Pd-dev June 2002

pd-dev@lists.iem.at

22 participants
30 discussions

External interaction speed + deconstructor methods
by Ben Bogart - FMPM/F1999 28 Jun '02

28 Jun '02

My SDL external blits and updates the render window each time the "blender_alpha" method is called (by the alpha selector). As far as I can tell the render speed is quite sufficient, but seems to lag behind what the number box is doing. Its like my number box goes through a pipe object before it gets to the external. Its just not quite as responsive as i would like. Is it more likely my SDL rendering speed or does an external have a built-in input buffer causing the lag? Here is my blender_setup function: void blender_setup(void) { myclass = class_new(gensym("blender"), /* symname is the symbolic name */ (t_newmethod)blender_new, /* Constructor Function */ 0, /* Destructor Function */ sizeof(mystruct), /* Size of the structure */ CLASS_DEFAULT, /* Graphical Representation */ 0); /* 0 Terminates Argument List */ class_addmethod(myclass, /* Add function as method */ (t_method)new_blender_window, /* Bind method to function */ gensym("create"), /* symbol to invoke method */ 0); /* method arguments */ class_addmethod(myclass, /* Add function as method */ (t_method)blender_close, /* Bind method to function */ gensym("destroy"), /* symbol to invoke method */ 0); /* method arguments */ class_addmethod(myclass, /* Add function as method */ (t_method)blender_alpha, /* Bind method to function */ gensym("alpha"), /* symbol to invoke method */ A_DEFFLOAT, 0); /* method arguments */ } While I'm on the topic, what is the proper way to create a deconstructor function? (to close my SDL process when PD closes) I tried passing (t_newmethod)sdl_close as the 3rd argument to my class_new function, but that causes the following compile error: blender.c: In function `blender_setup': blender.c:82: warning: passing arg 3 of `class_new' from incompatible pointer type Thanks! Ben B. Bogart ---------

3 2

Re: [PD-announce] pd 0.35 test 28!
by Krzysztof Czaja 28 Jun '02

28 Jun '02

hi Miller, . x is uninitialized in line 285 of x_net.c (netreceive is crashy): if (!x->x_udp) . there is a typo in line 673 of s_main.c: else if (!strcmp(*argv, "-23bit")) Krzysztof Miller Puckette wrote: ... > Test 28 is out... the main new "feature" is settable blocksize in ASIO. > I've managed to get latency down to about 10 msec for RME hammerfall in > ASIO. > > Some bug fixes too... I hope to get teh "final" 0.35 out in 2 weeks.

2 1

Using Shared libs in Externals
by Ben Bogart - FMPM/F1999 27 Jun '02

27 Jun '02

Hello all, I'm currently playing with the possibility of writing a SDL (www.libsdl.org) PD external. I've ported my code to the PD external format and can compile, but upon loading PD refuses to load the external because it can't find the SDL library functions (Undefined Symbol). So how can I access the SDL library functions from my external lib? How can I link my lib to the sdl lib? cc: -lSDL: linker input file unused since linking not done cc: -lm: linker input file unused since linking not done cc: -lpthread: linker input file unused since linking not done Here is my current Makefile: (based on the 6.externs example) current: echo make pd_linux, pd_nt, pd_irix5, or pd_irix6 clean: ; rm -f *.pd_linux *.o # ----------------------- LINUX i386 ----------------------- # Altered to compile SDL PD Externals pd_linux: ben.pd_linux .SUFFIXES: .pd_linux CFLAGS = -DPD -O2 -funroll-loops -fomit-frame-pointer \ -Wall -W -Wshadow -Wstrict-prototypes -Werror \ -Wno-unused -Wno-parentheses -Wno-switch INCLUDE = -I../../src LIBS = -lSDL -lm -lpthread .c.pd_linux: cc $(CFLAGS) $(LIBS) $(INCLUDE) -o $*.o -c $*.c ld -export_dynamic -shared -o $*.pd_linux $*.o -lc -lm strip --strip-unneeded $*.pd_linux rm $*.o I'm very new to C (been tinkering for only 2 weeks) so I'm still not quite clear on the whole linking/compiling process. Any pointers would be appriciated. If seeing my (ugly) sourcecode would help I'd be happy to send it along. Thanks Ben B. Bogart ---------

3 4

External CVS: Ready for phase I
by g�nter geiger 26 Jun '02

26 Jun '02

We (Juha and me) have been able to setup a first description how to work with the externals CVS on sourceforge. You can access it via http://pure-data.sourceforge.net. You'll see, its about content, not style, everyone who wants to improve this is welcome to do so :) For those who don't know what this is all about: ================================================ The Pure Data external repository is meant to be a place where developers of externals for the Pure Data computer music system can put their code. PD has a huge user base and we all enjoy working with PD. A lot of externals are being written and published on the web. As a consequence, it has become increasingly hard to have an overview of all the externals. A central code repository will offer this, as well as many other advantages. It will be easier to manage the code, update orphaned externals, distribute compiled versions for different systems, avoid duplication of effort, collaborate on improving the same external, and finally the users have the advantage of being able to download it from one single place. We could go on listing the advantages, but there is one big disadvantage. We have to work on it in order to make it a success. We have to learn CVS and we have to import our projects into the repository. I think it is worth this work, because you gain a lot too. If you don't have the time to get into this, but still agree that your external becomes part of the repository, just send an email to the list with the information how to download your code. Another drawback is that your code has to be GPL'd. If this is a problem for you, let us know. Happy CVSing Guenter

1 0

Re: [PD-announce] Unofficial pd-0.35-test26 for Debian
by Frank Barknecht 26 Jun '02

26 Jun '02

Hi, it took this to pd-dev, because I think, pd-announce should better be reserved for announcements only - or what do you guys think? günter geiger hat gesagt: // günter geiger wrote: > On Fri, 21 Jun 2002, Frank Barknecht wrote: > > As usual: Please _don't_ report packaging-related Bugs to the Debian > > BTS, this is packaged unofficially just for your convinience and not > > supported by Debian, although it uses the debian-configuration by > > Guenther Geiger. > > ahh, then I could use it directly to upload to Debian proper ? :) > Did you do lintian checks etc or fixed pending bugs, I would be > happy to give away packaging of some of my packages .... Of course I would be really proud to maintain PD on Debian/Agnula, and that's meant honest. For my package I didn't run any lintian checks, neither did I look at the pending bugs. I just copied over the debian-directory from pd-0.32 and made some minor arrangements, for example use dh_link to link /usr/lib/pd/doc/html to /usr/share/doc/pd/html and such. Also I did not follow debian-dev for about a year, so maybe there are some policy violations that I'm not aware of. And I didn't build correct source packages, *.dsc and *.changes yet, so for a NM-upload I would need to do some more work. But if you want to give away PD packaging and maybe lead me a bit through the new-maintainer queue, I would be willing to maintain PD or just do some Non-Maintainer-Uploads of better checked PD packages than the current one. ciao, -- Frank Barknecht _ ______footils.org__

2 1

PD and Qt?
by Bert Schiettecatte 25 Jun '02

25 Jun '02

Hi All, I was wondering if any of you have considered dropping Tk and using Qt for the GUI of PD? (see http://www.trolltech.com/) I think that would take care of any remaining GUI-problems and would make it easier to add new GUI externals to PD. Qt also has OpenGL support. If I had the time I would do the work, of course :-) Bert.

3 2

threads / soundfiler
by d_jj 25 Jun '02

25 Jun '02

i installed the two files (d_soundfile.c + s_inter.c) in my 0.35-test24 sources and compiled it, now it says some stuff on the console about child priorities but in the end i still get dropouts for fileloading? bst, xj [ydegoyon(a)free.fr]->[Fw: Re: [PD] we want threads !!!!]->[02-06-25 02:32] | | |> "ydegoyon(a)free.fr" schrieb: |> |> > so, if you like danger, you might try these.... |> > it's experimental still but with the help of some beta-testers, |> > it might get consistent soon. |> |> I've added some changes and now it works under win2k as well... get the files needed |> (and the binaries for Win) at http://www.akustische-kunst.de/puredata/winport.html |> | |yeh, that's the official place for these unofficials patches. |i won't set up my own place for a release. | |it's better cos it works on both linux and WXXXX? with |some recent PD release ( one patch ( inter ) was made over test24 and the |other one ( soundfiler ) is older, think it was made over test8 ). | |thanx, olaf. | -- x D u . o R g

3 4

OS X keyboard problems
by tigital 24 Jun '02

24 Jun '02

hmmm, ...now that I've got most of GEM working on OS X, I'm trying to go beyond just opening example patches, and starting to construct my own experiments...unfortunately, I'm having alot of trouble with entering text into object/message boxes, to say nothing about trying to bring up help files (I think it's known that mouse-modifier keys aren't working)... Is anyone else having similar problems? tanx, jamie

2 1

linking in the heat
by Krzysztof Czaja 22 Jun '02

22 Jun '02

hi all, it is so hot here, thinking is so hard and painful... are there cooler places on earth, where pd-dev people live? ...say, we have 150 classes using some amount of a common code. The size of this common code is rather small -- close to the average size of one class' code, or possibly 2 or 3 classes. Some of the options are: 1. All classes, and the common code, are statically linked in a one monolithic library. 2. Each class is kept as a separate external, linked with the common code (which means: there are as many common code copies, as the number of externals containing it, and loaded during a Pd session). 3. The common code is linked dynamically, and it has to be loaded prior to any of the classes (using an explicit -lib option). 4. Class names are prefixed with a library name, causing Pd to load the library with both the common code, and the classes (a feature not yet available, but already discussed on pd-list). 5. Like 4., but the library contains only the common code -- its setup routine has to know what particular class is demanded, and how to load it. 6. Each class uses a tiny ``stub code'', possibly a part of a future Pd api, which somehow checks if the common code is already loaded (currently it might be done by checking if something 'familiar' is bound to a particular symbol). If not, it loads the library, and obtains a set of pointers to the functions in the common code. Are there other options? Which option is best? Btw, perhaps due to my mind melting in the heat, I somehow tend to like the method 6 (if only it was a part of Pd api...) Btw, options 4/5 cannot be applied in case of the cyclone library. They might be in case of xeq or vexing (currently both are using a modified version of the first method, while a cyclone user has to choose between options 1 and 2). Btw, there are not as many classes in cyclone (yet:-?) just a bit over 40, but it keeps growing... Krzysztof

3 3

Re: dropouts when loading into arrays
by Juha Vehviläinen 22 Jun '02

22 Jun '02

Hi, Soundfiler should be threaded, shouldn't it? Juha > how do you avoid audio dropouts when loading new data into tables/arr

2 1

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

Pd-dev June 2002