Quick SBLive/AWE32 synth question

Quick SBLive/AWE32 synth question

[Richard Bown]

Lazy question - without searching through bug database or getting latest ALSA CVS
(0.9.0 beta12) is there any history or known bug when repeatedly playing the a note
of the same pitch through the SBLive or AWE32 onboard synth?

Given 4 crotchets in a bar playing through the ALSA seq - routing to a soft
synth (iiwusynth) or external MIDI device the notes sound correctly - routing
through the onboard wavetable synth the first note sounds correctly but the
following three (at the same pitch, remember) sound but have quaver-like durations.

Doesn't sound like an app problem as the interface is the same for any ALSA client
so I'm wondering if anyone has seen this too before I start scouring?

B
-- 
http://www.all-day-breakfast.com/rosegarden
http://www.bownie.com


-------------------------------------------------------
This sf.net email is sponsored by: Dice - The leading online job board
for high-tech professionals. Search and apply for tech jobs today!
http://seeker.dice.com/seeker.epl?rel_code=31

Re: Quick SBLive/AWE32 synth question

[Taupter]

Richard Bown wrote:
> Lazy question - without searching through bug database or getting latest ALSA CVS
> (0.9.0 beta12) is there any history or known bug when repeatedly playing the a note
> of the same pitch through the SBLive or AWE32 onboard synth?
> 
> Given 4 crotchets in a bar playing through the ALSA seq - routing to a soft
> synth (iiwusynth) or external MIDI device the notes sound correctly - routing
> through the onboard wavetable synth the first note sounds correctly but the
> following three (at the same pitch, remember) sound but have quaver-like durations.
> 
> Doesn't sound like an app problem as the interface is the same for any ALSA client
> so I'm wondering if anyone has seen this too before I start scouring?
> 
> B


I heard the same effect on a Soundblaster Live 5.1 loaded with the 
8mbgmsfx.sf2 bank and alsa-driver 0.9.0rc2. (MDK8.2 i586 kernel 
2.4.18-8.1mdk).


Taupter



-------------------------------------------------------
This sf.net email is sponsored by: Dice - The leading online job board
for high-tech professionals. Search and apply for tech jobs today!
http://seeker.dice.com/seeker.epl?rel_code=31

Re: Quick SBLive/AWE32 synth question

[Takashi Iwai]

At Tue, 30 Jul 2002 16:24:02 +0100,
Richard Bown wrote:
> 
> Lazy question - without searching through bug database or getting latest ALSA CVS
> (0.9.0 beta12) is there any history or known bug when repeatedly playing the a note
> of the same pitch through the SBLive or AWE32 onboard synth?
 
can you check whether this happens on SB Live only or both on SB Live
and AWE?

> Given 4 crotchets in a bar playing through the ALSA seq - routing to a soft
> synth (iiwusynth) or external MIDI device the notes sound correctly - routing
> through the onboard wavetable synth the first note sounds correctly but the
> following three (at the same pitch, remember) sound but have quaver-like durations.
 
a similar problem has been reported, but not the identical as i've
heard.  perhaps by the same reason, though.
that is, most likely a long-standing bug of wavetable routine on
emu10k1 driver...   (well, yeah, i know, it must be fixed :)

could you try to make a short midi clip which can cause the same
behavior, so that i can test to reproduce on my machine?

i can try, of couse, rosegarden (actually i did) but it's better to
have a smaller test environemnt :)


ciao,

Takashi


-------------------------------------------------------
This sf.net email is sponsored by: Dice - The leading online job board
for high-tech professionals. Search and apply for tech jobs today!
http://seeker.dice.com/seeker.epl?rel_code=31

Re: Quick SBLive/AWE32 synth question

[Richard Bown]

Takashi Iwai wrote:

> can you check whether this happens on SB Live only or both on SB Live
> and AWE?

Chris and I are both experiencing it - I've got an SBLive and he's the AWE32.
All of us (including Taupter who's the only other one to report this problem
so far on this list) are using Rosegarden so I'm suddenly very paranoid it's
us instead.  I'll double check probably tomorrow and get back to you.

I'm trying to reproduce the problem with another sequencer but to tell the
truth I'm having a problem finding one that does on-board ALSA synths too.
I've got Muse and Brahms but I can't get the former working properly and the
latter is aRts only - there must be a playmidi equivalent for ALSA but for the
moment my mind has gone blank.

> a similar problem has been reported, but not the identical as i've
> heard.  perhaps by the same reason, though.
> that is, most likely a long-standing bug of wavetable routine on
> emu10k1 driver...   (well, yeah, i know, it must be fixed :)

To me this sounds reasonable - from the dim and distant past I remember
fiddling with voice allocation maps when implementing synth output for OSS
and I know how tortuous it can all be.  I also know how flakey my code is
so YMMV.

> could you try to make a short midi clip which can cause the same
> behavior, so that i can test to reproduce on my machine?

I will do once I've convinced myself it's not us.

B


-------------------------------------------------------
This sf.net email is sponsored by: Dice - The leading online job board
for high-tech professionals. Search and apply for tech jobs today!
http://seeker.dice.com/seeker.epl?rel_code=31

Re: Quick SBLive/AWE32 synth question

[Taupter]

Richard Bown wrote:
> Takashi Iwai wrote:
> 
> 
>>can you check whether this happens on SB Live only or both on SB Live
>>and AWE?
> I'm trying to reproduce the problem with another sequencer but to tell the
> truth I'm having a problem finding one that does on-board ALSA synths too.

try noteedit.

http://rnvs.informatik.tu-chemnitz.de/~jan/noteedit/noteedit.html

to get ALSA MIDI output with it you'll need to install the TSE3 library 
by Peter Goodliffe.

http://tse3.sourceforge.net/

>>could you try to make a short midi clip which can cause the same
>>behavior, so that i can test to reproduce on my machine?
> I will do once I've convinced myself it's not us.

I would do the test right now if I could get my ALSA working... I 
reinstalled MDK8.2 and the kernel 2.4.18-8.1mdk. ALSA 0.9.0rc2 compiles 
fine, installs, but

service alsa start

tells me:

Starting ALSA version 
none:/lib/modules/2.4.18-8.1mdk/kernel/sound/acore/snd.o: unresolved 
symbol register_sound_special_R69a88294
/lib/modules/2.4.18-8.1mdk/kernel/sound/acore/snd.o: unresolved symbol 
unregister_sound_special_R99c95fa5
modprobe: insmod /lib/modules/2.4.18-8.1mdk/kernel/sound/acore/snd.o failed
modprobe: insmod snd failed


Any ideas?


Taupter



-------------------------------------------------------
This sf.net email is sponsored by: Dice - The leading online job board
for high-tech professionals. Search and apply for tech jobs today!
http://seeker.dice.com/seeker.epl?rel_code=31

Re: Quick SBLive/AWE32 synth question

[Richard Bown]

Taupter wrote:

[ALSA sequencers and that repeating note problem]

> try noteedit.

Of course!  Ok so, NoteEdit worked fine and so I reinspected our code and lo
and behold found a little bugette with our note off matrix.  Now fixed.

Sorry for the false alarm.

B


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

Re: Quick SBLive/AWE32 synth question

[Takashi Iwai]

At Thu, 1 Aug 2002 10:08:38 +0100,
Richard Bown wrote:
> 
> Taupter wrote:
> 
> [ALSA sequencers and that repeating note problem]
> 
> > try noteedit.
> 
> Of course!  Ok so, NoteEdit worked fine and so I reinspected our code and lo
> and behold found a little bugette with our note off matrix.  Now fixed.

well, so the problem doesn't lie on alsa side?


Takashi


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

Re: Quick SBLive/AWE32 synth question

[Richard Bown]

Takashi Iwai wrote:

> well, so the problem doesn't lie on alsa side?

No, it was in our app code.  I'm afraid this doesn't help you too much if you
suspect there's a bug in the emu10k synth driver anyway - I'll keep listening
out for one though.

B


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf