Csound Csound-dev Csound-tekno Search About

[Cs-dev] cs6 and pconvolve

Date2013-05-02 21:09
FromBen Hackbarth
Subject[Cs-dev] cs6 and pconvolve
AttachmentsNone  None  
hi all,

i think i've found a bug in cs6.  i've had a lot of trouble chasing it down, and so my apologies that i cannot make a more streamlined csd to demonstrate the problem.

http://pastebin.com/x11CXDkg

this csd runs fine in cs5.19.  in csound6, i get samples out of range and i believe that the problem is related to pconvolve.  certain values of ksmps make the problem go away.  and duplicating certain lines of the csd (for example #120) compound the problem.  samples of out range even change if i add extra carriage returns under the orchestra header.

i can post the impulse responses if need be -- clipping seems to be related to the duration of each impulse response.

i run on osx 10.7.5 built from git pulled a couple of days ago.
--  ben

Date2013-05-02 21:19
FromVictor Lazzarini
SubjectRe: [Cs-dev] cs6 and pconvolve
AttachmentsNone  None  
I think the impulse responses would be good.
On 2 May 2013, at 21:09, Ben Hackbarth wrote:

hi all,

i think i've found a bug in cs6.  i've had a lot of trouble chasing it down, and so my apologies that i cannot make a more streamlined csd to demonstrate the problem.

http://pastebin.com/x11CXDkg

this csd runs fine in cs5.19.  in csound6, i get samples out of range and i believe that the problem is related to pconvolve.  certain values of ksmps make the problem go away.  and duplicating certain lines of the csd (for example #120) compound the problem.  samples of out range even change if i add extra carriage returns under the orchestra header.

i can post the impulse responses if need be -- clipping seems to be related to the duration of each impulse response.

i run on osx 10.7.5 built from git pulled a couple of days ago.
--  ben
------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
tel.: +353 1 708 3545
Victor dot Lazzarini AT nuim dot ie




Date2013-05-02 21:27
FromVictor Lazzarini
SubjectRe: [Cs-dev] cs6 and pconvolve
AttachmentsNone  None  
The problem seems to be with the 

iImpulseAtten pow 10, -34/20

line
Victor
On 2 May 2013, at 21:19, Victor Lazzarini wrote:

I think the impulse responses would be good.
On 2 May 2013, at 21:09, Ben Hackbarth wrote:

hi all,

i think i've found a bug in cs6.  i've had a lot of trouble chasing it down, and so my apologies that i cannot make a more streamlined csd to demonstrate the problem.

http://pastebin.com/x11CXDkg

this csd runs fine in cs5.19.  in csound6, i get samples out of range and i believe that the problem is related to pconvolve.  certain values of ksmps make the problem go away.  and duplicating certain lines of the csd (for example #120) compound the problem.  samples of out range even change if i add extra carriage returns under the orchestra header.

i can post the impulse responses if need be -- clipping seems to be related to the duration of each impulse response.

i run on osx 10.7.5 built from git pulled a couple of days ago.
--  ben
------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
tel.: +353 1 708 3545
Victor dot Lazzarini AT nuim dot ie



------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
tel.: +353 1 708 3545
Victor dot Lazzarini AT nuim dot ie




Date2013-05-02 21:30
FromVictor Lazzarini
SubjectRe: [Cs-dev] cs6 and pconvolve
AttachmentsNone  None  
actually there is something funny with the multiplication.

Victor
On 2 May 2013, at 21:19, Victor Lazzarini wrote:

I think the impulse responses would be good.
On 2 May 2013, at 21:09, Ben Hackbarth wrote:

hi all,

i think i've found a bug in cs6.  i've had a lot of trouble chasing it down, and so my apologies that i cannot make a more st reamlined csd to demonstrate the problem.

http://pastebin.com/x11CXDkg

this csd runs fine in cs5.19.  in csound6, i get samples out of range and i believe that the problem is related to pconvolve.  certain values of ksmps make the problem go away.  and duplicating certain lines of the csd (for example #120) compound the problem.  samples of out range even change if i add extra carriage returns under the orchestra header.

i can post the impulse responses if need be -- clipping seems to be related to the duration of each impulse response.

i run on osx 10.7.5 built from git pulled a couple of days ago.
--  ben
------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
tel.: +353 1 708 3545
Victor dot Lazzarini AT nuim dot ie



------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
tel.: +353 1 708 3545
Victor dot Lazzarini AT nuim dot ie




Date2013-05-02 21:36
FromVictor Lazzarini
SubjectRe: [Cs-dev] cs6 and pconvolve
AttachmentsNone  None  
OK, it's something wrong with globals and multiplication. This variation on instr 100 seems to fix it

instr 100 ; EVERYTHING PASSES HERE -- CONVO REVERB ; AUTOMATICALLY ACTIVE
prints "------------------------------\n"
prints "SPATUTIL CONFIG - 2D - 2.0\n"
prints " 2 speakers\n"
prints "..............................\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ".1 2.\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints "..............................\n"
iImpulseAtten pow 10, -34/20
print iImpulseAtten

asig1 = gaWetSpk0
asig2 = gaWetSpk1
aWetSpk0 pconvolve asig1*iImpulseAtten, "/Users/victor/audio/pianoc2.wav", 1024; -90 degrees
aWetSpk1 pconvolve asig2*iImpulseAtten, "/Users/victor/audio/pianoc2.wav", 1024; 90 degrees
aDrySpk0 delay gaDrySpk0, 0.015
aDrySpk1 delay gaDrySpk1, 0.015
clear gaDrySpk0, gaWetSpk0, gaDrySpk1, gaWetSpk1
outs aWetSpk0+aDrySpk0, aWetSpk1+aDrySpk1
endin


On 2 May 2013, at 21:30, Victor Lazzarini wrote:

actually there is something funny with the multiplication.

Victor
On 2 May 2013, at 21:19, Victor Lazzarini wrote:

I think the impulse responses would be good.
On 2 May 2013, at 21:09, Ben Hackbarth wrote:

hi all,

i think i've found a bug in cs6.  i've had a lot of trouble chasing it down, and so my apologies that i cannot make a more streamlined csd to demonstrate the problem.

http://pastebin.com/x11CXDkg

this csd runs fine in cs5.19.  in csound6, i get samples out of range and i believe that the problem is related to pconvolve.  certain values of ksmps make the problem go away.  and duplicating certain lines of the csd (for example #120) compound the problem.  samples of out range even change if i add extra carriage returns under the orchestra header.

i can post the impulse responses if need be -- clipping seems to be related to the duration of each impulse response.

i run on osx 10.7.5 built from git pulled a couple of days ago.
--  ben
------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
tel.: +353 1 708 3545
Victor dot Lazzarini AT nuim dot ie



------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
tel.: +353 1 708 3545
Victor dot Lazzarini AT nuim dot ie



------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
tel.: +353 1 708 3545
Victor dot Lazzarini AT nuim dot ie




Date2013-05-02 21:45
Fromjpff@cs.bath.ac.uk
SubjectRe: [Cs-dev] cs6 and pconvolve
As the opcode does not know the argument is global this must be a parser
or type issue


> OK, it's something wrong with globals and multiplication. This variation
> on instr 100 seems to fix it
>
> instr 100 ; EVERYTHING PASSES HERE -- CONVO REVERB ; AUTOMATICALLY ACTIVE
> prints "------------------------------\n"
> prints "SPATUTIL CONFIG - 2D - 2.0\n"
> prints " 2 speakers\n"
> prints "..............................\n"
> prints ". .\n"
> prints ". .\n"
> prints ". .\n"
> prints ". .\n"
> prints ". .\n"
> prints ". .\n"
> prints ".1 2.\n"
> prints ". .\n"
> prints ". .\n"
> prints ". .\n"
> prints ". .\n"
> prints ". .\n"
> prints ". .\n"
> prints "..............................\n"
> iImpulseAtten pow 10, -34/20
> print iImpulseAtten
> asig1 = gaWetSpk0
> asig2 = gaWetSpk1
> aWetSpk0 pconvolve asig1*iImpulseAtten, "/Users/victor/audio/pianoc2.wav",
> 1024; -90 degrees
> aWetSpk1 pconvolve asig2*iImpulseAtten, "/Users/victor/audio/pianoc2.wav",
> 1024; 90 degrees
> aDrySpk0 delay gaDrySpk0, 0.015
> aDrySpk1 delay gaDrySpk1, 0.015
> clear gaDrySpk0, gaWetSpk0, gaDrySpk1, gaWetSpk1
> outs aWetSpk0+aDrySpk0, aWetSpk1+aDrySpk1
> endin
>
>
> On 2 May 2013, at 21:30, Victor Lazzarini wrote:
>
>> actually there is something funny with the multiplication.
>>
>> Victor
>> On 2 May 2013, at 21:19, Victor Lazzarini wrote:
>>
>>> I think the impulse responses would be good.
>>> On 2 May 2013, at 21:09, Ben Hackbarth wrote:
>>>
>>>> hi all,
>>>>
>>>> i think i've found a bug in cs6.  i've had a lot of trouble chasing it
>>>> down, and so my apologies that i cannot make a more streamlined csd to
>>>> demonstrate the problem.
>>>>
>>>> http://pastebin.com/x11CXDkg
>>>>
>>>> this csd runs fine in cs5.19.  in csound6, i get samples out of range
>>>> and i believe that the problem is related to pconvolve.  certain
>>>> values of ksmps make the problem go away.  and duplicating certain
>>>> lines of the csd (for example #120) compound the problem.  samples of
>>>> out range even change if i add extra carriage returns under the
>>>> orchestra header.
>>>>
>>>> i can post the impulse responses if need be -- clipping seems to be
>>>> related to the duration of each impulse response.
>>>>
>>>> i run on osx 10.7.5 built from git pulled a couple of days ago.
>>>> --  ben
>>>> ------------------------------------------------------------------------------
>>>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>>>> It's a free troubleshooting tool designed for production
>>>> Get down to code-level detail for bottlenecks, with <2% overhead.
>>>> Download for free and get started troubleshooting in minutes.
>>>> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
>>>> Csound-devel mailing list
>>>> Csound-devel@lists.sourceforge.net
>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>
>>> Dr Victor Lazzarini
>>> Senior Lecturer
>>> Dept. of Music
>>> NUI Maynooth Ireland
>>> tel.: +353 1 708 3545
>>> Victor dot Lazzarini AT nuim dot ie
>>>
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>>> It's a free troubleshooting tool designed for production
>>> Get down to code-level detail for bottlenecks, with <2% overhead.
>>> Download for free and get started troubleshooting in minutes.
>>> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
>>> Csound-devel mailing list
>>> Csound-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>
>> Dr Victor Lazzarini
>> Senior Lecturer
>> Dept. of Music
>> NUI Maynooth Ireland
>> tel.: +353 1 708 3545
>> Victor dot Lazzarini AT nuim dot ie
>>
>>
>>
>> ------------------------------------------------------------------------------
>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>> It's a free troubleshooting tool designed for production
>> Get down to code-level detail for bottlenecks, with <2% overhead.
>> Download for free and get started troubleshooting in minutes.
>> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
> Dr Victor Lazzarini
> Senior Lecturer
> Dept. of Music
> NUI Maynooth Ireland
> tel.: +353 1 708 3545
> Victor dot Lazzarini AT nuim dot ie
>
>
>
> ------------------------------------------------------------------------------
> Get 100% visibility into Java/.NET code with AppDynamics Lite
> It's a free troubleshooting tool designed for production
> Get down to code-level detail for bottlenecks, with <2% overhead.
> Download for free and get started troubleshooting in minutes.
> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel
>



------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Date2013-05-02 21:50
FromBen Hackbarth
SubjectRe: [Cs-dev] cs6 and pconvolve
AttachmentsNone  None  

--  ben


On Thu, May 2, 2013 at 10:30 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
actually there is something funny with the multiplication.

Victor

On 2 May 2013, at 21:19, Victor Lazzarini wrote:

I think the impulse responses would be good.
On 2 May 2013, at 21:09, Ben Hackbarth wrote:

hi all,

i think i've found a bug in cs6.  i've had a lot of trouble chasing it down, and so my apologies that i cannot make a more streamlined csd to demonstrate the problem.

http://pastebin.com/x11CXDkg

this csd runs fine in cs5.19.  in csound6, i get samples out of range and i believe that the problem is related to pconvolve.  certain values of ksmps make the problem go away.  and duplicating certain lines of the csd (for example #120) compound the problem.  samples of out range even change if i add extra carriage returns under the orchestra header.

i can post the impulse responses if need be -- clipping seems to be related to the duration of each impulse response.

i run on osx 10.7.5 built from git pulled a couple of days ago.
--  ben
------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
Victor dot Lazzarini AT nuim dot ie



------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
Victor dot Lazzarini AT nuim dot ie




------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel



Date2013-05-02 21:55
FromBen Hackbarth
SubjectRe: [Cs-dev] cs6 and pconvolve
AttachmentsNone  None  
hi victor,

this does not fix the problem on my end. 

--  ben


On Thu, May 2, 2013 at 10:36 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
OK, it's something wrong with globals and multiplication. This variation on instr 100 seems to fix it

instr 100 ; EVERYTHING PASSES HERE -- CONVO REVERB ; AUTOMATICALLY ACTIVE
prints "------------------------------\n"
prints "SPATUTIL CONFIG - 2D - 2.0\n"
prints " 2 speakers\n"
prints "..............................\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ".1 2.\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints "..............................\n"
iImpulseAtten pow 10, -34/20
print iImpulseAtten

asig1 = gaWetSpk0
asig2 = gaWetSpk1
aWetSpk0 pconvolve asig1*iImpulseAtten, "/Users/victor/audio/pianoc2.wav", 1024; -90 degrees
aWetSpk1 pconvolve asig2*iImpulseAtten, "/Users/victor/audio/pianoc2.wav", 1024; 90 degrees
aDrySpk0 delay gaDrySpk0, 0.015
aDrySpk1 delay gaDrySpk1, 0.015
clear gaDrySpk0, gaWetSpk0, gaDrySpk1, gaWetSpk1
outs aWetSpk0+aDrySpk0, aWetSpk1+aDrySpk1
endin


On 2 May 2013, at 21:30, Victor Lazzarini wrote:

actually there is something funny with the multiplication.

Victor
On 2 May 2013, at 21:19, Victor Lazzarini wrote:

I think the impulse responses would be good.
On 2 May 2013, at 21:09, Ben Hackbarth wrote:

hi all,

i think i've found a bug in cs6.  i've had a lot of trouble chasing it down, and so my apologies that i cannot make a more streamlined csd to demonstrate the problem.

http://pastebin.com/x11CXDkg

this csd runs fine in cs5.19.  in csound6, i get samples out of range and i believe that the problem is related to pconvolve.  certain values of ksmps make the problem go away.  and duplicating certain lines of the csd (for example #120) compound the problem.  samples of out range even change if i add extra carriage returns under the orchestra header.

i can post the impulse responses if need be -- clipping seems to be related to the duration of each impulse response.

i run on osx 10.7.5 built from git pulled a couple of days ago.
--  ben
------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
Victor dot Lazzarini AT nuim dot ie



------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
Victor dot Lazzarini AT nuim dot ie



------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
Victor dot Lazzarini AT nuim dot ie




------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel



Date2013-05-02 22:00
FromVictor Lazzarini
SubjectRe: [Cs-dev] cs6 and pconvolve
AttachmentsNone  None  
yes, it's inconsistent. I think this is a memory problem.
On 2 May 2013, at 21:55, Ben Hackbarth wrote:

hi victor,

this does not fix the problem on my end. 

--  ben


On Thu, May 2, 2013 at 10:36 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
OK, it's something wrong with globals and multiplication. This variation on instr 100 seems to fix it

instr 100 ; EVERYTHING PASSES HERE -- CONVO REVERB ; AUTOMATICALLY ACTIVE
prints "------------------------------\n"
prints "SPATUTIL CONFIG - 2D - 2.0\n"
prints " 2 speakers\n"
prints "..............................\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ".1 2.\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints ". .\n"
prints "..............................\n"
iImpulseAtten pow 10, -34/20
print iImpulseAtten

asig1 = gaWetSpk0
asig2 = gaWetSpk1
aWetSpk0 pconvolve asig1*iImpulseAtten, "/Users/victor/audio/pianoc2.wav", 1024; -90 degrees
aWetSpk1 pconvolve asig2*iImpulseAtten, "/Users/victor/audio/pianoc2.wav", 1024; 90 degrees
aDrySpk0 delay gaDrySpk0, 0.015
aDrySpk1 delay gaDrySpk1, 0.015
clear gaDrySpk0, gaWetSpk0, gaDrySpk1, gaWetSpk1
outs aWetSpk0+aDrySpk0, aWetSpk1+aDrySpk1
endin


On 2 May 2013, at 21:30, Victor Lazzarini wrote:

actually there is something funny with the multiplication.

Victor
On 2 May 2013, at 21:19, Victor Lazzarini wrote:

I think the impulse responses would be good.
On 2 May 2013, at 21:09, Ben Hackbarth wrote:

hi all,

i think i've found a bug in cs6.  i've had a lot of trouble chasing it down, and so my apologies that i cannot make a more streamlined csd to demonstrate the problem.

http://pastebin.com/x11CXDkg

this csd runs fine in cs5.19.  in csound6, i get samples out of range and i believe that the problem is related to pconvolve.  certain values of ksmps make the problem go away.  and duplicating certain lines of the csd (for example #120) compound the problem.  samples of out range even change if i add extra carriage returns under the orchestra header.

i can post the impulse responses if need be -- clipping seems to be related to the duration of each impulse response.

i run on osx 10.7.5 built from git pulled a couple of days ago.
--  ben
------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
Victor dot Lazzarini AT nuim dot ie



------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
Victor dot Lazzarini AT nuim dot ie



------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
Victor dot Lazzarini AT nuim dot ie




------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel


------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
tel.: +353 1 708 3545
Victor dot Lazzarini AT nuim dot ie




Date2013-05-02 22:01
FromVictor Lazzarini
SubjectRe: [Cs-dev] cs6 and pconvolve
Looks like a memory issue.
On 2 May 2013, at 21:45, jpff@cs.bath.ac.uk wrote:

> As the opcode does not know the argument is global this must be a parser
> or type issue
> 
> 
>> OK, it's something wrong with globals and multiplication. This variation
>> on instr 100 seems to fix it
>> 
>> instr 100 ; EVERYTHING PASSES HERE -- CONVO REVERB ; AUTOMATICALLY ACTIVE
>> prints "------------------------------\n"
>> prints "SPATUTIL CONFIG - 2D - 2.0\n"
>> prints " 2 speakers\n"
>> prints "..............................\n"
>> prints ". .\n"
>> prints ". .\n"
>> prints ". .\n"
>> prints ". .\n"
>> prints ". .\n"
>> prints ". .\n"
>> prints ".1 2.\n"
>> prints ". .\n"
>> prints ". .\n"
>> prints ". .\n"
>> prints ". .\n"
>> prints ". .\n"
>> prints ". .\n"
>> prints "..............................\n"
>> iImpulseAtten pow 10, -34/20
>> print iImpulseAtten
>> asig1 = gaWetSpk0
>> asig2 = gaWetSpk1
>> aWetSpk0 pconvolve asig1*iImpulseAtten, "/Users/victor/audio/pianoc2.wav",
>> 1024; -90 degrees
>> aWetSpk1 pconvolve asig2*iImpulseAtten, "/Users/victor/audio/pianoc2.wav",
>> 1024; 90 degrees
>> aDrySpk0 delay gaDrySpk0, 0.015
>> aDrySpk1 delay gaDrySpk1, 0.015
>> clear gaDrySpk0, gaWetSpk0, gaDrySpk1, gaWetSpk1
>> outs aWetSpk0+aDrySpk0, aWetSpk1+aDrySpk1
>> endin
>> 
>> 
>> On 2 May 2013, at 21:30, Victor Lazzarini wrote:
>> 
>>> actually there is something funny with the multiplication.
>>> 
>>> Victor
>>> On 2 May 2013, at 21:19, Victor Lazzarini wrote:
>>> 
>>>> I think the impulse responses would be good.
>>>> On 2 May 2013, at 21:09, Ben Hackbarth wrote:
>>>> 
>>>>> hi all,
>>>>> 
>>>>> i think i've found a bug in cs6.  i've had a lot of trouble chasing it
>>>>> down, and so my apologies that i cannot make a more streamlined csd to
>>>>> demonstrate the problem.
>>>>> 
>>>>> http://pastebin.com/x11CXDkg
>>>>> 
>>>>> this csd runs fine in cs5.19.  in csound6, i get samples out of range
>>>>> and i believe that the problem is related to pconvolve.  certain
>>>>> values of ksmps make the problem go away.  and duplicating certain
>>>>> lines of the csd (for example #120) compound the problem.  samples of
>>>>> out range even change if i add extra carriage returns under the
>>>>> orchestra header.
>>>>> 
>>>>> i can post the impulse responses if need be -- clipping seems to be
>>>>> related to the duration of each impulse response.
>>>>> 
>>>>> i run on osx 10.7.5 built from git pulled a couple of days ago.
>>>>> --  ben
>>>>> ------------------------------------------------------------------------------
>>>>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>>>>> It's a free troubleshooting tool designed for production
>>>>> Get down to code-level detail for bottlenecks, with <2% overhead.
>>>>> Download for free and get started troubleshooting in minutes.
>>>>> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
>>>>> Csound-devel mailing list
>>>>> Csound-devel@lists.sourceforge.net
>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>> 
>>>> Dr Victor Lazzarini
>>>> Senior Lecturer
>>>> Dept. of Music
>>>> NUI Maynooth Ireland
>>>> tel.: +353 1 708 3545
>>>> Victor dot Lazzarini AT nuim dot ie
>>>> 
>>>> 
>>>> 
>>>> ------------------------------------------------------------------------------
>>>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>>>> It's a free troubleshooting tool designed for production
>>>> Get down to code-level detail for bottlenecks, with <2% overhead.
>>>> Download for free and get started troubleshooting in minutes.
>>>> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
>>>> Csound-devel mailing list
>>>> Csound-devel@lists.sourceforge.net
>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>> 
>>> Dr Victor Lazzarini
>>> Senior Lecturer
>>> Dept. of Music
>>> NUI Maynooth Ireland
>>> tel.: +353 1 708 3545
>>> Victor dot Lazzarini AT nuim dot ie
>>> 
>>> 
>>> 
>>> ------------------------------------------------------------------------------
>>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>>> It's a free troubleshooting tool designed for production
>>> Get down to code-level detail for bottlenecks, with <2% overhead.
>>> Download for free and get started troubleshooting in minutes.
>>> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
>>> Csound-devel mailing list
>>> Csound-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>> 
>> Dr Victor Lazzarini
>> Senior Lecturer
>> Dept. of Music
>> NUI Maynooth Ireland
>> tel.: +353 1 708 3545
>> Victor dot Lazzarini AT nuim dot ie
>> 
>> 
>> 
>> ------------------------------------------------------------------------------
>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>> It's a free troubleshooting tool designed for production
>> Get down to code-level detail for bottlenecks, with <2% overhead.
>> Download for free and get started troubleshooting in minutes.
>> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>> 
> 
> 
> 
> ------------------------------------------------------------------------------
> Get 100% visibility into Java/.NET code with AppDynamics Lite
> It's a free troubleshooting tool designed for production
> Get down to code-level detail for bottlenecks, with <2% overhead.
> Download for free and get started troubleshooting in minutes.
> http://p.sf.net/sfu/appdyn_d2d_ap2
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
tel.: +353 1 708 3545
Victor dot Lazzarini AT nuim dot ie




------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Date2013-05-02 22:17
FromVictor Lazzarini
SubjectRe: [Cs-dev] cs6 and pconvolve
I think the lines with prints are causing this. I think it has to do with the string constants. 

On 2 May 2013, at 22:01, Victor Lazzarini wrote:

> Looks like a memory issue.
> On 2 May 2013, at 21:45, jpff@cs.bath.ac.uk wrote:
> 
>> As the opcode does not know the argument is global this must be a parser
>> or type issue
>> 
>> 
>>> OK, it's something wrong with globals and multiplication. This variation
>>> on instr 100 seems to fix it
>>> 
>>> instr 100 ; EVERYTHING PASSES HERE -- CONVO REVERB ; AUTOMATICALLY ACTIVE
>>> prints "------------------------------\n"
>>> prints "SPATUTIL CONFIG - 2D - 2.0\n"
>>> prints " 2 speakers\n"
>>> prints "..............................\n"
>>> prints ". .\n"
>>> prints ". .\n"
>>> prints ". .\n"
>>> prints ". .\n"
>>> prints ". .\n"
>>> prints ". .\n"
>>> prints ".1 2.\n"
>>> prints ". .\n"
>>> prints ". .\n"
>>> prints ". .\n"
>>> prints ". .\n"
>>> prints ". .\n"
>>> prints ". .\n"
>>> prints "..............................\n"
>>> iImpulseAtten pow 10, -34/20
>>> print iImpulseAtten
>>> asig1 = gaWetSpk0
>>> asig2 = gaWetSpk1
>>> aWetSpk0 pconvolve asig1*iImpulseAtten, "/Users/victor/audio/pianoc2.wav",
>>> 1024; -90 degrees
>>> aWetSpk1 pconvolve asig2*iImpulseAtten, "/Users/victor/audio/pianoc2.wav",
>>> 1024; 90 degrees
>>> aDrySpk0 delay gaDrySpk0, 0.015
>>> aDrySpk1 delay gaDrySpk1, 0.015
>>> clear gaDrySpk0, gaWetSpk0, gaDrySpk1, gaWetSpk1
>>> outs aWetSpk0+aDrySpk0, aWetSpk1+aDrySpk1
>>> endin
>>> 
>>> 
>>> On 2 May 2013, at 21:30, Victor Lazzarini wrote:
>>> 
>>>> actually there is something funny with the multiplication.
>>>> 
>>>> Victor
>>>> On 2 May 2013, at 21:19, Victor Lazzarini wrote:
>>>> 
>>>>> I think the impulse responses would be good.
>>>>> On 2 May 2013, at 21:09, Ben Hackbarth wrote:
>>>>> 
>>>>>> hi all,
>>>>>> 
>>>>>> i think i've found a bug in cs6.  i've had a lot of trouble chasing it
>>>>>> down, and so my apologies that i cannot make a more streamlined csd to
>>>>>> demonstrate the problem.
>>>>>> 
>>>>>> http://pastebin.com/x11CXDkg
>>>>>> 
>>>>>> this csd runs fine in cs5.19.  in csound6, i get samples out of range
>>>>>> and i believe that the problem is related to pconvolve.  certain
>>>>>> values of ksmps make the problem go away.  and duplicating certain
>>>>>> lines of the csd (for example #120) compound the problem.  samples of
>>>>>> out range even change if i add extra carriage returns under the
>>>>>> orchestra header.
>>>>>> 
>>>>>> i can post the impulse responses if need be -- clipping seems to be
>>>>>> related to the duration of each impulse response.
>>>>>> 
>>>>>> i run on osx 10.7.5 built from git pulled a couple of days ago.
>>>>>> --  ben
>>>>>> ------------------------------------------------------------------------------
>>>>>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>>>>>> It's a free troubleshooting tool designed for production
>>>>>> Get down to code-level detail for bottlenecks, with <2% overhead.
>>>>>> Download for free and get started troubleshooting in minutes.
>>>>>> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
>>>>>> Csound-devel mailing list
>>>>>> Csound-devel@lists.sourceforge.net
>>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>> 
>>>>> Dr Victor Lazzarini
>>>>> Senior Lecturer
>>>>> Dept. of Music
>>>>> NUI Maynooth Ireland
>>>>> tel.: +353 1 708 3545
>>>>> Victor dot Lazzarini AT nuim dot ie
>>>>> 
>>>>> 
>>>>> 
>>>>> ------------------------------------------------------------------------------
>>>>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>>>>> It's a free troubleshooting tool designed for production
>>>>> Get down to code-level detail for bottlenecks, with <2% overhead.
>>>>> Download for free and get started troubleshooting in minutes.
>>>>> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
>>>>> Csound-devel mailing list
>>>>> Csound-devel@lists.sourceforge.net
>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>> 
>>>> Dr Victor Lazzarini
>>>> Senior Lecturer
>>>> Dept. of Music
>>>> NUI Maynooth Ireland
>>>> tel.: +353 1 708 3545
>>>> Victor dot Lazzarini AT nuim dot ie
>>>> 
>>>> 
>>>> 
>>>> ------------------------------------------------------------------------------
>>>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>>>> It's a free troubleshooting tool designed for production
>>>> Get down to code-level detail for bottlenecks, with <2% overhead.
>>>> Download for free and get started troubleshooting in minutes.
>>>> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
>>>> Csound-devel mailing list
>>>> Csound-devel@lists.sourceforge.net
>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>> 
>>> Dr Victor Lazzarini
>>> Senior Lecturer
>>> Dept. of Music
>>> NUI Maynooth Ireland
>>> tel.: +353 1 708 3545
>>> Victor dot Lazzarini AT nuim dot ie
>>> 
>>> 
>>> 
>>> ------------------------------------------------------------------------------
>>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>>> It's a free troubleshooting tool designed for production
>>> Get down to code-level detail for bottlenecks, with <2% overhead.
>>> Download for free and get started troubleshooting in minutes.
>>> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
>>> Csound-devel mailing list
>>> Csound-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>> 
>> 
>> 
>> 
>> ------------------------------------------------------------------------------
>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>> It's a free troubleshooting tool designed for production
>> Get down to code-level detail for bottlenecks, with <2% overhead.
>> Download for free and get started troubleshooting in minutes.
>> http://p.sf.net/sfu/appdyn_d2d_ap2
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
> 
> Dr Victor Lazzarini
> Senior Lecturer
> Dept. of Music
> NUI Maynooth Ireland
> tel.: +353 1 708 3545
> Victor dot Lazzarini AT nuim dot ie
> 
> 
> 
> 
> ------------------------------------------------------------------------------
> Get 100% visibility into Java/.NET code with AppDynamics Lite
> It's a free troubleshooting tool designed for production
> Get down to code-level detail for bottlenecks, with <2% overhead.
> Download for free and get started troubleshooting in minutes.
> http://p.sf.net/sfu/appdyn_d2d_ap2
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
tel.: +353 1 708 3545
Victor dot Lazzarini AT nuim dot ie




------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Date2013-05-03 07:28
FromVictor Lazzarini
SubjectRe: [Cs-dev] cs6 and pconvolve
removing the prints lines seems to fix this, so it points to some memory issues with strings, but I could not locate where they are
with valgrind.

Maybe it's one for John or Steven?

On 2 May 2013, at 22:17, Victor Lazzarini wrote:

> I think the lines with prints are causing this. I think it has to do with the string constants. 
> 
> On 2 May 2013, at 22:01, Victor Lazzarini wrote:
> 
>> Looks like a memory issue.
>> On 2 May 2013, at 21:45, jpff@cs.bath.ac.uk wrote:
>> 
>>> As the opcode does not know the argument is global this must be a parser
>>> or type issue
>>> 
>>> 
>>>> OK, it's something wrong with globals and multiplication. This variation
>>>> on instr 100 seems to fix it
>>>> 
>>>> instr 100 ; EVERYTHING PASSES HERE -- CONVO REVERB ; AUTOMATICALLY ACTIVE
>>>> prints "------------------------------\n"
>>>> prints "SPATUTIL CONFIG - 2D - 2.0\n"
>>>> prints " 2 speakers\n"
>>>> prints "..............................\n"
>>>> prints ". .\n"
>>>> prints ". .\n"
>>>> prints ". .\n"
>>>> prints ". .\n"
>>>> prints ". .\n"
>>>> prints ". .\n"
>>>> prints ".1 2.\n"
>>>> prints ". .\n"
>>>> prints ". .\n"
>>>> prints ". .\n"
>>>> prints ". .\n"
>>>> prints ". .\n"
>>>> prints ". .\n"
>>>> prints "..............................\n"
>>>> iImpulseAtten pow 10, -34/20
>>>> print iImpulseAtten
>>>> asig1 = gaWetSpk0
>>>> asig2 = gaWetSpk1
>>>> aWetSpk0 pconvolve asig1*iImpulseAtten, "/Users/victor/audio/pianoc2.wav",
>>>> 1024; -90 degrees
>>>> aWetSpk1 pconvolve asig2*iImpulseAtten, "/Users/victor/audio/pianoc2.wav",
>>>> 1024; 90 degrees
>>>> aDrySpk0 delay gaDrySpk0, 0.015
>>>> aDrySpk1 delay gaDrySpk1, 0.015
>>>> clear gaDrySpk0, gaWetSpk0, gaDrySpk1, gaWetSpk1
>>>> outs aWetSpk0+aDrySpk0, aWetSpk1+aDrySpk1
>>>> endin
>>>> 
>>>> 
>>>> On 2 May 2013, at 21:30, Victor Lazzarini wrote:
>>>> 
>>>>> actually there is something funny with the multiplication.
>>>>> 
>>>>> Victor
>>>>> On 2 May 2013, at 21:19, Victor Lazzarini wrote:
>>>>> 
>>>>>> I think the impulse responses would be good.
>>>>>> On 2 May 2013, at 21:09, Ben Hackbarth wrote:
>>>>>> 
>>>>>>> hi all,
>>>>>>> 
>>>>>>> i think i've found a bug in cs6.  i've had a lot of trouble chasing it
>>>>>>> down, and so my apologies that i cannot make a more streamlined csd to
>>>>>>> demonstrate the problem.
>>>>>>> 
>>>>>>> http://pastebin.com/x11CXDkg
>>>>>>> 
>>>>>>> this csd runs fine in cs5.19.  in csound6, i get samples out of range
>>>>>>> and i believe that the problem is related to pconvolve.  certain
>>>>>>> values of ksmps make the problem go away.  and duplicating certain
>>>>>>> lines of the csd (for example #120) compound the problem.  samples of
>>>>>>> out range even change if i add extra carriage returns under the
>>>>>>> orchestra header.
>>>>>>> 
>>>>>>> i can post the impulse responses if need be -- clipping seems to be
>>>>>>> related to the duration of each impulse response.
>>>>>>> 
>>>>>>> i run on osx 10.7.5 built from git pulled a couple of days ago.
>>>>>>> --  ben
>>>>>>> ------------------------------------------------------------------------------
>>>>>>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>>>>>>> It's a free troubleshooting tool designed for production
>>>>>>> Get down to code-level detail for bottlenecks, with <2% overhead.
>>>>>>> Download for free and get started troubleshooting in minutes.
>>>>>>> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
>>>>>>> Csound-devel mailing list
>>>>>>> Csound-devel@lists.sourceforge.net
>>>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>>> 
>>>>>> Dr Victor Lazzarini
>>>>>> Senior Lecturer
>>>>>> Dept. of Music
>>>>>> NUI Maynooth Ireland
>>>>>> tel.: +353 1 708 3545
>>>>>> Victor dot Lazzarini AT nuim dot ie
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>> ------------------------------------------------------------------------------
>>>>>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>>>>>> It's a free troubleshooting tool designed for production
>>>>>> Get down to code-level detail for bottlenecks, with <2% overhead.
>>>>>> Download for free and get started troubleshooting in minutes.
>>>>>> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
>>>>>> Csound-devel mailing list
>>>>>> Csound-devel@lists.sourceforge.net
>>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>> 
>>>>> Dr Victor Lazzarini
>>>>> Senior Lecturer
>>>>> Dept. of Music
>>>>> NUI Maynooth Ireland
>>>>> tel.: +353 1 708 3545
>>>>> Victor dot Lazzarini AT nuim dot ie
>>>>> 
>>>>> 
>>>>> 
>>>>> ------------------------------------------------------------------------------
>>>>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>>>>> It's a free troubleshooting tool designed for production
>>>>> Get down to code-level detail for bottlenecks, with <2% overhead.
>>>>> Download for free and get started troubleshooting in minutes.
>>>>> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
>>>>> Csound-devel mailing list
>>>>> Csound-devel@lists.sourceforge.net
>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>> 
>>>> Dr Victor Lazzarini
>>>> Senior Lecturer
>>>> Dept. of Music
>>>> NUI Maynooth Ireland
>>>> tel.: +353 1 708 3545
>>>> Victor dot Lazzarini AT nuim dot ie
>>>> 
>>>> 
>>>> 
>>>> ------------------------------------------------------------------------------
>>>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>>>> It's a free troubleshooting tool designed for production
>>>> Get down to code-level detail for bottlenecks, with <2% overhead.
>>>> Download for free and get started troubleshooting in minutes.
>>>> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
>>>> Csound-devel mailing list
>>>> Csound-devel@lists.sourceforge.net
>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>> 
>>> 
>>> 
>>> 
>>> ------------------------------------------------------------------------------
>>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>>> It's a free troubleshooting tool designed for production
>>> Get down to code-level detail for bottlenecks, with <2% overhead.
>>> Download for free and get started troubleshooting in minutes.
>>> http://p.sf.net/sfu/appdyn_d2d_ap2
>>> _______________________________________________
>>> Csound-devel mailing list
>>> Csound-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>> 
>> Dr Victor Lazzarini
>> Senior Lecturer
>> Dept. of Music
>> NUI Maynooth Ireland
>> tel.: +353 1 708 3545
>> Victor dot Lazzarini AT nuim dot ie
>> 
>> 
>> 
>> 
>> ------------------------------------------------------------------------------
>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>> It's a free troubleshooting tool designed for production
>> Get down to code-level detail for bottlenecks, with <2% overhead.
>> Download for free and get started troubleshooting in minutes.
>> http://p.sf.net/sfu/appdyn_d2d_ap2
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
> 
> Dr Victor Lazzarini
> Senior Lecturer
> Dept. of Music
> NUI Maynooth Ireland
> tel.: +353 1 708 3545
> Victor dot Lazzarini AT nuim dot ie
> 
> 
> 
> 
> ------------------------------------------------------------------------------
> Get 100% visibility into Java/.NET code with AppDynamics Lite
> It's a free troubleshooting tool designed for production
> Get down to code-level detail for bottlenecks, with <2% overhead.
> Download for free and get started troubleshooting in minutes.
> http://p.sf.net/sfu/appdyn_d2d_ap2
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
tel.: +353 1 708 3545
Victor dot Lazzarini AT nuim dot ie




------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Date2013-05-03 09:01
FromVictor Lazzarini
SubjectRe: [Cs-dev] cs6 and pconvolve
No, that appears not to be the case too. But it definitely looks like a memory issue.
On 3 May 2013, at 07:28, Victor Lazzarini wrote:

> removing the prints lines seems to fix this, so it points to some memory issues with strings, but I could not locate where they are
> with valgrind.
> 
> Maybe it's one for John or Steven?
> 
> On 2 May 2013, at 22:17, Victor Lazzarini wrote:
> 
>> I think the lines with prints are causing this. I think it has to do with the string constants. 
>> 
>> On 2 May 2013, at 22:01, Victor Lazzarini wrote:
>> 
>>> Looks like a memory issue.
>>> On 2 May 2013, at 21:45, jpff@cs.bath.ac.uk wrote:
>>> 
>>>> As the opcode does not know the argument is global this must be a parser
>>>> or type issue
>>>> 
>>>> 
>>>>> OK, it's something wrong with globals and multiplication. This variation
>>>>> on instr 100 seems to fix it
>>>>> 
>>>>> instr 100 ; EVERYTHING PASSES HERE -- CONVO REVERB ; AUTOMATICALLY ACTIVE
>>>>> prints "------------------------------\n"
>>>>> prints "SPATUTIL CONFIG - 2D - 2.0\n"
>>>>> prints " 2 speakers\n"
>>>>> prints "..............................\n"
>>>>> prints ". .\n"
>>>>> prints ". .\n"
>>>>> prints ". .\n"
>>>>> prints ". .\n"
>>>>> prints ". .\n"
>>>>> prints ". .\n"
>>>>> prints ".1 2.\n"
>>>>> prints ". .\n"
>>>>> prints ". .\n"
>>>>> prints ". .\n"
>>>>> prints ". .\n"
>>>>> prints ". .\n"
>>>>> prints ". .\n"
>>>>> prints "..............................\n"
>>>>> iImpulseAtten pow 10, -34/20
>>>>> print iImpulseAtten
>>>>> asig1 = gaWetSpk0
>>>>> asig2 = gaWetSpk1
>>>>> aWetSpk0 pconvolve asig1*iImpulseAtten, "/Users/victor/audio/pianoc2.wav",
>>>>> 1024; -90 degrees
>>>>> aWetSpk1 pconvolve asig2*iImpulseAtten, "/Users/victor/audio/pianoc2.wav",
>>>>> 1024; 90 degrees
>>>>> aDrySpk0 delay gaDrySpk0, 0.015
>>>>> aDrySpk1 delay gaDrySpk1, 0.015
>>>>> clear gaDrySpk0, gaWetSpk0, gaDrySpk1, gaWetSpk1
>>>>> outs aWetSpk0+aDrySpk0, aWetSpk1+aDrySpk1
>>>>> endin
>>>>> 
>>>>> 
>>>>> On 2 May 2013, at 21:30, Victor Lazzarini wrote:
>>>>> 
>>>>>> actually there is something funny with the multiplication.
>>>>>> 
>>>>>> Victor
>>>>>> On 2 May 2013, at 21:19, Victor Lazzarini wrote:
>>>>>> 
>>>>>>> I think the impulse responses would be good.
>>>>>>> On 2 May 2013, at 21:09, Ben Hackbarth wrote:
>>>>>>> 
>>>>>>>> hi all,
>>>>>>>> 
>>>>>>>> i think i've found a bug in cs6.  i've had a lot of trouble chasing it
>>>>>>>> down, and so my apologies that i cannot make a more streamlined csd to
>>>>>>>> demonstrate the problem.
>>>>>>>> 
>>>>>>>> http://pastebin.com/x11CXDkg
>>>>>>>> 
>>>>>>>> this csd runs fine in cs5.19.  in csound6, i get samples out of range
>>>>>>>> and i believe that the problem is related to pconvolve.  certain
>>>>>>>> values of ksmps make the problem go away.  and duplicating certain
>>>>>>>> lines of the csd (for example #120) compound the problem.  samples of
>>>>>>>> out range even change if i add extra carriage returns under the
>>>>>>>> orchestra header.
>>>>>>>> 
>>>>>>>> i can post the impulse responses if need be -- clipping seems to be
>>>>>>>> related to the duration of each impulse response.
>>>>>>>> 
>>>>>>>> i run on osx 10.7.5 built from git pulled a couple of days ago.
>>>>>>>> --  ben
>>>>>>>> ------------------------------------------------------------------------------
>>>>>>>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>>>>>>>> It's a free troubleshooting tool designed for production
>>>>>>>> Get down to code-level detail for bottlenecks, with <2% overhead.
>>>>>>>> Download for free and get started troubleshooting in minutes.
>>>>>>>> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
>>>>>>>> Csound-devel mailing list
>>>>>>>> Csound-devel@lists.sourceforge.net
>>>>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>>>> 
>>>>>>> Dr Victor Lazzarini
>>>>>>> Senior Lecturer
>>>>>>> Dept. of Music
>>>>>>> NUI Maynooth Ireland
>>>>>>> tel.: +353 1 708 3545
>>>>>>> Victor dot Lazzarini AT nuim dot ie
>>>>>>> 
>>>>>>> 
>>>>>>> 
>>>>>>> ------------------------------------------------------------------------------
>>>>>>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>>>>>>> It's a free troubleshooting tool designed for production
>>>>>>> Get down to code-level detail for bottlenecks, with <2% overhead.
>>>>>>> Download for free and get started troubleshooting in minutes.
>>>>>>> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
>>>>>>> Csound-devel mailing list
>>>>>>> Csound-devel@lists.sourceforge.net
>>>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>>> 
>>>>>> Dr Victor Lazzarini
>>>>>> Senior Lecturer
>>>>>> Dept. of Music
>>>>>> NUI Maynooth Ireland
>>>>>> tel.: +353 1 708 3545
>>>>>> Victor dot Lazzarini AT nuim dot ie
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>> ------------------------------------------------------------------------------
>>>>>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>>>>>> It's a free troubleshooting tool designed for production
>>>>>> Get down to code-level detail for bottlenecks, with <2% overhead.
>>>>>> Download for free and get started troubleshooting in minutes.
>>>>>> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
>>>>>> Csound-devel mailing list
>>>>>> Csound-devel@lists.sourceforge.net
>>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>> 
>>>>> Dr Victor Lazzarini
>>>>> Senior Lecturer
>>>>> Dept. of Music
>>>>> NUI Maynooth Ireland
>>>>> tel.: +353 1 708 3545
>>>>> Victor dot Lazzarini AT nuim dot ie
>>>>> 
>>>>> 
>>>>> 
>>>>> ------------------------------------------------------------------------------
>>>>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>>>>> It's a free troubleshooting tool designed for production
>>>>> Get down to code-level detail for bottlenecks, with <2% overhead.
>>>>> Download for free and get started troubleshooting in minutes.
>>>>> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
>>>>> Csound-devel mailing list
>>>>> Csound-devel@lists.sourceforge.net
>>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>>>> 
>>>> 
>>>> 
>>>> 
>>>> ------------------------------------------------------------------------------
>>>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>>>> It's a free troubleshooting tool designed for production
>>>> Get down to code-level detail for bottlenecks, with <2% overhead.
>>>> Download for free and get started troubleshooting in minutes.
>>>> http://p.sf.net/sfu/appdyn_d2d_ap2
>>>> _______________________________________________
>>>> Csound-devel mailing list
>>>> Csound-devel@lists.sourceforge.net
>>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>> 
>>> Dr Victor Lazzarini
>>> Senior Lecturer
>>> Dept. of Music
>>> NUI Maynooth Ireland
>>> tel.: +353 1 708 3545
>>> Victor dot Lazzarini AT nuim dot ie
>>> 
>>> 
>>> 
>>> 
>>> ------------------------------------------------------------------------------
>>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>>> It's a free troubleshooting tool designed for production
>>> Get down to code-level detail for bottlenecks, with <2% overhead.
>>> Download for free and get started troubleshooting in minutes.
>>> http://p.sf.net/sfu/appdyn_d2d_ap2
>>> _______________________________________________
>>> Csound-devel mailing list
>>> Csound-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>> 
>> Dr Victor Lazzarini
>> Senior Lecturer
>> Dept. of Music
>> NUI Maynooth Ireland
>> tel.: +353 1 708 3545
>> Victor dot Lazzarini AT nuim dot ie
>> 
>> 
>> 
>> 
>> ------------------------------------------------------------------------------
>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>> It's a free troubleshooting tool designed for production
>> Get down to code-level detail for bottlenecks, with <2% overhead.
>> Download for free and get started troubleshooting in minutes.
>> http://p.sf.net/sfu/appdyn_d2d_ap2
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
> 
> Dr Victor Lazzarini
> Senior Lecturer
> Dept. of Music
> NUI Maynooth Ireland
> tel.: +353 1 708 3545
> Victor dot Lazzarini AT nuim dot ie
> 
> 
> 
> 
> ------------------------------------------------------------------------------
> Get 100% visibility into Java/.NET code with AppDynamics Lite
> It's a free troubleshooting tool designed for production
> Get down to code-level detail for bottlenecks, with <2% overhead.
> Download for free and get started troubleshooting in minutes.
> http://p.sf.net/sfu/appdyn_d2d_ap2
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
tel.: +353 1 708 3545
Victor dot Lazzarini AT nuim dot ie




------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Date2013-05-03 09:05
FromVictor Lazzarini
SubjectRe: [Cs-dev] cs6 and pconvolve
What I noticed is this: 1) commenting out the lines with prints does not stop the issue; 2) removing them seems to stop it.
So I am wondering whether this is some memory issue in the parser or corefil  functions.

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
tel.: +353 1 708 3545
Victor dot Lazzarini AT nuim dot ie




------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Date2013-05-03 10:03
FromBen Hackbarth
SubjectRe: [Cs-dev] cs6 and pconvolve
AttachmentsNone  None  
the issues also seems to stop when i remove any group of lines from the orchestra.  for instance, everything works if i remove an unused udo, such as "aStereoHRTFFilter".

--  ben


On Fri, May 3, 2013 at 10:05 AM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
What I noticed is this: 1) commenting out the lines with prints does not stop the issue; 2) removing them seems to stop it.
So I am wondering whether this is some memory issue in the parser or corefil  functions.

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
tel.: +353 1 708 3545
Victor dot Lazzarini AT nuim dot ie




------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel


Date2013-05-03 11:00
Fromjpff@cs.bath.ac.uk
SubjectRe: [Cs-dev] cs6 and pconvolve
Your original examole runs ok for me when I sunstitute for your aiff
files.  This is on Linux 64bit/doubles so it suggests an OSX problem.

==John ff

> hi all,
>
> i think i've found a bug in cs6.  i've had a lot of trouble chasing it
> down, and so my apologies that i cannot make a more streamlined csd to
> demonstrate the problem.
>
> http://pastebin.com/x11CXDkg
>
> this csd runs fine in cs5.19.  in csound6, i get samples out of range and
> i
> believe that the problem is related to pconvolve.  certain values of ksmps
> make the problem go away.  and duplicating certain lines of the csd (for
> example #120) compound the problem.  samples of out range even change if i
> add extra carriage returns under the orchestra header.
>
> i can post the impulse responses if need be -- clipping seems to be
> related
> to the duration of each impulse response.
>
> i run on osx 10.7.5 built from git pulled a couple of days ago.
> --  ben
> ------------------------------------------------------------------------------
> Get 100% visibility into Java/.NET code with AppDynamics Lite
> It's a free troubleshooting tool designed for production
> Get down to code-level detail for bottlenecks, with <2% overhead.
> Download for free and get started troubleshooting in minutes.
> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel
>



------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Date2013-05-03 11:33
FromVictor Lazzarini
SubjectRe: [Cs-dev] cs6 and pconvolve
But it does not happen all the time. You have to run it repeatedly to get an error.

Victor
On 3 May 2013, at 11:00, jpff@cs.bath.ac.uk wrote:

> Your original examole runs ok for me when I sunstitute for your aiff
> files.  This is on Linux 64bit/doubles so it suggests an OSX problem.
> 
> ==John ff
> 
>> hi all,
>> 
>> i think i've found a bug in cs6.  i've had a lot of trouble chasing it
>> down, and so my apologies that i cannot make a more streamlined csd to
>> demonstrate the problem.
>> 
>> http://pastebin.com/x11CXDkg
>> 
>> this csd runs fine in cs5.19.  in csound6, i get samples out of range and
>> i
>> believe that the problem is related to pconvolve.  certain values of ksmps
>> make the problem go away.  and duplicating certain lines of the csd (for
>> example #120) compound the problem.  samples of out range even change if i
>> add extra carriage returns under the orchestra header.
>> 
>> i can post the impulse responses if need be -- clipping seems to be
>> related
>> to the duration of each impulse response.
>> 
>> i run on osx 10.7.5 built from git pulled a couple of days ago.
>> --  ben
>> ------------------------------------------------------------------------------
>> Get 100% visibility into Java/.NET code with AppDynamics Lite
>> It's a free troubleshooting tool designed for production
>> Get down to code-level detail for bottlenecks, with <2% overhead.
>> Download for free and get started troubleshooting in minutes.
>> http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>> 
> 
> 
> 
> ------------------------------------------------------------------------------
> Get 100% visibility into Java/.NET code with AppDynamics Lite
> It's a free troubleshooting tool designed for production
> Get down to code-level detail for bottlenecks, with <2% overhead.
> Download for free and get started troubleshooting in minutes.
> http://p.sf.net/sfu/appdyn_d2d_ap2
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
tel.: +353 1 708 3545
Victor dot Lazzarini AT nuim dot ie




------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Date2013-05-03 12:14
Fromjpff@cs.bath.ac.uk
SubjectRe: [Cs-dev] cs6 and pconvolve
> But it does not happen all the time. You have to run it repeatedly to get
> an error.
>
> Victor

about 15 times so far without error



------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Date2013-05-03 16:52
FromVictor Lazzarini
SubjectRe: [Cs-dev] cs6 and pconvolve
But valgrind is giving you errors in Linux too, is it not? So the problem is there, it just did not manifest itself in the same way.
On 3 May 2013, at 12:14, jpff@cs.bath.ac.uk wrote:

>> But it does not happen all the time. You have to run it repeatedly to get
>> an error.
>> 
>> Victor
> 
> about 15 times so far without error
> 
> 
> 
> ------------------------------------------------------------------------------
> Get 100% visibility into Java/.NET code with AppDynamics Lite
> It's a free troubleshooting tool designed for production
> Get down to code-level detail for bottlenecks, with <2% overhead.
> Download for free and get started troubleshooting in minutes.
> http://p.sf.net/sfu/appdyn_d2d_ap2
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
tel.: +353 1 708 3545
Victor dot Lazzarini AT nuim dot ie




------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Date2013-05-03 21:25
FromVictor Lazzarini
SubjectRe: [Cs-dev] cs6 and pconvolve
AttachmentsNone  None  
Hi Ben,

I think this is fixed now. It was a hard one to track down (it made John and I together scratch our heads for a good long time today).
Thanks for finding, it was an important one to get right.

Regards

Victor

On 2 May 2013, at 21:09, Ben Hackbarth wrote:

hi all,

i think i've found a bug in cs6.  i've had a lot of trouble chasing it down, and so my apologies that i cannot make a more streamlined csd to demonstrate the problem.

http://pastebin.com/x11CXDkg

this csd runs fine in cs5.19.  in csound6, i get samples out of range and i believe that the problem is related to pconvolve.  certain values of ksmps make the problem go away.  and duplicating certain lines of the csd (for example #120) compound the problem.  samples of out range even change if i add extra carriage returns under the orchestra header.

i can post the impulse responses if need be -- clipping seems to be related to the duration of each impulse response.

i run on osx 10.7.5 built from git pulled a couple of days ago.
--  ben
------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
tel.: +353 1 708 3545
Victor dot Lazzarini AT nuim dot ie




Date2013-05-04 11:33
FromBen Hackbarth
SubjectRe: [Cs-dev] cs6 and pconvolve
AttachmentsNone  None  
just pulled from git and i can confirm that, for this example, cs6 now behaves identically to cs5. 


thanks to you both for your help !

--  ben


On Fri, May 3, 2013 at 10:25 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
Hi Ben,

I think this is fixed now. It was a hard one to track down (it made John and I together scratch our heads for a good long time today).
Thanks for finding, it was an important one to get right.

Regards

Victor

On 2 May 2013, at 21:09, Ben Hackbarth wrote:

hi all,

i think i've found a bug in cs6.  i've had a lot of trouble chasing it down, and so my apologies that i cannot make a more streamlined csd to demonstrate the problem.

http://pastebin.com/x11CXDkg

this csd runs fine in cs5.19.  in csound6, i get samples out of range and i believe that the problem is related to pconvolve.  certain values of ksmps make the problem go away.  and duplicating certain lines of the csd (for example #120) compound the problem.  samples of out range even change if i add extra carriage returns under the orchestra header.

i can post the impulse responses if need be -- clipping seems to be related to the duration of each impulse response.

i run on osx 10.7.5 built from git pulled a couple of days ago.
--  ben
------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
Victor dot Lazzarini AT nuim dot ie




------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel



Date2013-05-04 12:55
FromVictor Lazzarini
SubjectRe: [Cs-dev] cs6 and pconvolve
AttachmentsNone  None  
Thanks for confirming this, it's a great relief to us to hear ;)
On 4 May 2013, at 11:33, Ben Hackbarth wrote:

just pulled from git and i can confirm that, for this example, cs6 now behaves identically to cs5. 


thanks to you both for your help !

--  ben


On Fri, May 3, 2013 at 10:25 PM, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
Hi Ben,

I think this is fixed now. It was a hard one to track down (it made John and I together scratch our heads for a good long time today).
Thanks for finding, it was an important one to get right.

Regards

Victor

On 2 May 2013, at 21:09, Ben Hackbarth wrote:

hi all,

i think i've found a bug in cs6.  i've had a lot of trouble chasing it down, and so my apologies that i cannot make a more streamlined csd to demonstrate the problem.

http://pastebin.com/x11CXDkg

this csd runs fine in cs5.19.  in csound6, i get samples out of range and i believe that the problem is related to pconvolve.  certain values of ksmps make the problem go away.  and duplicating certain lines of the csd (for example #120) compound the problem.  samples of out range even change if i add extra carriage returns under the orchestra header.

i can post the impulse responses if need be -- clipping seems to be related to the duration of each impulse response.

i run on osx 10.7.5 built from git pulled a couple of days ago.
--  ben
------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
Victor dot Lazzarini AT nuim dot ie




------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel


------------------------------------------------------------------------------
Get 100% visibility into Java/.NET code with AppDynamics Lite
It's a free troubleshooting tool designed for production
Get down to code-level detail for bottlenecks, with <2% overhead.
Download for free and get started troubleshooting in minutes.
http://p.sf.net/sfu/appdyn_d2d_ap2_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-devel

Dr Victor Lazzarini
Senior Lecturer
Dept. of Music
NUI Maynooth Ireland
tel.: +353 1 708 3545
Victor dot Lazzarini AT nuim dot ie