Squeak-dev December 1999

squeak-dev@lists.squeakfoundation.org

146 participants
451 discussions

Start a nNew thread

Celeste and duplicates.
by Josh Flowers 01 Dec '99

01 Dec '99

If Celeste is set to leave messages on the server, is there a way to keep from downloading duplicates on every fetch?

1 0

RE: Life
by Norton, Chris 01 Dec '99

01 Dec '99

>>Does anybody have an implementation of life for Squeak? >I don't know if this is what you want, but here's a port to Squeak of my BitBlt Life program that first appeared in the August 1981 issue of Byte magazine. Thanks Dan! That was fun to play with; it works pretty fast on my windoze box. Cheers! ---==> Chris

1 0

[Bug] Scamper background
by Torsten.Bergmann＠phaidros.com 01 Dec '99

01 Dec '99

> Adds support for background color and background image to > Scamper. This is > an ALPHA version, it works, but I've tested it only for 15 minutes. Using InfiniteForm for displaying the background image will result in the following effect: The image always starts at the upper left corner of the screen, not at the upper left corner of the scamper web pane. Depending on the screen position of Scamper you will see the background image with different positions. See http://software.freepage.de/cronos/download/bgscampereffect.gif for an example. It's good for tiled images, but not for general background images. Bye Anoulak and Torsten

2 1

Life
by Gerardo Richarte 01 Dec '99

01 Dec '99

Does anybody have an implementation of life for Squeak? The point is: I've done one, using BitBlt, or Form, or Canvases or something like one of those (I don't remember which). The cells where directly pixels, and I used pixelValueAt: aPoint (I think) to access it. I tried this approach becouse I think that it'll faster to access, and ofcourse faster to draw on the screen... but it's still slow. My intention is to make an infinite life field (without bounds). So... does anybody have a fast implementation of this? Alive Bye! Richie++ -- A390 1BBA 2C58 D679 5A71 - 86F9 404F 4B53 3944 C2D0 Investigacion y Desarrollo - CoreLabs - Core SDI http://www.core-sdi.com --- For a personal reply use gera(a)core-sdi.com

2 1

Re: [FIX] Various fixes for Celeste
by Jan Theodore Galkowski 01 Dec '99

01 Dec '99

Here's a link to a site which sounds authoratative: http://www.magnet.ch/serendipity/hermetic/cfunlib/datlib/datedoc.htm --- ___________________________________________ demiourgos(a)smalltalk.org squeak.org/ home.stny.rr.com/algebraist/ www.smalltalk.org/ jtgalkowski(a)alum.mit.edu On Thu, 25 Nov 1999 13:14:46 David T. Lewis wrote: >On Thu, Nov 25, 1999 at 12:04:37PM +1100, Eric Ulevik wrote: >> >> Many systems use Julian day numbers. Days since noon GMT on 1 January 4713 >> BC in the proleptic Julian calendar. This would be preferable if we ever >> want historical dates. >> > >Is there a known way to calculate number of seconds between days since >noon GMT on 1 January 4713 BC in the proleptic Julian calendar, and >some other point in time (beginning of Unix time, or beginning of >Smalltalk time)? If so, can you give a pointer as to how to determine >this? > >> >> Hope this helps, >> >> Eric Ulevik >> >> > >Yes, thanks! > > Get your FREE, private e-mail account at http://www.salonmember.com.

1 0

RE: Some questions about Squeak
by Stephane Ducasse 01 Dec '99

01 Dec '99

Thanks So what is the relationship between World and Display then? For the morphic turtle I know but I would like to have something working for friday and I can only work on it before bedtime ;)

2 1

What is the current State of MIDI I/O?
by Daniel V. Oppenheim 01 Dec '99

01 Dec '99

--=====================_14947071==_.ALT Content-Type: text/plain; charset="us-ascii"; format=flowed Following are some initial observations on MIDI I/O. I would appreciate a pointer to more detailed information, future plans, etc. before I plunge in... I have been looking at the MIDI I/O primitives in Squeak 2.5, on Windows NT. It seems that there are two classes: SimpleMidiPort and MidiPrimTester. Seems that SimpleMidiPort is intended for public use. However, it lacks some of the primitives found in the MidiPrimTester. For example, I copied the primMidiGetClock primitive from the MidiPrimTester, hoping to be able to schedule events into the future. However, the time argument in primMIDIWritePort:from:at: is relative to the time the call is made. So at: 0 plays now, at: 500 in half a second. This means that scheduling has to be handled entirely by Squeak (whereas I was hoping I could get currentTime and then schedule events into the future relative to it using a much more relaxed algorithm). What is the buffer size for the MidiOut? Will the output buffer sort MIDI events if they are given out of time -sequence? If so, will sorting be smart, so say schedule controls and note-off events before note-on events with the same onset? From what I could tell, Midi input seems to lack in several respects: 1. no time stamps on incoming events. This can really hinder serious real-time interaction with Squeak. I'm not sure if one could make a MIDI recording that maintains correct expressive timing, especially if squeak is simultaneously working on other tasks and/or MIDI playback. 2. data comes in a byte stream, rather than already parsed into events. Extra work on the Squeak side. 3. no notification of midi input from system into squeak (using a semaphore for example). One must therefore implement a pooling loop when expecting input. 4. it would be nice to be able to flush the buffer. (this is picky, it can easily be done from squeak). Synchronized MIDI and Audio output. I'm pretty sure you can schedule wave audio from Squeak. Can MIDI and Audio play simultaneously? Can they be synchronized -- say if one has a score comprising both MIDI and SoundFiles events? If so, is this ability platform specific or does it work cross platform? Has anyone given any thought to adopting something like MidiShare (http://www.grame.fr/)? It is in the public domain, supported on most platforms, and quite rich in nifty and useful features. Thanks! Danny Oppenheim --- Dr. Daniel V. Oppenheim Computer Music Center IBM T.J. Watson Research Center phone: (914) 945-1989 P. O. Box 218 (or Route 134) fax: (914) 945-3434 Yorktown Heights, NY 10598 www.research.ibm.com/music --=====================_14947071==_.ALT Content-Type: text/html; charset="us-ascii" <html> Following are some initial observations on MIDI I/O. I would appreciate a pointer to more detailed information, future plans, etc. before I plunge in...<br> <br> I have been looking at the MIDI I/O primitives in Squeak 2.5, on Windows NT. It seems that there are two classes: SimpleMidiPort and MidiPrimTester. Seems that SimpleMidiPort is intended for public use. However, it lacks some of the primitives found in the MidiPrimTester.<br> <br> For example, I copied the primMidiGetClock primitive from the MidiPrimTester, hoping to be able to schedule events into the future. However, the time argument in primMIDIWritePort:from:at: is relative to the time the call is made. So at: 0 plays now, at: 500 in half a second. This means that scheduling has to be handled entirely by Squeak (whereas I was hoping I could get currentTime and then schedule events into the future relative to it using a much more relaxed algorithm). <br> <br> What is the buffer size for the MidiOut?<br> <br> Will the output buffer sort MIDI events if they are given out of time -sequence?<br> <br> If so, will sorting be smart, so say schedule controls and note-off events before note-on events with the same onset?<br> <br> From what I could tell, Midi input seems to lack in several respects:<br> <br> 1. no time stamps on incoming events. This can really hinder serious real-time interaction with Squeak. I'm not sure if one could make a MIDI recording that maintains correct expressive timing, especially if squeak is simultaneously  working on other tasks and/or MIDI playback.<br> 2. data comes in a byte stream, rather than already parsed into events. Extra work on the Squeak side.<br> 3. no notification of midi input from system into squeak (using a semaphore for example). One must therefore implement a pooling loop when expecting input.<br> 4. it would be nice to be able to flush the buffer. (this is picky, it can easily be done from squeak).<br> <br> Synchronized MIDI and Audio output. I'm pretty sure you can schedule wave audio from Squeak. Can MIDI and Audio play simultaneously? Can they be synchronized -- say if one has a score comprising both MIDI and SoundFiles events? If so, is this ability platform specific or does it work cross platform?<br> <br> Has anyone given any thought to adopting something like MidiShare (<a href="http://www.grame.fr/" eudora="autourl"><font color="#0000FF"><u>http://www.grame.fr/</a></u>)</font>? It is in the public domain, supported on most platforms, and quite rich in nifty and useful features.<br> <br> Thanks!<br> <br> <x-tab>        </x-tab>Danny Oppenheim<br> <br>  <br> <br> <div>---</div> <br> <div>Dr. Daniel V. Oppenheim                                                                                                                     </div> <div><x-tab>        </x-tab>Computer Music Center</div> <div><x-tab>        </x-tab>IBM T.J. Watson Research Center<x-tab> </x-tab>     phone: (914) 945-1989</div> <div><x-tab>        </x-tab>P. O. Box 218 (or Route 134)<x-tab>    </x-tab>     fax:   (914) 945-3434  </div> <div><x-tab>        </x-tab>Yorktown Heights, NY 10598<x-tab>      </x-tab>     <a href="http://www.research.ibm.com/music" EUDORA=AUTOURL>www.research.ibm.com/music</a>  </div> <br> </html> --=====================_14947071==_.ALT--

1 0

debugging broken in 2.6?
by Tim Olson 01 Dec '99

01 Dec '99

Has anyone else had difficulty debugging in 2.6? I used to insert a "self halt" at the point I wanted to drop into a debugger, but that doesn't seem to work quite right, anymore (has to do with the new Exception Handling package?) I tried "Halt signal", and that seems to stop at the right point, but the keys for step (t) and send (e) don't work, and sometimes stepping winds up far down the context stack from where I was. -- tim

4 3

Newbie: Enclose Command Key Actions
by Daniel V. Oppenheim 01 Dec '99

01 Dec '99

On Windows NT, some Command Key actions don't seem to work correctly. This is probably already well documented -- my apologies! 1. Some enclose keys don't all work. ' ', [ ], are fine (however, using ALT, not Command as specified in the help). But I can't get " ", or ( ) to work no matter what combination I try (Command, Shift, Alt). 2. Copy Cut and Paste use the Mac Alt key. No big deal, though I would like to use the standard Windows Control instead. Is there a simple way I can fix this? --- Dr. Daniel V. Oppenheim Computer Music Center IBM T.J. Watson Research Center phone: (914) 945-1989 P. O. Box 218 (or Route 134) fax: (914) 945-3434 Yorktown Heights, NY 10598 www.research.ibm.com/music

1 0

Re: Tetris
by Bob Arning 01 Dec '99

01 Dec '99

On Wed, 1 Dec 1999 02:27:48 -0600 Michael Donegan <invader(a)zilker.net> wrote: >The changes I had made up to this point not only made it faster, but also a >lot easier to understand. I'll give it another shot. Michael, If you want a somewhat simpler place to restart from, I have cleaned up the code a bit to remove some of the Java-nasties: http://www.charm.net/~arning/tetris2.01Dec0841.cs This does not really address performance issues, but it is easier to deal with than the original. Cheers, Bob

1 0

Jump to page: