On Sunday, May 16, 2004, at 13:01 US/Pacific, David W. Fenton wrote:
On 16 May 2004 at 12:29, Philip Aker wrote:
But to be fair, since it's generally acknowledged that Finale has features the competition doesn't, it stands to reason that the sheer number of object types supported in permutation and combination with the variety of means they may be placed will increase the number of calculations required by factorial leaps. That is part of the price one pays for flexibility.
Why should those complications be exposed at the display level?
Sometimes it's necessary to recalc everything and sometimes not.
Finale shouldn't repaint parts of the screen that have not been updated.
Unless there's been a change that affects at the global or regional level. For a change affecting only a local level (like dragging an articulation), then most certainly not.
It obviously isn't smart enough to tell the difference, so it repaints everything.
I would have to agree that there are some cases (like in the Expression Designer dialog on MacOS 9) where this is shown to be true. But that has naught to do with Finale keeping track of document draw state.
So, the problem is *not* in the interface between Finale and the graphics subsystem -- it's in what Finale is telling the graphics subsystem to do in the first place.
Previously I wrote:
"I believe it's more related to the use of the native drawing APIs (at least on Mac)."
On Mac (Carbon) I believe I can prove it because the Extract Lyrics plugin offers similar UI features to the Expression Designer on OS 9 but does not do unnecessary redraws (as occurs when switching between tab views). Problem is Finale's using just about the same code on OS X and only *looks* better because of native double buffering. It's still wasting cycles triple drawing.
On the other hand, is anyone complaining about the loss of redraw interrupt on Windows? I don't think so -- I certainly don't see it as an issue that I'd worry about, since the screen redraw in WinFin2K3 is fast enough to be no big deal (and I'm not on up-to-date hardware -- I'm using a 1999 PC).
I can remember trying to deal with a 1 Meg score on my SE/30 (25 Mhz). That was slow!
Philip Aker http://www.aker.ca
_______________________________________________ Finale mailing list [EMAIL PROTECTED] http://lists.shsu.edu/mailman/listinfo/finale
