etoys-dev November 2006

etoys-dev@lists.squeakfoundation.org

13 participants
32 discussions

[Etoys] Impressions on B1 Machine
by Yoshiki Ohshima 25 Dec '06

25 Dec '06

Hello, Since we (people at Viewpoints Research) got couples of B1 Machines, we've been experimenting with them. All in all, it is really nice to have the machine! First some comments and questions: * Unfortunately, the touch pad is not working well. To experiment various (software related) stuff on the machine, using an external mouse makes a lot of sense. * The infrastructure mode of wireless works with my 802.11b wireless router. However, we are having hard time to do it with our office network. It is to do with the router is 802.11g, the wep key is different or some other reasons. * Is there any way to set up ad hoc (mesh) network between units? I think I saw a wiki entry that says it is not working perfectly, but would like to try it. Now, a bit of my thought: Speaking of experiments, I would like to encourage the developers to use Etoys in various ways. It is pretty much the only thing that you can *make* something on the actual machine, currently. For application designers and developers, or content designers, it should give some sense on what it is like to do stuff on it. Its UI is different from the standard OLPC idioms, but the available functionality of Etoys and Squeak still let you do a lot of prototyping. For example, there is a paint tool in Etoys. If you would like to feel the painting experience on the machine, you can try it in Etoys. It comes will the real-time collaboration, so you can experiment a real-time collaborative painting, for example, with two B1 machines. (Once Sugar gets its own painting tool, we would like to use it, or at least modify the look of the Squeak's painting tool so that it imitates the native painting tool.) If you think that constructionist approach is a good idea, it would be good to think that kids should *make* stuff by themselves to learn stuff. For example, I think that the things like Memory game are something to be written by children. Also, there was some discussion on a clock application: waking up the machine is out of scope of Etoys, but the other part, making a functioning clock, is something an elementary school children can do in Etoys. Instead of giving a nicely done, professional looking clock from us, it would be great to see kids-made versions. This means that what we should write is to create the materials on how to make such games or things. We have a few examples in Etoys. Launch Etoys on it, and press "All" button. In the file list, go to "ExampleEtoys" and choose one. If you say that the curriculum is supposed to be something to be built by the local goverments and users. So, we should make an authoring tool to make such materials. Viewing source code was under discussion. The implementation langauge of Etoys, again, is different from the standard language on OLPC, so it wouldn't serve as a prototype or mockup. That said, Etoys already have built-in development facility that expose all source code in the system. One may get some idea from there. (Is somebody going to write a Python development environment for children?) For the basic concepts, take a look at: http://wiki.laptop.org/go/Sugar_EToys If you don't have your own projects, Etoys try to load a default project and it takes some time (too much time). Once you make one, the start up time is almost acceptable We are actively improving Etoys. To try the latest stuff please download: http://tinlizzie.org/olpc/etoys-2.0-1133.zip and copy/replace the etoys-2.0-1133.image to: /usr/share/etoys/etoys.image If you are interested in developing something in it, download: http://tinlizzie.org/olpc/etoys-dev-2.0-1133.zip and launch the image with a Squeak VM that you already have on your computer. Thank you! -- Yoshiki

6 10

[Etoys] OPLC video on youtube.com
by karl 14 Dec '06

14 Dec '06

I saw on Slashdot a link to a video of the UI of the OPLC: http://www.youtube.com/watch?v=DwzCsOFxT-U&mode=related&search= It was painful to watch the Squeak part of that video. I shows so many of the common mistakes people make with Squeak. And it shows a lot of quirks in Squeak and etoys. Should the event playback be modal so the user input will not disrupt the playback. What are do other people think ? Karl

6 8

[Etoys] Small progress of sound delay issue.
by Takashi Yamamiya 08 Dec '06

08 Dec '06

Hi, I'm still working with sound issue. There is not great progress though, I could find some direction how to deal with it. I guess latency is reduced almost from 0.5 ms to 0.3 ms with this patch. This is just current status. (you can use it to copy vm-sound-ALSA to /usr/lib/squeak/3.9-9/) What I did is; - Setting periods count to two. (more periods cause more latency). - Use "plughw:0,0" device instead of "default" The defect is the sound quality becomes bad! So, I will try another way. I think mmap is good to the purpose. API document said snd_pcm_avail_update() is used with mmap. http://www.alsa-project.org/alsa-doc/alsa-lib/group___p_c_m.html#g8bb836bd0… Cheers, - Takashi Index: platforms/unix/vm-sound-ALSA/sqUnixSoundALSA.c =================================================================== --- platforms/unix/vm-sound-ALSA/sqUnixSoundALSA.c (revision 1586) +++ platforms/unix/vm-sound-ALSA/sqUnixSoundALSA.c (working copy) @@ -33,7 +33,7 @@ #include <errno.h> -static char *sound_device = "default"; +static char *sound_device = "plughw:0,0"; static char *sound_playback = "Master"; static char *sound_capture = "Capture"; @@ -87,9 +87,10 @@ int err; snd_pcm_hw_params_t *hwparams; snd_pcm_sw_params_t *swparams; - snd_pcm_uframes_t frames; unsigned int uval; int dir; + int periodsCount = 2; + snd_pcm_uframes_t bufferSize; if (output_handle) sound_Stop(); @@ -104,10 +105,15 @@ snd_pcm_hw_params_set_channels(output_handle, hwparams, output_channels); uval= samplesPerSec; snd_pcm_hw_params_set_rate_near(output_handle, hwparams, &uval, &dir); - frames= frameCount; - snd_pcm_hw_params_set_period_size_near(output_handle, hwparams, &frames, &dir); + snd(pcm_hw_params_set_periods_near(output_handle, hwparams, &periodsCount, &dir), "pcm_hw_params_set_periods_near"); + bufferSize = frameCount * periodsCount; + snd(pcm_hw_params_set_buffer_size_near(output_handle, hwparams, &bufferSize), "pcm_hw_params_set_buffer_size_near"); + snd(pcm_hw_params(output_handle, hwparams), "sound_Start: snd_pcm_hw_params"); + fprintf(stderr, "Request: frames=%i, periods=%i\n", frameCount, 2); + fprintf(stderr, "Result: frames=%i, periods=%i\n", bufferSize / periodsCount, periodsCount); + snd_pcm_sw_params_alloca(&swparams); snd(pcm_sw_params_current(output_handle, swparams), "sound_Start: snd_pcm_sw_params_current"); snd(pcm_sw_params_set_start_threshold(output_handle, swparams, frameCount * SQ_SND_PLAY_START_THRESHOLD), "sound_Start: snd_pcm_sw_params_set_start_threshold");

2 4

[Etoys] Re: Camera support and ALSA diffs
by Ian Piumarta 02 Dec '06

02 Dec '06

Bert, ALSA diffs and camera support (renamed to VideoForLinuxPlugin) is committed. Please check that video and sound works in your image with these sources. Thanks! Cheers, Ian

2 4

[Etoys] sound player shutdown when initiating painting
by Scott Wallace 01 Dec '06

01 Dec '06

I just noticed that playback of sound using the EventRecorder stops abruptly if the event-recording includes a painting session. Just before the paint palette appears on playback, sound playback simply stops. I've tracked this down to the recent addition of WorldStethoscope code, in which a "SoundPlayer shutDown" call was added to PasteUpMorph >> stopRunningAll. So it seems that either #stopRunningAll should be restored to its original functionality, which was to suspend all ticking scripts but not to mess with sound playback, or else we can change the get-ready- to-paint code so that it no longer calls #stopRunningAll, but instead calls some to-be-written method that does the original intended work of #stopRunningAll, but without the additional clobbering of sound. What do you think? Cheers, -- Scott

3 3

[Etoys] Bug in fill in the blank
by karl 01 Dec '06

01 Dec '06

Seem like a bug in fill in the blank widget when there is a big text in it's heading. Look at attachment. karl

2 1

[Etoys] [FIX]ObjectsTool drag and drop
by karl 30 Nov '06

30 Nov '06

When you pull out a morph from the ObjectsTool you don't want to lose the morph because the you drop it on top of the ObjectsTool. Karl 'From OLPC2.0 of ''24 October 2006'' [latest update: #1133] on 1 December 2006 at 12:06:24 am'! !PartsBin methodsFor: 'dropping/grabbing' stamp: 'kfr 12/1/2006 00:04'! acceptDroppingMorph: dropped event: evt (self owner isKindOf: ObjectsTool) ifTrue:[^ self world acceptDroppingMorph: dropped event: evt.]. (dropped hasProperty: #beFullyVisibleAfterDrop) ifTrue: [dropped vanishAfterSlidingTo: dropped formerPosition event: evt. ^ self ]. ^ super acceptDroppingMorph: dropped event: evt. ! !

1 0

[Etoys] caveat on painting & evt recorder
by Scott Wallace 30 Nov '06

30 Nov '06

Nowadays (since update 1029fixPainter), the brush-size in force when the user exits from a painting session will be the brush-size selected when the user undertakes his next painting session. And likewise, the "currently selected color" is now remembered and restored from painting session to painting session. If you are making an event-recording that involves use of painting, this means that you can no longer know what the brush-size and selected-color are going to be when the paint palette comes up on replay. To assure intended brush-size and color-choice in playback, you should explicitly click on the desired brush-size and color, even if on your screen they are already set correctly, during any event recording of a painting session. Cheers, -- Scott

1 0

[Etoys] New bundle and RPM
by Bert Freudenberg 29 Nov '06

29 Nov '06

Hi folks, if you use git / sugar-jhbuild you will have noticed that I switched to bundle format now (http://wiki.laptop.org/go/Activity_Bundles). The bundle directory is in build/share/activities now, you will have to manually delete the old activity files in build/share/sugar/ activities. Remove the activity directory, as well as the two registration files. Besides, there now is a 2.0.1130 image and revamped activity code. The activity sends activation and deactivation commands to the image. For now, this only disables/enables the sound (see SugarLauncher>>#activate:). However, it requires a fixed ALSA plugin, but this has not yet been put into the SVN tree by Ian. I sent the patch around a few days ago, also attached, if you want to compile yourself. So I did not want to build a new VM RPM yet. I built one for etoys, but only pushed it to http://tinlizzie.org/olpc/sugar/yum/ and not yet to the public yum repository, I'll wait until I have a matching VM RPM, too. Would be nice to have this soon, to start testing, in particular on the b-test machines. I see strange behavior on my Linux install, but that may have to do with how ALSA is set up here. Btw, upgrading by RPM will clean up the old files automatically. - Bert -

1 0

[Etoys] Sound project
by Bert Freudenberg 28 Nov '06

28 Nov '06

While testing the sound deactivation code I came up with the attached project - nice and simple. If we could hook this up to CSound's synth we had something a lot like TamTam ;) Maybe mapping the distance from center to pitch would be even nicer than it is now (mapping from size). Enjoy! - Bert -

3 2

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

etoys-dev November 2006