[Cs-dev] ksmps=1 conditional issue
| Date | 2012-05-22 07:31 |
| From | Oeyvind Brandtsegg |
| Subject | [Cs-dev] ksmps=1 conditional issue |
Hi, I'm not sure what to make out of this. For pedagogical purposes, I tried to implement an explicit coding of hard clip (like the limit opcode). Using ksmps = 1, I did this: k1 downsamp a1 if k1 > iClip then k1 = iClip elseif k1 < -iClip then k1 = -iClip endif a1 upsamp k1 I can't seem to find anything wrong with the code, but it only limits the positive part of the waveform, the negative part remains untouched. If I replace k1 = -iClip with k1 = -0.5 it will limit, but obviously not the way I wanted. Could look like a bug ? best Oeyvind ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Csound-devel mailing list Csound-devel@lists.sourceforge.net |
| Date | 2012-05-22 10:14 |
| From | Victor Lazzarini |
| Subject | [Cs-dev] New Parser bug (csound 5) ( was Re: ksmps=1 conditional issue). |
It is a new parser bug (possibly optimisation). The assignment never occurs with i-vars, but it does with constants and k-vars, as demonstrated by the CSD below: |
| Date | 2012-05-24 12:50 |
| From | john ffitch |
| Subject | Re: [Cs-dev] New Parser bug (csound 5) ( was Re: ksmps=1 conditional issue). |
As you may be away I never liked the if...then syntax as it is unclear at what rate it is supposed to work. Can you describe the case that does not work? ==John ffitch ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Csound-devel mailing list Csound-devel@lists.sourceforge.net |
| Date | 2012-05-24 13:40 |
| From | jpff@cs.bath.ac.uk |
| Subject | Re: [Cs-dev] New Parser bug (csound 5) ( was Re: ksmps=1 conditional issue). |
Follow up 1)old parser is now broken Csound version 5.17 beta (double samples) May 24 2012 0dBFS level = 32768.0 sr = 44100, kr = 4410, ksmps = 7.410985e-322 error: invalid ksmps value 2) overwriting sr.kr/ksmps it works 3) Victor's example has 6 cases, one failing, which is ...if k1 < - iClip then .... Very strange > As you may be away I never liked the if...then syntax as it is unclear > at what rate it is supposed to work. > > Can you describe the case that does not work? > > ==John ffitch > > ------------------------------------------------------------------------------ > Live Security Virtual Conference > Exclusive live event will cover all the ways today's security and > threat landscape has changed and how IT managers can respond. Discussions > will include endpoint security, mobile security and the latest in malware > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ > _______________________________________________ > Csound-devel mailing list > Csound-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/csound-devel > > > ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Csound-devel mailing list Csound-devel@lists.sourceforge.net |
| Date | 2012-05-24 16:25 |
| From | Victor Lazzarini |
| Subject | Re: [Cs-dev] New Parser bug (csound 5) ( was Re: ksmps=1 conditional issue). |
yes, it's instrument 3 that is buggy. I think there is no ambiguity here, somehow the assignment is never done, and that is why I suggested it could be the optimisation. Regards Victor On 24 May 2012, at 13:40, jpff@cs.bath.ac.uk wrote: > Follow up > > 1)old parser is now broken > Csound version 5.17 beta (double samples) May 24 2012 > 0dBFS level = 32768.0 > sr = 44100, kr = 4410, ksmps = 7.410985e-322 > error: invalid ksmps value > > 2) overwriting sr.kr/ksmps it works > > 3) Victor's example has 6 cases, one failing, which is > ...if k1 < - iClip then .... > > Very strange > > >> As you may be away I never liked the if...then syntax as it is unclear >> at what rate it is supposed to work. >> >> Can you describe the case that does not work? >> >> ==John ffitch >> >> ------------------------------------------------------------------------------ >> Live Security Virtual Conference >> Exclusive live event will cover all the ways today's security and >> threat landscape has changed and how IT managers can respond. Discussions >> will include endpoint security, mobile security and the latest in malware >> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ >> _______________________________________________ >> Csound-devel mailing list >> Csound-devel@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/csound-devel >> >> >> > > > > ------------------------------------------------------------------------------ > Live Security Virtual Conference > Exclusive live event will cover all the ways today's security and > threat landscape has changed and how IT managers can respond. Discussions > will include endpoint security, mobile security and the latest in malware > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ > _______________________________________________ > Csound-devel mailing list > Csound-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/csound-devel Dr Victor Lazzarini Senior Lecturer Dept. of Music NUI Maynooth Ireland tel.: +353 1 708 3545 Victor dot Lazzarini AT nuim dot ie ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Csound-devel mailing list Csound-devel@lists.sourceforge.net |
| Date | 2012-05-24 17:25 |
| From | jpff@cs.bath.ac.uk |
| Subject | Re: [Cs-dev] New Parser bug (csound 5) ( was Re: ksmps=1 conditional issue). |
> yes, it's instrument 3 that is buggy. I think there is no ambiguity here, > somehow the assignment is never done, and that is why I suggested it could > be the optimisation. > > Regards > > Victor More information: note the line marked with a ********* It looks as if the assignment is "optimised" from itmp = (-1)*iClip k1 = itmp to k1 = k(itmp) which happens at i-time. I guess such anoptimisation is only valid if rate match. should optimise to k1 mult.ii -1, iClip which does not exist..... Not sure how to fix ==John Found IF statement Found if-then Creating boolean expression Operator Found: > (k i) Creating Synthetic T_IDENT: __synthetic_301 Engine/csound_orc_semantics.c(360) line = -1 Creating Synthetic label: __synthetic_301: Engine/csound_orc_semantics.c(360) line = -1 Creating Synthetic T_IDENT: __synthetic_300 Engine/csound_orc_semantics.c(360) line = -1 Creating simple goto token Engine/csound_orc_semantics.c(339) line = 16 Creating boolean expression HANDLING UNARY MINUS!SET_EXPRESSION_TYPE: mul.ii : #i0 Operator Found: < (k i) Creating Synthetic T_IDENT: __synthetic_302 Engine/csound_orc_semantics.c(360) line = -1 Creating Synthetic label: __synthetic_302: Engine/csound_orc_semantics.c(360) line = -1 Creating Synthetic T_IDENT: __synthetic_300 Engine/csound_orc_semantics.c(360) line = -1 Creating simple goto token Creating Synthetic label: __synthetic_300: Engine/csound_orc_semantics.c(360) line = -1 Found Statement. Handling Optional Args for opcode cngoto, 2, 2Found Statement. Found Assignment type: =.k Handling Optional Args for opcode =.k, 1, 1Found Statement. Handling Optional Args for opcode kgoto, 1, 1Found Statement. Handling Optional Args for opcode mul.ii, 2, 2Found Statement. Handling Optional Args for opcode <, 2, 2Found Statement. Handling Optional Args for opcode cngoto, 2, 2HANDLING UNARY MINUS!SET_EXPRESSIO N_TYPE: mul.ii : #i1 Engine/csound_orc_semantics.c(360) line = 16 Found Statement. Handling Optional Args for opcode k.i, 1, 1Found Statement. *********** Handling Optional Args for opcode kgoto, 1, 1Found Statement. Handling Optional Args for opcode printk, 2, 3Engine/csound_orc_semantics.c(360) line = 17 ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Csound-devel mailing list Csound-devel@lists.sourceforge.net |
| Date | 2012-05-24 17:50 |
| From | jpff@cs.bath.ac.uk |
| Subject | Re: [Cs-dev] New Parser bug (csound 5) ( was Re: ksmps=1 conditional issue). |
Fixed -- trivial really.... ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Csound-devel mailing list Csound-devel@lists.sourceforge.net |
| Date | 2012-05-24 17:57 |
| From | Victor |
| Subject | Re: [Cs-dev] New Parser bug (csound 5) ( was Re: ksmps=1 conditional issue). |
Nothing like really knowing your stuff... On 24 May 2012, at 17:50, jpff@cs.bath.ac.uk wrote: > Fixed -- trivial really.... > > > > ------------------------------------------------------------------------------ > Live Security Virtual Conference > Exclusive live event will cover all the ways today's security and > threat landscape has changed and how IT managers can respond. Discussions > will include endpoint security, mobile security and the latest in malware > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ > _______________________________________________ > Csound-devel mailing list > Csound-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/csound-devel ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Csound-devel mailing list Csound-devel@lists.sourceforge.net |