AltME groups: search

Whether that or not, it creates a problem of scalability.

If 1 in 100 of people who find the guest details are bad guys, then 
any public world has an easily estimated probably lifetime.


(The fact that worlds have been destroyed by hardware failure before 
the bad guys arrived doesn't invalidate that argument -- just highlights 
another security/scalability issue).


And I'm not suggesting that 1/100 of Internet users are bad guys 
-- the bad guys might work harder at finding things to damage than 
the good guys.

I still argue the hardware failure has nothing to do wtih AltME. 
 Since a double blind test shows all software fails this same test.

As to security, that is easy, don't have a guest account.


This "world" is like a Hostel.  People come and go, and if someone 
wants to come in and burn the books one night, they can.


There is no level of security that will stop this.  If I get your 
right password I can destroy your credit rating, blah blah blah.

None of this has anything to do with AltME.


However, we could add more locks to more doors, and that is a different 
issue.  We would thus be changing the philosophy here, which I'm 
open to, but it still needs to be clear this is a different issue 
than is be ascribed to AltME.

post, delete, read with each of these being 100% use private list, 
moderator only, others.  so you could have a list which is readable 
by all, postable only by private memembers and deletable only by 
the moderator.

Yeah, this happens to me too sometimes.  Although I use Word a clipboard 
(spell checking).  Sometimes I grab EVERYTHING and just paste somehere. 
 I have had to get in the habit of NOT hitting send anywhere (not 
just here, but o the web as a whole).

open your prefs.txt file and set:
sound-new-message: %wavefile.wav
It must be a simple wav format, not a compressed format.

Reichart -- glad to hear Altme is being actively worked on.  I'd 
got the impression from an earlier message of yours (on REBOL2, maybe) 
that Altme was what we used to call "functionally stabilised" (ie 
dead as a development)

Now about some interim releases? Create a sense of progress and momentum.

Actually quite the opposite.

AltME is being rebuilt form the ground up (which is why we don't 
try to fix every little thing on the current version).


We are building it to allow for the integration of applets, and a 
core messaging system to make it more compatible with external applications 
(for example).

I can state that the the team is in fact working day and night on 
this.  We just don't make a lot of noise when we are working.

You could reduce the name column size and give more space for chatting

How come when Reichart changed hs name to "R", it was updated everywhere 
and not "Jean-Françoi" ?

an->and*

FYI - My server crashed (win2k) just as I was starting Altme, somehow 
all world data was destroyed, and Altme refused to run.  Required 
a re-install.

And here is my first problem

and look at debug output in console

it still changes resolution to 8bit and channels to 2

unfortunately there's missing page 2 in the document. however if 
anybody is interested in synthesis, next version of sintezar is going 
to have five different oscilators, four filters, three waveshapers 
(these units are ready now) plus some other units (phaser is ready, 
i'm working on delay). I wrote some basic sequencer last week, so 
only some preset management system is missing and some work on GUI 
(drag and drop for connecting modules is ready, but I still have 
to rewrite all modules to support it)

Very nice looking ! May I criticise your knobs though ? They are 
a bit big. Before long, you won't have any room left ! :) I suggest 
make them half the size in width and height.

i love those fancy bars from winamp or wmp how they do that ? how 
they get the khz and hz and all that from the wav file ? suposing 
it is a wav ... (raw or pcm)

fast fourier transformation. no way in current rebol (to much computations). 
On my Athlon XP 1700+ can REBOL do some 3milions computations in 
second (this is very aproximate number, some computaions are more 
time-consuming than other ). May sound like a big number, but if 
sound's sample rate is 44100Hz, there's only some 70 computations 
for each sample. And this is the ideal situation (no GUI and other 
CPU-demanding stuff). So there are two possibilities. 1) Some faster 
math, or JIT REBOL compiler  and 2) wait for faster CPU's. Let's 
see what comes first.

Pekr yes, external library call can solve it (partially). I have 
to install some C/C++ SDK and start to learn it again...

Can you write prototype in rebol, and i try my c-skills?

I've got phase-distortion oscillator, FM oscillator, wavetable oscillator 
and my exponential oscillaor plus four types of filters and it's 
really time consuming in REBOL

osc: context [
	phase: 0
	AM: 1				;Amplitude modulation
	PWM: 0.5		;Pulse width modulation
	DCW: .5	; phase transition between sine and other wave [0 - 1]

 wave-length: 400	;wave-length is sample-rate / frequency - don't 
 wanna have samplerate here	
	algorithms: [
		saw [
			d: (1 - dcw) / 2
			cosine 360 * switch-range ph [
				0 d [ph * (pwm / d)]
				d 1 [(ph - d / (1  - d)) * pwm + pwm]
			]
		]
		square [
			d: 1 / max 1e-50 (1 - dcw)
			cosine 360 * switch-range ph reduce [
				0 .25 [ph * 4 ** d / 4]
				.25 .5 [(abs ph - .5) * 4 ** d / 4 + .5]
				.5 .75 [ph * 4 - 2 ** d / 4 + .5]
				.75 1 [(abs ph - 1) * 4 ** d / 4 + 1]
			]
		]
				
		triangle [
		;triangle is d~.588888
			d: 1 - (dcw * .41111) ;1 / max 1e-50 (1 - dcw)
			cosine 360 * switch-range ph reduce [
				0 .25 [ph * 4 ** d / 4]
				.25 .5 [(abs ph - .5) * 4 ** d / 4 + .5]
				.5 .75 [ph * 4 - 2 ** d / 4 + .5]
				.75 1 [(abs ph - 1) * 4 ** d / 4 + 1]
			]
		]
	]
	algorithm: 'saw
	curves: copy []
	forskip algorithms 2 [append curves to string! algorithms/1]
	phaser: func [][
		phase-per-byte: 1 / wave-length
		phase: phase + phase-per-byte
		if phase > 1 [phase: phase - 1]
		phase
	]
	wave: func [
	][
		ph: phaser
		do select algorithms algorithm
	]

and wave function - produces actual sound

I'm filling a buffer of 256 bytes so I can change parametres in realtime 
(buffer is filled, GUI works again, I can fiddle with knobs) and 
then I add buffer's content to some sound [block!]

but because there's no difference in speed of integer and FP math 
in REBOL, everything is done in FP sou it's needed to convert FP 
to 16-bit integer

I'm using this code (and I really hope there's some better solution): 
>> insert tail sample do rejoin ["#{" skip to-hex to integer! (sample-value 
+ 1) * 32767.5 4 "}"]

luca: take a look at http://www.rebol.com/docs/sound.html, it's 
googd start. there are examples how to synthetize simple sine wavew 
and how to apply some envelope on it.

two or three things still missing: WAV saver, Vibrato and probably 
something else. At least WAV saver must be finished before release.

Volker lot of things, really :) I'd like to finish features mentioned 
above and then translate all computations to C. I'll let you now, 
thank you.

You must create sound with stereo effect. Once you insert something 
to sound:// you lost control and cannot change even the volume.

And here's why:

Thanks Kru, I reread that part, and actually found Carl's explanation.

Now that's interesting ... 

if I paste your example into a console, I get a very short high beep

if I paste it into a script and do the script, I get a math error! 
(Math or number overflow)

Another thing - REBOL sounds different compared to other players. 
It's noisy and dull. I've got 44100Hz/16bit sound and it sound's 
more like 22050/8bit to my ear.

simply put, rebol sound quality and buggy behavior deserves to be 
put into trashcan ;-)

the question is - AGG for vectors, is one of two best free libraries 
- Cairo and AGG ... so - how is fmod quality wise? Is there any better, 
smaller library to have?

Btw, I get different values, and the stereo version plays a sound 
for me

Probably depends on sound hardware, because I've testing this previously 
and I was getting different results on my home computer and my work 
computer

Here under XP and REBOL/View 1.3.1.3.1, both example (mono and stereo) 
work, 2 sample/data is not changed after insert.

REBOL does definitely play sounds in 22kHz/8bit. You can hear the 
quantisation noise in low volume and everything sounds so filtered.

I've got some Intel chipset soundcard in work, but here at home I've 
got M-Audio Delta 1010LT and the results are same (sound results, 
sample/data looks different).

BTW, WAV saver is ready, so I'll do some demo-sounds and fixes and 
release it ASAP ;)

(don't expect that 'bell' sounds like bell, 'violin' like violin 
and so on. it's just a name :-)

I said I'll release Sintezar after Competition. Compo is over, so, 
I'm releasing it :) Download it from http://krutek.info/rebol/SintezarPM101.ZIP
. Unpack it somewhere and run %sintezar_pm-101.r I hope it should 
work.

Anyway, it works. In the bottom row, there are six buttons. Press 
second button (load sound) and go to %sounds/ directory. Select a 
sound and open it.

Then press fifth button (compute sound) and wait. Press last button 
(play sound, looks like triangle) and it should play a sound.

You can change waves, adjust modulation, filter, envelopes, wave 
shaper, bit crusher and so on...

There's also some unfinished documentation included. Things that 
does not work: init sound (first button), LFO (black square button 
right to envelope button) and playing empty sound. That's what I 
remember for sure. But editing sound, creating, saving to own format 
and exporting to WAV works. Have fun.

Oh, and the Melody button does not work too.

I got the data using a free MSX emulator (openMSX). The music come 
from the game "Auf wiedersehen Monty".

You can use a "dump PSG" functionality in openMSX to get the raw 
data (values of the registers of the PSG).

You can find lot of roms (games) for the msx wich contain cool musics 
and effects for the AY PSG on www.planetemu.net

Beware, these roms are not freeware even if you can download them 
freely.


Another way (i have not tried), is to use VortexTracker at http://bulba.at.kz/
(see http://www.youtube.com/watch?v=lYdIqcqgkPMfor a demo)

wich is well designed to compose an manipulate raw data for the AY-* 
PSG.
I think many samples are given too.

How are you creating them Alan, and what specs? Try generic 44.1Khz 
stereo.

I read that you could start it from teh command line and get it save 
in .wav but I could not get it to work..


THere are a lot of annyoing things about vista...but I have it on 
my wife's laptop and I'm forced to use it now and again..

doesn't even need an install  :-)  and it works the same on mac and 
linux.

ICarii, I just quickly tried your synth2.r. It plays the song here 
on Linux, and matches the score. I wonder, though, how you are going 
to scale the program. Rebol's sound port is currently underpowered, 
and Rebol isn't fast at number crunching which is necessary for audio 
work. I don't like the idea of waiting for hours for a full song 
to render.. I would be looking for an audio processing system to 
hook into rebol.

im going to port it to R3 tomorrow and rely on Call to play sounds

added an interpolate function for dual channel sound and works fine 
:)  still slow on initial build though :(
interpolate: func [series1 series2 chunksize /local ts1 ts2 c][
	ts1: copy series1
	ts2: copy series2
	c: copy #{}
	forall ts1 [
		insert tail c copy/part ts1 chunksize
		insert tail c copy/part at ts2 index? ts1 chunksize
		take/part ts1 chunksize - 1
	]
	c
]

next on the list of todo's is to separate top and bottom notes and 
timings and add in rests.

I was compiling the math functions to C and the result was about 
5-10x faster (not optimalized), but I cannot find that code right 
now.

i was thinking of prebuilding and saving envelope shapes and modifying 
the tone like a mask

I was also working on tracker-like program (not published). I had 
a simple format for notes there that was similar to sequencers - 
I was using tuple! for timing information, it may be better than 
having two blocks, one for notes and one for timing.

i cheated and used the number of waves

Steeve, thanks for the FFT, I'm really interested. I had only DFT 
running for my modular synth. And maybe iFFT also, I don't remember, 
I hope the sources are still somewhere in deeps of the Internets.

Anton, you asked why i had an FFT implementation...

Because, i made some tests to extract frequencies from wav sounds, 
to be able to play sounds on an old 8bit sound chip.
And also, i wanted to use FFT to make an image compressor.
Geez, so much projects vanished...

I needs real documentation, (beyond what  I have and source comments)

and I'm ironing out some details as I try to use it for GLASS.

4 if you differenciate dg and dag nodes.

stream is for filtering streams of data (like events)  in which you 
chain handlers and modify the stream as it flows from one node to 
another.

and another api is for aspect-oriented programming in which you define 
aspects and then let the engine compile results based on data you 
give it.   The engine allows you to speficy the rule for combining 
aspects and applying them to data sets.

I was really hoping for a release this week (it is ready codewise) 
but will not be able to document it properly, so it will be useless 
and will gobble time away, with me trying to explain how to use it 
!

Max, just add some really simple examples. How about a lazy evaluation 
example? Or a way to import an XLS sheet and have it calc by liquid?

hehe... I am working on the documentation and new steel site as I 
write this.

including the imminent liquid release on rebol.org.  but I want to 
finish a tutorial, which will serve as both a final test and a learning 
tool for new users.

(and me too  ;-)

I wanted to release liquid last week... I could have, but would not 
have been able to properly support the release itself and answer 
even basic questions.  so I am preparing more help around it, allowing 
people to get the basics by themselves... (I hope)

Hi Maxim, 
are there liquid pumps?

As an example, imagine a layout with 2 input values, and 2 outputvalues, 
which depend on both inputs, e.g. input 2 numbers, and get  product 
and sum.

Now if I change the value of 1 input field, the new value would flow 
to both output fields. As far as I understood liquid so far, I now 
need another way to notify those 2 output fields, that there are 
new values waiting. And they have to actively get the new data.

Now add a pump, which pumps the data into the system, so that dependent 
!plugs automatically are recalculated.

Is this possible now? What do you think about adding this?

all you do is tell the dependent plugs that they are "Stainless" 
 (setting stainless?: true) and a plug is never left dirty.

the nice thing is that in the same network, you can have a mix and 
match, since they all are lazy internally,  but the stainless are 
told to request their data right after being set dirty.

Where did you get the formula for teh Help and close button: http://www.pointillistic.com/open-REBOL/moa/steel/images/regraph/regraph-0-1-4-main.png

slim is about managing the internals ... both technically and politically. 
 its about providing a unique and consistent code model to allow 
tools to wrap around this.

so while people have been waiting 5 years for the promise of modules, 
I have been using them.  and when Carl released his spec... I did 
85% of it in exactly the way he suggests.  even in some details like 
'exposing code.

Robert tried slim and in 15 minutes adopted it.

So that's all that's needed ?

A problem with all these different module managers and include systems 
is that there is ignorance of what they actually do. When you post 
code like the above, I have no idea what it's going to do with the 
files, how they are going to be arranged on my harddisk etc., and 
what global words are going to be defined.

My old INCLUDE system and the new one I've been working on also suffer 
from this "gap of understanding".

I think the answer is a GUI option which shows in clear diagrams 
what is going on.

the module user exposes on demand, and even the context is unnamed, 
you assign it.... the above again misses code... man  its 2AM here 
I am sleepy.

I am saying it's better for us to integrate a clear image / animation 
of what's about to happen, and what's happening, within the user 
interface.

slim lets you put files where you want (in many dirs if you want) 
and even allows loaded-lib relative resource dirs.

if I say that a gui manager was needed... every reboler on this list 
would complain that its too complicated and would be scared that 
the lib manager does things behind the scenes.

If i understand correctly slim allows you reuse libraries or modules 
in other programs by opening them as opposed to cutting and pasting 
the module into the rebol script?

its more like what value it adds ;-)  but the core featues are:
-loading external code, but in a managed way

-path resolution, allows you to put libs in one or many folders, 
a part from code and reuse.
-version verification.

-enforcing of a standard, where all the code is wrapped in one context 
and the use of 'set is illegal (nothing gets bound to global context)

-allowing the user of a lib the choice of exposing what words he 
wants in his code.

and even automatic renaming of exposed words.

and many more features, like vprint, engine for advanced management 
of verbosity and powerfull indented tracing of your code.

ah... just wanted to announce a little milestone of liquid useage 
 :-)   In the current instance of liquidator, event handling and 
window refresh have been successfully decoupled  :-)  this means 
I can refresh based on a timer, instead of based on input events... 
 so I can now properly scale interactively of the view based on its 
weight without even removing one event from the view  :-)


you'd say... but we can already do that... well not exactly... yes 
 you can prevent the call to 'show in your mouse handling... but 
can you prevent the whole processing occurence when you are interactivally 
editing an item?  hehe .


because of liquid's lazy computing, the actual processing is delayed 
until needed, and the show will only occur after this.   so I can 
call update 30 times a second, and it will still only call show when 
things have realy changed, inversly, I can change the data 100 times 
a second, and only the real number of calls to update will actually 
cause a data refresh of all dependent data which changed (like the 
draw block :-)

that is the concept, but liquid will know about any atom of data 
which needs to refresh, and will only recompute what HAS changed, 
so in the example you give, not all draw element will actually get 
refreshed, probably only one or two, and then, the draw block is 
relinked... the smarter you are within the processing (as usually) 
the faster will be the refresh rate.

and connect several other nodes (plugs) to any node, you can even 
label and separate your connections,

the actual data you transport through the connections (links) is 
arbitrary, and can change at each data change.

then the linked nodes, will link to those containers, or other nodes, 
and the processing of each node, will determine what is output from 
the node as source data for another node or as actual processed data.

you create a new node, with the new line of text and link it AT the 
offset which you want , and then just ask the master node for its 
text, and it will give you the new output. (if that is what its processing 
does)

but if i insert a node between a and b, node c is not informed ?

and thus, even if you change any input node's values or link new 
stuff to node_c... none of that will actually call any processing, 
untill you ask for the content of  node_c

and now a more harder question