Hi again. In order to fix the remaining console input problems, I have just added a new "Command Interpreter" thing to the GUI (both branches again). The "Command Interpreter" is a dialog box, like this: +---------------------------------------+ | < text input field > | +---------------------------------------+ | Ok-button | Cancel-button | +---------------------------------------+ So the simple idea is that using this dialog, the user can type simple command strings like "add light" etc that are processed. The initial version of the interpreter now can understand two commands: 1) "help" 2) "add light" In the popup-menu(s) there is a new item "Enter a Command..." that will open the dialog. Well, at first sight, this does not seem very useful. But I think the most powerful way to use the interpreter is to use it with the popup-menu; now selecting "Render/Lights/Create New Light" will no longer ask any questions through the UNIX console, but it instead pops out a new Command Interpreter dialog, with a _default_ command string that will add the simplest light object once the user accepts the command string by clicking the "Ok" button. So, the user didn't have to use the UNIX console anymore, which is exactly the desing goal of this system. Ok, probably just writing nice and shiny dialog boxes would be better than using these commands, but the problem is that dialogs are expensive to write (in terms of time). The "add light" is just one of about dozen cases where UNIX console still is used, but using the command box it was very quickly moved under the GUI. Ok, I'll take this quick route away from console, and keep converting the "cin >> " stuff into these commands (so that a useful default command string could be show to the user in each case). Later, if we have time and interest, we still can convert them into more user-friendly dialogs. Regards, Tommi