[Cs-dev] Draft Release Notes
Date | 2012-02-03 15:17 |
From | jpff |
Subject | [Cs-dev] Draft Release Notes |
Slightly more upto date ==John ffitch Notes for 5.16 ============== The major change is that the new parser is now the default. The old parser is still available in case of difficulty but the new has been given extensive testing since the start of the year, including complete restructuring of macro expansion. A side effect is that the runtime of most orchestras is faster, although parsing is slower. There are a few optimisations implemented like constant folding in simple cases. Line numbers and file names are traced better than before. Some memory leaks also fixed New Opcodes: dust, dust2 fractalnoise gausstrig gendy ptable, like table but does not require a power-of-2 table ptable3, like table3 but does not require a power-of-2 table ptablei, like tablei but does not require a power-of-2 table ptablew, like tablew but does not require a power-of-2 table New Gen and Macros: Modified Opcodes and Gens: There was a fencepost problem in tab opcode that could falsely report a reference out of range GEN15 mis-called gens 13 and 14 internally, using uninitialised values voice amplitude problem fixed fmbell now takes an optional argument to control the sustain time Change to pvsbasic for tab to table conversions poscil is now polymorhic, allowing k- or a- rate amplitude and frequency p() and i() changed when argument at k-rate gen49 deferred now works gen23 now available deferred Utilities Checked foe use with the new parser/in memory files Bugs fixed: Many in new parser related to precedence and multicore. Better diagnostics when orchestra file/csd is missing csd file: fix CsFileB and CsSampleB fix score statement 'n' Fixed bug in diskin2 leading to infinite loop Fixed bug causing crossfade noise in hrtfmove Fixed unlikely buffer overflows in some utilities Avoid segfault in midicN Bug in mp3in in skip=0 case fixed r score statement fixed with respect to macros sndwarp could segfault System Changes: New code to transfer tables to/from PD (csoundapi~) preprocessor #if #else #endin working #includes depth now limited rather than infinite recursion really turn off all displays if --nodisplays or -d is used; fixes bug where using -d or --nodisplays would still cause the winFLTK.c csoundModuleInit to setup display callbacks; bug caused with python TK apps and CsoundYield_FLTK being called memory leak in mp3in and mp3len fixed API: Internal: Very, very, very many! And the new parser..... ======================================================================== Undocumented opcodes as of now: OSCrecv cogoto copy2ftab copy2ttab ctlchn midipgm pgmchn pitchac readks ------------------------------------------------------------------------------ Try before you buy = See our experts in action! The most comprehensive online learning library for Microsoft developers is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, Metro Style Apps, more. Free future releases when you subscribe now! http://p.sf.net/sfu/learndevnow-dev2 _______________________________________________ Csound-devel mailing list Csound-devel@lists.sourceforge.net |
Date | 2012-02-03 15:51 |
From | Rory Walsh |
Subject | Re: [Cs-dev] Draft Release Notes |
Attachments | None None |
This the most exciting release there has been in a while. The developers should take a bow! On Friday, 3 February 2012, jpff <jpff@cs.bath.ac.uk> wrote: > Slightly more upto date > ==John ffitch > > Notes for 5.16 > ============== > > The major change is that the new parser is now the default. The old > parser is still available in case of difficulty but the new has been > given extensive testing since the start of the year, including > complete restructuring of macro expansion. > A side effect is that the runtime of most orchestras is faster, > although parsing is slower. There are a few optimisations implemented > like constant folding in simple cases. Line numbers and file names > are traced better than before. > > Some memory leaks also fixed > > > > New Opcodes: > dust, dust2 > fractalnoise > gausstrig > gendy > ptable, like table but does not require a power-of-2 table > ptable3, like table3 but does not require a power-of-2 table > ptablei, like tablei but does not require a power-of-2 table > ptablew, like tablew but does not require a power-of-2 table > > New Gen and Macros: > > Modified Opcodes and Gens: > There was a fencepost problem in tab opcode that could falsely > report a reference out of range > GEN15 mis-called gens 13 and 14 internally, using uninitialised values > voice amplitude problem fixed > fmbell now takes an optional argument to control the sustain time > Change to pvsbasic for tab to table conversions > poscil is now polymorhic, allowing k- or a- rate amplitude and frequency > p() and i() changed when argument at k-rate > gen49 deferred now works > gen23 now available deferred > > Utilities > Checked foe use with the new parser/in memory files > > > Bugs fixed: > Many in new parser related to precedence and multicore. > Better diagnostics when orchestra file/csd is missing > csd file: fix CsFileB and CsSampleB > fix score statement 'n' > Fixed bug in diskin2 leading to infinite loop > Fixed bug causing crossfade noise in hrtfmove > Fixed unlikely buffer overflows in some utilities > Avoid segfault in midicN > Bug in mp3in in skip=0 case fixed > r score statement fixed with respect to macros > sndwarp could segfault > > System Changes: > New code to transfer tables to/from PD (csoundapi~) > preprocessor #if #else #endin working > #includes depth now limited rather than infinite recursion > really turn off all displays if --nodisplays or -d is used; fixes > bug where using -d or --nodisplays would still cause the > winFLTK.c csoundModuleInit to setup display callbacks; bug > caused with python TK apps and CsoundYield_FLTK being called > memory leak in mp3in and mp3len fixed > > API: > > Internal: > Very, very, very many! And the new parser..... > > > > > > > ======================================================================== > > Undocumented opcodes as of now: > OSCrecv > cogoto > copy2ftab > copy2ttab > ctlchn > midipgm > pgmchn > pitchac > readks > > ------------------------------------------------------------------------------ > Try before you buy = See our experts in action! > The most comprehensive online learning library for Microsoft developers > is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, > Metro Style Apps, more. Free future releases when you subscribe now! > http://p.sf.net/sfu/learndevnow-dev2 > _______________________________________________ > Csound-devel mailing list > Csound-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/csound-devel > |
Date | 2012-02-03 16:22 |
From | Victor.Lazzarini@nuim.ie |
Subject | Re: [Cs-dev] Draft Release Notes |
Attachments | None None None None |
Date | 2012-02-03 16:24 |
From | Victor.Lazzarini@nuim.ie |
Subject | Re: [Cs-dev] Draft Release Notes |
Attachments | None None None None |
Date | 2012-02-03 18:54 |
From | "Dr. Richard Boulanger" |
Subject | Re: [Cs-dev] Draft Release Notes |
Attachments | None None |
Awesome stuff and incredible work under the hood. What you might not know is that Tito has fed 18GB of .csd files from my latest and "almost complete" Csound Instrument Collection to the new parser to flesh out some of the critters and helped me find tons if my bugs too! Steven started the process by feeding the original Csound Catalog to the new parser and Tito continued. So.... A new and very very complete collection will be available soon... Still organizing and that may take a month, but they will All work with the new parser..... Thanks to All The Wonderful Developers. Dr. B. Sent from my iPhone.
|
Date | 2012-02-03 19:16 |
From | Rory Walsh |
Subject | Re: [Cs-dev] Draft Release Notes |
That's one heck of a way to launch the new parser. 18GBs of Csound instruments! On 3 February 2012 18:54, Dr. Richard Boulanger |
Date | 2012-02-03 19:54 |
From | Victor Lazzarini |
Subject | Re: [Cs-dev] Draft Release Notes |
yes, we need to thank John, Steven and now Tito for making the new parser a reality. On 3 Feb 2012, at 19:16, Rory Walsh wrote: > That's one heck of a way to launch the new parser. 18GBs of Csound instruments! > > On 3 February 2012 18:54, Dr. Richard Boulanger |
Date | 2012-02-03 20:28 |
From | joachim heintz |
Subject | Re: [Cs-dev] Draft Release Notes |
yes - many many thanks to all of you! it's already much more fun; for instance writing UDOs with strings as output. and as john said once, many more things are possible now ... hope you can go on - joachim Am 03.02.2012 20:54, schrieb Victor Lazzarini: > yes, we need to thank John, Steven and now Tito for making the new parser a reality. > On 3 Feb 2012, at 19:16, Rory Walsh wrote: > >> That's one heck of a way to launch the new parser. 18GBs of Csound instruments! >> >> On 3 February 2012 18:54, Dr. Richard Boulanger |
Date | 2012-02-03 20:46 |
From | Adam Puckett |
Subject | Re: [Cs-dev] Draft Release Notes |
Joachim, I thought writing UDOs was already possible - see Modular Instruments II: http://www.thumbuki.com/2007/07/02/modular-instruments-part-ii/ On 2/3/12, joachim heintz |
Date | 2012-02-03 21:10 |
From | "Dr. Richard Boulanger" |
Subject | Re: [Cs-dev] Draft Release Notes |
It's been "thoroughly" tested. Sent from my iPhone. On Feb 3, 2012, at 2:16 PM, Rory Walsh |
Date | 2012-02-03 21:20 |
From | joachim heintz |
Subject | Re: [Cs-dev] Draft Release Notes |
i meant UDOs with strings as output. this has not been possible some releases ago, as far as i know. best - joachim Am 03.02.2012 21:46, schrieb Adam Puckett: > Joachim, > > I thought writing UDOs was already possible - see Modular Instruments > II: http://www.thumbuki.com/2007/07/02/modular-instruments-part-ii/ > > > > On 2/3/12, joachim heintz |
Date | 2012-02-03 23:59 |
From | peiman khosravi |
Subject | Re: [Cs-dev] Draft Release Notes |
Attachments | None None |
I've been away from composing and therefore Csound for quite some time due to other, more pressing matters but this thread has excited me to get right back into it! This is truly impressive stuff. Thanks to all the developers, contributors and testers. Best, Peiman On 3 February 2012 21:20, joachim heintz <jh@joachimheintz.de> wrote: i meant UDOs with strings as output. this has not been possible some |