hi forum.

i dunno if announcements of this kind belong to the abstractions forum, the patches one or the news one (maybe even the pd-announce mailing list?) but for i wish to share my pack of abstractions with the community, i file it under abstract~.
for those of you who use pd mainly for music synthesis, my bagoftricks might be interesting, i announced it already (0.2.6) together with that shameful electro track i posted here (yes, i can do better .

here's the new version 0.2.8:
[url=https://www.puredata.org/Members/syntax_the_nerd ]https://www.puredata.org/Members/syntax_the_nerd

from the changes.txt (since 0.2.6):

-added bot.wshaper, capable of cosine, polynomic and some other transform
-fixed pitch bug in bot.blubber, much more usable on idm drums now
-added bot.chord, simple number to harmonic group of numbers converter
-repaired a bunch of ugly bugs in bot.plucker. should be in tune now and cpu
aware
-added bot.paraeq single 2p-bandpass wrapper
-slightly changed bot.distort's behaviour
-added a volume control to bot.blip
-changed bot.filter from using bot.synth2's vcf filterboxes to regular
lp/hp/bp's. saves a lot of cpu, they weren't signal controlled anyway
-bot.revplay can now play forward. idm retriggers hooray!
-added bot.kmute, a keyboard-controlled spigot scheduler
-added "please hurt" to the bot-song-archive
-nailed down some cpu leaks
-added some cpu leaks
-added bot.pong not really finished yet, though, but fun to watch (beware, it
drains some cpu cycles)
-added a new hihat sound to bot.minidrm
-some bugs added, some fixed
-extended bot.sq-piano somewhat
-crippled bot.sq, now depends heavily on bot.sq-piano


enjoy, and don't forget to mail comments, flames or tracks to me.
stn

yeah that should do the trick. i think that bag of tricks was made before the list function was introduced.

For the moment I solved the "iem_prepend_kernel" problem
by loading at startup the following pre-compiled libraries:

iem_t3_lib
iemlib1
iemlib2

downloaded from the iem website.
On long times I'd like to understand how this prepend from iemlib
works in the bot abstractions and possibly changing it with list prepend.

I definitely like the bot layout and simplicity. It seems to be very
simple to use and if it is easily expandable it will become my swiss knife for
composing music in PD...

Alberto

hi.
sorry for taking the zip offline, but some months ago i kinda abandoned the bagoftricks project. i wanted it to become my swiss army knife of pd composition and even if i did kinda well on the synthesis/effects part (maybe not in this release, i have newer, unreleased patches lying around here) my pain in the ass always was sequencing. i wrote several (around 8?) different sequencer/arranger solutions for bot, following different concepts, but none of them allowed me the quick workflow i wanted to achieve. having a song idea in my head always took me about 5 minutes to get some synths set up right and then another 10 minutes to write a (looping) sequence that is longer than 16 ticks. after those 15 minutes i forgot my original song idea, noodled around with long fx chains. all this never resulted in songs longer than 2 minutes, so for my part, the bot wasn't worth it, workflow-wise.
i've began a rewrite of the bot, reusing some dsp code, but taking different approaches on other aspects, but it now already suffers the same workflow problems the original bot had.
maybe if someone came up with a great, flexible, yet easy to deploy and easy to use sequencer/arranger solution i might take up the work again, but for now i'm heading elsewhere.

stn

The bot collection was a big influence on me and helped me work out my first pieces of Pd music, so a massive thanks for that!

Yes, the sequencing side sucks. It's difficult to come up with a flexible and extensible method in Pd I agree. I've tried a few things with lists and textfiles but never got that perfect and easy to understand/use combination for steptime compositions.

However, what you did with bot stands out on its own as an instantly usable grab bag for newbies to make Pd music. It's short of net-pd or other environments people have made, and that's still its strength imho. I would just loose the t3envs and any other non vanilla stuff, I never really heard the difference against an ordinary [vline~] to be honest. And sod the sequencer, just make everything be 2-tuple list controlled and connect them to bot.midi

What sort of music have you made lately btw?

best,

Andy

Yep... stn, the bot environment is a great immediate, self-contained PD
environment. I really like the examples (tutorial) you provided
with it - great sounds and clear examples.

Apart from sequencing side, I looked a bit into "expandibility" of bot
since I really like the approach
of bot pre-made (and candy colored tools like filters, synths, FX but
this is not really easy: for example
if you want to add a synth to the bot arsenal (hardoff's great juno synth comes to
my head in this moment this is not immediate, nor it is to take a FX out
of bot and use it into other environments (at least you need to make some - mmh, heavy? - mods). So at the end if you want some special features
you may feel a bit "narrow" between bot walls

On the other side I think that if one being used to it, bot *is* one of the much powerful environment I've seen in PD for direct, straigth music composing out-of-the-PD-box. (X Andy: looking into your songs I've the impression
of a slightly different approach: if in bot sequencer is completely separated from
the synth-effect part like a orchestra-score in csound, in your songs the
division is not so "clear", and the sequencing side seems a bit more integrated
with the synth-effect side: am I right? I still need to find my way

So when I first started with PD last year it was a great
discovery like: "wow, this bot is really cool!", which for a PD starter like I was
this helped me a lot in staying inside PD world. So, even if it is not complete,
even if you are not completely satisfied with it ... put it on-line again

I saw that you prepared later last year a tool called "mmm": are you still
developing around it? That was a different approach, and personally I had a
bit more difficulties in understanding it w.r.t. bot...

All the best,

Alberto

I'm workiing on a tracker for sequencing.

unfortunately due to pd's slow gui (especially on osx), it started to become very clunky once i added all the stuff i had planned, so i am now stripping it down to the bare necessities. will probably still be a month or two till i can put it up for people to use, but it's a nice way of sequencing.

@AlbertoZ said:

(X Andy: looking into your songs I've the impression
of a slightly different approach: if in bot sequencer is completely separated from
the synth-effect part like a orchestra-score in csound, in your songs the
division is not so "clear", and the sequencing side seems a bit more integrated
with the synth-effect side: am I right? I still need to find my way
Alberto

Yes, I like to experiment with all sorts of musical control structures that break the old sequencer/instruments way of doing things. If you like, one concept missing from Csound is the "performer". So building some intelligence into each instrument like having the bass become a "bassist" is one way I approach it. Also, building groups of instruments that have shared data, like a a "brass section".

@Hardoff

The problem with GUI elements getting clunky is one of the things I was thinking about. It's true that several people have built really innovative and well presented sequencers in Pd, but the problem is always that once you have about 10 or more flashing bang buttons or a moving time line the GUI brings the system to a halt. Guess that's why we have data-structures. In fact with tables and data-structures there's a world of interesting and powerful sequencing you can do with Pd, but not easily presentable as a beginners tool lsuch as would be suitable for bot.

for me the key is in having only a minimal few gui objects on screen, and then using radio objects to scroll between different instruments / settings

i would like to do more with data structures, but the main problem for me was that they are very dependant on the order of creation. so, if you want to quickly add a new data structure between structure F and structure G, it meant that you had to shuffle all the structure' values after structure G.

a data structure system where you could just name an entire group of structures, and send info to all of them regardless of creation order would be better i guess.

oh, and a moving time line on an old mac running os-x is impossible. it jumps in 200ms intervals.