Hello all,

I'm seeing more tables being generated from ftgenonce than I expected
again.  Csound version is 6.03.2 (double samples) Aug 21 2014, which
was built from the developer branch.

In the attached example, there are two tables being generated both
with gen7.  One is a very short envelope, and the other a triangle
wave form.  For some reason the triangle table will generate with each
note in this example.

In the project I'm working on, the behavior gets even weirder.  I have
a metronome instrument that generates beeps over a duration by adding
score entries using scoreline, it's envelope is the "ft_env" table
from my example, also generated with ftgenonce.  But if I add this
problematic triangle wave to a *completely* new instrument, the
envelope for the metronome instrument will generate a new ftable on
every single new note, even tho nothing had changed in that
instrument!

Here's the stripped down example:


; Select audio/midi flags here according to platform
-odac     ;;;realtime audio out
;-iadc    ;;;uncomment -iadc if realtime audio input is needed too
; For Non-realtime ouput leave only the line below:
; -o ftgenonce.wav -W ;;; for file output any platform



sr = 44100
ksmps = 32
nchnls = 2
0dbfs  = 1

instr 1
  ift_env ftgenonce 0, -99, 1024, 7, 0, 14, 1, 110, 0, 900, 0
endin

instr 2
  ift_wav ftgenonce 0, -99, 1024, 7, 0, 256, 1, 512, -1, 256, 0
endin




i 1 1 1
i 1 2 1
i 1 3 1
i 1 4 1

i 1 6 1
i 1 7 1
i 1 8 1
i 1 9 1

i 2 6 4
i 2 7 1
i 2 8 1
i 2 9 1
e

I could be wrong but my understanding is that ftgenonce will generate
a function table for each instance of the instrument. ftgen on the
other hand, when called in the orchestra header, will create only one
instance of the function table. I think the idea with ftgenonce is
that one can encapsulate a function table into an instrument,
therefore making it possible to include entire instruments using
#include statements.

On 25 August 2014 09:14, Kevin Welsh  wrote:
> Hello all,
>
> I'm seeing more tables being generated from ftgenonce than I expected
> again.  Csound version is 6.03.2 (double samples) Aug 21 2014, which
> was built from the developer branch.
>
> In the attached example, there are two tables being generated both
> with gen7.  One is a very short envelope, and the other a triangle
> wave form.  For some reason the triangle table will generate with each
> note in this example.
>
> In the project I'm working on, the behavior gets even weirder.  I have
> a metronome instrument that generates beeps over a duration by adding
> score entries using scoreline, it's envelope is the "ft_env" table
> from my example, also generated with ftgenonce.  But if I add this
> problematic triangle wave to a *completely* new instrument, the
> envelope for the metronome instrument will generate a new ftable on
> every single new note, even tho nothing had changed in that
> instrument!
>
> Here's the stripped down example:
> 
> 
> ; Select audio/midi flags here according to platform
> -odac     ;;;realtime audio out
> ;-iadc    ;;;uncomment -iadc if realtime audio input is needed too
> ; For Non-realtime ouput leave only the line below:
> ; -o ftgenonce.wav -W ;;; for file output any platform
> 
> 
>
> sr = 44100
> ksmps = 32
> nchnls = 2
> 0dbfs  = 1
>
> instr 1
>   ift_env ftgenonce 0, -99, 1024, 7, 0, 14, 1, 110, 0, 900, 0
> endin
>
> instr 2
>   ift_wav ftgenonce 0, -99, 1024, 7, 0, 256, 1, 512, -1, 256, 0
> endin
>
> 
> 
>
> i 1 1 1
> i 1 2 1
> i 1 3 1
> i 1 4 1
>
> i 1 6 1
> i 1 7 1
> i 1 8 1
> i 1 9 1
>
> i 2 6 4
> i 2 7 1
> i 2 8 1
> i 2 9 1
> e
> 
> 
>
>
> Send bugs reports to
>         https://github.com/csound/csound/issues
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"
>
>
>

I could easily be misunderstanding, but I'm pretty sure that's not the
case.  It should be generating a ftable for an instrument one time
only, and in fact, if another instrument tries to generate an
identical table, it should match that too.

Here's an example showing the proper behavior, it's a test file I made
a week or two ago to troubleshoot ftgenonce handling string values.
The two instrument will load the same wave file, you'll see that the
table only gets loaded once, even tho both instruments loading it, and
further, two instances of each instrument are active at a time due to
notes overlapping.



; Select audio/midi flags here according to platform
-odac     ;;;realtime audio out
;-iadc    ;;;uncomment -iadc if realtime audio input is needed too
; For Non-realtime ouput leave only the line below:
; -o ftgenonce.wav -W ;;; for file output any platform



sr = 44100
ksmps = 32
nchnls = 2
0dbfs  = 1

instr 1
  ift ftgenonce 0, -1, 0, -1, "kickroll.wav", 0, 0, 0
endin

instr 2
  Sfile = "kickroll.wav"
  ift ftgenonce 0, -1, 0, -1, Sfile, 0, 0, 0
endin




i 1 1 10
i 1 2 10
i 2 1 10
i 2 2 10
e



On Mon, Aug 25, 2014 at 6:31 AM, Rory Walsh  wrote:
> I could be wrong but my understanding is that ftgenonce will generate
> a function table for each instance of the instrument. ftgen on the
> other hand, when called in the orchestra header, will create only one
> instance of the function table. I think the idea with ftgenonce is
> that one can encapsulate a function table into an instrument,
> therefore making it possible to include entire instruments using
> #include statements.
>
> On 25 August 2014 09:14, Kevin Welsh  wrote:
>> Hello all,
>>
>> I'm seeing more tables being generated from ftgenonce than I expected
>> again.  Csound version is 6.03.2 (double samples) Aug 21 2014, which
>> was built from the developer branch.
>>
>> In the attached example, there are two tables being generated both
>> with gen7.  One is a very short envelope, and the other a triangle
>> wave form.  For some reason the triangle table will generate with each
>> note in this example.
>>
>> In the project I'm working on, the behavior gets even weirder.  I have
>> a metronome instrument that generates beeps over a duration by adding
>> score entries using scoreline, it's envelope is the "ft_env" table
>> from my example, also generated with ftgenonce.  But if I add this
>> problematic triangle wave to a *completely* new instrument, the
>> envelope for the metronome instrument will generate a new ftable on
>> every single new note, even tho nothing had changed in that
>> instrument!
>>
>> Here's the stripped down example:
>> 
>> 
>> ; Select audio/midi flags here according to platform
>> -odac     ;;;realtime audio out
>> ;-iadc    ;;;uncomment -iadc if realtime audio input is needed too
>> ; For Non-realtime ouput leave only the line below:
>> ; -o ftgenonce.wav -W ;;; for file output any platform
>> 
>> 
>>
>> sr = 44100
>> ksmps = 32
>> nchnls = 2
>> 0dbfs  = 1
>>
>> instr 1
>>   ift_env ftgenonce 0, -99, 1024, 7, 0, 14, 1, 110, 0, 900, 0
>> endin
>>
>> instr 2
>>   ift_wav ftgenonce 0, -99, 1024, 7, 0, 256, 1, 512, -1, 256, 0
>> endin
>>
>> 
>> 
>>
>> i 1 1 1
>> i 1 2 1
>> i 1 3 1
>> i 1 4 1
>>
>> i 1 6 1
>> i 1 7 1
>> i 1 8 1
>> i 1 9 1
>>
>> i 2 6 4
>> i 2 7 1
>> i 2 8 1
>> i 2 9 1
>> e
>> 
>> 
>>
>>
>> Send bugs reports to
>>         https://github.com/csound/csound/issues
>> 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
>         https://github.com/csound/csound/issues
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"
>
>
>

You are wrong. The manual is correct.

Regards,
Mike

So the table is only be recreated if some of its arguments change from
note to note. All clear now.

On 25 August 2014 12:52, Michael Gogins  wrote:
> You are wrong. The manual is correct.
>
> Regards,
> Mike
>
> On Aug 25, 2014 6:31 AM, "Rory Walsh"  wrote:
>>
>> I could be wrong but my understanding is that ftgenonce will generate
>> a function table for each instance of the instrument. ftgen on the
>> other hand, when called in the orchestra header, will create only one
>> instance of the function table. I think the idea with ftgenonce is
>> that one can encapsulate a function table into an instrument,
>> therefore making it possible to include entire instruments using
>> #include statements.
>>
>> On 25 August 2014 09:14, Kevin Welsh  wrote:
>> > Hello all,
>> >
>> > I'm seeing more tables being generated from ftgenonce than I expected
>> > again.  Csound version is 6.03.2 (double samples) Aug 21 2014, which
>> > was built from the developer branch.
>> >
>> > In the attached example, there are two tables being generated both
>> > with gen7.  One is a very short envelope, and the other a triangle
>> > wave form.  For some reason the triangle table will generate with each
>> > note in this example.
>> >
>> > In the project I'm working on, the behavior gets even weirder.  I have
>> > a metronome instrument that generates beeps over a duration by adding
>> > score entries using scoreline, it's envelope is the "ft_env" table
>> > from my example, also generated with ftgenonce.  But if I add this
>> > problematic triangle wave to a *completely* new instrument, the
>> > envelope for the metronome instrument will generate a new ftable on
>> > every single new note, even tho nothing had changed in that
>> > instrument!
>> >
>> > Here's the stripped down example:
>> > 
>> > 
>> > ; Select audio/midi flags here according to platform
>> > -odac     ;;;realtime audio out
>> > ;-iadc    ;;;uncomment -iadc if realtime audio input is needed too
>> > ; For Non-realtime ouput leave only the line below:
>> > ; -o ftgenonce.wav -W ;;; for file output any platform
>> > 
>> > 
>> >
>> > sr = 44100
>> > ksmps = 32
>> > nchnls = 2
>> > 0dbfs  = 1
>> >
>> > instr 1
>> >   ift_env ftgenonce 0, -99, 1024, 7, 0, 14, 1, 110, 0, 900, 0
>> > endin
>> >
>> > instr 2
>> >   ift_wav ftgenonce 0, -99, 1024, 7, 0, 256, 1, 512, -1, 256, 0
>> > endin
>> >
>> > 
>> > 
>> >
>> > i 1 1 1
>> > i 1 2 1
>> > i 1 3 1
>> > i 1 4 1
>> >
>> > i 1 6 1
>> > i 1 7 1
>> > i 1 8 1
>> > i 1 9 1
>> >
>> > i 2 6 4
>> > i 2 7 1
>> > i 2 8 1
>> > i 2 9 1
>> > e
>> > 
>> > 
>> >
>> >
>> > Send bugs reports to
>> >         https://github.com/csound/csound/issues
>> > 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
>>         https://github.com/csound/csound/issues
>> Discussions of bugs and features can be posted here
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>> csound"
>>
>>
>>
>

Right, at least theoretically... but in the first example (original
email) that is definitely not happening, and I can't see any reason as
to why!

Any suggestions are welcome, but based off the conversations Michael
and I had troubleshooting the string variable problem earlier, I'm
almost certain there's an underlying bug here.

On Mon, Aug 25, 2014 at 6:56 AM, Rory Walsh  wrote:
> So the table is only be recreated if some of its arguments change from
> note to note. All clear now.
>
> On 25 August 2014 12:52, Michael Gogins  wrote:
>> You are wrong. The manual is correct.
>>
>> Regards,
>> Mike
>>
>> On Aug 25, 2014 6:31 AM, "Rory Walsh"  wrote:
>>>
>>> I could be wrong but my understanding is that ftgenonce will generate
>>> a function table for each instance of the instrument. ftgen on the
>>> other hand, when called in the orchestra header, will create only one
>>> instance of the function table. I think the idea with ftgenonce is
>>> that one can encapsulate a function table into an instrument,
>>> therefore making it possible to include entire instruments using
>>> #include statements.
>>>
>>> On 25 August 2014 09:14, Kevin Welsh  wrote:
>>> > Hello all,
>>> >
>>> > I'm seeing more tables being generated from ftgenonce than I expected
>>> > again.  Csound version is 6.03.2 (double samples) Aug 21 2014, which
>>> > was built from the developer branch.
>>> >
>>> > In the attached example, there are two tables being generated both
>>> > with gen7.  One is a very short envelope, and the other a triangle
>>> > wave form.  For some reason the triangle table will generate with each
>>> > note in this example.
>>> >
>>> > In the project I'm working on, the behavior gets even weirder.  I have
>>> > a metronome instrument that generates beeps over a duration by adding
>>> > score entries using scoreline, it's envelope is the "ft_env" table
>>> > from my example, also generated with ftgenonce.  But if I add this
>>> > problematic triangle wave to a *completely* new instrument, the
>>> > envelope for the metronome instrument will generate a new ftable on
>>> > every single new note, even tho nothing had changed in that
>>> > instrument!
>>> >
>>> > Here's the stripped down example:
>>> > 
>>> > 
>>> > ; Select audio/midi flags here according to platform
>>> > -odac     ;;;realtime audio out
>>> > ;-iadc    ;;;uncomment -iadc if realtime audio input is needed too
>>> > ; For Non-realtime ouput leave only the line below:
>>> > ; -o ftgenonce.wav -W ;;; for file output any platform
>>> > 
>>> > 
>>> >
>>> > sr = 44100
>>> > ksmps = 32
>>> > nchnls = 2
>>> > 0dbfs  = 1
>>> >
>>> > instr 1
>>> >   ift_env ftgenonce 0, -99, 1024, 7, 0, 14, 1, 110, 0, 900, 0
>>> > endin
>>> >
>>> > instr 2
>>> >   ift_wav ftgenonce 0, -99, 1024, 7, 0, 256, 1, 512, -1, 256, 0
>>> > endin
>>> >
>>> > 
>>> > 
>>> >
>>> > i 1 1 1
>>> > i 1 2 1
>>> > i 1 3 1
>>> > i 1 4 1
>>> >
>>> > i 1 6 1
>>> > i 1 7 1
>>> > i 1 8 1
>>> > i 1 9 1
>>> >
>>> > i 2 6 4
>>> > i 2 7 1
>>> > i 2 8 1
>>> > i 2 9 1
>>> > e
>>> > 
>>> > 
>>> >
>>> >
>>> > Send bugs reports to
>>> >         https://github.com/csound/csound/issues
>>> > 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
>>>         https://github.com/csound/csound/issues
>>> 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
>         https://github.com/csound/csound/issues
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"
>
>
>

I should have tried your first example first :) For me it seems to
work fine. I'm using Csound version 6.03.2 (double samples) Jul 24
2014, Linux. The output is posted below. Each f-table only seems to be
generated onece?

SECTION 1:
B  0.000 ..  1.000 T  1.000 TT  1.000 M:  0.00000  0.00000
new alloc for instr 1:
ftable 101:
 calling draw callback
ftable 101:    1024 points, scalemax 1.000
 .
  '
   -
    -
     .
      _
       _

        '
_________-_____________________________________________________________________
ftgenonce: created new func: 101
B  1.000 ..  2.000 T  2.000 TT  2.000 M:  0.00000  0.00000
B  2.000 ..  3.000 T  3.000 TT  3.000 M:  0.00000  0.00000
B  3.000 ..  4.000 T  4.000 TT  4.000 M:  0.00000  0.00000
B  4.000 ..  6.000 T  6.000 TT  6.000 M:  0.00000  0.00000
new alloc for instr 2:
ftable 102:
 calling draw callback
ftable 102:    1024 points, scalemax 1.000
                  _-'.
                _-    '.
              _-        '.
            _-            '.
          _-                '.
        _-                    '.
      _-                        '_
    _-                            -_
  _-                                -_
_-____________________________________-________________________________________
                                        -_                                   _-
                                          -_                               _-
                                            -_                           _-
                                              -_                       _-
                                                -_                   _-
                                                  -_               _-
                                                    -_           _-
                                                      -_        -
                                                        -_    .'
                                                          -_.'
ftgenonce: created new func: 102
B  6.000 ..  7.000 T  7.000 TT  7.000 M:  0.00000  0.00000
new alloc for instr 2:
B  7.000 ..  8.000 T  8.000 TT  8.000 M:  0.00000  0.00000
B  8.000 ..  9.000 T  9.000 TT  9.000 M:  0.00000  0.00000
B  9.000 .. 10.000 T 10.000 TT 10.000 M:  0.00000  0.00000
Score finished in csoundPerform().
inactive allocs returned to freespace
end of score.           overall amps:  0.00000  0.00000
       overall samples out of range:        0        0
0 errors in performance
Elapsed time at end of performance: real: 9.970s, CPU: 0.555s
1723 512 sample blks of 64-bit floats written to dac
rory@lenny:~/Desktop$ csound --version
virtual_keyboard real time MIDI plugin for Csound
0dBFS level = 32768.0
Csound version 6.03.2 (double samples) Jul 24 2014
libsndfile-1.0.25
unknown long option: '--version'
end of score.           overall amps:      0.0
       overall samples out of range:        0
0 errors in performance
Elapsed time at end of performance: real: 0.000s, CPU: 0.000s
rory@lenny:~/Desktop$

On 25 August 2014 13:02, Kevin Welsh  wrote:
> Right, at least theoretically... but in the first example (original
> email) that is definitely not happening, and I can't see any reason as
> to why!
>
> Any suggestions are welcome, but based off the conversations Michael
> and I had troubleshooting the string variable problem earlier, I'm
> almost certain there's an underlying bug here.
>
> On Mon, Aug 25, 2014 at 6:56 AM, Rory Walsh  wrote:
>> So the table is only be recreated if some of its arguments change from
>> note to note. All clear now.
>>
>> On 25 August 2014 12:52, Michael Gogins  wrote:
>>> You are wrong. The manual is correct.
>>>
>>> Regards,
>>> Mike
>>>
>>> On Aug 25, 2014 6:31 AM, "Rory Walsh"  wrote:
>>>>
>>>> I could be wrong but my understanding is that ftgenonce will generate
>>>> a function table for each instance of the instrument. ftgen on the
>>>> other hand, when called in the orchestra header, will create only one
>>>> instance of the function table. I think the idea with ftgenonce is
>>>> that one can encapsulate a function table into an instrument,
>>>> therefore making it possible to include entire instruments using
>>>> #include statements.
>>>>
>>>> On 25 August 2014 09:14, Kevin Welsh  wrote:
>>>> > Hello all,
>>>> >
>>>> > I'm seeing more tables being generated from ftgenonce than I expected
>>>> > again.  Csound version is 6.03.2 (double samples) Aug 21 2014, which
>>>> > was built from the developer branch.
>>>> >
>>>> > In the attached example, there are two tables being generated both
>>>> > with gen7.  One is a very short envelope, and the other a triangle
>>>> > wave form.  For some reason the triangle table will generate with each
>>>> > note in this example.
>>>> >
>>>> > In the project I'm working on, the behavior gets even weirder.  I have
>>>> > a metronome instrument that generates beeps over a duration by adding
>>>> > score entries using scoreline, it's envelope is the "ft_env" table
>>>> > from my example, also generated with ftgenonce.  But if I add this
>>>> > problematic triangle wave to a *completely* new instrument, the
>>>> > envelope for the metronome instrument will generate a new ftable on
>>>> > every single new note, even tho nothing had changed in that
>>>> > instrument!
>>>> >
>>>> > Here's the stripped down example:
>>>> > 
>>>> > 
>>>> > ; Select audio/midi flags here according to platform
>>>> > -odac     ;;;realtime audio out
>>>> > ;-iadc    ;;;uncomment -iadc if realtime audio input is needed too
>>>> > ; For Non-realtime ouput leave only the line below:
>>>> > ; -o ftgenonce.wav -W ;;; for file output any platform
>>>> > 
>>>> > 
>>>> >
>>>> > sr = 44100
>>>> > ksmps = 32
>>>> > nchnls = 2
>>>> > 0dbfs  = 1
>>>> >
>>>> > instr 1
>>>> >   ift_env ftgenonce 0, -99, 1024, 7, 0, 14, 1, 110, 0, 900, 0
>>>> > endin
>>>> >
>>>> > instr 2
>>>> >   ift_wav ftgenonce 0, -99, 1024, 7, 0, 256, 1, 512, -1, 256, 0
>>>> > endin
>>>> >
>>>> > 
>>>> > 
>>>> >
>>>> > i 1 1 1
>>>> > i 1 2 1
>>>> > i 1 3 1
>>>> > i 1 4 1
>>>> >
>>>> > i 1 6 1
>>>> > i 1 7 1
>>>> > i 1 8 1
>>>> > i 1 9 1
>>>> >
>>>> > i 2 6 4
>>>> > i 2 7 1
>>>> > i 2 8 1
>>>> > i 2 9 1
>>>> > e
>>>> > 
>>>> > 
>>>> >
>>>> >
>>>> > Send bugs reports to
>>>> >         https://github.com/csound/csound/issues
>>>> > 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
>>>>         https://github.com/csound/csound/issues
>>>> 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
>>         https://github.com/csound/csound/issues
>> 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
>         https://github.com/csound/csound/issues
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"
>
>
>

Thanks for running the test Rory, that makes it sound like this might
be a regression introduced when fixing the string variable bug.

Here's my output for reference (Csound version is 6.03.2 (double
samples) Aug 21 2014)

SECTION 1:
B  0.000 ..  1.000 T  1.000 TT  1.000 M:  0.00000  0.00000
new alloc for instr 1:
ftable 101:
 calling draw callback
ftable 101:     1024 points, scalemax 1.000
 .
  '
   -
    -
     .
      _
       _

        '
_________-_____________________________________________________________________
ftgenonce: created new func: 101
B  1.000 ..  2.000 T  2.000 TT  2.000 M:  0.00000  0.00000
B  2.000 ..  3.000 T  3.000 TT  3.000 M:  0.00000  0.00000
B  3.000 ..  4.000 T  4.000 TT  4.000 M:  0.00000  0.00000
B  4.000 ..  6.000 T  6.000 TT  6.000 M:  0.00000  0.00000
new alloc for instr 2:
ftable 102:
 calling draw callback
ftable 102:     1024 points, scalemax 1.000
                  _-'.
                _-    '.
              _-        '.
            _-            '.
          _-                '.
        _-                    '.
      _-                        '_
    _-                            -_
  _-                                -_
_-____________________________________-________________________________________
                                        -_                                   _-
                                          -_                               _-
                                            -_                           _-
                                              -_                       _-
                                                -_                   _-
                                                  -_               _-
                                                    -_           _-
                                                      -_        -
                                                        -_    .'
                                                          -_.'
ftgenonce: created new func: 102
B  6.000 ..  7.000 T  7.000 TT  7.000 M:  0.00000  0.00000
ftable 103:
 calling draw callback
ftable 103:     1024 points, scalemax 1.000
                  _-'.
                _-    '.
              _-        '.
            _-            '.
          _-                '.
        _-                    '.
      _-                        '_
    _-                            -_
  _-                                -_
_-____________________________________-________________________________________
                                        -_                                   _-
                                          -_                               _-
                                            -_                           _-
                                              -_                       _-
                                                -_                   _-
                                                  -_               _-
                                                    -_           _-
                                                      -_        -
                                                        -_    .'
                                                          -_.'
ftgenonce: created new func: 103
B  7.000 ..  8.000 T  8.000 TT  8.000 M:  0.00000  0.00000
ftable 104:
 calling draw callback
ftable 104:     1024 points, scalemax 1.000
                  _-'.
                _-    '.
              _-        '.
            _-            '.
          _-                '.
        _-                    '.
      _-                        '_
    _-                            -_
  _-                                -_
_-____________________________________-________________________________________
                                        -_                                   _-
                                          -_                               _-
                                            -_                           _-
                                              -_                       _-
                                                -_                   _-
                                                  -_               _-
                                                    -_           _-
                                                      -_        -
                                                        -_    .'
                                                          -_.'
ftgenonce: created new func: 104
B  8.000 ..  9.000 T  9.000 TT  9.000 M:  0.00000  0.00000
ftable 105:
 calling draw callback
ftable 105:     1024 points, scalemax 1.000
                  _-'.
                _-    '.
              _-        '.
            _-            '.
          _-                '.
        _-                    '.
      _-                        '_
    _-                            -_
  _-                                -_
_-____________________________________-________________________________________
                                        -_                                   _-
                                          -_                               _-
                                            -_                           _-
                                              -_                       _-
                                                -_                   _-
                                                  -_               _-
                                                    -_           _-
                                                      -_        -
                                                        -_    .'
                                                          -_.'
ftgenonce: created new func: 105
B  9.000 .. 10.000 T 10.000 TT 10.000 M:  0.00000  0.00000
Score finished in csoundPerform().
inactive allocs returned to freespace
end of score.              overall amps:  0.00000  0.00000
           overall samples out of range:        0        0
0 errors in performance
Elapsed time at end of performance: real: 10.092s, CPU: 0.103s

On Mon, Aug 25, 2014 at 7:06 AM, Rory Walsh  wrote:
> I should have tried your first example first :) For me it seems to
> work fine. I'm using Csound version 6.03.2 (double samples) Jul 24
> 2014, Linux. The output is posted below. Each f-table only seems to be
> generated onece?
>
> SECTION 1:
> B  0.000 ..  1.000 T  1.000 TT  1.000 M:  0.00000  0.00000
> new alloc for instr 1:
> ftable 101:
>  calling draw callback
> ftable 101:    1024 points, scalemax 1.000
>  .
>   '
>    -
>     -
>      .
>       _
>        _
>
>         '
> _________-_____________________________________________________________________
> ftgenonce: created new func: 101
> B  1.000 ..  2.000 T  2.000 TT  2.000 M:  0.00000  0.00000
> B  2.000 ..  3.000 T  3.000 TT  3.000 M:  0.00000  0.00000
> B  3.000 ..  4.000 T  4.000 TT  4.000 M:  0.00000  0.00000
> B  4.000 ..  6.000 T  6.000 TT  6.000 M:  0.00000  0.00000
> new alloc for instr 2:
> ftable 102:
>  calling draw callback
> ftable 102:    1024 points, scalemax 1.000
>                   _-'.
>                 _-    '.
>               _-        '.
>             _-            '.
>           _-                '.
>         _-                    '.
>       _-                        '_
>     _-                            -_
>   _-                                -_
> _-____________________________________-________________________________________
>                                         -_                                   _-
>                                           -_                               _-
>                                             -_                           _-
>                                               -_                       _-
>                                                 -_                   _-
>                                                   -_               _-
>                                                     -_           _-
>                                                       -_        -
>                                                         -_    .'
>                                                           -_.'
> ftgenonce: created new func: 102
> B  6.000 ..  7.000 T  7.000 TT  7.000 M:  0.00000  0.00000
> new alloc for instr 2:
> B  7.000 ..  8.000 T  8.000 TT  8.000 M:  0.00000  0.00000
> B  8.000 ..  9.000 T  9.000 TT  9.000 M:  0.00000  0.00000
> B  9.000 .. 10.000 T 10.000 TT 10.000 M:  0.00000  0.00000
> Score finished in csoundPerform().
> inactive allocs returned to freespace
> end of score.           overall amps:  0.00000  0.00000
>        overall samples out of range:        0        0
> 0 errors in performance
> Elapsed time at end of performance: real: 9.970s, CPU: 0.555s
> 1723 512 sample blks of 64-bit floats written to dac
> rory@lenny:~/Desktop$ csound --version
> virtual_keyboard real time MIDI plugin for Csound
> 0dBFS level = 32768.0
> Csound version 6.03.2 (double samples) Jul 24 2014
> libsndfile-1.0.25
> unknown long option: '--version'
> end of score.           overall amps:      0.0
>        overall samples out of range:        0
> 0 errors in performance
> Elapsed time at end of performance: real: 0.000s, CPU: 0.000s
> rory@lenny:~/Desktop$
>
> On 25 August 2014 13:02, Kevin Welsh  wrote:
>> Right, at least theoretically... but in the first example (original
>> email) that is definitely not happening, and I can't see any reason as
>> to why!
>>
>> Any suggestions are welcome, but based off the conversations Michael
>> and I had troubleshooting the string variable problem earlier, I'm
>> almost certain there's an underlying bug here.
>>
>> On Mon, Aug 25, 2014 at 6:56 AM, Rory Walsh  wrote:
>>> So the table is only be recreated if some of its arguments change from
>>> note to note. All clear now.
>>>
>>> On 25 August 2014 12:52, Michael Gogins  wrote:
>>>> You are wrong. The manual is correct.
>>>>
>>>> Regards,
>>>> Mike
>>>>
>>>> On Aug 25, 2014 6:31 AM, "Rory Walsh"  wrote:
>>>>>
>>>>> I could be wrong but my understanding is that ftgenonce will generate
>>>>> a function table for each instance of the instrument. ftgen on the
>>>>> other hand, when called in the orchestra header, will create only one
>>>>> instance of the function table. I think the idea with ftgenonce is
>>>>> that one can encapsulate a function table into an instrument,
>>>>> therefore making it possible to include entire instruments using
>>>>> #include statements.
>>>>>
>>>>> On 25 August 2014 09:14, Kevin Welsh  wrote:
>>>>> > Hello all,
>>>>> >
>>>>> > I'm seeing more tables being generated from ftgenonce than I expected
>>>>> > again.  Csound version is 6.03.2 (double samples) Aug 21 2014, which
>>>>> > was built from the developer branch.
>>>>> >
>>>>> > In the attached example, there are two tables being generated both
>>>>> > with gen7.  One is a very short envelope, and the other a triangle
>>>>> > wave form.  For some reason the triangle table will generate with each
>>>>> > note in this example.
>>>>> >
>>>>> > In the project I'm working on, the behavior gets even weirder.  I have
>>>>> > a metronome instrument that generates beeps over a duration by adding
>>>>> > score entries using scoreline, it's envelope is the "ft_env" table
>>>>> > from my example, also generated with ftgenonce.  But if I add this
>>>>> > problematic triangle wave to a *completely* new instrument, the
>>>>> > envelope for the metronome instrument will generate a new ftable on
>>>>> > every single new note, even tho nothing had changed in that
>>>>> > instrument!
>>>>> >
>>>>> > Here's the stripped down example:
>>>>> > 
>>>>> > 
>>>>> > ; Select audio/midi flags here according to platform
>>>>> > -odac     ;;;realtime audio out
>>>>> > ;-iadc    ;;;uncomment -iadc if realtime audio input is needed too
>>>>> > ; For Non-realtime ouput leave only the line below:
>>>>> > ; -o ftgenonce.wav -W ;;; for file output any platform
>>>>> > 
>>>>> > 
>>>>> >
>>>>> > sr = 44100
>>>>> > ksmps = 32
>>>>> > nchnls = 2
>>>>> > 0dbfs  = 1
>>>>> >
>>>>> > instr 1
>>>>> >   ift_env ftgenonce 0, -99, 1024, 7, 0, 14, 1, 110, 0, 900, 0
>>>>> > endin
>>>>> >
>>>>> > instr 2
>>>>> >   ift_wav ftgenonce 0, -99, 1024, 7, 0, 256, 1, 512, -1, 256, 0
>>>>> > endin
>>>>> >
>>>>> > 
>>>>> > 
>>>>> >
>>>>> > i 1 1 1
>>>>> > i 1 2 1
>>>>> > i 1 3 1
>>>>> > i 1 4 1
>>>>> >
>>>>> > i 1 6 1
>>>>> > i 1 7 1
>>>>> > i 1 8 1
>>>>> > i 1 9 1
>>>>> >
>>>>> > i 2 6 4
>>>>> > i 2 7 1
>>>>> > i 2 8 1
>>>>> > i 2 9 1
>>>>> > e
>>>>> > 
>>>>> > 
>>>>> >
>>>>> >
>>>>> > Send bugs reports to
>>>>> >         https://github.com/csound/csound/issues
>>>>> > 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
>>>>>         https://github.com/csound/csound/issues
>>>>> 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
>>>         https://github.com/csound/csound/issues
>>> 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
>>         https://github.com/csound/csound/issues
>> 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
>         https://github.com/csound/csound/issues
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"
>
>
>

Looks like it.

On 25 August 2014 13:14, Kevin Welsh  wrote:
> Thanks for running the test Rory, that makes it sound like this might
> be a regression introduced when fixing the string variable bug.
>
> Here's my output for reference (Csound version is 6.03.2 (double
> samples) Aug 21 2014)
>
> SECTION 1:
> B  0.000 ..  1.000 T  1.000 TT  1.000 M:  0.00000  0.00000
> new alloc for instr 1:
> ftable 101:
>  calling draw callback
> ftable 101:     1024 points, scalemax 1.000
>  .
>   '
>    -
>     -
>      .
>       _
>        _
>
>         '
> _________-_____________________________________________________________________
> ftgenonce: created new func: 101
> B  1.000 ..  2.000 T  2.000 TT  2.000 M:  0.00000  0.00000
> B  2.000 ..  3.000 T  3.000 TT  3.000 M:  0.00000  0.00000
> B  3.000 ..  4.000 T  4.000 TT  4.000 M:  0.00000  0.00000
> B  4.000 ..  6.000 T  6.000 TT  6.000 M:  0.00000  0.00000
> new alloc for instr 2:
> ftable 102:
>  calling draw callback
> ftable 102:     1024 points, scalemax 1.000
>                   _-'.
>                 _-    '.
>               _-        '.
>             _-            '.
>           _-                '.
>         _-                    '.
>       _-                        '_
>     _-                            -_
>   _-                                -_
> _-____________________________________-________________________________________
>                                         -_                                   _-
>                                           -_                               _-
>                                             -_                           _-
>                                               -_                       _-
>                                                 -_                   _-
>                                                   -_               _-
>                                                     -_           _-
>                                                       -_        -
>                                                         -_    .'
>                                                           -_.'
> ftgenonce: created new func: 102
> B  6.000 ..  7.000 T  7.000 TT  7.000 M:  0.00000  0.00000
> ftable 103:
>  calling draw callback
> ftable 103:     1024 points, scalemax 1.000
>                   _-'.
>                 _-    '.
>               _-        '.
>             _-            '.
>           _-                '.
>         _-                    '.
>       _-                        '_
>     _-                            -_
>   _-                                -_
> _-____________________________________-________________________________________
>                                         -_                                   _-
>                                           -_                               _-
>                                             -_                           _-
>                                               -_                       _-
>                                                 -_                   _-
>                                                   -_               _-
>                                                     -_           _-
>                                                       -_        -
>                                                         -_    .'
>                                                           -_.'
> ftgenonce: created new func: 103
> B  7.000 ..  8.000 T  8.000 TT  8.000 M:  0.00000  0.00000
> ftable 104:
>  calling draw callback
> ftable 104:     1024 points, scalemax 1.000
>                   _-'.
>                 _-    '.
>               _-        '.
>             _-            '.
>           _-                '.
>         _-                    '.
>       _-                        '_
>     _-                            -_
>   _-                                -_
> _-____________________________________-________________________________________
>                                         -_                                   _-
>                                           -_                               _-
>                                             -_                           _-
>                                               -_                       _-
>                                                 -_                   _-
>                                                   -_               _-
>                                                     -_           _-
>                                                       -_        -
>                                                         -_    .'
>                                                           -_.'
> ftgenonce: created new func: 104
> B  8.000 ..  9.000 T  9.000 TT  9.000 M:  0.00000  0.00000
> ftable 105:
>  calling draw callback
> ftable 105:     1024 points, scalemax 1.000
>                   _-'.
>                 _-    '.
>               _-        '.
>             _-            '.
>           _-                '.
>         _-                    '.
>       _-                        '_
>     _-                            -_
>   _-                                -_
> _-____________________________________-________________________________________
>                                         -_                                   _-
>                                           -_                               _-
>                                             -_                           _-
>                                               -_                       _-
>                                                 -_                   _-
>                                                   -_               _-
>                                                     -_           _-
>                                                       -_        -
>                                                         -_    .'
>                                                           -_.'
> ftgenonce: created new func: 105
> B  9.000 .. 10.000 T 10.000 TT 10.000 M:  0.00000  0.00000
> Score finished in csoundPerform().
> inactive allocs returned to freespace
> end of score.              overall amps:  0.00000  0.00000
>            overall samples out of range:        0        0
> 0 errors in performance
> Elapsed time at end of performance: real: 10.092s, CPU: 0.103s
>
> On Mon, Aug 25, 2014 at 7:06 AM, Rory Walsh  wrote:
>> I should have tried your first example first :) For me it seems to
>> work fine. I'm using Csound version 6.03.2 (double samples) Jul 24
>> 2014, Linux. The output is posted below. Each f-table only seems to be
>> generated onece?
>>
>> SECTION 1:
>> B  0.000 ..  1.000 T  1.000 TT  1.000 M:  0.00000  0.00000
>> new alloc for instr 1:
>> ftable 101:
>>  calling draw callback
>> ftable 101:    1024 points, scalemax 1.000
>>  .
>>   '
>>    -
>>     -
>>      .
>>       _
>>        _
>>
>>         '
>> _________-_____________________________________________________________________
>> ftgenonce: created new func: 101
>> B  1.000 ..  2.000 T  2.000 TT  2.000 M:  0.00000  0.00000
>> B  2.000 ..  3.000 T  3.000 TT  3.000 M:  0.00000  0.00000
>> B  3.000 ..  4.000 T  4.000 TT  4.000 M:  0.00000  0.00000
>> B  4.000 ..  6.000 T  6.000 TT  6.000 M:  0.00000  0.00000
>> new alloc for instr 2:
>> ftable 102:
>>  calling draw callback
>> ftable 102:    1024 points, scalemax 1.000
>>                   _-'.
>>                 _-    '.
>>               _-        '.
>>             _-            '.
>>           _-                '.
>>         _-                    '.
>>       _-                        '_
>>     _-                            -_
>>   _-                                -_
>> _-____________________________________-________________________________________
>>                                         -_                                   _-
>>                                           -_                               _-
>>                                             -_                           _-
>>                                               -_                       _-
>>                                                 -_                   _-
>>                                                   -_               _-
>>                                                     -_           _-
>>                                                       -_        -
>>                                                         -_    .'
>>                                                           -_.'
>> ftgenonce: created new func: 102
>> B  6.000 ..  7.000 T  7.000 TT  7.000 M:  0.00000  0.00000
>> new alloc for instr 2:
>> B  7.000 ..  8.000 T  8.000 TT  8.000 M:  0.00000  0.00000
>> B  8.000 ..  9.000 T  9.000 TT  9.000 M:  0.00000  0.00000
>> B  9.000 .. 10.000 T 10.000 TT 10.000 M:  0.00000  0.00000
>> Score finished in csoundPerform().
>> inactive allocs returned to freespace
>> end of score.           overall amps:  0.00000  0.00000
>>        overall samples out of range:        0        0
>> 0 errors in performance
>> Elapsed time at end of performance: real: 9.970s, CPU: 0.555s
>> 1723 512 sample blks of 64-bit floats written to dac
>> rory@lenny:~/Desktop$ csound --version
>> virtual_keyboard real time MIDI plugin for Csound
>> 0dBFS level = 32768.0
>> Csound version 6.03.2 (double samples) Jul 24 2014
>> libsndfile-1.0.25
>> unknown long option: '--version'
>> end of score.           overall amps:      0.0
>>        overall samples out of range:        0
>> 0 errors in performance
>> Elapsed time at end of performance: real: 0.000s, CPU: 0.000s
>> rory@lenny:~/Desktop$
>>
>> On 25 August 2014 13:02, Kevin Welsh  wrote:
>>> Right, at least theoretically... but in the first example (original
>>> email) that is definitely not happening, and I can't see any reason as
>>> to why!
>>>
>>> Any suggestions are welcome, but based off the conversations Michael
>>> and I had troubleshooting the string variable problem earlier, I'm
>>> almost certain there's an underlying bug here.
>>>
>>> On Mon, Aug 25, 2014 at 6:56 AM, Rory Walsh  wrote:
>>>> So the table is only be recreated if some of its arguments change from
>>>> note to note. All clear now.
>>>>
>>>> On 25 August 2014 12:52, Michael Gogins  wrote:
>>>>> You are wrong. The manual is correct.
>>>>>
>>>>> Regards,
>>>>> Mike
>>>>>
>>>>> On Aug 25, 2014 6:31 AM, "Rory Walsh"  wrote:
>>>>>>
>>>>>> I could be wrong but my understanding is that ftgenonce will generate
>>>>>> a function table for each instance of the instrument. ftgen on the
>>>>>> other hand, when called in the orchestra header, will create only one
>>>>>> instance of the function table. I think the idea with ftgenonce is
>>>>>> that one can encapsulate a function table into an instrument,
>>>>>> therefore making it possible to include entire instruments using
>>>>>> #include statements.
>>>>>>
>>>>>> On 25 August 2014 09:14, Kevin Welsh  wrote:
>>>>>> > Hello all,
>>>>>> >
>>>>>> > I'm seeing more tables being generated from ftgenonce than I expected
>>>>>> > again.  Csound version is 6.03.2 (double samples) Aug 21 2014, which
>>>>>> > was built from the developer branch.
>>>>>> >
>>>>>> > In the attached example, there are two tables being generated both
>>>>>> > with gen7.  One is a very short envelope, and the other a triangle
>>>>>> > wave form.  For some reason the triangle table will generate with each
>>>>>> > note in this example.
>>>>>> >
>>>>>> > In the project I'm working on, the behavior gets even weirder.  I have
>>>>>> > a metronome instrument that generates beeps over a duration by adding
>>>>>> > score entries using scoreline, it's envelope is the "ft_env" table
>>>>>> > from my example, also generated with ftgenonce.  But if I add this
>>>>>> > problematic triangle wave to a *completely* new instrument, the
>>>>>> > envelope for the metronome instrument will generate a new ftable on
>>>>>> > every single new note, even tho nothing had changed in that
>>>>>> > instrument!
>>>>>> >
>>>>>> > Here's the stripped down example:
>>>>>> > 
>>>>>> > 
>>>>>> > ; Select audio/midi flags here according to platform
>>>>>> > -odac     ;;;realtime audio out
>>>>>> > ;-iadc    ;;;uncomment -iadc if realtime audio input is needed too
>>>>>> > ; For Non-realtime ouput leave only the line below:
>>>>>> > ; -o ftgenonce.wav -W ;;; for file output any platform
>>>>>> > 
>>>>>> > 
>>>>>> >
>>>>>> > sr = 44100
>>>>>> > ksmps = 32
>>>>>> > nchnls = 2
>>>>>> > 0dbfs  = 1
>>>>>> >
>>>>>> > instr 1
>>>>>> >   ift_env ftgenonce 0, -99, 1024, 7, 0, 14, 1, 110, 0, 900, 0
>>>>>> > endin
>>>>>> >
>>>>>> > instr 2
>>>>>> >   ift_wav ftgenonce 0, -99, 1024, 7, 0, 256, 1, 512, -1, 256, 0
>>>>>> > endin
>>>>>> >
>>>>>> > 
>>>>>> > 
>>>>>> >
>>>>>> > i 1 1 1
>>>>>> > i 1 2 1
>>>>>> > i 1 3 1
>>>>>> > i 1 4 1
>>>>>> >
>>>>>> > i 1 6 1
>>>>>> > i 1 7 1
>>>>>> > i 1 8 1
>>>>>> > i 1 9 1
>>>>>> >
>>>>>> > i 2 6 4
>>>>>> > i 2 7 1
>>>>>> > i 2 8 1
>>>>>> > i 2 9 1
>>>>>> > e
>>>>>> > 
>>>>>> > 
>>>>>> >
>>>>>> >
>>>>>> > Send bugs reports to
>>>>>> >         https://github.com/csound/csound/issues
>>>>>> > 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
>>>>>>         https://github.com/csound/csound/issues
>>>>>> 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
>>>>         https://github.com/csound/csound/issues
>>>> 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
>>>         https://github.com/csound/csound/issues
>>> 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
>>         https://github.com/csound/csound/issues
>> 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
>         https://github.com/csound/csound/issues
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"
>
>
>

Michael, this seems to have fixed the problem exhibited in the test,
but I'm seeing a lot of other regressions (I'm not sure if they're
form this change, or others).  I mentioned min/max and pset
regressions in another email.


But I also think ftgenonce also regressed with regard to the behavior
on string values.  My previous test for that was as follows:



; Select audio/midi flags here according to platform
-odac     ;;;realtime audio out
;-iadc    ;;;uncomment -iadc if realtime audio input is needed too
; For Non-realtime ouput leave only the line below:
; -o ftgenonce.wav -W ;;; for file output any platform



sr = 44100
ksmps = 32
nchnls = 2
0dbfs  = 1

instr 1
  ift ftgenonce 0, -1, 0, -1, "kickroll.wav", 0, 0, 0
endin

instr 2
  Sfile = "kickroll.wav"
  ift ftgenonce 0, -1, 0, -1, Sfile, 0, 0, 0
endin




i 1 1 10
i 1 2 10
i 2 1 10
i 2 2 10
e



This results in 3 tables being generated for kickroll.wav, when I
believe only one should have been.

On Thu, Aug 28, 2014 at 8:38 AM, Michael Gogins
 wrote:
> I think I have finally fixed this. My key comparator code was too
> complicated and had a bug I couldn't find, so after thinking about it, I
> just simplified the code. I have successfully tested with the attached
> files.
>
> The fix is now pushed to GIT.
>
> Regards,
> Mike
>
>
> -----------------------------------------------------
> Michael Gogins
> Irreducible Productions
> http://michaelgogins.tumblr.com
> Michael dot Gogins at gmail dot com
>
>
> On Wed, Aug 27, 2014 at 12:43 PM, Michael Gogins 
> wrote:
>>
>> Not in this case you can't. I already tried it. -ffast-math messes that up
>> too.
>>
>> Regards,
>> Mike
>>
>>
>> -----------------------------------------------------
>> Michael Gogins
>> Irreducible Productions
>> http://michaelgogins.tumblr.com
>> Michael dot Gogins at gmail dot com
>>
>>
>> On Wed, Aug 27, 2014 at 12:35 PM, Justin Smith 
>> wrote:
>>>
>>> You can also replace isnan with a != a (which can only be true if an
>>> object is NaN).
>>>
>>>
>>> On Wed, Aug 27, 2014 at 5:41 AM, Michael Gogins
>>>  wrote:
>>>>
>>>> There seem to be two problems. One was a bug in my code that I have
>>>> fixed. The other is that if -ffast-math is passed to gcc, isnan does not
>>>> work, but isnan is used by Csound to identify string pfields. Tonight I will
>>>> rebuild without -ffast-math and try again.
>>>>
>>>> Regards,
>>>> Mike
>>>>
>>>>
>>>> -----------------------------------------------------
>>>> Michael Gogins
>>>> Irreducible Productions
>>>> http://michaelgogins.tumblr.com
>>>> Michael dot Gogins at gmail dot com
>>>>
>>>>
>>>> On Mon, Aug 25, 2014 at 3:14 AM, Kevin Welsh  wrote:
>>>>>
>>>>> Hello all,
>>>>>
>>>>> I'm seeing more tables being generated from ftgenonce than I expected
>>>>> again.  Csound version is 6.03.2 (double samples) Aug 21 2014, which
>>>>> was built from the developer branch.
>>>>>
>>>>> In the attached example, there are two tables being generated both
>>>>> with gen7.  One is a very short envelope, and the other a triangle
>>>>> wave form.  For some reason the triangle table will generate with each
>>>>> note in this example.
>>>>>
>>>>> In the project I'm working on, the behavior gets even weirder.  I have
>>>>> a metronome instrument that generates beeps over a duration by adding
>>>>> score entries using scoreline, it's envelope is the "ft_env" table
>>>>> from my example, also generated with ftgenonce.  But if I add this
>>>>> problematic triangle wave to a *completely* new instrument, the
>>>>> envelope for the metronome instrument will generate a new ftable on
>>>>> every single new note, even tho nothing had changed in that
>>>>> instrument!
>>>>>
>>>>> Here's the stripped down example:
>>>>> 
>>>>> 
>>>>> ; Select audio/midi flags here according to platform
>>>>> -odac     ;;;realtime audio out
>>>>> ;-iadc    ;;;uncomment -iadc if realtime audio input is needed too
>>>>> ; For Non-realtime ouput leave only the line below:
>>>>> ; -o ftgenonce.wav -W ;;; for file output any platform
>>>>> 
>>>>> 
>>>>>
>>>>> sr = 44100
>>>>> ksmps = 32
>>>>> nchnls = 2
>>>>> 0dbfs  = 1
>>>>>
>>>>> instr 1
>>>>>   ift_env ftgenonce 0, -99, 1024, 7, 0, 14, 1, 110, 0, 900, 0
>>>>> endin
>>>>>
>>>>> instr 2
>>>>>   ift_wav ftgenonce 0, -99, 1024, 7, 0, 256, 1, 512, -1, 256, 0
>>>>> endin
>>>>>
>>>>> 
>>>>> 
>>>>>
>>>>> i 1 1 1
>>>>> i 1 2 1
>>>>> i 1 3 1
>>>>> i 1 4 1
>>>>>
>>>>> i 1 6 1
>>>>> i 1 7 1
>>>>> i 1 8 1
>>>>> i 1 9 1
>>>>>
>>>>> i 2 6 4
>>>>> i 2 7 1
>>>>> i 2 8 1
>>>>> i 2 9 1
>>>>> e
>>>>> 
>>>>> 
>>>>>
>>>>>
>>>>> Send bugs reports to
>>>>>         https://github.com/csound/csound/issues
>>>>> Discussions of bugs and features can be posted here
>>>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body
>>>>> "unsubscribe csound"
>>>>>
>>>>>
>>>>>
>>>>
>>>
>>
>

Weird, I use OS-X so it's probably not a .dll, but perhaps a .dylib?

I did "brew uninstall csound" to try to start fresh and clean.  After
doing that, I found these remaining on the system:

-rwxrwxr-x  1 root  wheel  353408 Jan 20  2013
/Library/Frameworks/CsoundLib.framework/Versions/5.2/Resources/Opcodes/libsignalflowgraph.dylib
-rwxrwxr-x  1 root  wheel  353408 Jan 20  2013
/Library/Frameworks/CsoundLib64.framework/Versions/5.2/Resources/Opcodes64/libsignalflowgraph.dylib
-rwxrwxr-x  1 root  wheel  461608 May 12 16:16
/Library/Frameworks/CsoundLib64.framework/Versions/6.0/Resources/Opcodes64/libsignalflowgraph.dylib
-rwxr-xr-x  1 tgrey  staff  224332 Aug 28 12:58
/Users/tgrey/Library/Frameworks/CsoundLib64.framework/Versions/6.0/Resources/Opcodes64/libsignalflowgraph.dylib

Perhaps I should try removing these files and building from scratch again?

On Thu, Aug 28, 2014 at 2:09 PM, Michael Gogins
 wrote:
> The other regressions are surely unrelated to the ftgenonce changes.
>
> For your test above, I get ONE ftable.
>
> For my test with 2 samples with slightly different filenames, I get TWO
> ftables.
>
> Do you have an older signalflowgraph.dll lying around somewhere, getting
> loaded first?
>
> Regards,
> Mike
>
>
> -----------------------------------------------------
> Michael Gogins
> Irreducible Productions
> http://michaelgogins.tumblr.com
> Michael dot Gogins at gmail dot com
>
>
> On Thu, Aug 28, 2014 at 1:50 PM, Kevin Welsh  wrote:
>>
>> Michael, this seems to have fixed the problem exhibited in the test,
>> but I'm seeing a lot of other regressions (I'm not sure if they're
>> form this change, or others).  I mentioned min/max and pset
>> regressions in another email.
>>
>>
>> But I also think ftgenonce also regressed with regard to the behavior
>> on string values.  My previous test for that was as follows:
>>
>> 
>> 
>> ; Select audio/midi flags here according to platform
>> -odac     ;;;realtime audio out
>> ;-iadc    ;;;uncomment -iadc if realtime audio input is needed too
>> ; For Non-realtime ouput leave only the line below:
>> ; -o ftgenonce.wav -W ;;; for file output any platform
>> 
>> 
>>
>> sr = 44100
>> ksmps = 32
>> nchnls = 2
>> 0dbfs  = 1
>>
>> instr 1
>>   ift ftgenonce 0, -1, 0, -1, "kickroll.wav", 0, 0, 0
>> endin
>>
>> instr 2
>>   Sfile = "kickroll.wav"
>>   ift ftgenonce 0, -1, 0, -1, Sfile, 0, 0, 0
>> endin
>>
>> 
>> 
>>
>> i 1 1 10
>> i 1 2 10
>> i 2 1 10
>> i 2 2 10
>> e
>> 
>> 
>>
>> This results in 3 tables being generated for kickroll.wav, when I
>> believe only one should have been.
>>
>> On Thu, Aug 28, 2014 at 8:38 AM, Michael Gogins
>>  wrote:
>> > I think I have finally fixed this. My key comparator code was too
>> > complicated and had a bug I couldn't find, so after thinking about it, I
>> > just simplified the code. I have successfully tested with the attached
>> > files.
>> >
>> > The fix is now pushed to GIT.
>> >
>> > Regards,
>> > Mike
>> >
>> >
>> > -----------------------------------------------------
>> > Michael Gogins
>> > Irreducible Productions
>> > http://michaelgogins.tumblr.com
>> > Michael dot Gogins at gmail dot com
>> >
>> >
>> > On Wed, Aug 27, 2014 at 12:43 PM, Michael Gogins
>> > 
>> > wrote:
>> >>
>> >> Not in this case you can't. I already tried it. -ffast-math messes that
>> >> up
>> >> too.
>> >>
>> >> Regards,
>> >> Mike
>> >>
>> >>
>> >> -----------------------------------------------------
>> >> Michael Gogins
>> >> Irreducible Productions
>> >> http://michaelgogins.tumblr.com
>> >> Michael dot Gogins at gmail dot com
>> >>
>> >>
>> >> On Wed, Aug 27, 2014 at 12:35 PM, Justin Smith 
>> >> wrote:
>> >>>
>> >>> You can also replace isnan with a != a (which can only be true if an
>> >>> object is NaN).
>> >>>
>> >>>
>> >>> On Wed, Aug 27, 2014 at 5:41 AM, Michael Gogins
>> >>>  wrote:
>> >>>>
>> >>>> There seem to be two problems. One was a bug in my code that I have
>> >>>> fixed. The other is that if -ffast-math is passed to gcc, isnan does
>> >>>> not
>> >>>> work, but isnan is used by Csound to identify string pfields. Tonight
>> >>>> I will
>> >>>> rebuild without -ffast-math and try again.
>> >>>>
>> >>>> Regards,
>> >>>> Mike
>> >>>>
>> >>>>
>> >>>> -----------------------------------------------------
>> >>>> Michael Gogins
>> >>>> Irreducible Productions
>> >>>> http://michaelgogins.tumblr.com
>> >>>> Michael dot Gogins at gmail dot com
>> >>>>
>> >>>>
>> >>>> On Mon, Aug 25, 2014 at 3:14 AM, Kevin Welsh 
>> >>>> wrote:
>> >>>>>
>> >>>>> Hello all,
>> >>>>>
>> >>>>> I'm seeing more tables being generated from ftgenonce than I
>> >>>>> expected
>> >>>>> again.  Csound version is 6.03.2 (double samples) Aug 21 2014, which
>> >>>>> was built from the developer branch.
>> >>>>>
>> >>>>> In the attached example, there are two tables being generated both
>> >>>>> with gen7.  One is a very short envelope, and the other a triangle
>> >>>>> wave form.  For some reason the triangle table will generate with
>> >>>>> each
>> >>>>> note in this example.
>> >>>>>
>> >>>>> In the project I'm working on, the behavior gets even weirder.  I
>> >>>>> have
>> >>>>> a metronome instrument that generates beeps over a duration by
>> >>>>> adding
>> >>>>> score entries using scoreline, it's envelope is the "ft_env" table
>> >>>>> from my example, also generated with ftgenonce.  But if I add this
>> >>>>> problematic triangle wave to a *completely* new instrument, the
>> >>>>> envelope for the metronome instrument will generate a new ftable on
>> >>>>> every single new note, even tho nothing had changed in that
>> >>>>> instrument!
>> >>>>>
>> >>>>> Here's the stripped down example:
>> >>>>> 
>> >>>>> 
>> >>>>> ; Select audio/midi flags here according to platform
>> >>>>> -odac     ;;;realtime audio out
>> >>>>> ;-iadc    ;;;uncomment -iadc if realtime audio input is needed too
>> >>>>> ; For Non-realtime ouput leave only the line below:
>> >>>>> ; -o ftgenonce.wav -W ;;; for file output any platform
>> >>>>> 
>> >>>>> 
>> >>>>>
>> >>>>> sr = 44100
>> >>>>> ksmps = 32
>> >>>>> nchnls = 2
>> >>>>> 0dbfs  = 1
>> >>>>>
>> >>>>> instr 1
>> >>>>>   ift_env ftgenonce 0, -99, 1024, 7, 0, 14, 1, 110, 0, 900, 0
>> >>>>> endin
>> >>>>>
>> >>>>> instr 2
>> >>>>>   ift_wav ftgenonce 0, -99, 1024, 7, 0, 256, 1, 512, -1, 256, 0
>> >>>>> endin
>> >>>>>
>> >>>>> 
>> >>>>> 
>> >>>>>
>> >>>>> i 1 1 1
>> >>>>> i 1 2 1
>> >>>>> i 1 3 1
>> >>>>> i 1 4 1
>> >>>>>
>> >>>>> i 1 6 1
>> >>>>> i 1 7 1
>> >>>>> i 1 8 1
>> >>>>> i 1 9 1
>> >>>>>
>> >>>>> i 2 6 4
>> >>>>> i 2 7 1
>> >>>>> i 2 8 1
>> >>>>> i 2 9 1
>> >>>>> e
>> >>>>> 
>> >>>>> 
>> >>>>>
>> >>>>>
>> >>>>> Send bugs reports to
>> >>>>>         https://github.com/csound/csound/issues
>> >>>>> 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
>>         https://github.com/csound/csound/issues
>> Discussions of bugs and features can be posted here
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>> csound"
>>
>>
>>
>