problem with "Preset," but I found "User 1, User 2, User 3, ..." a bit confusing. My first impression was that it was an attempt to set up user accounts within a controller dialog. Perhaps "User Setting 1, User Setting 2, User Setting 3, ..." would make clear what's going on here. >> * The function "QSpinBox.setAccelerated(bool)" would be a useful >> addition to spinBoxes, as it permits using a small step size for >> precision, without requiring the user to wait forever for the >> thing to crank around. > Good idea. I hadn't previously known about setAccelerated. I assume it > makes up/down arrows hyperbolic instead of linear? We could probably > use that on a lot of dialogs. When an arrow is held down, the spinBox performs normally for the first second or so, after which it kicks into a high-speed mode. > While I was at it, I changed the vibrato waveform to be a (more correct) > sine wave rather than the triangular wave we have been using. The > amplitude and frequency scalings are unchanged. Good idea. Tim Munro |