Hi,
I am now changing from pd "normal" to pd "extended" (latest test3) on
windows, and wanted to report some things:
- /bin/pthreadVC.dll was missing - pd started anyway after complaining. I
copied the file from the previous installation and everything was fine.
scale gem/maxlib clash (on a separate thread)
I'm used to having a .bat file for different configurations of pd
(different sound cards, etc.), so I find the registry parameters not so
flexible - in pd you can only "officially" set 10 paths/libraries. I
deleted all variables from the registry, as I won't use them. Maybe a
handy feature would be not just a more flexible/dynamic tcl/tk menu with
all options there, not just a flag field, but also the possibility to save
different configurations: pd -preset 1 ? It's always possible to do
different .bat files, but it would maybe cleaner that way (?).
- as usual, I keep an updated 0.INTRO.txt file. I will try to update it to
pd-ext asap, and leave a copy on my pured.org page (might take a while)
- the following dlls from unauthorized couldn't be loaded: playlist,
sonogram~*, stk*~
- in /doc/4.audio are all available .pd files there? In my other version
they went to K06, here only to G09
- the following externals couldn't be found: bartlett~, blackman~,
connes~, cosine~, gaussian~, hanning~, hamming~, kaiser~, lanczos~,
welch~, bthresher~, buzz~, the fftease library for pd, cverb~, exp~ (from
gge?)
- is there any option to define the start position of the pd window?
Joao
On May 6, 2006, at 4:17 PM, João Miguel Pais wrote:
Hi,
I am now changing from pd "normal" to pd "extended" (latest test3)
on windows, and wanted to report some things:
- /bin/pthreadVC.dll was missing - pd started anyway after
complaining. I copied the file from the previous installation and
everything was fine.
I recently created the Pd-extended category in the bug tracker.
Please add bugs like this there so they can be more easily tracked.
- scale gem/maxlib clash (on a separate thread)
Gem is still a multi-object/single-file object, so it still has the
old name clash issues. But you can always use [maxlib/scale] to get
that object.
- I'm used to having a .bat file for different configurations of pd
(different sound cards, etc.), so I find the registry parameters
not so flexible - in pd you can only "officially" set 10 paths/ libraries. I deleted all variables from the registry, as I won't
use them. Maybe a handy feature would be not just a more flexible/ dynamic tcl/tk menu with all options there, not just a flag field,
but also the possibility to save different configurations: pd - preset 1 ? It's always possible to do different .bat files, but it
would maybe cleaner that way (?).
The preference panes sorely need to be rewritten to make them work
better. GUI control over pref presets would be quite nice. Also,
Path and Lib lists with no length limit are important too. Any
volunteers? Ideally, these pref panes would be Pd patches that talk
to the core process to do the actual interfacing with the OS.
- as usual, I keep an updated 0.INTRO.txt file. I will try to
update it to pd-ext asap, and leave a copy on my pured.org page
(might take a while)
- the following dlls from unauthorized couldn't be loaded:
playlist, sonogram~*, stk*~
They need to be ported again to Windows. Carmen did the port a while
back, but the ported versions were mistakenly overwritten. To get
them working again, its a matter of comparing the two most recent
versions in CVS and putting the Windows porting stuff into the latest
version. Any volunteers?
- in /doc/4.audio are all available .pd files there? In my other
version they went to K06, here only to G09
- the following externals couldn't be found: bartlett~, blackman~,
connes~, cosine~, gaussian~, hanning~, hamming~, kaiser~, lanczos~,
welch~, bthresher~, buzz~, the fftease library for pd, cverb~, exp~
(from gge?)
Those look like the Windowing objects from Joe Sarlo. Joe doesn't
maintain his code in CVS, so they haven't been included. We could
import his code to include them. Any volunteers?
- is there any option to define the start position of the pd window?
Don't think so, except editing pd.tk perhaps. It wouldn't be so hard
to do.
.hc
Joao
-- Lugostr. 14 79100 Freiburg i. Br. Deutschland +49 (0)761 7074997 jmmmpais@googlemail.com http://www.puredata.org/Members/jmmmp
PD-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/ listinfo/pd-list
"Computer science is no more related to the computer than astronomy
is related to the telescope."
-Edsger
Dykstra
Hallo, Hans-Christoph Steiner hat gesagt: // Hans-Christoph Steiner wrote:
- scale gem/maxlib clash (on a separate thread)
Gem is still a multi-object/single-file object, so it still has the
old name clash issues. But you can always use [maxlib/scale] to get
that object.
maxlib/scale is trivial to clone with [expr]. I often use attached abstraction for linear scaling.
Ciao
Frank Barknecht _ ______footils.org_ __goto10.org__
now *these* are the kind of threads I like!
Frank's abstraction is quite useful in
[expr ($f1-$f2)*($f4-$f5)/($f2-$f3)+$f4]
I was happy to see this. It gives me more ways to think about this problem.
Also possibly useful could be an even simpler equation to represent a linear scaling operation
[expr ($f1*($f3-$f2)+$f2)]
where $f1 is your floating point input value in range 0-1, $f2 is the intended MIN value and $f3 is the intended MAX value. I use this quite frequently, although possibly less now that I see how Frank approaches it ;)
./d5
ps. Pd patch version of the equation introduced in this mail.
#N canvas 751 398 387 168 10; #X obj 97 74 expr ($f1*($f3-$f2)+$f2); #X floatatom 97 39 5 0 0 0 - - -; #X floatatom 177 39 5 0 0 0 - - -; #X floatatom 260 41 5 0 0 0 - - -; #X floatatom 98 115 5 0 0 0 - - -; #X text 180 19 min; #X text 266 19 max; #X text 19 18 input in range 0-1; #X connect 0 0 4 0; #X connect 1 0 0 0; #X connect 2 0 0 1; #X connect 3 0 0 2;
On May 7, 2006, at 11:13 AM, Frank Barknecht wrote:
Hans-Christoph Steiner hat gesagt: // Hans-Christoph Steiner wrote:
- scale gem/maxlib clash (on a separate thread)
Gem is still a multi-object/single-file object, so it still has the old name clash issues. But you can always use [maxlib/scale] to get that object.
maxlib/scale is trivial to clone with [expr]. I often use attached abstraction for linear scaling.
-
day 5 -
Frank Barknecht -
Hans-Christoph Steiner -
João Miguel Pais