This is what I expect to be in cs6.17

Any clarifications or suggestions please send to me soon as the release is
imminent.

==John ffitch



# CSOUND VERSION 6.17 RELEASE NOTES - DRAFT - DRAFT - DRAFT - DRAFT

Mainly a bug-fixing release but also a major re-organisation of the
libraries to move all opcodes with dependencies into a separate plugins
repository. The only exception to this is the OSC opcodes (which
depend on liblo). This is in part a preparation for Csound7. This is
the last release using the current parser (v.2).

-- The Developers

## USER-LEVEL CHANGES

### New opcodes

- scanmap and scansmap are like the xscanmap/xscansmap opcodes but
  work with the mainstream scan opcodes.

- trigexpseg, triglinseg are aliases for trigExpseg, trigLinseg.

- xscan opcodes are deprecated as they add nothing to the scan opcodes.

- bformdec2 provides more ambisonic decoding of bformat audio.

- metrobpm is like metro but the timing is in beats per minute and the
  signal can be offset by a proportion of the cycle.

- sequ is a csound version of an hardware sequencer.  It has many modes and options.

- gtadsr implements a gated ADSR envelope

### New gen

### Orchestra

- Message printing has been revised so -m16 suppresses all
    messages.

- add channel count to list_audio_devices (called when the flag
  --devices is used so that it can be parsed by frontends).

### Score

### Options

### Modified Opcodes and Gens

- event opcode does not bail out if the instrument called does not exist.

- Added an optional prefix to soundfont instruments printed via sfilist.

- lpslots reworked with better control.

### Utilities

- cvanal now uses the SADIR environment to look for analysis files.

### Frontends

### General Usage

## Bugs Fixed

- fareylen called a non-existent function leading to a crash.  Removed typo.

- turnoff could cause clicks in some cases; fixed.

- turnoff3 improved.

- cntReset fixed; had a false initialisation code.

- binary search in bpf fixed.

- pvscfs frame counting issue fixed.

- --sample-accurate fixed in a-rate form of tabsum.

- Problem in atssinnoi fixed; did read outside allocated memory.

- hrtfmove fixed; it could use the wrong value for sr.

- Named instruments could use wrong structure in redefinition.

- fixes in the jack backend relating to --get-system-sr.

- pvcross frequency warp mode fixed.

- lpcfilter/lpcanal bug where processing could not be restarted after
  a freeze fixed.

- the emugens collection of opcode, a large group incorporation mainly
  array operations, were not made available due to a small editing
  error.

- If using a float build (rather than a double) array access could be
  wrong.

- In some cases instr0 code could ignore loops after an incorrect
  warning.  Now fixed.

- Bugs in mp3len and similar fixed.

- Crash in syncgrain issue #1773 fixed

# SYSTEM LEVEL CHANGES

### System Changes

- winsound has been removed.

### Translations

### API

CreateThread2 is a new API function that allows threads to be given a
user-defined stack size.

### External Plugin Code

For a variety of reasons including licences, plugin opcodes that
allow csound to be a host for VST plugins is available from
https://github.com/gogins/csound-vst3-opcodes.git where installation
instructions can be found.  It has been shown to work on Linux, MacOS
and Windows.

### Platform Specific

==END==

Thnk it is fixed
==John ff

On Thu, 27 Jan 2022, Francois PINOT wrote:

> I just updated the what's new section of the manual accordingly. The manual
> entry for the gtadsr opcode has links to images/gtadsr.png and
> mgtadsr opcodes, but there are no such image and opcode.
> François
> 
> Le mer. 26 janv. 2022 à 18:53, John  a écrit :
>       This is what I expect to be in cs6.17
>
>       Any clarifications or suggestions please send to me soon as the
>       release is
>       imminent.
>
>       ==John ffitch
> 
> 
>
>       # CSOUND VERSION 6.17 RELEASE NOTES - DRAFT - DRAFT - DRAFT -
>       DRAFT
>
>       Mainly a bug-fixing release but also a major re-organisation of
>       the
>       libraries to move all opcodes with dependencies into a separate
>       plugins
>       repository. The only exception to this is the OSC opcodes (which
>       depend on liblo). This is in part a preparation for Csound7. This
>       is
>       the last release using the current parser (v.2).
>
>       -- The Developers
>
>       ## USER-LEVEL CHANGES
>
>       ### New opcodes
>
>       - scanmap and scansmap are like the xscanmap/xscansmap opcodes but
>         work with the mainstream scan opcodes.
>
>       - trigexpseg, triglinseg are aliases for trigExpseg, trigLinseg.
>
>       - xscan opcodes are deprecated as they add nothing to the scan
>       opcodes.
>
>       - bformdec2 provides more ambisonic decoding of bformat audio.
>
>       - metrobpm is like metro but the timing is in beats per minute and
>       the
>         signal can be offset by a proportion of the cycle.
>
>       - sequ is a csound version of an hardware sequencer.  It has many
>       modes and options.
>
>       - gtadsr implements a gated ADSR envelope
>
>       ### New gen
>
>       ### Orchestra
>
>       - Message printing has been revised so -m16 suppresses all
>           messages.
>
>       - add channel count to list_audio_devices (called when the flag
>         --devices is used so that it can be parsed by frontends).
>
>       ### Score
>
>       ### Options
>
>       ### Modified Opcodes and Gens
>
>       - event opcode does not bail out if the instrument called does not
>       exist.
>
>       - Added an optional prefix to soundfont instruments printed via
>       sfilist.
>
>       - lpslots reworked with better control.
>
>       ### Utilities
>
>       - cvanal now uses the SADIR environment to look for analysis
>       files.
>
>       ### Frontends
>
>       ### General Usage
>
>       ## Bugs Fixed
>
>       - fareylen called a non-existent function leading to a crash. 
>       Removed typo.
>
>       - turnoff could cause clicks in some cases; fixed.
>
>       - turnoff3 improved.
>
>       - cntReset fixed; had a false initialisation code.
>
>       - binary search in bpf fixed.
>
>       - pvscfs frame counting issue fixed.
>
>       - --sample-accurate fixed in a-rate form of tabsum.
>
>       - Problem in atssinnoi fixed; did read outside allocated memory.
>
>       - hrtfmove fixed; it could use the wrong value for sr.
>
>       - Named instruments could use wrong structure in redefinition.
>
>       - fixes in the jack backend relating to --get-system-sr.
>
>       - pvcross frequency warp mode fixed.
>
>       - lpcfilter/lpcanal bug where processing could not be restarted
>       after
>         a freeze fixed.
>
>       - the emugens collection of opcode, a large group incorporation
>       mainly
>         array operations, were not made available due to a small editing
>         error.
>
>       - If using a float build (rather than a double) array access could
>       be
>         wrong.
>
>       - In some cases instr0 code could ignore loops after an incorrect
>         warning.  Now fixed.
>
>       - Bugs in mp3len and similar fixed.
>
>       - Crash in syncgrain issue #1773 fixed
>
>       # SYSTEM LEVEL CHANGES
>
>       ### System Changes
>
>       - winsound has been removed.
>
>       ### Translations
>
>       ### API
>
>       CreateThread2 is a new API function that allows threads to be
>       given a
>       user-defined stack size.
>
>       ### External Plugin Code
>
>       For a variety of reasons including licences, plugin opcodes that
>       allow csound to be a host for VST plugins is available from
>       https://github.com/gogins/csound-vst3-opcodes.git where
>       installation
>       instructions can be found.  It has been shown to work on Linux,
>       MacOS
>       and Windows.
>
>       ### Platform Specific
>
>       ==END==
>
>       ========================================================================
> 
> 
>



*Warning*

This email originated from outside of Maynooth University's Mail System. Do not reply, click links or open attachments unless you recognise the sender and know the content is safe.

I just updated the what's new section of the manual accordingly. The manual entry for the gtadsr opcode has links to images/gtadsr.png and mgtadsr opcodes, but there are no such image and opcode.

François

Le mer. 26 janv. 2022 à 18:53, John <jpff@codemist.co.uk> a écrit :

This is what I expect to be in cs6.17

Any clarifications or suggestions please send to me soon as the release is
imminent.

==John ffitch



# CSOUND VERSION 6.17 RELEASE NOTES - DRAFT - DRAFT - DRAFT - DRAFT

Mainly a bug-fixing release but also a major re-organisation of the
libraries to move all opcodes with dependencies into a separate plugins
repository. The only exception to this is the OSC opcodes (which
depend on liblo). This is in part a preparation for Csound7. This is
the last release using the current parser (v.2).

-- The Developers

## USER-LEVEL CHANGES

### New opcodes

- scanmap and scansmap are like the xscanmap/xscansmap opcodes but
  work with the mainstream scan opcodes.

- trigexpseg, triglinseg are aliases for trigExpseg, trigLinseg.

- xscan opcodes are deprecated as they add nothing to the scan opcodes.

- bformdec2 provides more ambisonic decoding of bformat audio.

- metrobpm is like metro but the timing is in beats per minute and the
  signal can be offset by a proportion of the cycle.

- sequ is a csound version of an hardware sequencer.  It has many modes and options.

- gtadsr implements a gated ADSR envelope

### New gen

### Orchestra

- Message printing has been revised so -m16 suppresses all
    messages.

- add channel count to list_audio_devices (called when the flag
  --devices is used so that it can be parsed by frontends).

### Score

### Options

### Modified Opcodes and Gens

- event opcode does not bail out if the instrument called does not exist.

- Added an optional prefix to soundfont instruments printed via sfilist.

- lpslots reworked with better control.

### Utilities

- cvanal now uses the SADIR environment to look for analysis files.

### Frontends

### General Usage

## Bugs Fixed

- fareylen called a non-existent function leading to a crash.  Removed typo.

- turnoff could cause clicks in some cases; fixed.

- turnoff3 improved.

- cntReset fixed; had a false initialisation code.

- binary search in bpf fixed.

- pvscfs frame counting issue fixed.

- --sample-accurate fixed in a-rate form of tabsum.

- Problem in atssinnoi fixed; did read outside allocated memory.

- hrtfmove fixed; it could use the wrong value for sr.

- Named instruments could use wrong structure in redefinition.

- fixes in the jack backend relating to --get-system-sr.

- pvcross frequency warp mode fixed.

- lpcfilter/lpcanal bug where processing could not be restarted after
  a freeze fixed.

- the emugens collection of opcode, a large group incorporation mainly
  array operations, were not made available due to a small editing
  error.

- If using a float build (rather than a double) array access could be
  wrong.

- In some cases instr0 code could ignore loops after an incorrect
  warning.  Now fixed.

- Bugs in mp3len and similar fixed.

- Crash in syncgrain issue #1773 fixed

# SYSTEM LEVEL CHANGES

### System Changes

- winsound has been removed.

### Translations

### API

CreateThread2 is a new API function that allows threads to be given a
user-defined stack size.

### External Plugin Code

For a variety of reasons including licences, plugin opcodes that
allow csound to be a host for VST plugins is available from
https://github.com/gogins/csound-vst3-opcodes.git where installation
instructions can be found.  It has been shown to work on Linux, MacOS
and Windows.

### Platform Specific

==END==

========================================================================

i commited yesterday an extension of f-sigs to the UDO description of 
the manual:
https://github.com/csound/manual/pull/611/commits/885541c0002b5da5fbca307fdadf19c7ad176ad7

but i think i was wrong: it seems that f-sigs are only possible as 
inputs, not as outputs.

can you confirm? we should correct this then before release.

On 30/01/2022 19:02, Victor Lazzarini wrote:
> Are we ready for a manual release tomorrow? That needs to precede the 
> Csound release so the windows CI can build the package with the correct 
> manual version.
> 
> Prof. Victor Lazzarini
> Maynooth University
> Ireland
> 
>> On 27 Jan 2022, at 16:27, Francois PINOT  wrote:
>>
>> 
>>
>>
>>   *Warning*
>>
>> This email originated from outside of Maynooth University's Mail 
>> System. Do not reply, click links or open attachments unless you 
>> recognise the sender and know the content is safe.
>>
>> I just updated the what's new section of the manual accordingly. The 
>> manual entry for the gtadsr opcode has links to images/gtadsr.png and 
>> mgtadsr opcodes, but there are no such image and opcode.
>>
>> François
>>
>> Le mer. 26 janv. 2022 à 18:53, John > > a écrit :
>>
>>     This is what I expect to be in cs6.17
>>
>>     Any clarifications or suggestions please send to me soon as the
>>     release is
>>     imminent.
>>
>>     ==John ffitch
>>
>>
>>
>>     # CSOUND VERSION 6.17 RELEASE NOTES - DRAFT - DRAFT - DRAFT - DRAFT
>>
>>     Mainly a bug-fixing release but also a major re-organisation of the
>>     libraries to move all opcodes with dependencies into a separate
>>     plugins
>>     repository. The only exception to this is the OSC opcodes (which
>>     depend on liblo). This is in part a preparation for Csound7. This is
>>     the last release using the current parser (v.2).
>>
>>     -- The Developers
>>
>>     ## USER-LEVEL CHANGES
>>
>>     ### New opcodes
>>
>>     - scanmap and scansmap are like the xscanmap/xscansmap opcodes but
>>       work with the mainstream scan opcodes.
>>
>>     - trigexpseg, triglinseg are aliases for trigExpseg, trigLinseg.
>>
>>     - xscan opcodes are deprecated as they add nothing to the scan
>>     opcodes.
>>
>>     - bformdec2 provides more ambisonic decoding of bformat audio.
>>
>>     - metrobpm is like metro but the timing is in beats per minute and the
>>       signal can be offset by a proportion of the cycle.
>>
>>     - sequ is a csound version of an hardware sequencer.  It has many
>>     modes and options.
>>
>>     - gtadsr implements a gated ADSR envelope
>>
>>     ### New gen
>>
>>     ### Orchestra
>>
>>     - Message printing has been revised so -m16 suppresses all
>>         messages.
>>
>>     - add channel count to list_audio_devices (called when the flag
>>       --devices is used so that it can be parsed by frontends).
>>
>>     ### Score
>>
>>     ### Options
>>
>>     ### Modified Opcodes and Gens
>>
>>     - event opcode does not bail out if the instrument called does not
>>     exist.
>>
>>     - Added an optional prefix to soundfont instruments printed via
>>     sfilist.
>>
>>     - lpslots reworked with better control.
>>
>>     ### Utilities
>>
>>     - cvanal now uses the SADIR environment to look for analysis files.
>>
>>     ### Frontends
>>
>>     ### General Usage
>>
>>     ## Bugs Fixed
>>
>>     - fareylen called a non-existent function leading to a crash. 
>>     Removed typo.
>>
>>     - turnoff could cause clicks in some cases; fixed.
>>
>>     - turnoff3 improved.
>>
>>     - cntReset fixed; had a false initialisation code.
>>
>>     - binary search in bpf fixed.
>>
>>     - pvscfs frame counting issue fixed.
>>
>>     - --sample-accurate fixed in a-rate form of tabsum.
>>
>>     - Problem in atssinnoi fixed; did read outside allocated memory.
>>
>>     - hrtfmove fixed; it could use the wrong value for sr.
>>
>>     - Named instruments could use wrong structure in redefinition.
>>
>>     - fixes in the jack backend relating to --get-system-sr.
>>
>>     - pvcross frequency warp mode fixed.
>>
>>     - lpcfilter/lpcanal bug where processing could not be restarted after
>>       a freeze fixed.
>>
>>     - the emugens collection of opcode, a large group incorporation mainly
>>       array operations, were not made available due to a small editing
>>       error.
>>
>>     - If using a float build (rather than a double) array access could be
>>       wrong.
>>
>>     - In some cases instr0 code could ignore loops after an incorrect
>>       warning.  Now fixed.
>>
>>     - Bugs in mp3len and similar fixed.
>>
>>     - Crash in syncgrain issue #1773 fixed
>>
>>     # SYSTEM LEVEL CHANGES
>>
>>     ### System Changes
>>
>>     - winsound has been removed.
>>
>>     ### Translations
>>
>>     ### API
>>
>>     CreateThread2 is a new API function that allows threads to be given a
>>     user-defined stack size.
>>
>>     ### External Plugin Code
>>
>>     For a variety of reasons including licences, plugin opcodes that
>>     allow csound to be a host for VST plugins is available from
>>     https://github.com/gogins/csound-vst3-opcodes.git
>>     
>>     where installation
>>     instructions can be found.  It has been shown to work on Linux, MacOS
>>     and Windows.
>>
>>     ### Platform Specific
>>
>>     ==END==
>>
>>     ========================================================================

Looks like a bug me, it should work correctly or give an error. i'll have to investigate.

> On 30 Jan 2022, at 18:09, joachim heintz  wrote:
> 
> i commited yesterday an extension of f-sigs to the UDO description of the manual:
> https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fcsound%2Fmanual%2Fpull%2F611%2Fcommits%2F885541c0002b5da5fbca307fdadf19c7ad176ad7&data=04%7C01%7CVictor.Lazzarini%40mu.ie%7Cfdab319ecd1b489241a508d9e41b9dd2%7C1454f5ccbb354685bbd98621fd8055c9%7C0%7C0%7C637791629976492510%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=ct5Kw%2FVn4pvnDaPKgAWKGt2D%2F%2FLANV45pXn3VCcaJoI%3D&reserved=0
> 
> but i think i was wrong: it seems that f-sigs are only possible as inputs, not as outputs.
> 
> can you confirm? we should correct this then before release.
> 
> On 30/01/2022 19:02, Victor Lazzarini wrote:
>> Are we ready for a manual release tomorrow? That needs to precede the Csound release so the windows CI can build the package with the correct manual version.
>> Prof. Victor Lazzarini
>> Maynooth University
>> Ireland
>>> On 27 Jan 2022, at 16:27, Francois PINOT  wrote:
>>> 
>>> 
>>> 
>>> 
>>>  *Warning*
>>> 
>>> This email originated from outside of Maynooth University's Mail System. Do not reply, click links or open attachments unless you recognise the sender and know the content is safe.
>>> 
>>> I just updated the what's new section of the manual accordingly. The manual entry for the gtadsr opcode has links to images/gtadsr.png and mgtadsr opcodes, but there are no such image and opcode.
>>> 
>>> François
>>> 
>>> Le mer. 26 janv. 2022 à 18:53, John > a écrit :
>>> 
>>>    This is what I expect to be in cs6.17
>>> 
>>>    Any clarifications or suggestions please send to me soon as the
>>>    release is
>>>    imminent.
>>> 
>>>    ==John ffitch
>>> 
>>> 
>>> 
>>>    # CSOUND VERSION 6.17 RELEASE NOTES - DRAFT - DRAFT - DRAFT - DRAFT
>>> 
>>>    Mainly a bug-fixing release but also a major re-organisation of the
>>>    libraries to move all opcodes with dependencies into a separate
>>>    plugins
>>>    repository. The only exception to this is the OSC opcodes (which
>>>    depend on liblo). This is in part a preparation for Csound7. This is
>>>    the last release using the current parser (v.2).
>>> 
>>>    -- The Developers
>>> 
>>>    ## USER-LEVEL CHANGES
>>> 
>>>    ### New opcodes
>>> 
>>>    - scanmap and scansmap are like the xscanmap/xscansmap opcodes but
>>>      work with the mainstream scan opcodes.
>>> 
>>>    - trigexpseg, triglinseg are aliases for trigExpseg, trigLinseg.
>>> 
>>>    - xscan opcodes are deprecated as they add nothing to the scan
>>>    opcodes.
>>> 
>>>    - bformdec2 provides more ambisonic decoding of bformat audio.
>>> 
>>>    - metrobpm is like metro but the timing is in beats per minute and the
>>>      signal can be offset by a proportion of the cycle.
>>> 
>>>    - sequ is a csound version of an hardware sequencer.  It has many
>>>    modes and options.
>>> 
>>>    - gtadsr implements a gated ADSR envelope
>>> 
>>>    ### New gen
>>> 
>>>    ### Orchestra
>>> 
>>>    - Message printing has been revised so -m16 suppresses all
>>>        messages.
>>> 
>>>    - add channel count to list_audio_devices (called when the flag
>>>      --devices is used so that it can be parsed by frontends).
>>> 
>>>    ### Score
>>> 
>>>    ### Options
>>> 
>>>    ### Modified Opcodes and Gens
>>> 
>>>    - event opcode does not bail out if the instrument called does not
>>>    exist.
>>> 
>>>    - Added an optional prefix to soundfont instruments printed via
>>>    sfilist.
>>> 
>>>    - lpslots reworked with better control.
>>> 
>>>    ### Utilities
>>> 
>>>    - cvanal now uses the SADIR environment to look for analysis files.
>>> 
>>>    ### Frontends
>>> 
>>>    ### General Usage
>>> 
>>>    ## Bugs Fixed
>>> 
>>>    - fareylen called a non-existent function leading to a crash.     Removed typo.
>>> 
>>>    - turnoff could cause clicks in some cases; fixed.
>>> 
>>>    - turnoff3 improved.
>>> 
>>>    - cntReset fixed; had a false initialisation code.
>>> 
>>>    - binary search in bpf fixed.
>>> 
>>>    - pvscfs frame counting issue fixed.
>>> 
>>>    - --sample-accurate fixed in a-rate form of tabsum.
>>> 
>>>    - Problem in atssinnoi fixed; did read outside allocated memory.
>>> 
>>>    - hrtfmove fixed; it could use the wrong value for sr.
>>> 
>>>    - Named instruments could use wrong structure in redefinition.
>>> 
>>>    - fixes in the jack backend relating to --get-system-sr.
>>> 
>>>    - pvcross frequency warp mode fixed.
>>> 
>>>    - lpcfilter/lpcanal bug where processing could not be restarted after
>>>      a freeze fixed.
>>> 
>>>    - the emugens collection of opcode, a large group incorporation mainly
>>>      array operations, were not made available due to a small editing
>>>      error.
>>> 
>>>    - If using a float build (rather than a double) array access could be
>>>      wrong.
>>> 
>>>    - In some cases instr0 code could ignore loops after an incorrect
>>>      warning.  Now fixed.
>>> 
>>>    - Bugs in mp3len and similar fixed.
>>> 
>>>    - Crash in syncgrain issue #1773 fixed
>>> 
>>>    # SYSTEM LEVEL CHANGES
>>> 
>>>    ### System Changes
>>> 
>>>    - winsound has been removed.
>>> 
>>>    ### Translations
>>> 
>>>    ### API
>>> 
>>>    CreateThread2 is a new API function that allows threads to be given a
>>>    user-defined stack size.
>>> 
>>>    ### External Plugin Code
>>> 
>>>    For a variety of reasons including licences, plugin opcodes that
>>>    allow csound to be a host for VST plugins is available from
>>>    https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fgogins%2Fcsound-vst3-opcodes.git&data=04%7C01%7CVictor.Lazzarini%40mu.ie%7Cfdab319ecd1b489241a508d9e41b9dd2%7C1454f5ccbb354685bbd98621fd8055c9%7C0%7C0%7C637791629976492510%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=LC803t%2B9EMwMPvtR9a53Gi0FljYrEoVGPT%2FihF9nuKg%3D&reserved=0
>>>    
>>>    where installation
>>>    instructions can be found.  It has been shown to work on Linux, MacOS
>>>    and Windows.
>>> 
>>>    ### Platform Specific
>>> 
>>>    ==END==
>>> 
>>>    ============================================

Fixed now, fsigs as output are working.

> On 31 Jan 2022, at 06:43, Victor Lazzarini  wrote:
> 
> Looks like a bug me, it should work correctly or give an error. i'll have to investigate.
> 
>> On 30 Jan 2022, at 18:09, joachim heintz  wrote:
>> 
>> i commited yesterday an extension of f-sigs to the UDO description of the manual:
>> https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fcsound%2Fmanual%2Fpull%2F611%2Fcommits%2F885541c0002b5da5fbca307fdadf19c7ad176ad7&data=04%7C01%7CVictor.Lazzarini%40mu.ie%7Cbb7a3180390c43695de308d9e48513bd%7C1454f5ccbb354685bbd98621fd8055c9%7C0%7C0%7C637792082487936790%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=B9K9%2BeZV00rWiwIu%2Bq%2FvDwG0M7zeMdQB83rZMnQoQQY%3D&reserved=0
>> 
>> but i think i was wrong: it seems that f-sigs are only possible as inputs, not as outputs.
>> 
>> can you confirm? we should correct this then before release.
>> 
>> On 30/01/2022 19:02, Victor Lazzarini wrote:
>>> Are we ready for a manual release tomorrow? That needs to precede the Csound release so the windows CI can build the package with the correct manual version.
>>> Prof. Victor Lazzarini
>>> Maynooth University
>>> Ireland
>>>> On 27 Jan 2022, at 16:27, Francois PINOT  wrote:
>>>> 
>>>> 
>>>> 
>>>> 
>>>> *Warning*
>>>> 
>>>> This email originated from outside of Maynooth University's Mail System. Do not reply, click links or open attachments unless you recognise the sender and know the content is safe.
>>>> 
>>>> I just updated the what's new section of the manual accordingly. The manual entry for the gtadsr opcode has links to images/gtadsr.png and mgtadsr opcodes, but there are no such image and opcode.
>>>> 
>>>> François
>>>> 
>>>> Le mer. 26 janv. 2022 à 18:53, John > a écrit :
>>>> 
>>>>   This is what I expect to be in cs6.17
>>>> 
>>>>   Any clarifications or suggestions please send to me soon as the
>>>>   release is
>>>>   imminent.
>>>> 
>>>>   ==John ffitch
>>>> 
>>>> 
>>>> 
>>>>   # CSOUND VERSION 6.17 RELEASE NOTES - DRAFT - DRAFT - DRAFT - DRAFT
>>>> 
>>>>   Mainly a bug-fixing release but also a major re-organisation of the
>>>>   libraries to move all opcodes with dependencies into a separate
>>>>   plugins
>>>>   repository. The only exception to this is the OSC opcodes (which
>>>>   depend on liblo). This is in part a preparation for Csound7. This is
>>>>   the last release using the current parser (v.2).
>>>> 
>>>>   -- The Developers
>>>> 
>>>>   ## USER-LEVEL CHANGES
>>>> 
>>>>   ### New opcodes
>>>> 
>>>>   - scanmap and scansmap are like the xscanmap/xscansmap opcodes but
>>>>     work with the mainstream scan opcodes.
>>>> 
>>>>   - trigexpseg, triglinseg are aliases for trigExpseg, trigLinseg.
>>>> 
>>>>   - xscan opcodes are deprecated as they add nothing to the scan
>>>>   opcodes.
>>>> 
>>>>   - bformdec2 provides more ambisonic decoding of bformat audio.
>>>> 
>>>>   - metrobpm is like metro but the timing is in beats per minute and the
>>>>     signal can be offset by a proportion of the cycle.
>>>> 
>>>>   - sequ is a csound version of an hardware sequencer.  It has many
>>>>   modes and options.
>>>> 
>>>>   - gtadsr implements a gated ADSR envelope
>>>> 
>>>>   ### New gen
>>>> 
>>>>   ### Orchestra
>>>> 
>>>>   - Message printing has been revised so -m16 suppresses all
>>>>       messages.
>>>> 
>>>>   - add channel count to list_audio_devices (called when the flag
>>>>     --devices is used so that it can be parsed by frontends).
>>>> 
>>>>   ### Score
>>>> 
>>>>   ### Options
>>>> 
>>>>   ### Modified Opcodes and Gens
>>>> 
>>>>   - event opcode does not bail out if the instrument called does not
>>>>   exist.
>>>> 
>>>>   - Added an optional prefix to soundfont instruments printed via
>>>>   sfilist.
>>>> 
>>>>   - lpslots reworked with better control.
>>>> 
>>>>   ### Utilities
>>>> 
>>>>   - cvanal now uses the SADIR environment to look for analysis files.
>>>> 
>>>>   ### Frontends
>>>> 
>>>>   ### General Usage
>>>> 
>>>>   ## Bugs Fixed
>>>> 
>>>>   - fareylen called a non-existent function leading to a crash.     Removed typo.
>>>> 
>>>>   - turnoff could cause clicks in some cases; fixed.
>>>> 
>>>>   - turnoff3 improved.
>>>> 
>>>>   - cntReset fixed; had a false initialisation code.
>>>> 
>>>>   - binary search in bpf fixed.
>>>> 
>>>>   - pvscfs frame counting issue fixed.
>>>> 
>>>>   - --sample-accurate fixed in a-rate form of tabsum.
>>>> 
>>>>   - Problem in atssinnoi fixed; did read outside allocated memory.
>>>> 
>>>>   - hrtfmove fixed; it could use the wrong value for sr.
>>>> 
>>>>   - Named instruments could use wrong structure in redefinition.
>>>> 
>>>>   - fixes in the jack backend relating to --get-system-sr.
>>>> 
>>>>   - pvcross frequency warp mode fixed.
>>>> 
>>>>   - lpcfilter/lpcanal bug where processing could not be restarted after
>>>>     a freeze fixed.
>>>> 
>>>>   - the emugens collection of opcode, a large group incorporation mainly
>>>>     array operations, were not made available due to a small editing
>>>>     error.
>>>> 
>>>>   - If using a float build (rather than a double) array access could be
>>>>     wrong.
>>>> 
>>>>   - In some cases instr0 code could ignore loops after an incorrect
>>>>     warning.  Now fixed.
>>>> 
>>>>   - Bugs in mp3len and similar fixed.
>>>> 
>>>>   - Crash in syncgrain issue #1773 fixed
>>>> 
>>>>   # SYSTEM LEVEL CHANGES
>>>> 
>>>>   ### System Changes
>>>> 
>>>>   - winsound has been removed.
>>>> 
>>>>   ### Translations
>>>> 
>>>>   ### API
>>>> 
>>>>   CreateThread2 is a new API function that allows threads to be given a
>>>>   user-defined stack size.
>>>> 
>>>>   ### External Plugin Code
>>>> 
>>>>   For a variety of reasons including licences, plugin opcodes that
>>>>   allow csound to be a host for VST plugins is available from
>>>>   https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fgogins%2Fcsound-vst3-opcodes.git&data=04%7C01%7CVictor.Lazzarini%40mu.ie%7Cbb7a3180390c43695de308d9e48513bd%7C1454f5ccbb354685bbd98621fd8055c9%7C0%7C0%7C637792082487936790%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=sTCOLtCPFeJcLQ5%2Bdmj%2Fo3aRnkUsteUo8Y67QpzGAG0%3D&reserved=0
>>>>   
>>>>   where installation
>>>>   instructions can be found.  It has been shown to work on Linux, MacOS
>>>>   and Windows.
>>>> 
>>>>   ### Platform Specific
>>>> 
>>>>   ==END==
>>>> 
>>>>   =======================================================

yes i can confirm this.  great --- i will submit an examples for it.
	joachim


On 31/01/2022 10:06, Victor Lazzarini wrote:
> Fixed now, fsigs as output are working.
> 
>> On 31 Jan 2022, at 06:43, Victor Lazzarini  wrote:
>>
>> Looks like a bug me, it should work correctly or give an error. i'll have to investigate.
>>
>>> On 30 Jan 2022, at 18:09, joachim heintz  wrote:
>>>
>>> i commited yesterday an extension of f-sigs to the UDO description of the manual:
>>> https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fcsound%2Fmanual%2Fpull%2F611%2Fcommits%2F885541c0002b5da5fbca307fdadf19c7ad176ad7&data=04%7C01%7CVictor.Lazzarini%40mu.ie%7Cbb7a3180390c43695de308d9e48513bd%7C1454f5ccbb354685bbd98621fd8055c9%7C0%7C0%7C637792082487936790%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=B9K9%2BeZV00rWiwIu%2Bq%2FvDwG0M7zeMdQB83rZMnQoQQY%3D&reserved=0
>>>
>>> but i think i was wrong: it seems that f-sigs are only possible as inputs, not as outputs.
>>>
>>> can you confirm? we should correct this then before release.
>>>
>>> On 30/01/2022 19:02, Victor Lazzarini wrote:
>>>> Are we ready for a manual release tomorrow? That needs to precede the Csound release so the windows CI can build the package with the correct manual version.
>>>> Prof. Victor Lazzarini
>>>> Maynooth University
>>>> Ireland
>>>>> On 27 Jan 2022, at 16:27, Francois PINOT  wrote:
>>>>>
>>>>> 
>>>>>
>>>>>
>>>>> *Warning*
>>>>>
>>>>> This email originated from outside of Maynooth University's Mail System. Do not reply, click links or open attachments unless you recognise the sender and know the content is safe.
>>>>>
>>>>> I just updated the what's new section of the manual accordingly. The manual entry for the gtadsr opcode has links to images/gtadsr.png and mgtadsr opcodes, but there are no such image and opcode.
>>>>>
>>>>> François
>>>>>
>>>>> Le mer. 26 janv. 2022 à 18:53, John > a écrit :
>>>>>
>>>>>    This is what I expect to be in cs6.17
>>>>>
>>>>>    Any clarifications or suggestions please send to me soon as the
>>>>>    release is
>>>>>    imminent.
>>>>>
>>>>>    ==John ffitch
>>>>>
>>>>>
>>>>>
>>>>>    # CSOUND VERSION 6.17 RELEASE NOTES - DRAFT - DRAFT - DRAFT - DRAFT
>>>>>
>>>>>    Mainly a bug-fixing release but also a major re-organisation of the
>>>>>    libraries to move all opcodes with dependencies into a separate
>>>>>    plugins
>>>>>    repository. The only exception to this is the OSC opcodes (which
>>>>>    depend on liblo). This is in part a preparation for Csound7. This is
>>>>>    the last release using the current parser (v.2).
>>>>>
>>>>>    -- The Developers
>>>>>
>>>>>    ## USER-LEVEL CHANGES
>>>>>
>>>>>    ### New opcodes
>>>>>
>>>>>    - scanmap and scansmap are like the xscanmap/xscansmap opcodes but
>>>>>      work with the mainstream scan opcodes.
>>>>>
>>>>>    - trigexpseg, triglinseg are aliases for trigExpseg, trigLinseg.
>>>>>
>>>>>    - xscan opcodes are deprecated as they add nothing to the scan
>>>>>    opcodes.
>>>>>
>>>>>    - bformdec2 provides more ambisonic decoding of bformat audio.
>>>>>
>>>>>    - metrobpm is like metro but the timing is in beats per minute and the
>>>>>      signal can be offset by a proportion of the cycle.
>>>>>
>>>>>    - sequ is a csound version of an hardware sequencer.  It has many
>>>>>    modes and options.
>>>>>
>>>>>    - gtadsr implements a gated ADSR envelope
>>>>>
>>>>>    ### New gen
>>>>>
>>>>>    ### Orchestra
>>>>>
>>>>>    - Message printing has been revised so -m16 suppresses all
>>>>>        messages.
>>>>>
>>>>>    - add channel count to list_audio_devices (called when the flag
>>>>>      --devices is used so that it can be parsed by frontends).
>>>>>
>>>>>    ### Score
>>>>>
>>>>>    ### Options
>>>>>
>>>>>    ### Modified Opcodes and Gens
>>>>>
>>>>>    - event opcode does not bail out if the instrument called does not
>>>>>    exist.
>>>>>
>>>>>    - Added an optional prefix to soundfont instruments printed via
>>>>>    sfilist.
>>>>>
>>>>>    - lpslots reworked with better control.
>>>>>
>>>>>    ### Utilities
>>>>>
>>>>>    - cvanal now uses the SADIR environment to look for analysis files.
>>>>>
>>>>>    ### Frontends
>>>>>
>>>>>    ### General Usage
>>>>>
>>>>>    ## Bugs Fixed
>>>>>
>>>>>    - fareylen called a non-existent function leading to a crash.     Removed typo.
>>>>>
>>>>>    - turnoff could cause clicks in some cases; fixed.
>>>>>
>>>>>    - turnoff3 improved.
>>>>>
>>>>>    - cntReset fixed; had a false initialisation code.
>>>>>
>>>>>    - binary search in bpf fixed.
>>>>>
>>>>>    - pvscfs frame counting issue fixed.
>>>>>
>>>>>    - --sample-accurate fixed in a-rate form of tabsum.
>>>>>
>>>>>    - Problem in atssinnoi fixed; did read outside allocated memory.
>>>>>
>>>>>    - hrtfmove fixed; it could use the wrong value for sr.
>>>>>
>>>>>    - Named instruments could use wrong structure in redefinition.
>>>>>
>>>>>    - fixes in the jack backend relating to --get-system-sr.
>>>>>
>>>>>    - pvcross frequency warp mode fixed.
>>>>>
>>>>>    - lpcfilter/lpcanal bug where processing could not be restarted after
>>>>>      a freeze fixed.
>>>>>
>>>>>    - the emugens collection of opcode, a large group incorporation mainly
>>>>>      array operations, were not made available due to a small editing
>>>>>      error.
>>>>>
>>>>>    - If using a float build (rather than a double) array access could be
>>>>>      wrong.
>>>>>
>>>>>    - In some cases instr0 code could ignore loops after an incorrect
>>>>>      warning.  Now fixed.
>>>>>
>>>>>    - Bugs in mp3len and similar fixed.
>>>>>
>>>>>    - Crash in syncgrain issue #1773 fixed
>>>>>
>>>>>    # SYSTEM LEVEL CHANGES
>>>>>
>>>>>    ### System Changes
>>>>>
>>>>>    - winsound has been removed.
>>>>>
>>>>>    ### Translations
>>>>>
>>>>>    ### API
>>>>>
>>>>>    CreateThread2 is a new API function that allows threads to be given a
>>>>>    user-defined stack size.
>>>>>
>>>>>    ### External Plugin Code
>>>>>
>>>>>    For a variety of reasons including licences, plugin opcodes that
>>>>>    allow csound to be a host for VST plugins is available from
>>>>>    https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fgogins%2Fcsound-vst3-opcodes.git&data=04%7C01%7CVictor.Lazzarini%40mu.ie%7Cbb7a3180390c43695de308d9e48513bd%7C1454f5ccbb354685bbd98621fd8055c9%7C0%7C0%7C637792082487936790%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=sTCOLtCPFeJcLQ5%2Bdmj%2Fo3aRnkUsteUo8Y67QpzGAG0%3D&reserved=0
>>>>>    
>>>>>    where installation
>>>>>    instructions can be found.  It has been shown to work on Linux, MacOS
>>>>>    and Windows.
>>>>>
>>>>>    ### Platform Specific
>>>>>
>>>>>    ==END==
>>>>>
>>>>>    ========================================================================
>>>>>
>>