Pd-list January 2003

pd-list@lists.iem.at

96 participants
150 discussions

Re: [PD] some basic questions
by CK 01 Feb '03

01 Feb '03

I read: > 1) is there a way to switch off dsp computing in a subpatcher only, and how? > (i think i should know that) |switch~ | > 2) i remember a discussion thread about max-like poly (or even better - > poly~) objects a while ago... has there been any external code written in an > attempt to realize that? AFAIK there are pixes attempts to do it dyamically (i.e. using internal messages, maybe there was an external involved ... let pix answer that one) anyway -> http://pix.test.at/pd/nqpoly/index.html > 3) is it guaranteed (or at least intended - and under which circumstances) > that - contrary to Max/MSP - opening a new patcher and adding objects to the > signal tree doesn't result in an overall crackle? At least in my experiments > i could open fairly complex and cpu-intense patches in PD without audio > interruption. hmmm good one, sometimes there's trouble sometimes there is none ... > 4) i think to recall vaguely that there has been a project featuring a > tcl/tk-less pd patcher system (i think with GTK). If this is true, does that > still reside somewhere? that was pdgui by karl macmillan: http://gigue.peabody.jhu.edu/~karlmac/download/pd/ regards, x -- chris(a)lo-res.org Postmodernism is german romanticism with better http://pilot.fm/ special effects. (Jeff Keuss / via ctheory.com)

4 5

1d cellular automata patch
by wwcarpen 31 Jan '03

31 Jan '03

hello, just thought i would share the cellular automata patch i was working on wwcarpenCA.pd is the main patch give it some time to load, because there is a loadbang that creates 1250 objects in a subwindow. it also requires the gem and zexy libraries overall, i am happy that i got it to do exactly what i set out to do. but i'm dissapointed that it is pretty complex for what is supposed the be the simplest program capable of creating complex behavior. fundamentally i think that it would be much easier if i was using gridflow, (ahem, windows .dll, somebody, anyone? pleeeeease?) because much of the extra code seems to be coming from dealing with my manually made grid matrix. am i correct in thinking that i could make a much bigger window with gridflow without having to worry much about using cpu? my laptop starts to stutter if i use much more than 1250 squares in gem. as far as what i have right now, one way to speed up the scrolling at least 49x would be to have it analyze just the last line of squares rather than the whole window after all the square information is transposed up a line. I spent a couple days trying to get this to work, but i cant seem to locate the problem, (if anyone is brave enough to help me here, what i thought would work is to have a separate 'calculate' bang that is triggered after the 'scroll down' bang which in turn sends a 1226 msg to the int object rather than 23.) the purpose of the demux objects is to handle line wrapping correctly. wwcarpen

3 3

[threshold~]
by Alex from Idoia 31 Jan '03

31 Jan '03

Hello, I would like some precisions regarding the arguments of that object: - how the arguments 1 and 3 (trigger threshold and trigger rest) are expressed ? I guess is not in db ? I tried to use -15 0 -15 0 to trigger a bang with a signal over -15db but it does not seem to work that way. - If the arguments 2 and 4 are set to 0 that means the bang threshold and the rest threshold can be trigger almost simultaneously ? (I am thinking in a case of picks of signal) - what are the right inlets used for then ? Thanks

2 1

Mac OS X and the -rt switch
by Michael McGonagle 31 Jan '03

31 Jan '03

Hello all, I was just looking thru the commandline switch options and noticed that the -rt switch "needs root privledges". Well, if this is true, I did try a couple of examples (from Adam Lindsay, in conjunction with Paradiddle) that used the -rt switch, but it did not ask me for a password. Does this just ignore the -rt switch? Or is it that because my user is the main (and only) user, does it automatically get root access? I am currently running my Mac with a "rootless" system (no password for root), and all the times I have entered a password for root, I end up entering my user password in its place. Do I need to "enable" my root user (give it a password) and run PD as root to get the best realtime performance? Or, is the -rt switch even implemented on Mac OS X? Thanks, Mike

5 6

Re: [PD-dev] [GEM] Further CVS changes
by Christian Klippel 30 Jan '03

30 Jan '03

hi, Am Donnerstag, 30. Januar 2003 20:39 schrieb Tom Schouten: > i got some (at first glance) counterintuitive results using mmx in pdp too. > i guess a lot of this strangeness has to do with memory bandwidth. simple > operations like add or scale are not much faster than their scalar integer > c counterparts. i did get a lot of speedup for the more compute intensive > stuff like the biquad filters, iterated convolution and basicly anything > that needs to do a lot of clipping. also i try to limit the data copying to > a minimum in pdp, this seems to help too.. > > the general rule seems to be: keep your memory accesses local and your data > size small: do as much as possible inside the pixel loop, or iterate > several times over 1 scanline instead of the whole image. > > tom a while ago i did extensive searches about mmx coding styles. there are some important issues when writing mmx code, i also had less efficient code (compared to c) on the first try. first, the cpu has two instructions pipes. not all mmx instructions can be executed in any pipe. so you have some that can use boot, and some that can only use one of them. you need to pair your mmx instructions to take care of that. yes, mmx is cache sensitive. when you first load a register from memory, a whole chunk gets loaded in the cache. best is to use as much mmx regs as possible at once, process them, and write them back. this improves the cahce hit/miss ration, since the cache is invalid when written back. some mmx instructions need more than one cycle, so if another instruction depends on such operation, you get stalls and penality cycles. there is a good page on that topic: http://www.ce.unipr.it/~tommesa/Pixel64.html it does not tell about the basics of mmx or so, but it shows optimization issues on some practical examples and explains it a lot. that page helped me a lot. sometimes the server seems down, i have a local copy of the site, so if one needs i can put it on my site for you to grab. you can also take a look in the mmx code i have written so far. a lot of functions are done and i tried to make them as fast as possible. they are for rgba (=32 bit) pixels. they are intended to work on chunks of a image, but you may try them with a whole image as well. my guess is that, if you do serveral ops on a image, it is faster to do chunk1->op1->op2->result1, chunk2->op1->op2->result2.... than image->op1->op2->result, but i may be wrong (didnt tested it) the data to the routines however need to be a multiple of 64 byte, as some routines load all regs with subsequent pixels at once. hope that helped a little, greets, chris

1 0

MIDI tools in PD
by Michael McGonagle 30 Jan '03

30 Jan '03

Hello, I have put together a system with an SL880 as the main keyboard controller, and as the "front panel" on the beast leaves a little to be desired (I got it for a good price and I like the keybed), I was hoping to create some tools in PD that would modify an incoming MIDI stream. I am interested in working with alternate tunings, and as such would be interested in any tools for PD that would handle "on-the-fly" retuning of a MIDI stream. From what I have seen of other implementations, the MIDI stream needs to be split up into different channels, where each channel has a different pitch bend assigned to it (or they could be doled out in a round robin fashion). Anyway, the point is to translate the incoming MIDI, and based on the current tuning system selected, and produce a multi-channel MIDI stream with pitch bend info to effect the tuning. From working with the program 'scala', the definitions it has for what it takes to create a "keyboard map" for a particular tuning system would be a good place to start. While this is not a complete description, the file contains info to define a range of MIDI notes to translate, the base reference MIDI note number, and the frequency that is produced by the base reference note. There are other things that allow the "removal" of keys from the MIDI stream that are not used in the particular keyboard map. Thanks, Mike

2 1

Re: [PD-dev] [GEM] Further CVS changes
by Christian Klippel 30 Jan '03

30 Jan '03

hi chris and all, Am Donnerstag, 30. Januar 2003 20:36 schrieb chris clepper: > It is possible to do a runtime architecture > check for Altivec, is it possible for MMX? yes, you can retrieve those flags. they will tell you what cpu and what extensions, like mmx or 3d-now. look at the startup code from my video stuff. i can also post it here when needed. there is a "lib" that lets you include mmx as c-functions in your code for gcc, its kust an include file with defines for the opcodes and register types. afaik, that is available for sse as well. guess its called mmxlib or the like, but im not sure for now. sse is not mmx ! sse contains a lot more functions to apply on much wider registers. compared to sse (or sse2) mmx looks like a nice toy ;) also sse lets you use floating point ops as well in sse registers, whereas mmx doesnt. in fact, you have to instruct the cpu that youre not going to use any more mmx code so it can use floating point now (mmx shares the fp registers). mmx is of course the "common sense" of acceleration on x86 style systems. otoh, if you have a sse(2) cpu, it will do mmx as well. greets, chris

1 0

beginner question about input/output of fiddle
by Jie Ma 30 Jan '03

30 Jan '03

dear everyone just have one of this tiny question when playing around with fiddle -- that is, how could i send a file as an input to fiddle, and write the output of fiddle into a file (say, a txt file)? writesf~ only works for unix. i've also try to use send/receive/openpanel etc. but seems could hardly get those opjects working together as desired. i am sure for you it's just a simple line of answers. thanks a lot. warmest regards Jay

2 1

Object List (documentation question)
by steven yi 30 Jan '03

30 Jan '03

hi all, after the flurry of downloading externals yesterday and not having the mind enough to remember what exactly was included in all of them, i was looking to see if there was an object list. however, i didn't find one. and probably the usefulness of that would be limited, just knowing the name of the object without any explanation. now, for finding information about an object there's the help patch system. but what if you don't even remember the name of the object? taking a look at the help documentation (the one from clicking help on the canvas), i thought it a great memory jogger to see the one liners of object names and a quick description. i'll probably go ahead and edit my 0.INTRO.txt file to include a list of all of the external objects to jog my own memory. i was curious to ask if others are doing the same or if there is another method people are using to look up objects. also, if this is deemed a good method, would it be good practice to recommend externals writers to supply a oneliner doc for their object? thanks, steven

2 1

alias.mine file for OSX
by Alex from Idoia 30 Jan '03

30 Jan '03

Hello, In the pd doc, is said that we can create an alias for pd to avoid to type the complete path everytime we open a terminal window. It is also said to add the alias command line in the ~/.tcshrc file to keep track of the alias for further uses. Is that something to do with an alias.mine file ? How to create a such file ? I could not find the ~/.tcshrc. Thanks

6 7

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 January 2003