Hi List,

Lately I've been working a lot with this device called the lemur that 
transmits OSC [1]. I have reached the point where I need more than 23 
arguments to output from this device's interface.

I've discovered what could potentially be an abnormality in the UDO 
mechanism [0].

Can someone please tell me why the commented out instrument in example 
[0] does not compile ? It is *identical* in every respect to the 
version that does compile, however as soon as the output args exceed 
23, Csound is erroring out simply with 'too many output args' - is 
there a limitation of <= 23 output args in UDO's ??? Is this intended ? 
Is there a way to possibly extend this to 50 or maybe 100-200 (like 
p-fields) ??? I would really find it useful if it were possible to have 
 >= 24 output arguments in a a UDO ! Hopefully this is just user error 
of some sort ... any help is appreciated in understanding this behavior 
!




-David




[0]





sr		=	44100
kr		=	441
ksmps	=	100
nchnls	=	2

/*

	opcode	outtest, kkkkkkkkkkkkkkkkkkkkkkkk, 0 ; does not work, why ?

; is it only because there's 24 outlets ???

arnd	rnd31 1, -0.5

krn01	downsamp	arnd
krn02	downsamp	arnd
krn03	downsamp	arnd
krn04	downsamp	arnd
krn05	downsamp	arnd
krn06	downsamp	arnd
krn07	downsamp	arnd
krn08	downsamp	arnd
krn09	downsamp	arnd
krn10	downsamp	arnd
krn11	downsamp	arnd
krn12	downsamp	arnd
krn13	downsamp	arnd
krn14	downsamp	arnd
krn15	downsamp	arnd
krn16	downsamp	arnd
krn17	downsamp	arnd
krn18	downsamp	arnd
krn19	downsamp	arnd
krn20	downsamp	arnd
krn21	downsamp	arnd
krn22	downsamp	arnd
krn23	downsamp	arnd
krn24	downsamp	arnd

	xout	krn01, krn02, krn03, krn04, krn05, krn06,\
		krn07, krn08, krn09, krn10, krn11, krn12,\
		krn13, krn14, krn15, krn16, krn17, krn18,\
		krn19, krn20, krn21, krn22, krn23, krn24

	endop

*/

	opcode	outtest_work, kkkkkkkkkkkkkkkkkkkkkkk, 0 ; only difference is 
two args removed

arnd	rnd31 1, -0.5

krn01	downsamp	arnd
krn02	downsamp	arnd
krn03	downsamp	arnd
krn04	downsamp	arnd
krn05	downsamp	arnd
krn06	downsamp	arnd
krn07	downsamp	arnd
krn08	downsamp	arnd
krn09	downsamp	arnd
krn10	downsamp	arnd
krn11	downsamp	arnd
krn12	downsamp	arnd
krn13	downsamp	arnd
krn14	downsamp	arnd
krn15	downsamp	arnd
krn16	downsamp	arnd
krn17	downsamp	arnd
krn18	downsamp	arnd
krn19	downsamp	arnd
krn20	downsamp	arnd
krn21	downsamp	arnd
krn22	downsamp	arnd
krn23	downsamp	arnd
;krn24	downsamp	arnd

	xout	krn01, krn02, krn03, krn04, krn05, krn06,\
		krn07, krn08, krn09, krn10, krn11, krn12,\
		krn13, krn14, krn15, krn16, krn17, krn18,\
		krn19, krn20, krn21, krn22, krn23;, krn24

	endop

/*--- ---*/

		instr	1

kout1, kout2, kout3, kout4, kout5, kout6,\
kout7, kout8, kout9, kout10, kout11, kout12,\
kout13, kout14, kout15, kout16, kout17, kout18,\
kout19, kout20, kout21, kout22, kout23 outtest_work

	printk	1/2, kout1

		endin

/*---

		instr	2

kout1, kout2, kout3, kout4, kout5, kout6,\
kout7, kout8, kout9, kout10, kout11, kout12,\
kout13, kout14, kout15, kout16, kout17, kout18,\
kout19, kout20, kout21, kout22, kout23, kout24	outtest

	printk	1/2, kout1

		endin

---*/



i1	0	3600
i2	0	3600

e







[1]




sr=44100
ksmps=100
nchnls=2

0dbfs = 1
	zakinit	2, 32

giport	OSCinit	47120

ga1	init	0
ga2	init	0
gkvol	init	0
gicount	init	0

itmp	ftgen	11, 0, 4, -2, 7.02, 7.04, 8.02, 8.07
itmp	ftgen	12, 0, 4, -2, 7.09, 7.11, 8.09, 9.02
itmp	ftgen	13, 0, 4, -2, 8.04, 8.06, 9.04, 9.09
itmp	ftgen	14, 0, 4, -2, (2^(-7/12)), (2^(0/12)), (2^(7/12)), 
(2^(14/12))

/*--- ---*/

		opcode		lemur_scanned, kkkkkkkkkkkkkkkkkkkkkk, 0

krax1		init	0
kray1		init	0
krax2		init	0
kray2		init	0
krax3		init	0
kray3		init	0
kpd1		init	0
kpd2		init	0
kpd3		init	0
kpd4		init	0
kpd5		init	0
kfd1		init	0
kfd2		init	0
kfd3		init	0
kfd4		init	0
kfd5		init	0
kfd6		init	0
kfd7		init	0
kmbx1	init	0
kmby1	init	0
kmbx2	init	0
kmby2	init	0
kmbx3	init	0
kmby3	init	0

krx01	OSClisten	giport, "/RingArea/x", "f", krax1
kry01	OSClisten	giport, "/RingArea/y", "f", kray1
krx02	OSClisten	giport, "/RingArea2/x", "f", krax2
kry02	OSClisten	giport, "/RingArea2/y", "f", kray2
krx03	OSClisten	giport, "/RingArea3/x", "f", krax3
kry03	OSClisten	giport, "/RingArea3/y", "f", kray3
kp01		OSClisten	giport, "/Pads/x", "f", kpd1
kp02		OSClisten	giport, "/Pads2/x", "f", kpd2
kp03		OSClisten	giport, "/Pads3/x", "f", kpd3
kp04		OSClisten	giport, "/Pads4/x", "f", kpd4
kp05		OSClisten	giport, "/Pads5/x", "f", kpd5
kf01		OSClisten	giport, "/Fader/x", "f", kfd1
kf02		OSClisten	giport, "/Fader2/x", "f", kfd2
kf03		OSClisten	giport, "/Fader3/x", "f", kfd3
kf04		OSClisten	giport, "/Fader4/x", "f", kfd4
kf05		OSClisten	giport, "/Fader5/x", "f", kfd5
kf06		OSClisten	giport, "/Fader6/x", "f", kfd6
kf07		OSClisten	giport, "/Fader7/x", "f", kfd7
kmbx		OSClisten	giport, "/Multiball/x", "fff", kmbx1, kmbx2, kmbx3
kmbx		OSClisten	giport, "/Multiball/y", "fff", kmby1, kmby2, kmby3

	xout	krax1, kray1, krax2, kray2, krax3, kray3, kpd1, kpd2, kpd3, kfd1, 
kfd2, kfd3, kfd4, kfd5, kfd6, kfd7, kmbx1, kmby1, kmbx2, kmby2, kmbx3, 
kmby3

		endop

/*--- ---*/

	opcode	scale, k, kkk

kval, kmin, kmax	xin

kscl	=	kval * (kmax - kmin) + kmin

	xout	kscl

	endop

/*--- ---*/

	opcode	LogCurve, k, kk

kfloat, ksteep	xin

if (ksteep > 1) then

	ksteep = ksteep

elseif (ksteep < 1) then

	ksteep = 1.00001

endif

kout	=	(log(kfloat * (ksteep-1)+1)/log(ksteep))

	xout	kout

		endop

/*--- ---*/

		instr	1

krax1, kray1, krax2, kray2, krax3, kray3,\
kpd1, kpd2, kpd3, kpd4, kpd5,\
kfd1, kfd2, kfd3, kfd4, kfd5, kfd6, kfd7,\
kmbx1, kmby1,\
kmbx2, kmby2,\
kmbx3, kmby3	lemur_scanned

/*
	printk2	kpd1
	printk2	kpd2
	printk2	kpd3
*/

	zkw	krax1, 1
	zkw	kray1, 2
	zkw	krax2, 3
	zkw	kray2, 4
	zkw	krax3, 5
	zkw	kray3, 6
	zkw	kpd1, 7
	zkw	kpd2, 8
	zkw	kpd3, 9
	zkw	kpd4, 10
	zkw	kpd5, 11
	zkw	kfd1, 12
	zkw	kfd2, 13
	zkw	kfd3, 14
	zkw	kfd4, 15
	zkw	kfd5, 16
	zkw	kfd6, 17
	zkw	kfd7, 18
	zkw	kmbx1, 19
	zkw	kmby1, 20
	zkw	kmbx2, 21
	zkw	kmby2, 22
	zkw	kmbx3, 23
	zkw	kmby3, 24

		endin

/*--- ---*/

	instr 2	;Ambient Scanned Synth

ain1	=	0

krax1	zkr	1
kray1	zkr	2
krax2	zkr	3
kray2	zkr	4
krax3	zkr	5
kray3	zkr	6
kpd1	zkr	7
kpd2	zkr	8
kpd3	zkr	9
kpd4	zkr	10
kpd5	zkr	11
kfd1	zkr	12
kfd2	zkr	13
kfd3	zkr	14
kfd4	zkr	15
kfd5	zkr	16
kfd6	zkr	17
kfd7	zkr	18
kmbx1	zkr	19
kmby1	zkr	20
kmbx2	zkr	21
kmby2	zkr	22
kmbx3	zkr	23
kmby3	zkr	24

;kch	changed	kpd4
;khc	changed	kpd5

;if	(kch == 1) then
;gicount	=	(gicount + 1) % 3
;elseif	(khc == 1) then
;gicount	=	(gicount - 1) % 3
;else
;	endif

irate	=	0.01

kpos	=	(krax1+kray1)/2
kstr	=	(krax2+kray2)/2
kstif	=	(krax3+kray3)/2
kstif scale kstif, .01, 1
kam	=	kfd1
gkvol	=	gkvol + kfd5

kpos scale kpos, 0, 1
kdamp		scale	kfd2, -1, 0.5
kcentr	scale	kfd3, 0, 1
kmass		scale	kfd4, 1.1, 20

;kpan1	=	kfd2
;kpan2	=	kfd3
;kpan3	=	kfd4

	scanu	1, .01, 6, 2, 3, 4, 5, 127/74+kmass, 127/64+kstif, 
127/64+kcentr, 127/74+kdamp, .3, .1, 127/64+kpos, 0.01+kstr, ain1, 0, 2
/*
ctrlinit 1, 74,64, 71,64, 81,74, 91,64, 7,24 ; preset1

kc16 midic7 74, .01, 1      ; stiffness
kc17 midic7 71,   0, 1      ; centering
kc18 midic7 81,  -1, .5     ; damping
kc19 midic7 91,   0, 1      ; position
kc73 midic7 7,   1.1, 20   ; mass
	a0	=				0

; icondfunc, scnrat, ivelf, imasf, ispgmatrxf, icntrf, idampf, kmas, 
kspgstf, kcntr, kdamp, ilplkpos, irplkpos, kpos, kstrngth, ain, disp, 
id
    scanu  1,    .01,     6,     2,          3,      4,      5, kc73,    
kc16,  kc17,  kc18,       .3,       .1, kc19,        0,  a0,    1,  0
*/


kamp	LogCurve	kam, 1.00001
kamp1	scale	kamp, 0, .0000001

kph1	scale	kmby1, 0, ftlen(11)
kph2	scale	kmby2, 0, ftlen(12)
kph3	scale	kmby3, 0, ftlen(13)

kpch1	table	kph1, 11
kpch2	table	kph2, 12
kpch3 table	kph3, 13

kspd1	scale	kmbx1, 0.01, 4
kspd2	scale	kmbx2, 0.01, 4
kspd3	scale	kmbx3, 0.01, 4

ktrn	=	kfd6

ktrns	table	ktrn, 14, 1

a1	scans 	kamp1, (cpspch(kpch1)-cpspch(1.00))*ktrns, 7, 2
a2	scans		kamp1, (cpspch(kpch2)-cpspch(1.00))*ktrns, 7, 2
a3	scans		kamp1, (cpspch(kpch3)-cpspch(1.00))*ktrns, 7, 2

a1	=	a1 * kpd1
a2	=	a2 * kpd2
a3	=	a3 * kpd3

arez1	rezzy	a1, 358, 0.01
arez2	rezzy	a2, 358, 0.01
arez3	rezzy	a3, 358, 0.01

kenv1	phasor	(1 / kspd1)
kenv2	phasor (1 / kspd2)
kenv3	phasor	(1 / kspd3)

isz	tableng	9

kenv1	=	kenv1 * isz
kenv2	=	kenv2 * isz
kenv3	=	kenv3 * isz

ken1v	table	kenv1, 9
ken2v	table	kenv2, 9
ken3v	table	kenv3, 9

kpan1	=	0.5
kpan2	=	0.5
kpan3	=	0.5

a0l1	=	(a1*(1-kpan1))*ken1v
a0r1	=	(a1*(kpan1))*ken1v

a0l2	=	(a2*(1-kpan2))*ken2v
a0r2	=	(a2*(kpan2))*ken2v

a0l3	=	(a3*(1-kpan3))*ken3v
a0r3	=	(a3*(kpan3))*ken3v

al	=	(a0l1+a0l2+a0l3)/3
ar	=	(a0r1+a0r2+a0r3)/3

kfco	scale	kfd7, 100, 2000

a0l	rezzy	al, kfco, 0.77
a0r	rezzy	ar, kfco, 0.77

a1l	dcblock	a0l
a1r	dcblock	a0r

ga1	=	ga1 + a1l
ga2	=	ga2 + a1r

	outs	a1l, a1r
	zkcl	1, 32

		endin

/*--- ---*/

		instr	25

ail	buthp	ga1, 222
air	buthp	ga2, 222

al, ar	reverbsc	ail, air, 0.89, 22050

al	=	al * gkvol
ar	=	ar * gkvol

	outs	al, ar

ga1	=	0
ga2	=	0
gkvol	=	0

		endin

/*--- ---*/


;; Scanned Synth parameters
; Initial condition
f1 0 128 20 9 1

; Masses
f2 0 128 -7 1 128 1

; Spring matrices
f3 0 16384 -23 "/Users/dB/Desktop/circularstring-128.matrix"

; Centering force
f4  0 128 -7 0 128 2

; Damping
f5 0 128 -7 .8 128 1

; Initial velocity
f6 0 128 -7 0 128 .5

; Trajectories
f7 0 128 -5 .001 128 128
; Trajectories
f77 0 128 -5 128 128 0.001
; Trajectories
f777 0 128 -5 0.001 64 128 64 0.001
;;;;

f8 0 16384 11 1	; cosine wave

f9 0 16384 20 9 1	; sync window

f10 0 16384 10 1 .5 .333333	; siney wave

f 181 0 8192 -7 0.00000 573.016 13.02521 509.347 20.37815 350.176 
32.14286 180.394 \
49.78991 201.617 34.45378 435.067 27.52101 615.461 20.37815 668.518 
26.26050 488.124 \
43.90756 275.896 75.42017 371.399 67.64706 530.570 52.94117 488.124 
27.94117 509.347 \
42.22689 339.565 53.78151 583.627 44.74790 509.347 26.26050 265.285 
15.33613 297.119 0.00000  ; amp for crowd noise

t0	37.5

i1	0	3600
i2	0	3600
i25	0	3600

e






-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Well, the good thing about most of these hard-coded limits in Csound is that
they can be easily changed, right?  There is probably no need to modify the
main codebase since David knows how to compile Csound and could modify the
#define for max opcode outputs and recompile his own "expanded" version.

Or is it more complicated than that?

Anthony Kozar
anthonykozar AT sbcglobal DOT net


Istvan Varga wrote on 5/4/06 4:52 AM:

> While it is possible to increase the limit, the space for arguments is
> of static size (due to the way opcodes work in Csound), 



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

For the record, I tried this solution and it works however I reverted 
any changes made to the local source tree on this machine simply out of 
fear that I would make something useful but it would not be portable 
with subsequent canonical revisions of the language.

After some tests to see exactly what the magnitude is for the statics 
on orchestra compilation / score sort times, it seems that the current 
limit (23 outargs) is actually compiling things a little slower than a 
power of two value - I tried 16, 32, 48, 64.

16 and 64 resulted in the fastest compile times.

If other users would find this useful and could confirm this behaviour, 
then I think it might be useful to extend the maximum number of outargs 
in the UDO mechanism to 64. It might also be preferable to come up with 
a better solution to the statics in general but I know that this is a 
non-trivial endeavor.

As it is now, I'm leaving the limit of 23 until a more official 
decision is made and using an instr with zak k-space.

Thanks again for your help !


-David

On May 4, 2006, at 9:46 AM, Anthony Kozar wrote:

> Well, the good thing about most of these hard-coded limits in Csound 
> is that
> they can be easily changed, right?  There is probably no need to 
> modify the
> main codebase since David knows how to compile Csound and could modify 
> the
> #define for max opcode outputs and recompile his own "expanded" 
> version.
>
> Or is it more complicated than that?
>
> Istvan Varga wrote on 5/4/06 4:52 AM:
>
>> While it is possible to increase the limit, the space for arguments is
>> of static size (due to the way opcodes work in Csound), 



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

If there is not a significant difference at this point, it would be  
great to increase to 64 at this time.

We really need them for our prototyping with the Radio Baton and  
Lemur and P5 Glove Controllers..... all amazing with Csound5 -  
especially the LEMUR!

-Dr. B.

On May 5, 2006, at 10:11 AM, Istvan Varga wrote:

> I do not think the amount of space for arguments should make a
> difference at performance time - if it seems to do for you, it may
> be the result of random variation and inaccuracy in the measured
> times, or minor differences due to data alignment etc.
> Where it does make a difference is memory usage and the time needed
> for note allocations, and there more is worse (i.e. more space
> wasted, longer time to clear the allocated memory to zero, etc.).
> It may not be a large difference, but it can add up if large numbers
> of opcodes are used.
> Of course, increasing the limit by a small amount is not a real
> solution anyway: what if someone needs 65 outlets ?
>
> On Friday 05 May 2006 16:57, David Akbari wrote:
>
>> For the record, I tried this solution and it works however I reverted
>> any changes made to the local source tree on this machine simply  
>> out of
>> fear that I would make something useful but it would not be portable
>> with subsequent canonical revisions of the language.
>>
>> After some tests to see exactly what the magnitude is for the statics
>> on orchestra compilation / score sort times, it seems that the  
>> current
>> limit (23 outargs) is actually compiling things a little slower  
>> than a
>> power of two value - I tried 16, 32, 48, 64.
>>
>> 16 and 64 resulted in the fastest compile times.
>>
>> If other users would find this useful and could confirm this  
>> behaviour,
>> then I think it might be useful to extend the maximum number of  
>> outargs
>> in the UDO mechanism to 64. It might also be preferable to come up  
>> with
>> a better solution to the statics in general but I know that this is a
>> non-trivial endeavor.
>>
>> As it is now, I'm leaving the limit of 23 until a more official
>> decision is made and using an instr with zak k-space.
>>
>> Thanks again for your help !
>
>
> -------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services,  
> security?
> Get stuff done quickly with pre-integrated technology to make your  
> job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache  
> Geronimo
> http://sel.as-us.falkag.net/sel? 
> cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Please support 64 thank you.

I would prefer to be able to have more arguments in my UDO - thank you.

This... would be conceptually clearer and simpler and allow us to  
expand on current designs
rather than re-conceptualize and re-impliment.

so....... since - as you have indicated that more variables
would not introduce any significant overhead issues... we would  
surely appreciate your
or john ffithc's or one of the core developers increasing the number  
of UDO variables.

thanks .

Dr.B.
On May 5, 2006, at 11:27 AM, Istvan Varga wrote:

> I already have the stack opcodes, but have not decided whether they
> should be added to the CVS or not. But if those opcodes are added,
> then it may be unnecessary to change the UDO code.
>
> On Friday 05 May 2006 18:24, Dr. Richard Boulanger wrote:
>
>> If there is not a significant difference at this point, it would be
>> great to increase to 64 at this time.
>>
>> We really need them for our prototyping with the Radio Baton and
>> Lemur and P5 Glove Controllers..... all amazing with Csound5 -
>> especially the LEMUR!
>
>
> -------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services,  
> security?
> Get stuff done quickly with pre-integrated technology to make your  
> job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache  
> Geronimo
> http://sel.as-us.falkag.net/sel? 
> cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

John Ffitch, Micheal Gogins, Matt Ingalls, Victor Lazzarini... will  
one of you please change this nunber for me and for my work with Csound?

I would reall appreciate it.

Thanks

Dr. B.

On May 5, 2006, at 4:04 PM, Istvan Varga wrote:

> On Friday 05 May 2006 22:52, Dr. Richard Boulanger wrote:
>
>> so....... since - as you have indicated that more variables
>> would not introduce any significant overhead issues...
>
> I did mean it introduces an overhead, and even if it is not very
> significant, I think UDOs are already too heavy and much less
> efficient than they should be; considering that increasing the
> static limit is both relatively rarely useful and at the same
> time still not always enough, it is just not worth changing.
>
>
> -------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services,  
> security?
> Get stuff done quickly with pre-integrated technology to make your  
> job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache  
> Geronimo
> http://sel.as-us.falkag.net/sel? 
> cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

If I understand correctly, the increased time to zero memory will only
impact the allocation of new notes, which even in long compositions with
many score events does not typically happen a lot.  But I haven't examined
the code on this issue, so maybe I am wrong.

So, it does not seem like a big deal to me to increase to 64.

On the other hand, it is possible to compile a custom version of Csound to
raise this limit, and David said he was already successful in doing so.  Is
there a reason that this is not an acceptable solution for Dr. B's research?

Anthony


Istvan Varga wrote on 5/5/06 10:11 AM:

> Where it does make a difference is memory usage and the time needed
> for note allocations, and there more is worse (i.e. more space
> wasted, longer time to clear the allocated memory to zero, etc.).
> It may not be a large difference, but it can add up if large numbers
> of opcodes are used.



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

The way I see it, I would hate to spin my wheels making something 
musical in a UDO with 64 outargs that will not compile with future 
versions of canonical Csound.

I sincerely hope that if the number of outargs will not increase from 
the present limit of 23, that we can look forward to a better solution 
which will lend itself to a flexible expandability of UDOs without a 
significant increase in the time required to allocate new notes in 
instruments containing UDOs with > 23 outargs. Of course, in a future 
version increment of the language as this is a problem that will 
require a bit of thought to write code that will replace the current 
static definitions and is non trivial.



-David

On May 6, 2006, at 12:50 PM, Anthony Kozar wrote:

> Is there a reason that this is not an acceptable solution for Dr. B's 
> research?
>
> Anthony



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

I would rather not have to work with a custom version.  I can of course.
David is the person developing the UDOs for me, but....  over a  
millisecond in performance
time, there is something else at point here that is really pushing me  
the wrong way.

i think it was determined that have arguments that were powers of two  
would speed things up.

so - is 24 faster than 64 actually?


On May 6, 2006, at 12:50 PM, Anthony Kozar wrote:

> If I understand correctly, the increased time to zero memory will only
> impact the allocation of new notes, which even in long compositions  
> with
> many score events does not typically happen a lot.  But I haven't  
> examined
> the code on this issue, so maybe I am wrong.
>
> So, it does not seem like a big deal to me to increase to 64.
>
> On the other hand, it is possible to compile a custom version of  
> Csound to
> raise this limit, and David said he was already successful in doing  
> so.  Is
> there a reason that this is not an acceptable solution for Dr. B's  
> research?
>
> Anthony
>
>
> Istvan Varga wrote on 5/5/06 10:11 AM:
>
>> Where it does make a difference is memory usage and the time needed
>> for note allocations, and there more is worse (i.e. more space
>> wasted, longer time to clear the allocated memory to zero, etc.).
>> It may not be a large difference, but it can add up if large numbers
>> of opcodes are used.
>
>
>
> -------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services,  
> security?
> Get stuff done quickly with pre-integrated technology to make your  
> job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache  
> Geronimo
> http://sel.as-us.falkag.net/sel? 
> cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Well, the only way to know for sure how much slower it is, etc. is to do
some real profiling at the code level.  (Just running with "time" on Unix
does not really tell you much in this case).

I think that we should go ahead and try it for awhile with the limit
increased to 64.  _IF_ it actually has a measurable effect on performance
that is considered undesirable for real-time use or something, then we could
back it down some.

I personally doubt that anyone will notice the difference in speed.  But
clearly the difference in functionality is important.

Anthony Kozar
anthonykozar AT sbcglobal DOT net


Dr.Richard Boulanger wrote on 5/6/06 5:17 PM:

> I would rather not have to work with a custom version.  I can of course.
> David is the person developing the UDOs for me, but....  over a
> millisecond in performance
> time, there is something else at point here that is really pushing me
> the wrong way.



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

I am still seeing this value as 24 in CVS:

/* max number of input/output args for user defined opcodes */
#define OPCODENUMOUTS   24

I am planning to change it tomorrow to 64 unless presented with good reasons
as to why not or complaints from several other developers.

Anthony Kozar
anthonykozar AT sbcglobal DOT net

Anthony Kozar wrote on 5/7/06 2:44 PM:

> I think that we should go ahead and try it for awhile with the limit increased
> to 64.  _IF_ it actually has a measurable effect on performance that is
> considered undesirable for real-time use or something, then we could back it
> down some.



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

because 24 is not enough for my work and 64 will make all that I am  
currently developing easier
for those helping with my development.

and

i would appreciate it very much.

-dB

On May 14, 2006, at 4:48 AM, Istvan Varga wrote:

> On Sunday 14 May 2006 05:25, Anthony Kozar wrote:
>
>> /* max number of input/output args for user defined opcodes */
>> #define OPCODENUMOUTS   24
>>
>> I am planning to change it tomorrow to 64 unless presented with  
>> good reasons
>> as to why not or complaints from several other developers.
>
> Why change it ? There are very rare cases when 24 is not enough, and
> you can already use the stack opcodes for passing large number of  
> values
> (even more than 64) as well as strings.
>
>
> -------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services,  
> security?
> Get stuff done quickly with pre-integrated technology to make your  
> job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache  
> Geronimo
> http://sel.as-us.falkag.net/sel? 
> cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

UDO's are quite new and many Csound users have yet to realize their  
potential.

MIDI AND OSC controllers - such at the iCube, MIDItron, Lightning,  
Thunder, Lemur, Radio Baton,
p5-Glove and Trigger Finger - all systems that I personally use with  
Csound and all systems that I
teach my students to use with Csound - put out more than 24 streams  
of data and thus.... i and
my students needed this limit increased.

and as I though you said... THERE IS NO REASON NOT TO INCREASE THIS  
LIMIT
- except perhaps for the fact that, once again, you are trying to  
exercise and demonstrate
some sort of power - control - i am the head developer - the i decide  
what's in and out - the i rule Csound THING
- i decide how many opcodes is enough, i decide how many arguments is  
enough, i decide how many p-fields
is enough,  i decide if we need a new parser or not

YOU TOLD US THAT THERE IS A SPEED INCREASE BY USING POWERS OF 2
- well.... is 24 a power of 2 or a multiple of 2?  I think that 64 is  
a power of 2 and would
thus result in a speed increase as well as satisfying my current  
developmental need.

Since you are all developers here, I hope you can appreciate my  
sensitivity
here when I say to you Istvan - CUT THE FUCKING SHIT

If you and ALL of the Csound Developers would like, I will be happy  
to shut down cSounds.com,
save the time and money that is cost me personally, and stop  
communicating with this list and this
online community completely - this sort of BULLSHIT is what caused  
Paris Smaragdis to leave this
community some years ago and that was a serious loss.  We also lost  
Greg Thompson this way as
well - incredible Csounders and Incredibly Musician/Developers.    
INCREDIBLE - and INCOMPARABLE.
gone.... And I think this is why Barry Vercoe himself does not read  
the list....  He would rather do his own
thing, and develop his own versions of the program and not deal with  
all this CRAP.

Well...  I'm getting there too.  Thank you very much.

I need more than 24 - that's it.  And I don't want a special version  
of the program.  I want the program to be improved
and support controllers such as these in the form of UDO's that are  
being developed for them by my current collaborator
David Akbari - unless we'd like him to leave the community too.  He  
told me he needed more than 24 to get a full
Lemur prototype in place for me and support for the new version of  
the wireless Radio Baton. (yes there is only one
of these in the world right now.... and i will have it in a few  
weeks...  I'd like to use it with Csound!)

So....

Thanks for the stack opcodes.  Happy about them.  But, conceptually,  
they are
not as simple as what we were trying to do with UDO's.

This is such petty CRAP.  Are there any developers in this community  
who could increase
this limitation in the next release of Csound5?  I would really  
appreciate it.

Thank you.

(and to those of you I have offended, please forgive me, but... it's  
been a difficult year
and I have little patience left for this sort of BULLSHIT)

Dr. B.


On May 15, 2006, at 5:57 PM, Istvan Varga wrote:

> I already explained that it is possible and easy to work around the
> limitation, and I do not like the idea of changing the limit for a
> small minority of uses (note that no one complained about the 24
> arguments for more than 3 years).
> Maybe in the future a more scalable (albeit also more complex)
> implementation will be developed.
>
> On Monday 15 May 2006 14:04, Dr. Richard Boulanger wrote:
>
>> because 24 is not enough for my work and 64 will make all that I am
>> currently developing easier for those helping with my development.
>>
>> and
>>
>> i would appreciate it very much.
>
>
> -------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services,  
> security?
> Get stuff done quickly with pre-integrated technology to make your  
> job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache  
> Geronimo
> http://sel.as-us.falkag.net/sel? 
> cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

I did increase the max number of outputs to 64.  We can consider this an
experiment and lower the number in the future if necessary.

Anthony Kozar
anthonykozar AT sbcglobal DOT net


Anthony Kozar wrote on 5/13/06 11:25 PM:

> #define OPCODENUMOUTS   24
> 
> I am planning to change it tomorrow to 64



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

At 02:12 16/05/2006, you wrote:
>If you and ALL of the Csound Developers would like, I will be happy
>to shut down cSounds.com,
>save the time and money that is cost me personally, and stop
>communicating with this list and this


Well I don't want that. I have not really been following this discussion, as
but just noticed that it has been going on and on. I only read this e-mail
because it had three little red chilli peppers beside and I wondered
what that meant. Never realised that there was any trouble at all...

In any case, my opinion is that Csound development, whenever possible,
should be driven by users' needs. It is a shame to miss the
opportunity to improve it giving more possibilities to everyone.

It's a shame that Paris, Greg, Barry, Maurizio and many others do not
participate here, because of some issues and differences of opinion.
No one wants any splits in the
community and we all would like the development to be smooth.
However, the difference between the canonical csound  and the
customised versions around that people have been developing
on their own is plain for all to see. With more contributors a much
better and more complete software has been created. So it is
a shame that people have left, we do not want that, but they are
also missing a very exciting thing.

What made this software much, much better overall than the competition
is the contribution of the community. We do not want to miss out
on that. We do not want to lose cSounds.com and any single
individual user, as he/she is contributing to make it better by using
it, creating music with it, developing synthesis and processing designs,
suggesting things etc (and finding bugs!)

Last month at the Linux Audio Conference, I saw that most of the music
presented at the concerts had been made with Csound. This is an
incredible feat, considering all the wealth of Linux-based software. It is
a tribute to the system and its community. We don't want this to be
lost.

Victor


Victor Lazzarini
Music Technology Laboratory
Music Department
National University of Ireland, Maynooth 



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Hi,

I'm a bit perplexed as to what's going on. Can someone enlighten me?

Cheers,

Chris

Istvan Varga wrote:
> OK, it seems someone has made the decision instead of me, and
> changed code that I wrote and is under my copyright against my
> expressed will. That is fine, though, and is perfectly legal,
> because the code is under the LGPL. However, from now on, I
> consider user defined opcodes to be deprecated; this means that
> I will not use them in any orchestra code or examples I write,
> will not fix any bugs or add any new features or maintain the
> relevant code in any way, and will not answer any questions, be
> it usage or development related, about user defined opcodes.

I can't.

At 12:03 16/05/2006, you wrote:
>Hi,
>
>I'm a bit perplexed as to what's going on. Can someone enlighten me?
>
>Cheers,
>
>Chris
>
>Istvan Varga wrote:
>>OK, it seems someone has made the decision instead of me, and
>>changed code that I wrote and is under my copyright against my
>>expressed will. That is fine, though, and is perfectly legal,
>>because the code is under the LGPL. However, from now on, I
>>consider user defined opcodes to be deprecated; this means that
>>I will not use them in any orchestra code or examples I write,
>>will not fix any bugs or add any new features or maintain the
>>relevant code in any way, and will not answer any questions, be
>>it usage or development related, about user defined opcodes.
>--
>Send bugs reports to this list.
>To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk

Victor Lazzarini
Music Technology Laboratory
Music Department
National University of Ireland, Maynooth 

Anthony,

Thank you very much.  In fact, if powers of two don't make a  
difference, then I think we actually only needed about 30.
When David gets home and back on the Lemur development, then I will  
let you know.

Dr. B.

On May 15, 2006, at 10:36 PM, Anthony Kozar wrote:

> I did increase the max number of outputs to 64.  We can consider  
> this an
> experiment and lower the number in the future if necessary.
>
> Anthony Kozar
> anthonykozar AT sbcglobal DOT net
>
>
> Anthony Kozar wrote on 5/13/06 11:25 PM:
>
>> #define OPCODENUMOUTS   24
>>
>> I am planning to change it tomorrow to 64
>
>
>
> -------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services,  
> security?
> Get stuff done quickly with pre-integrated technology to make your  
> job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache  
> Geronimo
> http://sel.as-us.falkag.net/sel? 
> cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

On May 16, 2006, at 7:48 AM, Istvan Varga wrote:

> On Tuesday 16 May 2006 03:12, Dr. Richard Boulanger wrote:
>
>> UDO's are quite new and many Csound users have yet to realize their
>> potential.
>
> I think it would not be really wise to realize their "potential"
> given that they are now considered to be deprecated, and may be

what do you mean?



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Istvan Varga wrote on 5/16/06 7:25 AM:

> I did explain that I cannot do anything to prevent changes I do not
> agree with, and that making such changes is definitely legal under
> the terms of the LGPL; it does not really feel good, however, if my
> work is changed against my explicit request (i.e. not just changing
> without asking but also not knowing whether the changes will be liked
> or not).

FIrst, I want to apologize profusely to Istvan and the other developers for
making this change without there being sufficient time for them to comment
on it.  I did send a "warning message" and then waited two days before
changing the code, but I did not realize the developer list was taking 3 or
4 days to deliver messages!!

So, I'm sorry for causing an uproar -- my intention was to get feedback
before committing the change.

To defend my reasoning for the change, I was responding to Dr. B's repeated
requests for the increase in output args.  There was assent from one other
developer that the change was OK, and there was Istvan's dissent.  I felt
that we should at least try the higher number with the express warning that
this change might not be permanent.

Istvan Varga wrote on 5/16/06 5:05 AM:

> On Tuesday 16 May 2006 03:12, Dr. Richard Boulanger wrote:
> 
>> YOU TOLD US THAT THERE IS A SPEED INCREASE BY USING POWERS OF 2
> 
> I definitely did not say this, and there is no real evidence or
> explanation why this would be true in general.

David made the comment that powers of 2 might be faster.  However, this
observation was not based on any valid method for measuring the effects of
this type of change.

>> here when I say to you Istvan - CUT THE FUCKING SHIT

This attitude really disturbs me.

While we (the Csound developers) try to respond to reasonable (and polite)
requests for features and changes, I do not feel that we should ever have to
deal with or respond to this kind of abuse.  When I read this comment, _I_
felt like immediately quitting Csound development because I do not want to
work on a project where users hurl these kinds of reactions at the
volunteers working for their benefit.

I am sorry now that I made the change.  I apologize again for upsetting
Istvan.  I may not always agree with Istvan, but I do greatly respect the
work that he does.  Without Istvan, there WOULD NOT CURRENTLY BE ANY CSOUND
5!  (Maybe a year from now we would have gotten to the point that we are
currently at).

I think apologies from other people are also in order.

If the other developers want to revert OPCODENUMOUTS back to 24 or something
lower like 32, then that is fine with me.


Anthony Kozar
anthonykozar AT sbcglobal DOT net



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

[ as we say in california:  DUH! ]
of course i know what the word means --

but what i am asking is what
do you mean by claiming that UDOs are deprecated?
is there some other functionality replacing them?
or is just the implementation going to change
[ which i would then not say they are deprecated ]

are you labeling them deprecated just because they are a hack?
just because something is a hack doesn't mean
it is unusable!

-m

On May 16, 2006, at 8:23 AM, Istvan Varga wrote:

> On Tuesday 16 May 2006 18:16, matt ingalls wrote:
>
>> On May 16, 2006, at 7:48 AM, Istvan Varga wrote:
>>
>>> On Tuesday 16 May 2006 03:12, Dr. Richard Boulanger wrote:
>>>
>>>> UDO's are quite new and many Csound users have yet to realize their
>>>> potential.
>>>
>>> I think it would not be really wise to realize their "potential"
>>> given that they are now considered to be deprecated, and may be
>>
>> what do you mean?
>
> Deprecated means old stuff that is no longer maintained or recommended
> to use, and may degrade in quality (stability, speed, etc.) or even
> disappear at some time in the future, even if not the very soon.
>
>
> -------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services,  
> security?
> Get stuff done quickly with pre-integrated technology to make your  
> job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache  
> Geronimo
> http://sel.as-us.falkag.net/sel? 
> cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel
>



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

OK - so i wouldn't call that deprecated - as far as users are concerned
i think they should have no fear using them, whether the internal
implementation changes or not or even the syntax changes a bit
is not their concern.

i think the functionality is essential [ & that's why i started  
hacking the
named instruments predecessor!! ] and removing it would be a huge
step backwards.

On May 16, 2006, at 8:57 AM, Istvan Varga wrote:

> On Tuesday 16 May 2006 18:43, matt ingalls wrote:
>
>> are you labeling them deprecated just because they are a hack?
>> just because something is a hack doesn't mean
>> it is unusable!
>
> It can still be a pain to maintain and update for any major
> internal change (e.g. a new orchestra parser), though. It remains
> to be seen if a replacement is developed, the current syntax is
> kept but the code is simplified and cleaned up (at the expense
> of reduced efficiency), or everything is just left as it is but
> no longer improved.
>
>
> -------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services,  
> security?
> Get stuff done quickly with pre-integrated technology to make your  
> job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache  
> Geronimo
> http://sel.as-us.falkag.net/sel? 
> cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel
>



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

I agree, UDOs are much too useful to consider removing.

- Dave

Steven Yi wrote:
> Hi Istvan,
> 
> I really have liked UDO's since there introduction by you, and even
> though the code to support them is "hackish" by your standards, it
> works brilliantly for me, and the feature does allow for better
> expression of ideas in a safer and easier to debug way than macro's
> (in the same way macro's are eschewed by C++ people for reasons of
> type safety, IMO).
> 
> I can understand not wanting to work on it considering the latest
> emails, but I would think that UDO code shouldn't be labeled
> deprecated as I believe it has become a core part of the Csound
> experience and I think should absolutely be supported for any new
> parser. I think even if you were not to work on it again, others would
> and should continue to develop and maintain it.
> 
> steven
> 
> On 5/16/06, Istvan Varga  wrote:
>> On Tuesday 16 May 2006 18:43, matt ingalls wrote:
>>
>> > are you labeling them deprecated just because they are a hack?
>> > just because something is a hack doesn't mean
>> > it is unusable!
>>
>> It can still be a pain to maintain and update for any major
>> internal change (e.g. a new orchestra parser), though. It remains
>> to be seen if a replacement is developed, the current syntax is
>> kept but the code is simplified and cleaned up (at the expense
>> of reduced efficiency), or everything is just left as it is but
>> no longer improved.
>>
>>
>> -------------------------------------------------------
>> Using Tomcat but need to do more? Need to support web services, security?
>> Get stuff done quickly with pre-integrated technology to make your job 
>> easier
>> Download IBM WebSphere Application Server v.1.0.1 based on Apache 
>> Geronimo
>> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>
> 
> 
> -------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job 
> easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=k&kid0709&bid&3057&dat1642
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel
> 
> 



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Once my doctoral dissertation is completed - which with any luck will be
in the next couple of weeks - I would gladly volunteer for this task.

Yours truly,
David

Istvan Varga wrote:
> On Tuesday 16 May 2006 23:12, Steven Yi wrote:
> 
>> Would it be possible to find a solution that could satisfy the UDO
>> internal code being rewritten while maintaining the public interface
>> of using opcode-endop and all that
> 
> It may be possible, but I will surely not waste my time on it again.
> If someone feels like writing a compatible implementation from scratch,
> that should be fine. However, I still consider the current code that I
> wrote to be deprecated and subject to eventual removal.
> 
> 
> -------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel


-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Istvan Varga wrote:

> On Tuesday 16 May 2006 23:51, David Ogborn wrote:
> 
> > Once my doctoral dissertation is completed - which with any luck will be
> > in the next couple of weeks - I would gladly volunteer for this task.
> 
> Does this mean that I can safely remove all the old code after 5.02
> is released ?

Istvan, please, settle down.

If you want to wash your hands of that code then fine. Understood
loud and clear, but do let other people use and/or maintain it.
I'm sure that there are a number of people here who would be happy
to maintain it until something better comes along.

As for a good time to remove the old code, that  would be after 
David has a new implementation and everyone has tested it and is 
happy with it. 

Cheers,
Erik
-- 
+-----------------------------------------------------------+
  Erik de Castro Lopo
+-----------------------------------------------------------+
'Unix beats Windows' - says Microsoft! 
http://blogs.zdnet.com/Murphy/index.php?p=459


-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

No

>>>>> "Istvan" == Istvan Varga  writes:

 Istvan> Does this mean that I can safely remove all the old code after 5.02
 Istvan> is released ?

The point of Csound is to serve the composers and musicians.  If they
fins it useful, even if the developers think it is cr*p, we have to
leave it until it can be safely replaced.  There are many parts of
csound that I have written that I think are pointless, but others
assure me they are needed.

==John ffitch


-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

This is a real shame. Is there anyway this situation can be sorted out?

Rory.



Istvan Varga wrote:
> On Wednesday 17 May 2006 02:21, Steven Yi wrote:
> 
>> I can certainly restore the code myself as any other developer here
>> can, but I would hope you would see that it serves no purpose to have
>> taken it out and would restore it yourself.
> 
> Sorry, I will not. Actually, I quit developing Csound.
> 
> 
> -------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel
> 



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Well someone will have to restore the UDO code before we do the 5.02
release. We can't release it without UDOs otherwise the rule of not
breaking existing orchestras will not be followed.

Victor

At 11:01 17/05/2006, you wrote:
>This is a real shame. Is there anyway this situation can be sorted out?
>
>Rory.
>
>
>
>Istvan Varga wrote:
>>On Wednesday 17 May 2006 02:21, Steven Yi wrote:
>>
>>>I can certainly restore the code myself as any other developer here
>>>can, but I would hope you would see that it serves no purpose to have
>>>taken it out and would restore it yourself.
>>Sorry, I will not. Actually, I quit developing Csound.
>>
>>-------------------------------------------------------
>>Using Tomcat but need to do more? Need to support web services, security?
>>Get stuff done quickly with pre-integrated technology to make your job easier
>>Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
>>http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
>>_______________________________________________
>>Csound-devel mailing list
>>Csound-devel@lists.sourceforge.net
>>https://lists.sourceforge.net/lists/listinfo/csound-devel
>
>
>
>-------------------------------------------------------
>Using Tomcat but need to do more? Need to support web services, security?
>Get stuff done quickly with pre-integrated technology to make your job easier
>Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
>http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
>_______________________________________________
>Csound-devel mailing list
>Csound-devel@lists.sourceforge.net
>https://lists.sourceforge.net/lists/listinfo/csound-devel

Victor Lazzarini
Music Technology Laboratory
Music Department
National University of Ireland, Maynooth 



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Yes; code restored

==John




-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Istvan Varga wrote:

>> I can certainly restore the code myself as any other developer here
>> can, but I would hope you would see that it serves no purpose to have
>> taken it out and would restore it yourself.
> 
> Sorry, I will not. Actually, I quit developing Csound.

Seems to me we have been here before. Istvan, it is a given that 
sometimes programmers have to allow the purity of their code to defer to 
the needs of users (to say nothing of the needs of other developers). 
That is all that has happened here; as you have noted yourself, this is 
enshrined in the use of licenses such as the LGPL. It is a testament 
indeed, to the success and popularity of UDOs that users inevitably, and 
quite properly, push all such new initiatives to the limit and beyond, 
as they seek out the full scope they offer.  Saying that "nobody has 
complained for three years" is meaningless - code can never be 
immutable, and it can take time for the full opportunities a new 
technology offers to be realised; especially if that involves parallel 
development of other facilities such as OSC, and the appearance of new 
hardware controllers.  Of course, my opinion remains that the parser 
should have been allowed to be developed first, but that is another story.

A threat to remove code, or to quit development, therefore is a threat 
with little force, as by the natire of CVS and the LGPL anybody else can 
restore the code. If that is truly your desire, go in peace, but if it 
is meant to be rhetorical, a gesture of pique, I suggest you will in the 
end feel happier just standing up, accepting the small changes that were 
requested (which will give UDOs considerable prominence amongst a group 
of users who are being head-hunted by Hollywood producers, among other 
things!), and continue to make the contributions to Csound that have 
made you such a vital member of the development community.

You may recall you expressed yourself extremely harshly to me years ago, 
and as it happens I have effectively stopped contributing code to Csound 
since that time (I am not even registered as a developer), so I know 
what it is like to receive harsh comments in a public forum. You may 
well feel that was a wise decision on my part - but it it equally a wise 
decision on yours?


Richard Dobson

We all would love to have you back, if you could spare some time!

At 11:24 17/05/2006, Richard Dobson wrote:
>You may recall you expressed yourself extremely harshly to me years ago, 
>and as it happens I have effectively stopped contributing code to Csound 
>since that time (I am not even registered as a developer), so I know what 
>it is like to receive harsh comments in a public forum. You may well feel 
>that was a wise decision on my part - but it it equally a wise decision on 
>yours?

Victor Lazzarini
Music Technology Laboratory
Music Department
National University of Ireland, Maynooth 

Istvan Varga wrote:

> Actually, I quit developing Csound.

I'm sorry to hear that but fully understand that it is your choice
to make. I'm sure everyone here would like to thank you for your
contributions and wish you luck in all your future software and
music making ventures.

Regards,
Erik
-- 
+-----------------------------------------------------------+
  Erik de Castro Lopo
+-----------------------------------------------------------+
Virgins? What virgins? Its 72 raisins!
http://www.guardian.co.uk/religion/Story/0,2763,631357,00.html


-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Istvan Varga wrote:

> Actually, I quit developing Csound.

I tried to understand what's going on following this thread, but don't 
really get it.

All I know is that you, Istvan, is a very valuable member of the 
community (I believe you also contributed alot to the code, but since 
I'm not a developer, I'll leave that to others). You helped me out alot 
of times when I was either brain dead, blind or ignorant. You always 
provide quick, in-depth explanations to problems.

As I said I don't know the background for this heated debate, and there 
might be more to it. But the bottom line for me is: Please stay onboard, 
Istvan.

-- 
peace, love & harmony
Atte

http://www.atte.dk      | quartet:      http://www.anagrammer.dk
http://www.atte.dk/gps  | compositions: http://www.atte.dk/compositions

Sad Istvan,

but all of these responses from you are not really a surprise.

(Including your deciding to spam the regular Csound list with this
*developer* issue - you have pulled that prank before too... right..
with john ffitch and micheal gogins in the past... trying to smear their
fantastic reputations as developers and contributors to this  
*COMMUNITY*.)

Sad because I was lookiing forward to learning, using, and teaching  
the stack opcodes
because they too, like the UDOs and Macros,  seemed to be part of a  
family
that would help me take my students closer to thinking of themselves  
as *programmers and
developers*


On May 16, 2006, at 10:58 AM, Istvan Varga wrote:

> On Tuesday 16 May 2006 03:12, Dr. Richard Boulanger wrote:
>
>> Thanks for the stack opcodes.  Happy about them.  But, conceptually,
>> they are not as simple as what we were trying to do with UDO's.
>
> The stack opcodes are now removed, as there are less reasons for
> having them, and most feedback from users (e.g. Steven Yi or John
> ffitch) has not been really positive.
>
>
> -------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services,  
> security?
> Get stuff done quickly with pre-integrated technology to make your  
> job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache  
> Geronimo
> http://sel.as-us.falkag.net/sel? 
> cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Victor Lazzarini wrote:
> We all would love to have you back, if you could spare some time!
> 

I expect it will happen sometime. These days I have joined the Dark Side 
and am busy with commercial work (CDP etc), which pretty much swallows 
all my attention (gotta pay those bills...). I would like to contribute 
to the new parser, once things settle down enough for that to happen 
(Csound 6?).

In the meantime, I can take this opportunity to say how delighted I am 
that you have taken the pvs stuff and sprinted with it. I had all sorts 
of plans for further opcodes, much of which of course you have 
manifested, plus plenty I had not thought of. So I know Csound is in 
good hands while I hover around the outer fringes!


Richard Dobson

To Anthony Kozar and All Developers,

Anthony, I want to thank you for increasing the number of arguments for
the purposes of my research and development and I am sorry that
all this controversy has resulted from your effort to support my work.

David Akbari was developing demonstration materials for a one-day visit
from Max Mathews to Berklee. David Akbari was graduating from Berklee
and leaving Boston with his family. It was a time critical task.  And we
were working around the clock, during finals week, to get the project/ 
demos
done when we came across these limitations that were not allowing us
to directly map some of the coolest multi-ball control from the new  
Lemur Controller
into Csound5.  (My goal was to inspire Max about both Csound5 - which  
we did
- and about ways to improve and expand his new Radio Baton Controller.)

I asked and asked and asked and it seemed like an arbitrary decision
on the part of Istvan, not to up the number.  I did and do  
appreciated his
work-around solution (the stack opcodes) and with the appropriate  
headspace,
it seems that this would make some sense too.  but... i was also  
frustrated by the
fact that it seemed so arbitrary for Istvan to make this decision and  
shut  down my
current time-critical work.  Given the pressure I was under these  
past few weeks,
it seems to me the  equivalent of Michael Gogins telling all  
Csounders that they only
need a few p-fields, or John ffitch deciding that my instruments  
should never
use more than 6 oscillators.  And so finally.... after begging and  
begging,

I snapped...

Still - there is no excuse for brutalizing all of you - or Istvan for  
that matter -

Tough day, tough week, tough year in my life... all led me to the  
point where
- if I was sitting in a room with him - and all of you developers - I  
would have
said exactly the same thing - and then some.   I am sorry that I  
didn't just send
Istvan a private email.  Rather, I should have stormed out of the room
and gone for a jog or reached for another scotch....

and Most importantly...

- ISTVAN...  I AM SORRY
- ANTHONY... I AM SORRY
- John, Michael, Matt, Richard, Steven, and all members of the  
developer's list

I.... AM.... TRULY... SORRY....

and I do hope that you can all forgive me this momentary lapse of  
control
and reason.... especially Istvan - who's many contributions to  
Csound5, we all
so greatly appreciate.

Dr. B.


On May 16, 2006, at 12:25 PM, Anthony Kozar wrote:

> Istvan Varga wrote on 5/16/06 7:25 AM:
>
>> I did explain that I cannot do anything to prevent changes I do not
>> agree with, and that making such changes is definitely legal under
>> the terms of the LGPL; it does not really feel good, however, if my
>> work is changed against my explicit request (i.e. not just changing
>> without asking but also not knowing whether the changes will be liked
>> or not).
>
> FIrst, I want to apologize profusely to Istvan and the other  
> developers for
> making this change without there being sufficient time for them to  
> comment
> on it.  I did send a "warning message" and then waited two days before
> changing the code, but I did not realize the developer list was  
> taking 3 or
> 4 days to deliver messages!!
>
> So, I'm sorry for causing an uproar -- my intention was to get  
> feedback
> before committing the change.
>
> To defend my reasoning for the change, I was responding to Dr. B's  
> repeated
> requests for the increase in output args.  There was assent from  
> one other
> developer that the change was OK, and there was Istvan's dissent.   
> I felt
> that we should at least try the higher number with the express  
> warning that
> this change might not be permanent.
>
> Istvan Varga wrote on 5/16/06 5:05 AM:
>
>> On Tuesday 16 May 2006 03:12, Dr. Richard Boulanger wrote:
>>
>>> YOU TOLD US THAT THERE IS A SPEED INCREASE BY USING POWERS OF 2
>>
>> I definitely did not say this, and there is no real evidence or
>> explanation why this would be true in general.
>
> David made the comment that powers of 2 might be faster.  However,  
> this
> observation was not based on any valid method for measuring the  
> effects of
> this type of change.
>
>>> here when I say to you Istvan - CUT THE FUCKING SHIT
>
> This attitude really disturbs me.
>
> While we (the Csound developers) try to respond to reasonable (and  
> polite)
> requests for features and changes, I do not feel that we should  
> ever have to
> deal with or respond to this kind of abuse.  When I read this  
> comment, _I_
> felt like immediately quitting Csound development because I do not  
> want to
> work on a project where users hurl these kinds of reactions at the
> volunteers working for their benefit.
>
> I am sorry now that I made the change.  I apologize again for  
> upsetting
> Istvan.  I may not always agree with Istvan, but I do greatly  
> respect the
> work that he does.  Without Istvan, there WOULD NOT CURRENTLY BE  
> ANY CSOUND
> 5!  (Maybe a year from now we would have gotten to the point that  
> we are
> currently at).
>
> I think apologies from other people are also in order.
>
> If the other developers want to revert OPCODENUMOUTS back to 24 or  
> something
> lower like 32, then that is fine with me.
>
>
> Anthony Kozar
> anthonykozar AT sbcglobal DOT net
>



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Well, this alone is worth any code change!

At 14:09 17/05/2006, you wrote:
>(My goal was to inspire Max about both Csound5 - which
>we did
>- and about ways to improve and expand his new Radio Baton Controller.)

Victor Lazzarini
Music Technology Laboratory
Music Department
National University of Ireland, Maynooth 



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

 From and educational standpoint - UDOs are an amazingly important  
teaching and
an incredible developmental tool as well - especially for Csound  
Instrument Designers
who might not be C programmers.

Once again, I am unclear.... this feature will be removed from Csound  
because
*you* are deciding not do work on them anymore or because they are  
not being
used by sound designers and students or members of the Csound Community
of users and developers.

You would be breaking a lot of orchestras and compositions by  
removing this feature.

Do you use Max/MSP or PD at all?  Well all of my students do - as do  
many other
computer musicians, and the UDO aspect of Csound (both 4 and 5) is  
very much
like some of the *abstraction* concept in Max/MSP which add  
significant power to
that programming enviroment - the parallel in Csound (via UDO's) is  
significant
as well - and a point I very much like to make in my teaching.

Dr. B.


On May 16, 2006, at 10:48 AM, Istvan Varga wrote:

> On Tuesday 16 May 2006 03:12, Dr. Richard Boulanger wrote:
>
>> UDO's are quite new and many Csound users have yet to realize their
>> potential.
>
> I think it would not be really wise to realize their "potential"
> given that they are now considered to be deprecated, and may be
> removed at some point in the future (most probably when the new
> parser is introduced, as I am aware of serious difficulties related
> to adding support for UDOs in the new parser). I always found this
> feature a bit hackish (and the actual implementation a lot more
> than just a bit hackish), and now really and seriously regret ever
> adding it, and wasting unknown number of hours of work.
>
>
> -------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services,  
> security?
> Get stuff done quickly with pre-integrated technology to make your  
> job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache  
> Geronimo
> http://sel.as-us.falkag.net/sel? 
> cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

This suggestions of Istvan's strikes me as quite petty and absurd.

Do warning messages such as these slow down the operation of the
program in any way? Does it take an extra second to get the audio
running or launch the program as a result?

If not, then given my recent heart attack, I am wondering if we could  
also add warning messages
about  cigarette smoking, and high cholesterol diets, trans-fats,  
vegetables, etc.
when Csound is launched - or just some healthy lifestyle advice and  
tips each
time the program is launched?

Please Istvan - stop this... you are being rediculous.

Dr. B.


On May 16, 2006, at 3:28 PM, Istvan Varga wrote:

> On Tuesday 16 May 2006 19:30, Steven Yi wrote:
>
>> I can understand not wanting to work on it considering the latest
>> emails, but I would think that UDO code shouldn't be labeled
>> deprecated as I believe it has become a core part of the Csound
>> experience and I think should absolutely be supported for any new
>> parser.
>
> I do not think it is necessary to support them in a new parser if the
> old one is still kept as an option for use with legacy orchestra code.
> I also like the idea of reworking the current UDO code to be simpler
> even if at the expense of worse performance and closing the way for
> future improvements. In any case, I think warning messages about the
> use of these opcodes not being recommended should be added that are
> printed at initialization time.
>
>
> -------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services,  
> security?
> Get stuff done quickly with pre-integrated technology to make your  
> job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache  
> Geronimo
> http://sel.as-us.falkag.net/sel? 
> cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

NO.

On May 16, 2006, at 5:13 PM, Istvan Varga wrote:

> On Tuesday 16 May 2006 23:51, David Ogborn wrote:
>
>> Once my doctoral dissertation is completed - which with any luck  
>> will be
>> in the next couple of weeks - I would gladly volunteer for this task.
>
> Does this mean that I can safely remove all the old code after 5.02
> is released ?
>
>
> -------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services,  
> security?
> Get stuff done quickly with pre-integrated technology to make your  
> job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache  
> Geronimo
> http://sel.as-us.falkag.net/sel? 
> cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

I agree that UDOs are extremely important to everyone - that's why I  
asked to increase the number
of arguments.  they are amazing and add so much to the language.

Dr. B.

On May 16, 2006, at 6:31 PM, Steven Yi wrote:

> So now it seems that you're interested to remove that body of code and
> the language feature outright just so that someone else can redo that
> code to implement the very same thing?  I still don't understand the
> technical reasons for requiring a new non-compatible implementation
> nor the need to remove a useful feature, but I'd ask that you simply
> leave the code in and let others deal with it since you seem no longer
> interested.
>
> For what it's worth, I certainly don't consider your time wasted at
> all on it, as it has been one of the major feature additions to the
> Csound language in the past 7 years since I've been using Csound, IMO.
> The only way I could see it being a waste was if it was removed, as
> it would serverely hinder projects like Cabel (which would have to be
> rewritten as it uses UDO's to represent its modules), break many
> Csound works and libraries of code, and waste all the time that
> everyone has spent to learn and use UDO's.  (I know it will have then
> wasted a great deal of my time too as there are many features in blue
> that are built around UDO's, as well as most of my musical work uses
> them now too).
>
> steven
>
>
> On 5/16/06, Istvan Varga  wrote:
>> On Tuesday 16 May 2006 23:12, Steven Yi wrote:
>>
>> > Would it be possible to find a solution that could satisfy the UDO
>> > internal code being rewritten while maintaining the public  
>> interface
>> > of using opcode-endop and all that
>>
>> It may be possible, but I will surely not waste my time on it again.
>> If someone feels like writing a compatible implementation from  
>> scratch,
>> that should be fine. However, I still consider the current code  
>> that I
>> wrote to be deprecated and subject to eventual removal.
>>
>>
>> -------------------------------------------------------
>> Using Tomcat but need to do more? Need to support web services,  
>> security?
>> Get stuff done quickly with pre-integrated technology to make your  
>> job easier
>> Download IBM WebSphere Application Server v.1.0.1 based on Apache  
>> Geronimo
>> http://sel.as-us.falkag.net/sel? 
>> cmd=lnk&kid=120709&bid=263057&dat=121642
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>
>
>
> -------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services,  
> security?
> Get stuff done quickly with pre-integrated technology to make your  
> job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache  
> Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid0709&bid&3057&dat1642
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid0709&bid&3057&dat1642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

amazing.

On May 17, 2006, at 3:17 AM, Istvan Varga wrote:

> On Wednesday 17 May 2006 02:21, Steven Yi wrote:
>
>> I can certainly restore the code myself as any other developer here
>> can, but I would hope you would see that it serves no purpose to have
>> taken it out and would restore it yourself.
>
> Sorry, I will not. Actually, I quit developing Csound.
>
>
> -------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services,  
> security?
> Get stuff done quickly with pre-integrated technology to make your  
> job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache  
> Geronimo
> http://sel.as-us.falkag.net/sel? 
> cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel



-------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

I agree with Victor .... that we all miss Richard Dobson's many  
contributions - even this one to the list.

I hope that he will find the time now to rejoin the developer's  
ranks.  If Istvan has left, then there is
truly a large seat to fill.

As I stated earlier today - I am sorry about the language and  
intensity of my comments and
absolutely had not hoped that it would lead to this.  but....  like  
many of you....

I have also not been happy at all with Istvan's responses, removal of  
code, additions of warning messages, and now with his current decision.

This is a sad loss, but...


On May 17, 2006, at 6:31 AM, Victor Lazzarini wrote:

> We all would love to have you back, if you could spare some time!
>
> At 11:24 17/05/2006, Richard Dobson wrote:
>> You may recall you expressed yourself extremely harshly to me  
>> years ago, and as it happens I have effectively stopped  
>> contributing code to Csound since that time (I am not even  
>> registered as a developer), so I know what it is like to receive  
>> harsh comments in a public forum. You may well feel that was a  
>> wise decision on my part - but it it equally a wise decision on  
>> yours?
>
> Victor Lazzarini
> Music Technology Laboratory
> Music Department
> National University of Ireland, Maynooth
> -- 
> Send bugs reports to this list.
> To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk

Lazzarini in Ireland, Irish and Italian mix.... oh it's hopeless! But then
again, it did work, at least for a while....

Take a moment and realize that through it all, good, bad, nice, nasty, and
everything in between you have achieved a unique pinacle in the universe of
audio.

No other application, free or otherwise, is as capiable.

No other application, free or otherwise, is as accepted.

No other application, free or otherwise, has the heritage of every great
music algorithm ever thought of, and lots of new ones every day.

It is an amazing accomplishment. Stop a minute and smell the roses. You have
a free garden to play in, and a global community of like minded individuals
and individualists to share it with.

It is really not surprising that a few people have left from time to time,
what is amazing is that so many have joined and stayed.

Shalom

Michael

> -----Original Message-----
> From: Dr. Richard Boulanger [mailto:rboulanger@berklee.edu]
> Sent: Wednesday, May 17, 2006 9:01 AM
> To: csound@lists.bath.ac.uk
> Subject: Re: [Csnd] [Cs-dev] Max number of outlets in a UDO ?
>
>
> I agree with Victor .... that we all miss Richard Dobson's many
> contributions - even this one to the list.
>
> I hope that he will find the time now to rejoin the developer's
> ranks.  If Istvan has left, then there is
> truly a large seat to fill.
>
> As I stated earlier today - I am sorry about the language and
> intensity of my comments and
> absolutely had not hoped that it would lead to this.  but....  like
> many of you....
>
> I have also not been happy at all with Istvan's responses, removal of
> code, additions of warning messages, and now with his current decision.
>
> This is a sad loss, but...
>
>
> On May 17, 2006, at 6:31 AM, Victor Lazzarini wrote:
>
> > We all would love to have you back, if you could spare some time!
> >
> > At 11:24 17/05/2006, Richard Dobson wrote:
> >> You may recall you expressed yourself extremely harshly to me
> >> years ago, and as it happens I have effectively stopped
> >> contributing code to Csound since that time (I am not even
> >> registered as a developer), so I know what it is like to receive
> >> harsh comments in a public forum. You may well feel that was a
> >> wise decision on my part - but it it equally a wise decision on
> >> yours?
> >
> > Victor Lazzarini
> > Music Technology Laboratory
> > Music Department
> > National University of Ireland, Maynooth
> > --
> > Send bugs reports to this list.
> > To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk
>
> --
> Send bugs reports to this list.
> To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk

Wow! I leave for 5 minutes and this is what happens!

Internet. Serious business.


Anyway, I think that we all appreciate Istvan's work in creating the 
stack, push, and pop opcodes that allow a scalable use of memory. This 
idea, like all of Istvan's other ideas is an excellent one.

Although in this case, it almost seems like killing a mosquito with a 
cannonball - a little overkill. I think these opcodes should remain in 
the language however because I find them to be actually of more use in 
subinstrs. It makes the idea that one of my colleagues Greg Thompson 
said "try to make the whole ensemble of instruments in one instr so 
it's more like a C program" much more plausible.

I think Istvan was absolutely thinking for the benefit and future of 
Csound when thinking of a solution that reaches beyond the current 
static number of 24 outargs.

Maybe a better solution could be just like a UDOinit opcode or 
something? Add an optional i-rate argument to the opcode that could 
specify a handler taken from something like


giudo1	UDOinit	36
giudo2	UDOinit	3
giudo3	UDOinit	1

	opcode	thirtysixout	kkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkk, 0, giudo1
...
	endop

	opcode	threeout	kkk, 0, giudo2
...
	endop

	opcode	oneout	k, 0, giudo3
...
	endop


so that not only are we able to go beyond the current limit of 24, but 
it could also increase performance and note allocation time for 
instruments with far less arguments than 24.

Regarding the increase in the static number of args in UDO, I tend to 
think that that is an ad-hoc solution that would've worked quite well 
for the demos for Max Mathews but also does not think to the larger 
picture like Istvan had mentioned.

Most of the "tests" I did with the increase in the static args were all 
with instrs with this type of score

i1	0	3600

e

... so not many note allocations. It "felt" faster (no benchmark 
information printed to stdout) but I did not test with instruments with 
large scores.


One thing, as a user that I'd really like to see is the addition of 
f-type signals as input and outputs to the UDO mechanism. String 
variables would be really nice.

I don't think the UDO part of the language should be deprecated because 
local ksmps is such a huge thing for doing filters and such using only 
equations.

To remove this important part of the language is to ignore an important 
pedagogical element to the users as well as forcing users to use 
ksmps=1 and macros which can be markedly less efficient but still 
possible now that there is multiple instantiation of the Csound lib in 
separate memory space. However this might hinge on the Jack driver, 
another important contribution from Istvan.

One of the great things about the open source community is that 
developers are free to do what they want when they want to the sources. 
If Istvan has truly left, then I wish nothing but the best for him, as 
he will undoubtedly be successful in whatever he chooses to do. I, for 
one, will continue to use and study his ideas. In the hope of one day 
improving the Csound language at least a fraction of what he has done.



-David

Istvan Varga wrote:

> It appears that I am now locked out from canonical Csound CVS access, but

Istvan, you stated that you wouldn't be working on Csound any more. 
Given that statement, isn't it reasonable that your CVS write access 
was removed?

However, if you do indeed intend to return to development then I'm
sure your CVS write access will be reinstated.

> I am creating an alternate "unofficial" project that can be used for
> releasing such improvements and new features if there is enough interest.

That seems like such a pity. Why not work with the rest of the developers
to improve csound for everyone.

Cheers,
Erik
-- 
+-----------------------------------------------------------+
  Erik de Castro Lopo
+-----------------------------------------------------------+
"Perl as a language has less a design than a thousand special features
flying in close formation."
-- From the c2 wiki

All you need is to ask John and he will put you back on the
canonical. I think it is imperative that we do not have splits
and provide an unified csound. Unofficial alternative projects
only confuse things and make things less efficient.

Please don't split the development. If you wish to be back onto
csound development, just do it. I am sure John will restore
your developer status, as this is the wish of the community.

If you would like to be back, we really need to re-take the
release date discussions. As I said before, the 5th is bad for me,
but 6/06 is workable.

Victor

At 10:52 19/05/2006, you wrote:
>On Thursday 18 May 2006 23:27, David Akbari wrote:
>
> > I think Istvan was absolutely thinking for the benefit and future of
> > Csound when thinking of a solution that reaches beyond the current
> > static number of 24 outargs.
>
>Actually, I already have ideas about how the current UDO implementation
>can be changed to efficiently handle up to something like 256 input and/or
>output arguments; adding the new types (S and f) is trickier, but may also
>be possible at a small cost in performance.
>It appears that I am now locked out from canonical Csound CVS access, but
>I am creating an alternate "unofficial" project that can be used for
>releasing such improvements and new features if there is enough interest.
>--
>Send bugs reports to this list.
>To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk

Victor Lazzarini
Music Technology Laboratory
Music Department
National University of Ireland, Maynooth 

I agree with Victor, it seems the best thing about the current state of 
Csound is the fact that it is unified. Users only have to choose a 
platform rather than a *version* of Csound. I'm sure that Istvan has 
great plans for future improvements to Csound so please Istvan, can you 
continue to work within the canonical version rather than working on 
your own version and causing a split?

Rory.



Victor Lazzarini wrote:
> All you need is to ask John and he will put you back on the
> canonical. I think it is imperative that we do not have splits
> and provide an unified csound. Unofficial alternative projects
> only confuse things and make things less efficient.
> 
> Please don't split the development. If you wish to be back onto
> csound development, just do it. I am sure John will restore
> your developer status, as this is the wish of the community.
> 
> If you would like to be back, we really need to re-take the
> release date discussions. As I said before, the 5th is bad for me,
> but 6/06 is workable.
> 
> Victor
> 
> At 10:52 19/05/2006, you wrote:
>> On Thursday 18 May 2006 23:27, David Akbari wrote:
>>
>> > I think Istvan was absolutely thinking for the benefit and future of
>> > Csound when thinking of a solution that reaches beyond the current
>> > static number of 24 outargs.
>>
>> Actually, I already have ideas about how the current UDO implementation
>> can be changed to efficiently handle up to something like 256 input 
>> and/or
>> output arguments; adding the new types (S and f) is trickier, but may 
>> also
>> be possible at a small cost in performance.
>> It appears that I am now locked out from canonical Csound CVS access, but
>> I am creating an alternate "unofficial" project that can be used for
>> releasing such improvements and new features if there is enough interest.
>> -- 
>> Send bugs reports to this list.
>> To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk
> 
> Victor Lazzarini
> Music Technology Laboratory
> Music Department
> National University of Ireland, Maynooth

I don't think we need or want this. The CVS sources can be as experimental
as we'd like, and the releases stable. IMHO any type of forking or splitting
will only cause confusion.

For instance, Istvan does not have a OSX machine. He might implement things
that don't work there and there would be no way of checking. I can check and
if necessary fix things. But I don't want to be checking out more than one 
source
tree for the same software.

Victor

At 23:45 19/05/2006, Iain Duncan wrote:

> >>I am creating an alternate "unofficial" project that can be used for
> >>releasing such improvements and new features if there is enough interest.
> >
> >
> > That seems like such a pity. Why not work with the rest of the developers
> > to improve csound for everyone.
>
>Well, a split does not *need* to be a fork. Some projects keep multiple
>dev branches going so that individual developers have a way of doing
>rapid experimentation without worrying too much about breaking code. If
>this were to be a dev experimental branch and not a full fork, I could
>see a properly managed split being a good solution for Istvan. It does
>not mean that work from his branch can't be incorporated back into
>"csound5 stable" when he has figured out how he wants to do things.
>
>However, Istvan, for such a thing to be useful, I believe effort would
>need to be made on both your end and canonical to keep those in sync at
>reasonably regular intervals.
>
>Iain
>--
>Send bugs reports to this list.
>To unsubscribe, send email to csound-unsubscribe@lists.bath.ac.uk

Victor Lazzarini
Music Technology Laboratory
Music Department
National University of Ireland, Maynooth 

>>I am creating an alternate "unofficial" project that can be used for
>>releasing such improvements and new features if there is enough interest.
> 
> 
> That seems like such a pity. Why not work with the rest of the developers
> to improve csound for everyone.

Well, a split does not *need* to be a fork. Some projects keep multiple
dev branches going so that individual developers have a way of doing
rapid experimentation without worrying too much about breaking code. If
this were to be a dev experimental branch and not a full fork, I could
see a properly managed split being a good solution for Istvan. It does
not mean that work from his branch can't be incorporated back into
"csound5 stable" when he has figured out how he wants to do things.

However, Istvan, for such a thing to be useful, I believe effort would
need to be made on both your end and canonical to keep those in sync at
reasonably regular intervals.

Iain