hu.
just installed pd-0.25TEST7 (which in fact shows TEST8 when started verbose .. ;) right now, on a redhaat 5.2, it refused to come up with audio, because there was no /dev/dsp0, which it was looking for. a link did it. and on all three machines i hab the same problem with the ggext pkg. so, before any of the objects in ggext work, i have to call an object 'ggext' itself, then it displays some init info pd: ** pd: loaded ggext version 0.5 (C) 1999 Guenter Geiger pd: ** , then i all the other objects work.
another thing: my laptop seems to have serious trouble with both receiving or transmitting audio via streamin/out. sounds like everything's coming in bursts or pulses , then there'S loads of collissions on the ethernet and it also disturbs the other streams.. maybe this is some sideeffect of ((too) old) pcmcia drivers? i tried it with 2 different cards, both with the same result. so its fine from pc to pc, but with either from or to laptop it sounds the same, .. anybody experienced this?
yet another thing: did i already ask about pd outputting pcm data to standard out by flipping a switch?
and a sub problem: i wanted to test 'stream' from yet another pc now (old SuSe lnx) , but pd didnt compile on it with:
cd ../obj; cc -Wl,-export-dynamic -o ../bin/pd-oss g_canvas.o g_graph.o g_text.o g_rtext.o g_array.o g_template.o g_io.o g_scalar.o g_traversal.o m_pd.o m_class.o m_obj.o m_atom.o m_memory.o m_binbuf.o m_conf.o m_glob.o m_sched.o s_main.o s_inter.o s_unix.o s_file.o s_print.o s_loader.o s_path.o s_entry.o d_delay.o x_arithmetic.o x_connective.o x_interface.o x_midi.o x_misc.o x_time.o x_acoustics.o x_net.o x_qlist.o x_gui.o d_ugen.o d_ctl.o d_arithmetic.o d_osc.o d_filter.o d_dac.o d_misc.o d_math.o d_fft.o d_mayer_fft.o d_fftroutine.o d_array.o d_global.o s_linux.o linux_midiparse.o linux_midi_io.o m_stamp.o
-ldl -lm
x_arithmetic.o: In function binop1_pow_bang': /home/ost/pd/pd-0.25TEST7/src/x_arithmetic.c:134: undefined reference topowf'
x_arithmetic.o: In function binop1_pow_float': /home/ost/pd/pd-0.25TEST7/src/x_arithmetic.c:141: undefined reference topowf'
x_arithmetic.o: In function sin_float': /home/ost/pd/pd-0.25TEST7/src/x_arithmetic.c:443: undefined reference tosinf'
x_arithmetic.o: In function cos_float': /home/ost/pd/pd-0.25TEST7/src/x_arithmetic.c:457: undefined reference tocosf'
x_arithmetic.o: In function tan_float': /home/ost/pd/pd-0.25TEST7/src/x_arithmetic.c:471: undefined reference tocosf'
/home/ost/pd/pd-0.25TEST7/src/x_arithmetic.c:472: undefined reference to sinf' x_arithmetic.o: In functionatan_float':
/home/ost/pd/pd-0.25TEST7/src/x_arithmetic.c:487: undefined reference to atanf' x_arithmetic.o: In functionatan2_float':
/home/ost/pd/pd-0.25TEST7/src/x_arithmetic.c:508: undefined reference to atan2f' x_arithmetic.o: In functionsqrt_float':
/home/ost/pd/pd-0.25TEST7/src/x_arithmetic.c:523: undefined reference to sqrtf' x_arithmetic.o: In functionlog_float':
/home/ost/pd/pd-0.25TEST7/src/x_arithmetic.c:538: undefined reference to logf' x_arithmetic.o: In functionexp_float':
/home/ost/pd/pd-0.25TEST7/src/x_arithmetic.c:560: undefined reference to `expf'
make: *** [../bin/pd-oss] Error 1
?
onk, thts about it. .so
-- bzorstogipptude
17.hzV.tRL.478 writes:
hu.
just installed pd-0.25TEST7 (which in fact shows TEST8 when started verbose .. ;) right now, on a redhaat 5.2, it refused to come up with audio, because there was no /dev/dsp0, which it was looking for. a link did it. and on all three machines i hab the same problem with the ggext pkg. so, before any of the objects in ggext work, i have to call an object 'ggext' itself, then it displays some init info pd: ** pd: loaded ggext version 0.5 (C) 1999 Guenter Geiger pd: ** , then i all the other objects work.
my externals are packed into a library now (just like GEM). You have to load them with the -lib switch, or do the thing with the ggext object. Actually, using the ggext object makes sense, because you have the dependency on the library directly written into the patch.
another thing: my laptop seems to have serious trouble with both receiving or transmitting audio via streamin/out. sounds like everything's coming in bursts or pulses , then there'S loads of collissions on the ethernet and it also disturbs the other streams.. maybe this is some sideeffect of ((too) old) pcmcia drivers? i tried it with 2 different cards, both with the same result. so its fine from pc to pc, but with either from or to laptop it sounds the same, .. anybody experienced this?
:( hmm , don't know why .. sorry. Is this on a "real" ethernet or direct (no other hosts on the ethernet) ?
yet another thing: did i already ask about pd outputting pcm data to standard out by flipping a switch?
Best thing would be a sfwrite. Then use it on a named pipe and read the pipe from any program you like. --> I put it on a TODO list for my xternals.
and a sub problem: i wanted to test 'stream' from yet another pc now (old SuSe lnx) , but pd didnt compile on it with:
old libc5 problem. remove the "#ifdef" NT and it's corresponding "#endif" from the top of the file.
Guenter
|-< my externals are packed into a library now (just like GEM). |-< You have to load them with the -lib switch, or do the thing with the |-< ggext object. Actually, using the ggext object makes sense, because |-< you have the dependency on the library directly written into the |-< patch.
oh, i c. ;)) so it would also work with Gem that way? thats cool, so you dont need to care wether you re starting pd with this or that -lib ...
|-< :( hmm , don't know why .. sorry. Is this on a "real" ethernet or |-< direct (no other hosts on the ethernet) ?
its on a 'real' net. but when i testet it last time, it was on an isolated net, on which were only two machines. it showed the same effect. only, i couldnt tell by then it only appeared when streaming from the laptop, since i only had 1 pc and the schleppi to test it. ;) when it compiles on the other pc, i can probably remove all doubt in the evil laptop's guilt. ;)
|-< Best thing would be a sfwrite. Then use it on a named pipe and read |-< the pipe from any program you like. |-< --> I put it on a TODO list for my xternals.
jodel!
|-< > and a sub problem: |-< > i wanted to test 'stream' from yet another pc now (old SuSe lnx) , but pd |-< > didnt compile on it with: |-< > |-< |-< old libc5 problem. |-< remove the |-< "#ifdef" NT and it's corresponding "#endif" from the top of the file.
onk, i m trying. it just takes ages to compile on an already loaded P90 .. ;)))
ok, thanks so far. .so
17.hzV.tRL.478 writes:
oh, i c. ;)) so it would also work with Gem that way?
it does.
thats cool, so you dont need to care wether you re starting pd with this or that -lib ...
Yes, but that's only functional under Linux at the moment. (Which has some notion of path handling for externals). For the other OS' you have to make sure that the library you want to load is in the current directory.
when it compiles on the other pc, i can probably remove all doubt in the evil laptop's guilt. ;)
let me know, under linux (and SGI?) I can tune networking a little bit, so there may still be a solution for the problem.
|-< Best thing would be a sfwrite. Then use it on a named pipe and read |-< the pipe from any program you like. |-< --> I put it on a TODO list for my xternals.
jodel!
don't jodel to early .. the TODO list is long ... :)
Guenter
Guenter Geiger discourseth:
|-< Best thing would be a sfwrite. Then use it on a named pipe and read |-< the pipe from any program you like. |-< --> I put it on a TODO list for my xternals.
jodel!
don't jodel to early .. the TODO list is long ... :)
Ahh..well fifo i/o for both midi and pcm are critical for me..if pd proves as usable as I hope, I'll hack these in if you don't. :)
In other new: my full-duplex problems of a couple of months ago are gone now that I've upgraded to a 2.2.7 kernel. Are there any non-GPL restrictions on the use of your PCI128 driver patches?
Many Thanks,
Eric
est@hyperreal.org writes:
Guenter Geiger discourseth:
|-< Best thing would be a sfwrite. Then use it on a named pipe and read |-< the pipe from any program you like. |-< --> I put it on a TODO list for my xternals.
jodel!
don't jodel to early .. the TODO list is long ... :)
Ahh..well fifo i/o for both midi and pcm are critical for me..if pd proves as usable as I hope, I'll hack these in if you don't. :)
Yeah, .. shouldn't be too hard .. let me know if yiu have any success ..
In other new: my full-duplex problems of a couple of months ago are gone now that I've upgraded to a 2.2.7 kernel. Are there any non-GPL restrictions on the use of your PCI128 driver patches?
No, of course not, as I just made a patch, all my changes are under the same restrictions as the original code was. (I wouldn't be allowed to put any other licensing on them.) I hope they will be included in the kernel sooner or later.
Guenter
Many Thanks,
Eric
-
17.hzV.tRL.478 -
est＠hyperreal.org -
Guenter Geiger