[Csnd] windows asio audio problem
Date | 2009-02-27 21:31 |
From | Libero Mureddu |
Subject | [Csnd] windows asio audio problem |
Hi, I cannot get a clean sound with the command line options I'm using, I always get clicks: csound -d -b1024 -B1024 -W -+rtaudio=PortAudio -odac6 (my mackie interface) band.csd I tried different combinations of -B and -b but it doesn't help. The version I'm using is: Csound version 5.10 (double samples) Jan 9 2009 Another problem is that even if I specify dac6, when I hit enter it always asks about a not found MOTU interface: Sometimes I use a MOTU traveler, so I'd like to keep its drivers installed. Same happens with pure-data, but once I select the right device, the warning popup window doesn't show anymore. Finally, the http://www.cockos.com/wiki/index.php/ReaRoute rearoute asio driver from reaper is absent from the devices list. However, in pure-data, using asio via portaudio, it's there. Windows xp home sp3 Intel Core 2 Quad Q6600 / 2.4 GH Mackie satellite audio interface Any suggestion? I'm not really a frequent user of csound, so it's quite possible that I'm missing something really obvious. thanks a lot (especially for qutecsound, looks really great!) Libero Mureddu -- View this message in context: http://www.nabble.com/windows-asio-audio-problem-tp22254858p22254858.html Sent from the Csound - General mailing list archive at Nabble.com. |
Date | 2009-03-01 14:31 |
From | Oeyvind Brandtsegg |
Subject | [Csnd] Re: windows asio audio problem |
Just some suggestions, maybe it helps maybe not. The ksmps value, -b and -B should be in integer ratios, ksmps can be == -b -B should be at least two times -b for example ksmps = 100 -b = 400 -B = 800 for really slow drivers: ksmps = 100 -b = 1000 -B = 2000 It might also be easier to help you if you include a console print from your Csound session, so we can verify that it indeed does what you expect (uses the correct drivers etc.). best Oeyvind 2009/2/27 Libero Mureddu |
Date | 2009-03-03 15:13 |
From | Libero Mureddu |
Subject | [Csnd] Re: windows asio audio problem |
Oeyvind Brandtsegg-2 wrote: > > > > It might also be easier to help you if you include a console print > from your Csound session, > > 2009/2/27 Libero Mureddu |
Date | 2009-03-03 15:21 |
From | Oeyvind Brandtsegg |
Subject | [Csnd] Re: Re: windows asio audio problem |
I have never tested the Mackie audio interface, but it seems you should use much larger buffers as a start. Did you try -b1024 -B2048 ? This will give terrible latency, you should be able to use -b128 but clearly this does not currently work on your system, so you could try starting higher. best Oeyvind 2009/3/3 Libero Mureddu |
Date | 2009-03-03 16:17 |
From | Libero Mureddu |
Subject | [Csnd] Re: Re: windows asio audio problem |
Oeyvind Brandtsegg-2 wrote: > > Did you try -b1024 -B2048 ? > This will give terrible latency, you should be able to use -b128 but > clearly this does not currently work on your system, so you could try > starting higher. > > best > Oeyvind > Hi, yes, I forgot to mention that I tried different values but it didn't help (in qutecsound, it just slightly reduced the clicks). However, starting from the command line or from csound5gui everything seems to work, now that you explained me the -b/-B/ksmps relation, so it might be a problem related with qutecsound. But then, i cannot find the source of the problem: very simple examples like the toot2 (one oscil playing 4 notes), gives lots of clicks, while bigger ones (like "trapped" from csound example folder) works, even in high resolution. if it can help, I can add that I don't usually have any buffer issue with my setup, even working with really low latencies (64 samples or even 32), with quite heavy audio projects. thanks a lot and best regards, Libero -- View this message in context: http://www.nabble.com/windows-asio-audio-problem-tp22254858p22311874.html Sent from the Csound - General mailing list archive at Nabble.com. |
Date | 2009-03-03 16:22 |
From | joachim heintz |
Subject | [Csnd] Re: Re: Re: windows asio audio problem |
Did you choose "Run in seperate thread" (in Configuration -> Run -> Csound execution)? Am 03.03.2009 um 17:17 schrieb Libero Mureddu: > > > > Oeyvind Brandtsegg-2 wrote: >> >> Did you try -b1024 -B2048 ? >> This will give terrible latency, you should be able to use -b128 but >> clearly this does not currently work on your system, so you could try >> starting higher. >> >> best >> Oeyvind >> > Hi, > yes, I forgot to mention that I tried different values but it didn't > help > (in qutecsound, it just slightly reduced the clicks). However, > starting from > the command line or from csound5gui everything seems to work, now > that you > explained me the -b/-B/ksmps relation, so it might be a problem > related with > qutecsound. > But then, i cannot find the source of the problem: very simple > examples like > the toot2 (one oscil playing 4 notes), gives lots of clicks, while > bigger > ones (like "trapped" from csound example folder) works, even in high > resolution. > if it can help, I can add that I don't usually have any buffer > issue with > my setup, even working with really low latencies (64 samples or even > 32), > with quite heavy audio projects. > > thanks a lot and best regards, > > Libero > -- > View this message in context: http://www.nabble.com/windows-asio-audio-problem-tp22254858p22311874.html > Sent from the Csound - General mailing list archive at Nabble.com. > > > > Send bugs reports to this list. > To unsubscribe, send email sympa@lists.bath.ac.uk with body > "unsubscribe csound" |
Date | 2009-03-03 18:02 |
From | Libero Mureddu |
Subject | [Csnd] Re: Re: Re: windows asio audio problem |
joachim heintz wrote: > > Did you choose "Run in seperate thread" (in Configuration -> Run -> > Csound execution)? > > Yes, it was already selected, I tried with and without it but it didn't change. But I just discovered that the problematic files work if I select "run in term" from qutecsound interface. Checking from the task manager, i discovered that when I run csound from qutecsound, the CPU used is between 25/30% while, running from the terminal, it barely reaches 0.1%. Also the memory use seems quite high (40508 KB). My system is: Windows xp home sp3 Intel Core 2 Quad Q6600 / 2.4 GHz Mackie satellite audio interface -- View this message in context: http://www.nabble.com/windows-asio-audio-problem-tp22254858p22314250.html Sent from the Csound - General mailing list archive at Nabble.com. |
Date | 2009-03-04 03:07 |
From | Andres Cabrera |
Subject | [Csnd] Re: Re: Re: Re: windows asio audio problem |
Hi, On my superficial tests on Linux, the performance hit of running in QuteCsound instead of on a terminal are very small. Can you try running on a terminal from QuteCsound, to make sure the exact same arguments are used? If you see adequate performance on the command line or Csound5GUI, you should be able to get comparable performance usign QuteCsound. Cheers, Andrés On Tue, Mar 3, 2009 at 1:02 PM, Libero Mureddu |
Date | 2009-03-04 13:56 |
From | Libero Mureddu |
Subject | [Csnd] Re: Re: Re: Re: windows asio audio problem |
Andres Cabrera wrote: > > Hi, > > On my superficial tests on Linux, the performance hit of running in > QuteCsound instead of on a terminal are very small. Can you try > running on a terminal from QuteCsound, to make sure the exact same > arguments are used? If you see adequate performance on the command > line or Csound5GUI, you should be able to get comparable performance > usign QuteCsound. > Cheers, > Andrés > Hi, maybe it wasn't clear from my previous post, sorry about that. In fact the problem I'm reporting (the cpu issue) happens inside qutecsound, from the command line it works. to resume the situation: Same example: toot2 from qutecsound example folder (see attachment) http://www.nabble.com/file/p22330451/Toot2.csd Toot2.csd 1)inside qutecsound - run from terminal: cpu 0.1% (see attachment) http://www.nabble.com/file/p22330451/qutecsound-output-run-in-term.txt qutecsound-output-run-in-term.txt - run from "run"button: cpu 25/30% (see attachment) http://www.nabble.com/file/p22330451/qutecsound-output-run.txt qutecsound-output-run.txt 2) command line cpu 0%(doesn't even show 0.1) 3) csound5gui cpu 0.1/02% Windows xp home sp3 Intel Core 2 Quad Q6600 / 2.4 GHz Mackie satellite audio interface Ciao libero -- View this message in context: http://www.nabble.com/windows-asio-audio-problem-tp22254858p22330451.html Sent from the Csound - General mailing list archive at Nabble.com. |
Date | 2009-03-04 13:56 |
From | Libero Mureddu |
Subject | [Csnd] Re: Re: Re: Re: windows asio audio problem |
Andres Cabrera wrote: > > Hi, > > On my superficial tests on Linux, the performance hit of running in > QuteCsound instead of on a terminal are very small. Can you try > running on a terminal from QuteCsound, to make sure the exact same > arguments are used? If you see adequate performance on the command > line or Csound5GUI, you should be able to get comparable performance > usign QuteCsound. > Cheers, > Andrés > Hi, maybe it wasn't clear from my previous post, sorry about that. In fact the problem I'm reporting (the cpu issue) happens inside qutecsound, from the command line it works. to resume the situation: Same example: toot2 from qutecsound example folder (see attachment) http://www.nabble.com/file/p22330462/Toot2.csd Toot2.csd 1)inside qutecsound - run from terminal: cpu 0.1% (see attachment) http://www.nabble.com/file/p22330462/qutecsound-output-run-in-term.txt qutecsound-output-run-in-term.txt - run from "run"button: cpu 25/30% (see attachment) http://www.nabble.com/file/p22330462/qutecsound-output-run.txt qutecsound-output-run.txt 2) command line cpu 0%(doesn't even show 0.1) 3) csound5gui cpu 0.1/02% Windows xp home sp3 Intel Core 2 Quad Q6600 / 2.4 GHz Mackie satellite audio interface Ciao libero -- View this message in context: http://www.nabble.com/windows-asio-audio-problem-tp22254858p22330462.html Sent from the Csound - General mailing list archive at Nabble.com. |
Date | 2009-03-04 15:28 |
From | Andres Cabrera |
Subject | [Csnd] Re: Re: Re: Re: Re: windows asio audio problem |
Thanks for the detailed report. I'll investigate. Cheers, Andrés On Wed, Mar 4, 2009 at 9:56 AM, Libero Mureddu |