[Cs-dev] csound6 alpha --sample-accurate error
| Date | 2013-02-28 00:54 |
| From | peiman khosravi |
| Subject | [Cs-dev] csound6 alpha --sample-accurate error |
| Attachments | barmodel.csd None None |
It looks like there are some issues with --sample-accurate. The attached csd (using 'barmodel') works in normal mode but prints 2 errors with --sample-accurate (see below). I hadn't caught this before because my previus csd just had one note in them. It sounds like discontinuities when there is a new note. Best, Peiman SECTION 1: new alloc for instr 1: >>>> no_end=14 (1379,16,22050.000000,0) B 0.000 .. 0.500 T 0.500 TT 0.500 M: 16380.7 0.0 new alloc for instr 1:
INIT ERROR in instr 1: No data to continue aq barmodel 1 1 p4 0.001 0.23 5 p5 p6 p7
B 0.500 - note deleted. i1 had 1 init errors B 0.500 .. 1.000 T 1.000 TT 1.000 M: 595.8 0.0 >>>> no_end=10 (1379,16,22050.000000,4)
B 1.000 .. 1.500 T 1.500 TT 1.500 M: 49213.5 0.0 number of samples out of range: 615 0 INIT ERROR in instr 1: No data to continue
aq barmodel 1 1 p4 0.001 0.23 5 p5 p6 p7
B 1.500 - note deleted. i1 had 1 init errors B 1.500 .. 1.500 T 1.500 TT 1.500 M: 3.0 0.0 Score finished in csoundPerformKsmps().
inactive allocs returned to freespace end of score. overall amps: 49213.5 0.0 overall samples out of range: 615 0
2 errors in performance Elapsed time at end of performance: real: 1.570s, CPU: 0.153s 130 1024 sample blks of 64-bit floats written to dac
|
| Date | 2013-02-28 00:55 |
| From | peiman khosravi |
| Subject | Re: [Cs-dev] csound6 alpha --sample-accurate error |
| Attachments | None None |
| Here is the csd in the email body. <CsoundSynthesizer> <CsOptions> -odac -d -A -3 -b1024 -B1024 </CsOptions>
<CsInstruments> sr = 44100 ksmps = 16 nchnls = 2 ; Instrument #1. instr 1 aq barmodel 1, 1, p4, 0.001, 0.23, 5, p5, p6, p7
out aq endin </CsInstruments> <CsScore> i1 0.0 0.5 3 0.2 500 0.05 i1 0.5 0.5 -3 0.3 1000 0.05
i1 1.0 0.5 -3 0.4 1000 0.1 i1 1.5 4.0 -3 0.5 800 0.05 e /* barmodel */ </CsScore> </CsoundSynthesizer>
On 28 February 2013 00:54, peiman khosravi <peimankhosravi@gmail.com> wrote:
|
| Date | 2013-02-28 00:59 |
| From | Victor Lazzarini |
| Subject | Re: [Cs-dev] csound6 alpha --sample-accurate error |
It's the tied notes I think. We'll need to fix them. Victor On 28 Feb 2013, at 00:54, peiman khosravi wrote: > It looks like there are some issues with --sample-accurate. > > The attached csd (using 'barmodel') works in normal mode but prints 2 errors with --sample-accurate (see below). > > I hadn't caught this before because my previus csd just had one note in them. It sounds like discontinuities when there is a new note. > > Best, > Peiman > > SECTION 1: > new alloc for instr 1: > >>>> no_end=14 (1379,16,22050.000000,0) > B 0.000 .. 0.500 T 0.500 TT 0.500 M: 16380.7 0.0 > new alloc for instr 1: > INIT ERROR in instr 1: No data to continue > aq barmodel 1 1 p4 0.001 0.23 5 p5 p6 p7 > B 0.500 - note deleted. i1 had 1 init errors > B 0.500 .. 1.000 T 1.000 TT 1.000 M: 595.8 0.0 > >>>> no_end=10 (1379,16,22050.000000,4) > B 1.000 .. 1.500 T 1.500 TT 1.500 M: 49213.5 0.0 > number of samples out of range: 615 0 > INIT ERROR in instr 1: No data to continue > aq barmodel 1 1 p4 0.001 0.23 5 p5 p6 p7 > B 1.500 - note deleted. i1 had 1 init errors > B 1.500 .. 1.500 T 1.500 TT 1.500 M: 3.0 0.0 > Score finished in csoundPerformKsmps(). > inactive allocs returned to freespace > end of score. overall amps: 49213.5 0.0 > overall samples out of range: 615 0 > 2 errors in performance > Elapsed time at end of performance: real: 1.570s, CPU: 0.153s > 130 1024 sample blks of 64-bit floats written to dac > |
| Date | 2013-02-28 01:01 |
| From | Victor Lazzarini |
| Subject | Re: [Cs-dev] csound6 alpha --sample-accurate error |
yes, negative p3 seems to be a problem. On 28 Feb 2013, at 00:54, peiman khosravi wrote: > It looks like there are some issues with --sample-accurate. > > The attached csd (using 'barmodel') works in normal mode but prints 2 errors with --sample-accurate (see below). > > I hadn't caught this before because my previus csd just had one note in them. It sounds like discontinuities when there is a new note. > > Best, > Peiman > > SECTION 1: > new alloc for instr 1: > >>>> no_end=14 (1379,16,22050.000000,0) > B 0.000 .. 0.500 T 0.500 TT 0.500 M: 16380.7 0.0 > new alloc for instr 1: > INIT ERROR in instr 1: No data to continue > aq barmodel 1 1 p4 0.001 0.23 5 p5 p6 p7 > B 0.500 - note deleted. i1 had 1 init errors > B 0.500 .. 1.000 T 1.000 TT 1.000 M: 595.8 0.0 > >>>> no_end=10 (1379,16,22050.000000,4) > B 1.000 .. 1.500 T 1.500 TT 1.500 M: 49213.5 0.0 > number of samples out of range: 615 0 > INIT ERROR in instr 1: No data to continue > aq barmodel 1 1 p4 0.001 0.23 5 p5 p6 p7 > B 1.500 - note deleted. i1 had 1 init errors > B 1.500 .. 1.500 T 1.500 TT 1.500 M: 3.0 0.0 > Score finished in csoundPerformKsmps(). > inactive allocs returned to freespace > end of score. overall amps: 49213.5 0.0 > overall samples out of range: 615 0 > 2 errors in performance > Elapsed time at end of performance: real: 1.570s, CPU: 0.153s > 130 1024 sample blks of 64-bit floats written to dac > |
| Date | 2013-02-28 01:11 |
| From | peiman khosravi |
| Subject | Re: [Cs-dev] csound6 alpha --sample-accurate error |
| Attachments | None None |
| But there are no negative p3s is that csd. Is it the same as the note terminating and starting at the same time? On 28 February 2013 01:01, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote: yes, negative p3 seems to be a problem. |
| Date | 2013-02-28 01:19 |
| From | Victor Lazzarini |
| Subject | Re: [Cs-dev] csound6 alpha --sample-accurate error |
| Attachments | None None |
| Sorry, there were negative p4s. I made them positive and the problem disappeared. Something else then. On 28 Feb 2013, at 01:11, peiman khosravi wrote: But there are no negative p3s is that csd. Is it the same as the note terminating and starting at the same time? Dr Victor Lazzarini Senior Lecturer Dept. of Music NUI Maynooth Ireland tel.: +353 1 708 3545 Victor dot Lazzarini AT nuim dot ie |
| Date | 2013-02-28 01:21 |
| From | Victor Lazzarini |
| Subject | Re: [Cs-dev] csound6 alpha --sample-accurate error |
| Attachments | None None |
| it's actually to do with tied notes (see barmodel manual), I think because there is a slight overlap, then it doesn't work. On 28 Feb 2013, at 01:11, peiman khosravi wrote: But there are no negative p3s is that csd. Is it the same as the note terminating and starting at the same time? Dr Victor Lazzarini Senior Lecturer Dept. of Music NUI Maynooth Ireland tel.: +353 1 708 3545 Victor dot Lazzarini AT nuim dot ie |
| Date | 2013-02-28 01:25 |
| From | peiman khosravi |
| Subject | Re: [Cs-dev] csound6 alpha --sample-accurate error |
| Attachments | None None |
| ahh I see. Yes. This makes sense. I think this is related to the tied note issue. P4 represents: iK -- dimensionless stiffness parameter. If this parameter is negative then the initialisation is skipped and the previous state of the bar is continued.
So it's practically a tied not. On 28 February 2013 01:19, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
|
| Date | 2013-02-28 01:25 |
| From | peiman khosravi |
| Subject | Re: [Cs-dev] csound6 alpha --sample-accurate error |
| Attachments | None None |
| Yes, that makes sense. On 28 February 2013 01:21, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
|
| Date | 2013-03-01 09:50 |
| From | Victor Lazzarini |
| Subject | Re: [Cs-dev] csound6 alpha --sample-accurate error |
| Attachments | None None |
OK, I fixed this. If there are overlaps, ties are not allowed for this opcode. Performance does not break up anymore. On 28 Feb 2013, at 01:25, peiman khosravi wrote: Yes, that makes sense. Dr Victor Lazzarini Senior Lecturer Dept. of Music NUI Maynooth Ireland tel.: +353 1 708 3545 Victor dot Lazzarini AT nuim dot ie |
| Date | 2013-03-02 01:19 |
| From | peiman khosravi |
| Subject | Re: [Cs-dev] csound6 alpha --sample-accurate error |
| Attachments | None None |
| Thanks Victor. Is this available as a binary? I was running a script today to test all the manual CSDs but my laptop started to play up. I got some errors but there wasn't time to check them properly. I'll give it another go tomorrow and compile a list of all the errors. In fact I'll run it over night.
Best, Peiman
On 1 March 2013 09:50, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
|
| Date | 2013-03-02 07:22 |
| From | Victor Lazzarini |
| Subject | Re: [Cs-dev] csound6 alpha --sample-accurate error |
| Attachments | None None |
Not yet, but you can carry on your tests with other opcodes. Thanks a lot for doing this. Victor On 2 Mar 2013, at 01:19, peiman khosravi wrote: Thanks Victor. Is this available as a binary? Dr Victor Lazzarini Senior Lecturer Dept. of Music NUI Maynooth Ireland tel.: +353 1 708 3545 Victor dot Lazzarini AT nuim dot ie |
| Date | 2013-03-02 20:13 |
| From | peiman khosravi |
| Subject | Re: [Cs-dev] csound6 alpha --sample-accurate error |
| Attachments | csound6 csds Errors.rtf None None |
| I've run most of the manual csds with only a few left o do. Attached is a text file with a list of all the problematics ones. A lot of these (e.g. ATS) are caused by missing data files or midi devices, as well as fltk and STK opcodes. So all in all it looks generally promising with a few errors that I need to check out. These where all rendered with the sample-accurate option. I'll investigate and let you know but here's the list in case someone beats me to it. Best, Peiman
On 2 March 2013 07:22, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
|
| Date | 2013-03-02 20:29 |
| From | Victor Lazzarini |
| Subject | Re: [Cs-dev] csound6 alpha --sample-accurate error |
| Attachments | None None |
| Thanks a lot. The FLTK and other plugin opcodes were not working because of a bug that made them load too late for syntax checks. I fixed that already. Victor On 2 Mar 2013, at 20:13, peiman khosravi wrote: I've run most of the manual csds with only a few left o do. Attached is a text file with a list of all the problematics ones. A lot of these (e.g. ATS) are caused by missing data files or midi devices, as well as fltk and STK opcodes. So all in all it looks generally promising with a few errors that I need to check out. These where all rendered with the sample-accurate option. Dr Victor Lazzarini Senior Lecturer Dept. of Music NUI Maynooth Ireland tel.: +353 1 708 3545 Victor dot Lazzarini AT nuim dot ie |
| Date | 2013-03-02 20:40 |
| From | Victor Lazzarini |
| Subject | Re: [Cs-dev] csound6 alpha --sample-accurate error |
| Attachments | None None |
| I had a quick look and nothing seems at first to be iffy. All plugins would not be recognised, so the accounts for many of the errors. The opcodes that depend on input files seem to fail because of missing files. The only one I thought was strange was rewindscore.csd. I ran it here and it works fine. Victor On 2 Mar 2013, at 20:13, peiman khosravi wrote: I've run most of the manual csds with only a few left o do. Attached is a text file with a list of all the problematics ones. A lot of these (e.g. ATS) are caused by missing data files or midi devices, as well as fltk and STK opcodes. So all in all it looks generally promising with a few errors that I need to check out. These where all rendered with the sample-accurate option. Dr Victor Lazzarini Senior Lecturer Dept. of Music NUI Maynooth Ireland tel.: +353 1 708 3545 Victor dot Lazzarini AT nuim dot ie |
| Date | 2013-03-02 21:16 |
| From | peiman khosravi |
| Subject | Re: [Cs-dev] csound6 alpha --sample-accurate error |
| Attachments | None None |
Thanks Victor, I'll go ahead and check the remaining opcodes.
Best, Peiman
On 2 March 2013 20:40, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
|
| Date | 2013-03-02 21:23 |
| From | Victor Lazzarini |
| Subject | Re: [Cs-dev] csound6 alpha --sample-accurate error |
| Attachments | None None |
| Thanks, OK. Just remember that with the binaries I made, plugins will not be seen. I'll see if I can update them early next week. Victor On 2 Mar 2013, at 21:16, peiman khosravi wrote:
Dr Victor Lazzarini Senior Lecturer Dept. of Music NUI Maynooth Ireland tel.: +353 1 708 3545 Victor dot Lazzarini AT nuim dot ie |
| Date | 2013-03-02 22:13 |
| From | peiman khosravi |
| Subject | Re: [Cs-dev] csound6 alpha --sample-accurate error |
| Attachments | None None |
| Sure, I'll bear that in mind. But this all seems very positive! Looks like csound6 is well on the way! Thanks very much Peiman
On 2 March 2013 21:23, Victor Lazzarini <Victor.Lazzarini@nuim.ie> wrote:
|