Csound Csound-dev Csound-tekno Search About

Re: PVSfilter bug?

Date2007-10-17 09:04
FromVictor Lazzarini
SubjectRe: PVSfilter bug?
I'll look into it. Do you mean crashing
csound? Does it happen often? Can you give
us a CSD that does not depend on your
soundfiles?

Victor

>
>
> Hi all,
>
> It has just been brought to my attention that pvsfilter
> freezes csound. Running the csd bellow I get a random set
> of numbers printed in the terminal, very strange and I am
> heart broken as this is one of my favorite opcodes ever!
> This is on OS X (intel) by the way.
>
> Thanks
> Peiman
>
> 
>
> 
> sr=44100
> ksmps=10
> nchnls=2
>
> instr 1
> ar1, ar2    diskin
> "/Users/peimankhosravi/Desktop/newsounds_2006/china_impuls
> es3.aif", 1, 0, 1 ain1, ain2    diskin
> "/Users/peimankhosravi/Desktop/newsounds_2006/grains_screa
> m.aif", 1, 0, 1
>
> ftps1  pvsanal   ar1, 1024, 1024/4, 1024*2, 1 ; analyse it
> ftps2  pvsanal   ain1, 1024, 1024/4, 1024*2, 1 ; analyse
> it
>
> fsig1     pvsfilter     ftps2, ftps1, .5, .2
>
> aout2  pvsynth  fsig1
> outs aout2, aout2
> endin
>
> 
>
> 
>
> i1 0 10
>
>
>
> e
>
> 
>
> 
>
>
> This is printed in the terminal:
>
> diskin: opened
> '/Users/peimankhosravi/Desktop/newsounds_2006/china_impuls
> es3.aif':
>         44100 Hz, 2 channel(s), 126389 sample frames
> diskin: opened
> '/Users/peimankhosravi/Desktop/newsounds_2006/grains_screa
> m.aif':
>         44100 Hz, 2 channel(s), 67472 sample frames
> kdepth=0.500000 dirgain=0.000000
> 0: 0.000000; 0.000000 -> 0.000000
> 1: 0.000000; 0.000000 -> 0.000000
> 2: 0.000000; 0.000000 -> 0.000000
> 3: 0.000000; 0.000000 -> 0.000000
> 364: 20.260473; 0.028343 -> 3.704309
> 365: 17.096979; 0.003454 -> 3.083362
> 366: 16.842743; 0.036720 -> 3.093540
> 367: 12.612227; 0.041961 -> 2.323123
> 368: 14.283969; 0.166731 -> 2.809273
> 369: 14.587306; 0.313576 -> 3.083138
> 370: 11.687224; 0.072398 -> 2.188313
> 371: 9.055784; 0.055306 -> 1.680125
> 372: 11.423029; 0.078322 -> 2.145612
> and so on....
> --
> View this message in context:
>
http://www.nabble.com/PVSfilter-bug--tf4637157.html#a13243686
> Sent from the Csound - General mailing list archive at
> Nabble.com.
>
> --
> Send bugs reports to this list.
> To unsubscribe, send email to
> csound-unsubscribe@lists.bath.ac.uk

Date2007-10-17 09:53
Frompeiman
SubjectRe: PVSfilter bug?
Thanks very much,

Bellow is a csd that doesn't depend on any sound files. When I run pvsfilter
in blue it completely crashes, outside blue csound starts the performance
but doesn't seem to end the performance and crashes: I don't get the usual
messages like "Score finished in csoundPerform()" and so on. Instead after
the sound (which is very very glitchy) has stopped all these numbers are
printed for a while (it is this that must be overloading blue maybe). All in
all the effect is very much like a cpu overload (in fact just checking on
the activity monitor, it is a CPU overload!). 

Best
Peiman   
 



sr=44100
ksmps=10
nchnls=2


instr 1

ar1 oscil 5000, 440, 1
ain1 noise 5000, .5

ftps1  pvsanal   ar1, 1024, 1024/4, 1024*2, 1 ; analyse it
ftps2  pvsanal   ain1, 1024, 1024/4, 1024*2, 1 ; analyse it

fsig1 	pvsfilter	 ftps2, ftps1, .5, .2
                
aout2  pvsynth  fsig1
outs aout2, aout2
endin

instr 2

ar1 oscil 5000, 440, 1
ain1 noise 5000, .5

outs ar1, ar1
endin





f 1 0 4096 10 0 1
;i2 0 5
i1 0 5
e





Victor Lazzarini wrote:
> 
> I'll look into it. Do you mean crashing
> csound? Does it happen often? Can you give
> us a CSD that does not depend on your
> soundfiles?
> 
> Victor
> 
>>
>>
>> Hi all,
>>
>> It has just been brought to my attention that pvsfilter
>> freezes csound. Running the csd bellow I get a random set
>> of numbers printed in the terminal, very strange and I am
>> heart broken as this is one of my favorite opcodes ever!
>> This is on OS X (intel) by the way.
>>
>> Thanks
>> Peiman
>>
>> 
>>
>> 
>> sr=44100
>> ksmps=10
>> nchnls=2
>>
>> instr 1
>> ar1, ar2    diskin
>> "/Users/peimankhosravi/Desktop/newsounds_2006/china_impuls
>> es3.aif", 1, 0, 1 ain1, ain2    diskin
>> "/Users/peimankhosravi/Desktop/newsounds_2006/grains_screa
>> m.aif", 1, 0, 1
>>
>> ftps1  pvsanal   ar1, 1024, 1024/4, 1024*2, 1 ; analyse it
>> ftps2  pvsanal   ain1, 1024, 1024/4, 1024*2, 1 ; analyse
>> it
>>
>> fsig1     pvsfilter     ftps2, ftps1, .5, .2
>>
>> aout2  pvsynth  fsig1
>> outs aout2, aout2
>> endin
>>
>> 
>>
>> 
>>
>> i1 0 10
>>
>>
>>
>> e
>>
>> 
>>
>> 
>>
>>
>> This is printed in the terminal:
>>
>> diskin: opened
>> '/Users/peimankhosravi/Desktop/newsounds_2006/china_impuls
>> es3.aif':
>>         44100 Hz, 2 channel(s), 126389 sample frames
>> diskin: opened
>> '/Users/peimankhosravi/Desktop/newsounds_2006/grains_screa
>> m.aif':
>>         44100 Hz, 2 channel(s), 67472 sample frames
>> kdepth=0.500000 dirgain=0.000000
>> 0: 0.000000; 0.000000 -> 0.000000
>> 1: 0.000000; 0.000000 -> 0.000000
>> 2: 0.000000; 0.000000 -> 0.000000
>> 3: 0.000000; 0.000000 -> 0.000000
>> 364: 20.260473; 0.028343 -> 3.704309
>> 365: 17.096979; 0.003454 -> 3.083362
>> 366: 16.842743; 0.036720 -> 3.093540
>> 367: 12.612227; 0.041961 -> 2.323123
>> 368: 14.283969; 0.166731 -> 2.809273
>> 369: 14.587306; 0.313576 -> 3.083138
>> 370: 11.687224; 0.072398 -> 2.188313
>> 371: 9.055784; 0.055306 -> 1.680125
>> 372: 11.423029; 0.078322 -> 2.145612
>> and so on....
>> --
>> View this message in context:
>>
> http://www.nabble.com/PVSfilter-bug--tf4637157.html#a13243686
>> Sent from the Csound - General mailing list archive at
>> Nabble.com.
>>
>> --
>> 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
> 
> 

-- 
View this message in context: http://www.nabble.com/PVSfilter-bug--tf4637157.html#a13249328
Sent from the Csound - General mailing list archive at Nabble.com.

Date2007-10-17 10:26
FromRichard Dobson
SubjectRe: PVSfilter bug?
You are on an intel mac now? Could this be a denormal problem? A 
characteristic of that is that the numbers go haywire and the CPU 
overloads, when audio input falls to zero.

Richard Dobson


peiman wrote:
> Thanks very much,
> 
> Bellow is a csd that doesn't depend on any sound files. When I run pvsfilter
> in blue it completely crashes, outside blue csound starts the performance
> but doesn't seem to end the performance and crashes: I don't get the usual
> messages like "Score finished in csoundPerform()" and so on. Instead after
> the sound (which is very very glitchy) has stopped all these numbers are
> printed for a while (it is this that must be overloading blue maybe). All in
> all the effect is very much like a cpu overload (in fact just checking on
> the activity monitor, it is a CPU overload!). 
>