Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - rwdobson

Pages: 1 2 [3] 4 5
31
Announce / Re: Can't get pvoc anal to work - ERROR: INVALID DATA
« on: March 13, 2017, 04:36:11 PM »
Yes, the .ana extension is required; CDP uses a fixed set of custom file formats for all its frequency domain data, each identified by its extension, e.g. .ana, .env, .pch etc. Internally they are all WAVE format files.

The reported error suggests that this is an earlier version of the program which requires the environment variable CDP_SOUND_EXT to be set (to wav). The recently updated v7.1 sources on Github have removed this requirement. You can test this by setting the environment variable directly at the console ("export CDP_SOUND_EXT=wav") and then running pvoc. If the progam then runs, you know that it is an older version. The variable is effectively obsolete these days now that all Apple machines are Intel-based.



32
Announce / Re: CDP 7.1 sources now on github
« on: January 18, 2017, 08:14:09 AM »
It is all fully cross-platform (32/64bit compatible), intended that way from the outset. John Fitch updated the distro now on github after checking through and adjusting a couple of things to make sure it all builds on Linux. So if for whatever reason it is not, please let me know with as much detail as possible.

33
Announce / CDP 7.1 sources now on github
« on: December 21, 2016, 02:23:12 PM »
Get it in the usual way by cloning a new repository

https://github.com/ComposersDesktop/CDP7

With thanks to John ffitch for getting it all set up.

The recommended build system uses CMake, which will need to be installed separately if not already available.

Building for Windows requires the Mingw command line environment with gcc etc.

34
General Board / Re: Grainmill error
« on: July 24, 2016, 11:46:28 AM »
It will be helpful to know what version of Windows is presenting the problem. I only have XP and Win7 machines at home.

The first thing to check is if the path to the temporary file is correct/legal - this is shown in the Settings dialog. It should show something like: C:\Users\Richard\AppData\Local\Temp; but can be set to anything suitable (best to avoid paths with spaces in, but I can't state offhand whether that actually matters for Grainmill - in theory it should not).

35
General Board / Re: Can I get 24-bit output?
« on: May 24, 2016, 10:38:27 PM »
24bit operation is definitely supposed to happen. A bug somehow got into the sound filing system routines some while ago (probably because of general code updates getting crossed in the post), which made all output revert to 16 bit as reported. I have completed a full rebuild of all programs which resolves this, restoring the intended behaviour, and I hope to have these available within the next couple of weeks (I am about to be away for a week) on the CDP website, with revised sources in due course on github.

The intended behaviour is for the output file to match the format of the input file, so that e.g. 24bit input gives 24bit output; with the option to force the output to floats using the -f prefix.

36
I would be the one to reply to this, but as it happens I am about to go away for a week and cannot run the tests until i get back.  However, I would not expect floats to be bitwise identical after any rescaling operation (even a seemingly "exact" inverse may not prove to be so in practice, in floating-point - the lowest f/p bits are very flaky!), as there will inevitably be a cocktail of truncation and rounding errors. The precision of 32bit floats is nominally just one bit better than 24bit integers (or 6/7 decimal digits), so there will always be likely to be more or less correlated errors in the lowest bit(s). Basic conversion float/int should result in bit-identical values, but using "pure" floats I would not expect always to "get away with it" in terms of bit-identical results. The differences will need helium-cooled hardware (and ears) to detect.

37
Ah indeed; I will need to check the version, as it is ~supposed~ to write a float file! It was a facility incorporated many years ago; unfortunately a collision of sources at one point resulted in a regression defaulting to 16bit, and it had to be un-regressed, which was done about a year ago (it is a generic facility for all applicable output files). Somehow you have one of the older versions. It is too late to look at that tonight, but I will investigate further tomorrow.


38
It needs to be literally prefixed, without a space: -ffile.wav

39
Unfortunately it is not really possible to find the true net amplitude of an analysis frame other than by resynthesising it. Bins are not truly independent (may reinforce/cancel mutually), and in principle the contribution of the analysis and resynthesis windows should also be taken into account. Any computation to find the level would probably take longer than the already efficient resynthesis.

One solution is indeed to pre-normalise the input, best to a level below digital peak, e.g. -3dB or -6dB, process, and hope. The alternative is to resynthesise to a floating point sound file (on the command line, prepend -f to the outfile name), which will preserve over-range samples without clipping, and then rescale as required.

40
Mac installation / Re: Problems installing on 10.6
« on: February 18, 2016, 02:03:26 PM »
It will be helpful if you can just confirm (using Finder) the programs really are there, inside the _cdprogs folder. If they are, it is definitely puzzling. Please also run the following commands in Terminal, to confirm the hidden file .sloomrc exists and gives the correct path:

cd ~
cat .sloomrc

The first is only needed if you have Terminal open in some directory other than your home directory. Terminal always starts up there.
the second command just writes the contents of the file to the console. It shoudl give the full path to the _cdp folder (should be inside "cdpr7").

What version of Soundloom do you have? It is stated on the opening flash window.

The other aspect worth checking is file and folder permissions. This can be done in the finder by selecting the _cdp folder and then "Show Info" from the File menu. If the "read only" box is shown checked, it will need to be unchecked, as Soundloom may be prevented from updating its files, running programs, etc.




41
Mac installation / Re: Problems installing on 10.6
« on: February 17, 2016, 10:17:13 PM »
The .cdp files are plain text files (mostly very small) used behind the scenes by Soundloom. There is no need in the normal run of things to open and read them directly. Have you also installed the CDP programs? Soundloom is a "front end" GUI for them, and has no functionality without them - it may appear to fail "silently".

If you used the program "Unarchiver" to unpack things, it is probable that file permissions have got messed up; the standard native Mac tools should be used.

42
General Board / Re: Issue in soundshaper with put formant function
« on: December 17, 2015, 11:30:00 AM »
Just to check - do you have the environment variable CDP_SOUND_EXT defined (to wav)? If not, it is likely there was an error writing a file, which would break the scheme whereby Soundshaper splits stereo files into multiple mono files (with suffix _c1, _c2 etc)  before running frequency domain processes, all of which (including pvoc) only process mono files.


43
General Board / Re: Changing WORKSPACE Location in Sound Loom
« on: June 29, 2015, 12:03:55 PM »
I am pretty sure the answer is No, but I have passed this on to Trevor Wishart, in case he has anything further or different to say. He may be busy and/or away, so a quick reply may not be possible.

44
Breakpoint files are not in any way functionally or semantically connected to a soundfile (it would be very common for the same file to be used with many different soundfiles). If the soundfile happens to be longer than the final duration of a breakpoint file, the final value is sustained. All CDP programs parse the breakpoint file to ensure time values are always increasing. I suppose, given a basic utility to report a duration,  a shell script, perl program etc could be used to detect a placeholder string such as "VAL" and replace it with a value extracted from the console output of "sfprops" or "sndinfo len", or from whatever new utility was written to do the job.

45
General Board / Re: non-linear data in bpf, for example pitches?
« on: June 29, 2015, 11:43:19 AM »
Breakpoint files do not have a single format, and are strictly speaking not self-describing - each program determines what they should contain. thus while some programs expect frequencies, or raw amplitudes, others may expect MIDI numbers or dB; a few (e.g. mixfiles for "submix" etc) allow you to use either amplitude or dB. Many years ago Martin Atkins developed an 'extended" breakpoint file format when creating the Windows GUI program "BrkEdit" (which as I recall runs very nicely under Wine on Linux); this includes the ability to specify logarithmic or exponential spans between points, with a parameter for slope. In the event it has not gained wider currency – Soundloom for example does not use it. Soundloom uses a very complex program "cdparse" behind the scenes to scan any breakpoint file and try to work out which programs would use it, so that they can be enabled in the Process panel.

For modifying breakpoint files (which are mostly space-separated columns of numbers; a few require some text as well) the workhorse program is "columns". Again this it not designed to convert a single value and return the result to stdout unix-style, but to take a breakpoint file as input, split it into single column files if necessary, and generate a new file as necessary. It was also designed (as I note elsewhere) very much for SoundLoom, and may need scripts to be written to drive it comfortably from the command line.

Pages: 1 2 [3] 4 5