Hi all,
http://crca.ucsd.edu/~msp/software.html now offers Pd version 0.35 test 7...
Bug fixes, plus, for Linux, the "feature" that, if an audio or MIDI device is opened for reading and writing, Pd opens it a single time, hopefully making it work better with certain drivers.
cheers Miller
hi Miller,
I have just tried (hard) to understand array flipping in test7, but failed. The Y-bounds in [table] object's properties are flipped, to ensure proper graphing into [table]. But the regular array shows up with -1 to 1 Y-bounds (as before), and the drawing is inverted.
Are we to flip array's Y-bounds manually, or are you going to change this somehow?
Krzysztof
hi again,
...and also, while freeing a graph in test7, there is an attempt to unbind "pd-graph<n>". There is no such symbol, because graphs appear to be bound to symbols without leading "pd-". Well, at least it looks like that, because after freeing a graph, and sending anything to "graph<n>", Pd crashes...
Krzysztof
hi yet again,
I would prefer pdtk_openpanel to use enquoted $filename.
32-bit float WAVEs are loading fine, but still 16-byte header
is assumed, which may be enough for a while (e.g. all my files have 16-byte header) -- but cf ``floating point soundfiles'' thread, esp. Richard Dobson's detailed explanation.
- WAVE chunk skipping still does not work, at least with WaveLab
files. As I reported on Wed, 09 Jan 2002, the (temporary!) fix is in line 266 of d_soundfile.c:
long skip = swap4(((t_wave *)buf)->w_datachunksize, swap) + 4;
long skip = swap4(((t_wave *)buf)->w_datachunksize, swap);
and in lines 274-5:
headersize += skip;if (read(fd, buf, sizeof(t_wave)) < (int)sizeof(t_wave))
headersize += skip + 8;if (read(fd, ((t_wave *)buf)->w_datachunkid, 8) < 8)
This fix is temporary, because it uses the current t_wave structure, which needs to be adjusted in order to work with non 16-byte header lengths.
Krzysztof
On Tue, Jan 29, 2002 at 04:07:57PM +0100, Krzysztof Czaja wrote:
hi yet again,
- I would prefer pdtk_openpanel to use enquoted $filename.
yep. Thanks.
- 32-bit float WAVEs are loading fine, but still 16-byte header
is assumed, which may be enough for a while (e.g. all my files have 16-byte header) -- but cf ``floating point soundfiles'' thread, esp. Richard Dobson's detailed explanation.
- WAVE chunk skipping still does not work, at least with WaveLab
files. As I reported on Wed, 09 Jan 2002, the (temporary!) fix is in line 266 of d_soundfile.c:
long skip = swap4(((t_wave *)buf)->w_datachunksize, swap) + 4;
long skip = swap4(((t_wave *)buf)->w_datachunksize, swap);and in lines 274-5:
headersize += skip;if (read(fd, buf, sizeof(t_wave)) < (int)sizeof(t_wave))
headersize += skip + 8;if (read(fd, ((t_wave *)buf)->w_datachunkid, 8) < 8)This fix is temporary, because it uses the current t_wave structure, which needs to be adjusted in order to work with non 16-byte header lengths.
I updated my copy; I'll put non-16-byte headers off for a calmer time.
thanks Miller
Krzysztof
-
Krzysztof Czaja -
Miller Puckette