Gary, That's great to know! There are natural questions, such as whether the rectangle is in screen or morph-relative coordinates (I fear the former), and how to get the same information when rendering (to allow non-visible objects to be skipped), but I'm guessing those answers will flow from browsing beginning with #invalidRect:.
At the last steps of this journey, I'm going to need every clock cycle I can get :( Thanks! Bill ________________________________________ From: [email protected] [[email protected]] On Behalf Of Gary Chambers [[email protected]] Sent: Friday, June 11, 2010 10:47 AM To: [email protected] Subject: Re: [Pharo-project] GUI Polymorph Take a look at Morph>>invalidRect: as opposed to #changed for more optimal screen updates if required ;-) Nice job on the tutorial Hilaire btw. Regards, Gary ----- Original Message ----- From: "Schwab,Wilhelm K" <[email protected]> To: <[email protected]> Sent: Friday, June 11, 2010 2:59 PM Subject: Re: [Pharo-project] GUI Polymorph Hilaire, I will indeed consider it, but it will take me time to get to your level of understanding of Polymorph. My priorities are networking, serial communications, printing (which I am dreading) and then finally GUI construction. The last two are backwards, but the GUI does me no good if I can't print on Linux. Somewhere in there I hope to pull a big thorn out of my foot by moving various machines to Linux. With a working GUI+printing, the rest of the boxes go to Linux, and the drinks are on me :) Understood about your objectives, but I think they SHOULD include teaching efficient updates. When MS did this correctly, they did the simplest thing that would work, and then separately modified it for efficiency. Smalltalk is not the fastest language in existence, Pharo is not the fastest among its peers, and even if it were, we should be doing things better and more efficiently than the rest of the world. It is nuts that we need 2GB of RAM to boot an OS and read email; while that sloppiness arguably begins in Redmond, it can be attenuated here. Bill ________________________________________ From: [email protected] [[email protected]] On Behalf Of Hilaire Fernandes [[email protected]] Sent: Friday, June 11, 2010 9:46 AM To: [email protected] Subject: Re: [Pharo-project] GUI Polymorph Schwab,Wilhelm K a écrit : > Hilaire, > > It was interesting to watch MS' position on GUI examples change over time. > Visual C++ started out teaching how to create efficient software. Then > one day those examples were gone and all they could talk about was "smooth > scrolling" which turned out to be a way to make the machine do the > inefficient thing four times instead of once. > > The idea of altering the model, finding out what is invalidated, marking > that as closely as can be done within bounds of common sense (or until > diminishing returns), and then drawing just the changed parts is > non-intuitive and VERY important. You have done a wonderful job so far, > but optimization deserves attention. Please fell free to search and to document it. As I wrote in my previous email, my objective writing this documentation was not how to efficiently implement a paint canvas. Hilaire _______________________________________________ Pharo-project mailing list [email protected] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project _______________________________________________ Pharo-project mailing list [email protected] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project _______________________________________________ Pharo-project mailing list [email protected] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project _______________________________________________ Pharo-project mailing list [email protected] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
