[Csnd] pvsbuffer issue

[Csnd] pvsbuffer issue

Date	2011-07-07 15:35
From	peiman khosravi
Subject	[Csnd] pvsbuffer issue
	Hello, As the csd below demonstrates there seems to be something fishy going on with pvsbuffer. It sounds as if some bins are zeroed out. The result is drastically different from a straight resynthesis of the fsig. Am I doing something wrong? Thanks Peiman -odac -d sr = 96000 ksmps = 100 nchnls = 2 0dbfs = 1 instr 1 asig1 noise .2, 0 fsig1 pvsanal asig1,1024,256,1024,1 ibuf1,kt1 pvsbuffer fsig1, 10 ; 10-sec buf with fsig1 khan = ibuf1 fsb pvsbufread kt1, khan, 0, sr/2 ; read buffer ar1 pvsynth fsb ar2 pvsynth fsig1 outs ar1, ar2 endin i1 0 10 Send bugs reports to the Sourceforge bug tracker https://sourceforge.net/tracker/?group_id=81968&atid=564599 Discussions of bugs and features can be posted here To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Date	2011-07-07 15:40
From	peiman khosravi
Subject	[Csnd] Re: pvsbuffer issue
	It turns out if ksmps=1 then it sounds fine. Should I file it as a bug? That's a real pity, I had great plans for that opcode - my Sunday roast is ruined now :-) Best, Peiman On 07/07/2011 15:35, peiman khosravi wrote: > Hello, > > As the csd below demonstrates there seems to be something fishy going > on with pvsbuffer. It sounds as if some bins are zeroed out. The > result is drastically different from a straight resynthesis of the fsig. > > Am I doing something wrong? > > Thanks > > Peiman > > > > > > -odac -d > > > > > sr = 96000 > ksmps = 100 > nchnls = 2 > 0dbfs = 1 > > instr 1 > > asig1 noise .2, 0 > > fsig1 pvsanal asig1,1024,256,1024,1 > > > ibuf1,kt1 pvsbuffer fsig1, 10 ; 10-sec buf with fsig1 > > khan = ibuf1 > > fsb pvsbufread kt1, khan, 0, sr/2 ; read buffer > > ar1 pvsynth fsb > ar2 pvsynth fsig1 > > outs ar1, ar2 > endin > > > > > > i1 0 10 > > > > Send bugs reports to the Sourceforge bug tracker https://sourceforge.net/tracker/?group_id=81968&atid=564599 Discussions of bugs and features can be posted here To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Date	2011-07-07 15:49
From	Steven Yi
Subject	Re: [Csnd] Re: pvsbuffer issue
	Does this work if you wrap it in a UDO and use setksmps 1? On Thu, Jul 7, 2011 at 10:40 AM, peiman khosravi wrote: > It turns out if ksmps=1 then it sounds fine. Should I file it as a bug? > That's a real pity, I had great plans for that opcode - my Sunday roast is > ruined now :-) > > Best, > Peiman > > On 07/07/2011 15:35, peiman khosravi wrote: >> >> Hello, >> >> As the csd below demonstrates there seems to be something fishy going on >> with pvsbuffer. It sounds as if some bins are zeroed out. The result is >> drastically different from a straight resynthesis of the fsig. >> >> Am I doing something wrong? >> >> Thanks >> >> Peiman >> >> >> >> >> >> -odac -d >> >> >> >> >> sr = 96000 >> ksmps = 100 >> nchnls = 2 >> 0dbfs = 1 >> >> instr 1 >> >> asig1 noise .2, 0 >> >> fsig1 pvsanal asig1,1024,256,1024,1 >> >> >> ibuf1,kt1 pvsbuffer fsig1, 10 ; 10-sec buf with fsig1 >> >> khan = ibuf1 >> >> fsb pvsbufread kt1, khan, 0, sr/2 ; read buffer >> >> ar1 pvsynth fsb >> ar2 pvsynth fsig1 >> >> outs ar1, ar2 >> endin >> >> >> >> >> >> i1 0 10 >> >> >> >> > > > > Send bugs reports to the Sourceforge bug tracker > https://sourceforge.net/tracker/?group_id=81968&atid=564599 > Discussions of bugs and features can be posted here > To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe > csound" > > Send bugs reports to the Sourceforge bug tracker https://sourceforge.net/tracker/?group_id=81968&atid=564599 Discussions of bugs and features can be posted here To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Date	2011-07-07 15:57
From	peiman khosravi
Subject	Re: [Csnd] Re: pvsbuffer issue
	Thanks Steven, Good idea, I will try it in a sec. But is it supposed to behave like this? Thanks Peiman On 07/07/2011 15:49, Steven Yi wrote: > Does this work if you wrap it in a UDO and use setksmps 1? > > On Thu, Jul 7, 2011 at 10:40 AM, peiman khosravi > wrote: >> It turns out if ksmps=1 then it sounds fine. Should I file it as a bug? >> That's a real pity, I had great plans for that opcode - my Sunday roast is >> ruined now :-) >> >> Best, >> Peiman >> >> On 07/07/2011 15:35, peiman khosravi wrote: >>> Hello, >>> >>> As the csd below demonstrates there seems to be something fishy going on >>> with pvsbuffer. It sounds as if some bins are zeroed out. The result is >>> drastically different from a straight resynthesis of the fsig. >>> >>> Am I doing something wrong? >>> >>> Thanks >>> >>> Peiman >>> >>> >>> >>> >>> >>> -odac -d >>> >>> >>> >>> >>> sr = 96000 >>> ksmps = 100 >>> nchnls = 2 >>> 0dbfs = 1 >>> >>> instr 1 >>> >>> asig1 noise .2, 0 >>> >>> fsig1 pvsanal asig1,1024,256,1024,1 >>> >>> >>> ibuf1,kt1 pvsbuffer fsig1, 10 ; 10-sec buf with fsig1 >>> >>> khan = ibuf1 >>> >>> fsb pvsbufread kt1, khan, 0, sr/2 ; read buffer >>> >>> ar1 pvsynth fsb >>> ar2 pvsynth fsig1 >>> >>> outs ar1, ar2 >>> endin >>> >>> >>> >>> >>> >>> i1 0 10 >>> >>> >>> >>> >> >> >> Send bugs reports to the Sourceforge bug tracker >> https://sourceforge.net/tracker/?group_id=81968&atid=564599 >> Discussions of bugs and features can be posted here >> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe >> csound" >> >> > > Send bugs reports to the Sourceforge bug tracker > https://sourceforge.net/tracker/?group_id=81968&atid=564599 > Discussions of bugs and features can be posted here > To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound" > Send bugs reports to the Sourceforge bug tracker https://sourceforge.net/tracker/?group_id=81968&atid=564599 Discussions of bugs and features can be posted here To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Date	2011-07-07 16:12
From	peiman khosravi
Subject	Re: [Csnd] Re: pvsbuffer issue
	Actually UDO is not an option as the writing and reading into buffer is done in two different instruments and you cannot pass fsig to a UDO :-( Thanks Peiman On 07/07/2011 15:49, Steven Yi wrote: > Does this work if you wrap it in a UDO and use setksmps 1? > > On Thu, Jul 7, 2011 at 10:40 AM, peiman khosravi > wrote: >> It turns out if ksmps=1 then it sounds fine. Should I file it as a bug? >> That's a real pity, I had great plans for that opcode - my Sunday roast is >> ruined now :-) >> >> Best, >> Peiman >> >> On 07/07/2011 15:35, peiman khosravi wrote: >>> Hello, >>> >>> As the csd below demonstrates there seems to be something fishy going on >>> with pvsbuffer. It sounds as if some bins are zeroed out. The result is >>> drastically different from a straight resynthesis of the fsig. >>> >>> Am I doing something wrong? >>> >>> Thanks >>> >>> Peiman >>> >>> >>> >>> >>> >>> -odac -d >>> >>> >>> >>> >>> sr = 96000 >>> ksmps = 100 >>> nchnls = 2 >>> 0dbfs = 1 >>> >>> instr 1 >>> >>> asig1 noise .2, 0 >>> >>> fsig1 pvsanal asig1,1024,256,1024,1 >>> >>> >>> ibuf1,kt1 pvsbuffer fsig1, 10 ; 10-sec buf with fsig1 >>> >>> khan = ibuf1 >>> >>> fsb pvsbufread kt1, khan, 0, sr/2 ; read buffer >>> >>> ar1 pvsynth fsb >>> ar2 pvsynth fsig1 >>> >>> outs ar1, ar2 >>> endin >>> >>> >>> >>> >>> >>> i1 0 10 >>> >>> >>> >>> >> >> >> Send bugs reports to the Sourceforge bug tracker >> https://sourceforge.net/tracker/?group_id=81968&atid=564599 >> Discussions of bugs and features can be posted here >> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe >> csound" >> >> > > Send bugs reports to the Sourceforge bug tracker > https://sourceforge.net/tracker/?group_id=81968&atid=564599 > Discussions of bugs and features can be posted here > To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound" > Send bugs reports to the Sourceforge bug tracker https://sourceforge.net/tracker/?group_id=81968&atid=564599 Discussions of bugs and features can be posted here To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Date	2011-07-07 16:20
From	Steven Yi
Subject	Re: [Csnd] Re: pvsbuffer issue
	Hm, well another possibility to is put all of your instrument code in a UDO, and have your instrument just call that. :P Other than that, I guess perhaps someone will need to modify the opcode to work without ksmps=1. On Thu, Jul 7, 2011 at 11:12 AM, peiman khosravi wrote: > Actually UDO is not an option as the writing and reading into buffer is done > in two different instruments and you cannot pass fsig to a UDO :-( > > Thanks > Peiman > > On 07/07/2011 15:49, Steven Yi wrote: >> >> Does this work if you wrap it in a UDO and use setksmps 1? >> >> On Thu, Jul 7, 2011 at 10:40 AM, peiman khosravi >> wrote: >>> >>> It turns out if ksmps=1 then it sounds fine. Should I file it as a bug? >>> That's a real pity, I had great plans for that opcode - my Sunday roast >>> is >>> ruined now :-) >>> >>> Best, >>> Peiman >>> >>> On 07/07/2011 15:35, peiman khosravi wrote: >>>> >>>> Hello, >>>> >>>> As the csd below demonstrates there seems to be something fishy going on >>>> with pvsbuffer. It sounds as if some bins are zeroed out. The result is >>>> drastically different from a straight resynthesis of the fsig. >>>> >>>> Am I doing something wrong? >>>> >>>> Thanks >>>> >>>> Peiman >>>> >>>> >>>> >>>> >>>> >>>> -odac -d >>>> >>>> >>>> >>>> >>>> sr = 96000 >>>> ksmps = 100 >>>> nchnls = 2 >>>> 0dbfs = 1 >>>> >>>> instr 1 >>>> >>>> asig1 noise .2, 0 >>>> >>>> fsig1 pvsanal asig1,1024,256,1024,1 >>>> >>>> >>>> ibuf1,kt1 pvsbuffer fsig1, 10 ; 10-sec buf with fsig1 >>>> >>>> khan = ibuf1 >>>> >>>> fsb pvsbufread kt1, khan, 0, sr/2 ; read buffer >>>> >>>> ar1 pvsynth fsb >>>> ar2 pvsynth fsig1 >>>> >>>> outs ar1, ar2 >>>> endin >>>> >>>> >>>> >>>> >>>> >>>> i1 0 10 >>>> >>>> >>>> >>>> >>> >>> >>> Send bugs reports to the Sourceforge bug tracker >>> https://sourceforge.net/tracker/?group_id=81968&atid=564599 >>> Discussions of bugs and features can be posted here >>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe >>> csound" >>> >>> >> >> Send bugs reports to the Sourceforge bug tracker >> https://sourceforge.net/tracker/?group_id=81968&atid=564599 >> Discussions of bugs and features can be posted here >> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe >> csound" >> > > > > Send bugs reports to the Sourceforge bug tracker > https://sourceforge.net/tracker/?group_id=81968&atid=564599 > Discussions of bugs and features can be posted here > To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe > csound" > > Send bugs reports to the Sourceforge bug tracker https://sourceforge.net/tracker/?group_id=81968&atid=564599 Discussions of bugs and features can be posted here To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Date	2011-07-07 17:02
From	peiman khosravi
Subject	Re: [Csnd] Re: pvsbuffer issue
	I think I found the problem. Increasing the FFT size seems to fix it somehow. Mhh it's strange though P On 07/07/2011 16:20, Steven Yi wrote: > Hm, well another possibility to is put all of your instrument code in > a UDO, and have your instrument just call that. :P Other than that, I > guess perhaps someone will need to modify the opcode to work without > ksmps=1. > > On Thu, Jul 7, 2011 at 11:12 AM, peiman khosravi > wrote: >> Actually UDO is not an option as the writing and reading into buffer is done >> in two different instruments and you cannot pass fsig to a UDO :-( >> >> Thanks >> Peiman >> >> On 07/07/2011 15:49, Steven Yi wrote: >>> Does this work if you wrap it in a UDO and use setksmps 1? >>> >>> On Thu, Jul 7, 2011 at 10:40 AM, peiman khosravi >>> wrote: >>>> It turns out if ksmps=1 then it sounds fine. Should I file it as a bug? >>>> That's a real pity, I had great plans for that opcode - my Sunday roast >>>> is >>>> ruined now :-) >>>> >>>> Best, >>>> Peiman >>>> >>>> On 07/07/2011 15:35, peiman khosravi wrote: >>>>> Hello, >>>>> >>>>> As the csd below demonstrates there seems to be something fishy going on >>>>> with pvsbuffer. It sounds as if some bins are zeroed out. The result is >>>>> drastically different from a straight resynthesis of the fsig. >>>>> >>>>> Am I doing something wrong? >>>>> >>>>> Thanks >>>>> >>>>> Peiman >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> -odac -d >>>>> >>>>> >>>>> >>>>> >>>>> sr = 96000 >>>>> ksmps = 100 >>>>> nchnls = 2 >>>>> 0dbfs = 1 >>>>> >>>>> instr 1 >>>>> >>>>> asig1 noise .2, 0 >>>>> >>>>> fsig1 pvsanal asig1,1024,256,1024,1 >>>>> >>>>> >>>>> ibuf1,kt1 pvsbuffer fsig1, 10 ; 10-sec buf with fsig1 >>>>> >>>>> khan = ibuf1 >>>>> >>>>> fsb pvsbufread kt1, khan, 0, sr/2 ; read buffer >>>>> >>>>> ar1 pvsynth fsb >>>>> ar2 pvsynth fsig1 >>>>> >>>>> outs ar1, ar2 >>>>> endin >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> i1 0 10 >>>>> >>>>> >>>>> >>>>> >>>> >>>> Send bugs reports to the Sourceforge bug tracker >>>> https://sourceforge.net/tracker/?group_id=81968&atid=564599 >>>> Discussions of bugs and features can be posted here >>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe >>>> csound" >>>> >>>> >>> Send bugs reports to the Sourceforge bug tracker >>> https://sourceforge.net/tracker/?group_id=81968&atid=564599 >>> Discussions of bugs and features can be posted here >>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe >>> csound" >>> >> >> >> Send bugs reports to the Sourceforge bug tracker >> https://sourceforge.net/tracker/?group_id=81968&atid=564599 >> Discussions of bugs and features can be posted here >> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe >> csound" >> >> > > Send bugs reports to the Sourceforge bug tracker > https://sourceforge.net/tracker/?group_id=81968&atid=564599 > Discussions of bugs and features can be posted here > To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound" > Send bugs reports to the Sourceforge bug tracker https://sourceforge.net/tracker/?group_id=81968&atid=564599 Discussions of bugs and features can be posted here To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Date	2011-07-07 19:04
From	Victor Lazzarini
Subject	Re: [Csnd] Re: pvsbuffer issue
	It could be to do with the interpolation of windows, which pvsbufread does. Using a ksmps that is pow-of-two seems to sort it (ksmps=64) (at least in my crappy laptop speakers) Victor On 7 Jul 2011, at 17:02, peiman khosravi wrote: > I think I found the problem. > Increasing the FFT size seems to fix it somehow. Mhh it's strange > though > > P > > On 07/07/2011 16:20, Steven Yi wrote: >> Hm, well another possibility to is put all of your instrument code in >> a UDO, and have your instrument just call that. :P Other than >> that, I >> guess perhaps someone will need to modify the opcode to work without >> ksmps=1. >> >> On Thu, Jul 7, 2011 at 11:12 AM, peiman khosravi >> wrote: >>> Actually UDO is not an option as the writing and reading into >>> buffer is done >>> in two different instruments and you cannot pass fsig to a UDO :-( >>> >>> Thanks >>> Peiman >>> >>> On 07/07/2011 15:49, Steven Yi wrote: >>>> Does this work if you wrap it in a UDO and use setksmps 1? >>>> >>>> On Thu, Jul 7, 2011 at 10:40 AM, peiman khosravi >>>> wrote: >>>>> It turns out if ksmps=1 then it sounds fine. Should I file it as >>>>> a bug? >>>>> That's a real pity, I had great plans for that opcode - my >>>>> Sunday roast >>>>> is >>>>> ruined now :-) >>>>> >>>>> Best, >>>>> Peiman >>>>> >>>>> On 07/07/2011 15:35, peiman khosravi wrote: >>>>>> Hello, >>>>>> >>>>>> As the csd below demonstrates there seems to be something fishy >>>>>> going on >>>>>> with pvsbuffer. It sounds as if some bins are zeroed out. The >>>>>> result is >>>>>> drastically different from a straight resynthesis of the fsig. >>>>>> >>>>>> Am I doing something wrong? >>>>>> >>>>>> Thanks >>>>>> >>>>>> Peiman >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> -odac -d >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> sr = 96000 >>>>>> ksmps = 100 >>>>>> nchnls = 2 >>>>>> 0dbfs = 1 >>>>>> >>>>>> instr 1 >>>>>> >>>>>> asig1 noise .2, 0 >>>>>> >>>>>> fsig1 pvsanal asig1,1024,256,1024,1 >>>>>> >>>>>> >>>>>> ibuf1,kt1 pvsbuffer fsig1, 10 ; 10-sec buf with fsig1 >>>>>> >>>>>> khan = ibuf1 >>>>>> >>>>>> fsb pvsbufread kt1, khan, 0, sr/2 ; read buffer >>>>>> >>>>>> ar1 pvsynth fsb >>>>>> ar2 pvsynth fsig1 >>>>>> >>>>>> outs ar1, ar2 >>>>>> endin >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> i1 0 10 >>>>>> >>>>>> >>>>>> >>>>>> >>>>> >>>>> Send bugs reports to the Sourceforge bug tracker >>>>> https://sourceforge.net/tracker/?group_id=81968&atid=564599 >>>>> Discussions of bugs and features can be posted here >>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body >>>>> "unsubscribe >>>>> csound" >>>>> >>>>> >>>> Send bugs reports to the Sourceforge bug tracker >>>> https://sourceforge.net/tracker/?group_id=81968&atid=564599 >>>> Discussions of bugs and features can be posted here >>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body >>>> "unsubscribe >>>> csound" >>>> >>> >>> >>> Send bugs reports to the Sourceforge bug tracker >>> https://sourceforge.net/tracker/? >>> group_id=81968&atid=564599 >>> Discussions of bugs and features can be posted here >>> To unsubscribe, send email sympa@lists.bath.ac.uk with body >>> "unsubscribe >>> csound" >>> >>> >> >> Send bugs reports to the Sourceforge bug tracker >> https://sourceforge.net/tracker/?group_id=81968&atid=564599 >> Discussions of bugs and features can be posted here >> To unsubscribe, send email sympa@lists.bath.ac.uk with body >> "unsubscribe csound" >> > > > > Send bugs reports to the Sourceforge bug tracker > https://sourceforge.net/tracker/?group_id=81968&atid=564599 > Discussions of bugs and features can be posted here > To unsubscribe, send email sympa@lists.bath.ac.uk with body > "unsubscribe csound" > Dr Victor Lazzarini Senior Lecturer Dept. of Music NUI Maynooth Ireland tel.: +353 1 708 3545 Victor dot Lazzarini AT nuim dot ie Send bugs reports to the Sourceforge bug tracker https://sourceforge.net/tracker/?group_id=81968&atid=564599 Discussions of bugs and features can be posted here To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Date	2011-07-07 21:41
From	Victor Lazzarini
Subject	Re: [Csnd] Re: pvsbuffer issue
	Ok, it was a bug in the buffer position counting of pvsbufread, which only showed with ksmps not pow of 2 (or 1). Now fixed in GIT. Victor On 7 Jul 2011, at 19:04, Victor Lazzarini wrote: > It could be to do with the interpolation of windows, which > pvsbufread does. > Using a ksmps that is pow-of-two seems to sort it (ksmps=64) (at > least in my crappy laptop speakers) > > Victor > On 7 Jul 2011, at 17:02, peiman khosravi wrote: > >> I think I found the problem. >> Increasing the FFT size seems to fix it somehow. Mhh it's strange >> though >> >> P >> >> On 07/07/2011 16:20, Steven Yi wrote: >>> Hm, well another possibility to is put all of your instrument code >>> in >>> a UDO, and have your instrument just call that. :P Other than >>> that, I >>> guess perhaps someone will need to modify the opcode to work without >>> ksmps=1. >>> >>> On Thu, Jul 7, 2011 at 11:12 AM, peiman khosravi >>> wrote: >>>> Actually UDO is not an option as the writing and reading into >>>> buffer is done >>>> in two different instruments and you cannot pass fsig to a UDO :-( >>>> >>>> Thanks >>>> Peiman >>>> >>>> On 07/07/2011 15:49, Steven Yi wrote: >>>>> Does this work if you wrap it in a UDO and use setksmps 1? >>>>> >>>>> On Thu, Jul 7, 2011 at 10:40 AM, peiman khosravi >>>>> wrote: >>>>>> It turns out if ksmps=1 then it sounds fine. Should I file it >>>>>> as a bug? >>>>>> That's a real pity, I had great plans for that opcode - my >>>>>> Sunday roast >>>>>> is >>>>>> ruined now :-) >>>>>> >>>>>> Best, >>>>>> Peiman >>>>>> >>>>>> On 07/07/2011 15:35, peiman khosravi wrote: >>>>>>> Hello, >>>>>>> >>>>>>> As the csd below demonstrates there seems to be something >>>>>>> fishy going on >>>>>>> with pvsbuffer. It sounds as if some bins are zeroed out. The >>>>>>> result is >>>>>>> drastically different from a straight resynthesis of the fsig. >>>>>>> >>>>>>> Am I doing something wrong? >>>>>>> >>>>>>> Thanks >>>>>>> >>>>>>> Peiman >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> -odac -d >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> sr = 96000 >>>>>>> ksmps = 100 >>>>>>> nchnls = 2 >>>>>>> 0dbfs = 1 >>>>>>> >>>>>>> instr 1 >>>>>>> >>>>>>> asig1 noise .2, 0 >>>>>>> >>>>>>> fsig1 pvsanal asig1,1024,256,1024,1 >>>>>>> >>>>>>> >>>>>>> ibuf1,kt1 pvsbuffer fsig1, 10 ; 10-sec buf with fsig1 >>>>>>> >>>>>>> khan = ibuf1 >>>>>>> >>>>>>> fsb pvsbufread kt1, khan, 0, sr/2 ; read buffer >>>>>>> >>>>>>> ar1 pvsynth fsb >>>>>>> ar2 pvsynth fsig1 >>>>>>> >>>>>>> outs ar1, ar2 >>>>>>> endin >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> i1 0 10 >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>> >>>>>> Send bugs reports to the Sourceforge bug tracker >>>>>> https://sourceforge.net/tracker/?group_id=81968&atid=564599 >>>>>> Discussions of bugs and features can be posted here >>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body >>>>>> "unsubscribe >>>>>> csound" >>>>>> >>>>>> >>>>> Send bugs reports to the Sourceforge bug tracker >>>>> https://sourceforge.net/tracker/?group_id=81968&atid=564599 >>>>> Discussions of bugs and features can be posted here >>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body >>>>> "unsubscribe >>>>> csound" >>>>> >>>> >>>> >>>> Send bugs reports to the Sourceforge bug tracker >>>> https://sourceforge.net/tracker/? >>>> group_id=81968&atid=564599 >>>> Discussions of bugs and features can be posted here >>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body >>>> "unsubscribe >>>> csound" >>>> >>>> >>> >>> Send bugs reports to the Sourceforge bug tracker >>> https://sourceforge.net/tracker/?group_id=81968&atid=564599 >>> Discussions of bugs and features can be posted here >>> To unsubscribe, send email sympa@lists.bath.ac.uk with body >>> "unsubscribe csound" >>> >> >> >> >> Send bugs reports to the Sourceforge bug tracker >> https://sourceforge.net/tracker/?group_id=81968&atid=564599 >> Discussions of bugs and features can be posted here >> To unsubscribe, send email sympa@lists.bath.ac.uk with body >> "unsubscribe csound" >> > > Dr Victor Lazzarini > Senior Lecturer > Dept. of Music > NUI Maynooth Ireland > tel.: +353 1 708 3545 > Victor dot Lazzarini AT nuim dot ie > > > > > > Send bugs reports to the Sourceforge bug tracker > https://sourceforge.net/tracker/?group_id=81968&atid=564599 > Discussions of bugs and features can be posted here > To unsubscribe, send email sympa@lists.bath.ac.uk with body > "unsubscribe csound" > Dr Victor Lazzarini Senior Lecturer Dept. of Music NUI Maynooth Ireland tel.: +353 1 708 3545 Victor dot Lazzarini AT nuim dot ie Send bugs reports to the Sourceforge bug tracker https://sourceforge.net/tracker/?group_id=81968&atid=564599 Discussions of bugs and features can be posted here To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Date	2011-07-07 23:11
From	peiman khosravi
Subject	Re: [Csnd] Re: pvsbuffer issue
	Thanks Victor, Best, Peiman On 7 July 2011 21:41, Victor Lazzarini wrote: > Ok, it was a bug in the buffer position counting of pvsbufread, which only > showed with ksmps not pow of 2 (or 1). Now fixed in GIT. > > Victor > > On 7 Jul 2011, at 19:04, Victor Lazzarini wrote: > >> It could be to do with the interpolation of windows, which pvsbufread >> does. >> Using a ksmps that is pow-of-two seems to sort it (ksmps=64) (at least in >> my crappy laptop speakers) >> >> Victor >> On 7 Jul 2011, at 17:02, peiman khosravi wrote: >> >>> I think I found the problem. >>> Increasing the FFT size seems to fix it somehow. Mhh it's strange though >>> >>> P >>> >>> On 07/07/2011 16:20, Steven Yi wrote: >>>> >>>> Hm, well another possibility to is put all of your instrument code in >>>> a UDO, and have your instrument just call that. :P Other than that, I >>>> guess perhaps someone will need to modify the opcode to work without >>>> ksmps=1. >>>> >>>> On Thu, Jul 7, 2011 at 11:12 AM, peiman khosravi >>>> wrote: >>>>> >>>>> Actually UDO is not an option as the writing and reading into buffer is >>>>> done >>>>> in two different instruments and you cannot pass fsig to a UDO :-( >>>>> >>>>> Thanks >>>>> Peiman >>>>> >>>>> On 07/07/2011 15:49, Steven Yi wrote: >>>>>> >>>>>> Does this work if you wrap it in a UDO and use setksmps 1? >>>>>> >>>>>> On Thu, Jul 7, 2011 at 10:40 AM, peiman khosravi >>>>>> wrote: >>>>>>> >>>>>>> It turns out if ksmps=1 then it sounds fine. Should I file it as a >>>>>>> bug? >>>>>>> That's a real pity, I had great plans for that opcode - my Sunday >>>>>>> roast >>>>>>> is >>>>>>> ruined now :-) >>>>>>> >>>>>>> Best, >>>>>>> Peiman >>>>>>> >>>>>>> On 07/07/2011 15:35, peiman khosravi wrote: >>>>>>>> >>>>>>>> Hello, >>>>>>>> >>>>>>>> As the csd below demonstrates there seems to be something fishy >>>>>>>> going on >>>>>>>> with pvsbuffer. It sounds as if some bins are zeroed out. The result >>>>>>>> is >>>>>>>> drastically different from a straight resynthesis of the fsig. >>>>>>>> >>>>>>>> Am I doing something wrong? >>>>>>>> >>>>>>>> Thanks >>>>>>>> >>>>>>>> Peiman >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> -odac -d >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> sr = 96000 >>>>>>>> ksmps = 100 >>>>>>>> nchnls = 2 >>>>>>>> 0dbfs = 1 >>>>>>>> >>>>>>>> instr 1 >>>>>>>> >>>>>>>> asig1 noise .2, 0 >>>>>>>> >>>>>>>> fsig1 pvsanal asig1,1024,256,1024,1 >>>>>>>> >>>>>>>> >>>>>>>> ibuf1,kt1 pvsbuffer fsig1, 10 ; 10-sec buf with fsig1 >>>>>>>> >>>>>>>> khan = ibuf1 >>>>>>>> >>>>>>>> fsb pvsbufread kt1, khan, 0, sr/2 ; read buffer >>>>>>>> >>>>>>>> ar1 pvsynth fsb >>>>>>>> ar2 pvsynth fsig1 >>>>>>>> >>>>>>>> outs ar1, ar2 >>>>>>>> endin >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> i1 0 10 >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>> >>>>>>> Send bugs reports to the Sourceforge bug tracker >>>>>>> https://sourceforge.net/tracker/?group_id=81968&atid=564599 >>>>>>> Discussions of bugs and features can be posted here >>>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body >>>>>>> "unsubscribe >>>>>>> csound" >>>>>>> >>>>>>> >>>>>> Send bugs reports to the Sourceforge bug tracker >>>>>> https://sourceforge.net/tracker/?group_id=81968&atid=564599 >>>>>> Discussions of bugs and features can be posted here >>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body >>>>>> "unsubscribe >>>>>> csound" >>>>>> >>>>> >>>>> >>>>> Send bugs reports to the Sourceforge bug tracker >>>>> https://sourceforge.net/tracker/?group_id=81968&atid=564599 >>>>> Discussions of bugs and features can be posted here >>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body >>>>> "unsubscribe >>>>> csound" >>>>> >>>>> >>>> >>>> Send bugs reports to the Sourceforge bug tracker >>>> https://sourceforge.net/tracker/?group_id=81968&atid=564599 >>>> Discussions of bugs and features can be posted here >>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe >>>> csound" >>>> >>> >>> >>> >>> Send bugs reports to the Sourceforge bug tracker >>> https://sourceforge.net/tracker/?group_id=81968&atid=564599 >>> Discussions of bugs and features can be posted here >>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe >>> csound" >>> >> >> Dr Victor Lazzarini >> Senior Lecturer >> Dept. of Music >> NUI Maynooth Ireland >> tel.: +353 1 708 3545 >> Victor dot Lazzarini AT nuim dot ie >> >> >> >> >> >> Send bugs reports to the Sourceforge bug tracker >> https://sourceforge.net/tracker/?group_id=81968&atid=564599 >> Discussions of bugs and features can be posted here >> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe >> csound" >> > > Dr Victor Lazzarini > Senior Lecturer > Dept. of Music > NUI Maynooth Ireland > tel.: +353 1 708 3545 > Victor dot Lazzarini AT nuim dot ie > > > > > > Send bugs reports to the Sourceforge bug tracker > https://sourceforge.net/tracker/?group_id=81968&atid=564599 > Discussions of bugs and features can be posted here > To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe > csound" > > Send bugs reports to the Sourceforge bug tracker https://sourceforge.net/tracker/?group_id=81968&atid=564599 Discussions of bugs and features can be posted here To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"