> The Windows Task manager will display that a program is not responding, even > though it is still running perfectly fine. I see this on other programs as > well. Perhaps a more informed expert on windows can explain, but I believe it > has to do with the program not accepting input interupts from windows while > processing.
While R:Base (or any other disk or processor intensive program) is off "doing its thing" Windows has no idea of what it's doing or whether it's still "alive". Any "messages" that Windows sends to the application -- in particular the "part of your window has been obscured or revealed, so please paint yourself" message -- are ignored by the application. In such cases, it's possible for the application to periodically set aside the work it's doing to process the messages it has received from Windows. If it does this frequently enough it will appear to be "normal". If it does it only occasionally you get a "white window" that suddenly repaints, or the jerky window-drag effect you get with some programs. Of course, taking the time to process these messages means that the main work that the application was doing is going to take longer. For that reason, you don't want to do this too often. If I had to guess, I would say that probably what's happening is that work that happens entirely in the R:Base engine causes a suspension of all Windows message processing until control returns to the front-end. If it's really a problem for you, consider breaking up the operation into smaller chunks that return control to your program in between (for instance, appending 100 or 1000 rows at a time, then updating a status message). -- Larry
