Pd-list December 2008

pd-list@lists.iem.at

112 participants
182 discussions

[OT] Hardware suggestions - Webcam for GEM/pdp/Linux
by Martin Schied 03 Dec '08

03 Dec '08

Hey List! I'm looking for a webcam wich I can use with 640x480@30fps, preferable with v4l (v4l2 support is still difficult with some apps), usb preferred. It has to be a camera with ccd (no cmos chip). maybe there are *no* cameras with v4l and VGA@30fps - perhaps someone knows? Did anyone try a philips spc900nc camera? It should support 90fps at lower resolutions but I didn't find one single positive mention about that with linux. not that important, but nice to have: it would be great if the infrared blocking filter was removable when the camera is opened, because I have to use it in IR range for tracking. If it's not removable it doesn't really matter because usually it's possible to unscrew and replace the lens with an IR-compatible lens from other webcams. I already tried 2 cams: Logitech Quickcam Pro 5000: vga@30fps, very low latency, v4l2 driver (no v4l compatibility), raw-mode with 640x480 or 640x380 (didn't get other resolutions to work), lot of dead pixels (10 dead pixels on 2 different models each, seems to be a secret linux-feature, hidden on windows-driver), not very crisp but usable image quality, IR-Filter not removable from Lens, whole Lens can be easily unscrewed when camera is opened, easy to open, no adjustable focus when not opened. Philips Vesta Pro (PCVC680K): vga@15fps, v4l driver, slow, less light sensitive than quickcam but nice crisp image, old (only few used models available), removable IR-Filter (pop off the first "ring" on the object lens to remove), tripod mount, rugged housing. both lens-mounts are compatible, seems to be an undefined "standard" sorry for posting that much - perhaps it's useful for someone? kind regards, and thanks... Martin

1 0

[PD-announce] NIME concert @ Exit Art NYC, Monday, Dec 15th
by Hans-Christoph Steiner 03 Dec '08

03 Dec '08

Come see some Pd instruments in action: > NIME @ Exit Art > http://itp.nyu.edu/nime/show/ > 475 Tenth Avenue > New York, NY > > Monday, December 15, 2008 > Performances 8PM-11PM > Doors open at 7PM > > NIME: New Interfaces for Musical Expression. > NIME: creating new performance tools for digital music. > NIME: a graduate course at NYU’s Interactive Telecommunications > Program (ITP). > > In the seventh annual NIME end-of-semester performance, students > will perform on a series of newly designed electronic instruments > that aim to keep the “live” in live performance of digital music. > The NIME performances are presented by ITP instructor Hans- > Christoph Steiner. > > Computer music is usually played with a keyboard and mouse. Laptop > musicians often sit at a desk and give performances that are little > more than watching someone engage in “office gestures.” The idea > behind NIME is to go beyond the mouse and keyboard, beyond even > piano keyboards and drum pads, and develop performance tools that > make the most out of the new opportunities that digital music > offers. NIME students answer questions like: > - "What will the next generation of musical instruments look like?" > - "What will they be able to do that traditional instruments can’t > already do?" > - "What aspects of traditional instruments will we want to retain > in digital instruments?" > > Over the course of this year’s 14-week course, students are > developing projects such as a collection of autonomous solar- > powered robots, a augmented flutist, plate-spinning sampler, sing- > along animatronics, and a host of others. > > NIME at ITP > http://itp.nyu.edu/nime > Hans-Christoph Steiner, +1-718-360-4872, hansi(a)nyu.edu > > ITP > http://itp.nyu.edu > George Agudow +1-212-998-1891, george.agudow(a)nyu.edu > > EXIT ART > 475 Tenth Avenue (at 36th Street) > New York, NY 10018 > (212) 966-7745 > info(a)exitart.org > http://www.exitart.org ------------------------------------------------------------------------ ---- Computer science is no more related to the computer than astronomy is related to the telescope. -Edsger Dykstra _______________________________________________ Pd-announce mailing list Pd-announce(a)iem.at http://lists.puredata.info/listinfo/pd-announce

1 0

Re: [PD] list-sort
by Matt Barber 03 Dec '08

03 Dec '08

For what it's worth, I just ran one possible benchmark on the [list-*sort] abstractions submitted recently, using [list-random] and [realtime]. I think overall the overhead for [list-sort] is much less than for [list-shellsort] -- for short lists (<100) it takes about half the time. [list-shellsort] is a bit quicker when the variance of the elements is smaller (which makes sense because the more alike the elements are the less the innermost loop will have to run) -- but [list-sort] still beats it. I have a feeling that some of the overhead in [list-shellsort] is in the [list-swap] function and other list-abs objects, which are a little too feature-rich and/or idiot-proof for this particular use. =o) However, after some trial and error with [list-sort] I found I am unable to sort a list of more than exactly 121 elements without freezing Pd -- 122 and up will freeze it. This is because the main loop is not controlled with an [until] -- I think you can see the structure of [list-drip] for similar reasoning (the [until] is not "needed" logically, but helps with long lists to do it in steps). After implementing the [until] loop, I found that [list-sort] is even quicker than before, but the efficiency of the [list-shellsort] algorithm take over for list sizes >450 -- above that [list-shellsort] is substantially quicker (and quicker yet if the variance is low). It beats [list-sort] by half for a list about 1300-1400 elements long or so (but not within reason -- it still takes about 2 seconds!). I attached the new [list-sort] and the goofy little diagnostic. Matt > Date: Tue, 2 Dec 2008 13:59:08 +0100 > From: Frank Barknecht <fbar(a)footils.org> > Subject: Re: [PD] list-sort > To: pd-list(a)iem.at > Message-ID: <20081202125908.GJ18987(a)fliwatut.scifi> > Content-Type: text/plain; charset="us-ascii" > > Hallo, > > oh, and I also now committed two more abstractions that Matt Barber > sent me offlist, of which one is a sorting abstraction as well. Matt's > [list-shellsort] uses the Shell sorting algorithm: > http://en.wikipedia.org/wiki/Shell_sort which generally is a bit > faster than insertion sort, but I didn't benchmark the two Pd > implementations (the speed in Pd of course also depends on how much > element shuffling and list-dripping is needed) > > Anyway, currently list-shellsort only does ascending sorting, so I > just decided to include both Michal's list-sort, which probably is > easier to understand, and Matt's list-shellsort in the current SVN's > [list]-abs collection. Choose your poison. ;) > > Ciao > -- > Frank > > Michal Seta hat gesagt: // Michal Seta wrote: > >> Hi all, >> >> It is amazing how we take things for granted. Most programming >> languages provide some sort of list sorting function/method. >> Surprisingly (or not) pd does not (or my search skills are null, or I >> am not bleeding edge enough). I needed a solution that works with a >> vanilla pd. >> >> I was almost going to do the academia move and announce a pd exam, and >> have little pd-bees come up with a solution but I needed it *now* else >> I would not sleep or have terrible nightmares. So here it is. Thank >> heavens (but give your offerings to fbar's footils shrine) for >> list-abs. >> >> Busy pd-bees, should you care to improve on my solution, please be my >> guest, I am sure there are better ways of accomplishing this trivial >> task. In any case, go forth and sort the world around (or within) >> you. >> >> ./MiS

3 6

Peripheral USB devices
by Andrew Faraday 03 Dec '08

03 Dec '08

Okay, this is probably quite basic. But I want to use signals from unrelated USB devices (such as the controllers for the PS2 game buzz) to control PD. How would I go about recognising these devices and logging commands from them? _________________________________________________________________ Are you a PC? Upload your PC story and show the world http://clk.atdmt.com/UKM/go/122465942/direct/01/

4 3

Peak envelope follower?
by Bill Gribble 03 Dec '08

03 Dec '08

Is there a way within pd-extended to do the equivalent of [env~], but following peak value rather than RMS? I have pulled my hair out trying to implement this with patching but I have had no luck. Thanks, Bill Gribble

6 5

Anyone using Lua also using other Lua extentions?
by Mike McGonagle 03 Dec '08

03 Dec '08

Hello all, I have been wondering about what I would need to do to use other Lua extentions along with [lua]? Is it just a matter of compiling those linked libraries, and then doing a require "my-extention'' in the code? Does anything in the C portions of PdLua need to be recompiled? Thanks, Mike -- Peace may sound simple—one beautiful word— but it requires everything we have, every quality, every strength, every dream, every high ideal. —Yehudi Menuhin (1916–1999), musician

3 2

Fwd: probability table ?
by volker böhm 03 Dec '08

03 Dec '08

sorry, this was not intended to go to you privately, but to the list. could resolve one issue. list-wrandom and list-random load fine in pd 0.39.3-extended after adding them to the extra folder (they were missing before). but they weren't missing in the pd 0.40.3 version. i can't get this version to load any other externals. after startup i get "creb: can't load library cxc: can't load library iemlib: can't load library list-abs: can't load library mapping: can't load library markex: can't load library maxlib: can't load library ..." so one question remains: isn't the package contents automatically part of pd's search path? if not, how can i change the search path, as with the pd preferences menu i can't seem to navigate inside an application folder? well, actually pd seems to find the libraries, but it can't load them... what could cause this error? thanks for any pointers. volker. Begin forwarded message: > From: volker böhm <vboehm(a)gmx.ch> > Date: Wed 3 Dec 2008 13:16:58 GMT+01:00 > To: Frank Barknecht <fbar(a)footils.org> > Subject: Re: [PD] probability table ? > > > On 02 Dec 2008, at 18:25, Frank Barknecht wrote: >> Hallo, >> volker b?hm hat gesagt: // volker b?hm wrote: >> >>> i'm looking again for a max equivalent in pd. >>> when sending a bang to a table object in max you can use it as a >>> probability table*. >>> >>> is there something similar in pd? >> >> [list-wrandom] from the [list]-abs collection is similar. >> >> Use [list-tabdump] or Zexy's [tabdump] to get the values from a table >> into a list to feed into [list-wrandom]. >> > > thanks, frank, that's what i'm looking for. > > but i ran into some strange filepath issue again on my computer > (osx 10.4.11, pd 0.39.3-extended). > i can't instantiate the list-wrandom abstraction, although it's > sitting in the right place (inside the pd package in resources/ > extra/list-abs) > funny thing is, i can load all the other list-abs, but not list- > random and list-wrandom. > > why would that be? > isn't the package contents automatically part of pd's search path? > if not, how can i change the search path, as with the pd > preferences menu i can't seem to navigate inside an application > folder? > > if i copy the abstraction with its helpfile to some folder and load > the help file, pd finds and loads the abstraction. > so the current folder is part of the search path. > > any help would be appreciated. > volker. > >

1 0

Gem: pix_write and pix_multiimage
by Max 03 Dec '08

03 Dec '08

it's a pity pix_write and pix_multiimage don't follow the same file naming paradigm. pix write makes image like this: image00000.jpg to image99999.jpg while pix_multiimage does: image0.jpg to image99999.jpg

1 0

Re: [PD] text2d for japanese, persan and other languages
by IOhannes m zmoelnig 03 Dec '08

03 Dec '08

please always include the mailing list when replying. Renaud RUBIANO wrote: >> oh, and is there a special reason to use [text2d] rather than [text3d]? >> the latter usually performs better... > Oh no, but I doesn't need 3D options, so I guessed that 2d was GPU > less-consuming ... so ... i see. nevertheless this is usually not true with Gem. Gem is openGL based, meaning it uses hardware acceleration. off-the-shelf gfx hardware has been highly pushed by aming industries, which made 3d operations be more optimized than 2d operations... so: [text3d] should be better than [text2d] in most cases (the latter sometimes is needed when you want to use antialiasing) [pix_draw] is usually way less efficient than [pix_texture]+[rectangle] .... fgmasdr IOhannes

2 2

Re: [PD] Pd-list Digest, Vol 45, Issue 15
by Collin Oldham 03 Dec '08

03 Dec '08

1 0

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

Pd-list December 2008