On Mon, 29 Jan 2007 13:06:34 -0800 (PST) Bill Unruh <unruh@xxxxxxxxxxxxxx> wrote: > > I said "observe", not "hear". I am saying that to roughly estimate that > > a signal contains a 1Hz spectral component one has to observe it (with > > his/her eyes, or scope, supersensitive to infra low frequencies ears) for > > at least one second. > > It is certainly true that a bandpass filter rings. Not sure I would call > that a delay. If I hit a bell, is the sound that comes out a "delay" on > the hit. I suppose you could say that. > I insist on that. Take Octave/Scilab/MatLab and simulate this: an burst (i.e. a limited number of sine wave periods) going through an oscillating loop - the DELAY, i.e. the moment at which output amplitude more or less reaches input one, will be abot Q periods - I'm talking about a burst of more or less resonance frequency. > But this is true always. And I assume that the reason he needs to form the > response is to make up for deficiencies ( in the room, the speaker, > whatever) that already have that ringing in them. And if the ringing of the > filter is out of phase with that of say the speaker, then the two will > cancel. > > I have however no idea whatsoever why he wants what he claims he wants. > What I was responding to was his desire to replace analog circuits with > digital Linux driven filters. I assume he already has low frequency narrow > band filters that he wants to replace ( or maybe avoid buying). Any > comments you make about the digital is equally true for his analog system. > Thus IF he want those kinds of specs for his analog system, then the > digital system will give them to him as well. It is not latency. it is the > natural response of the system, analog or digital. Latency is the delay in > the digital processing of signals, which is completely irrelevant at low > frequencies. > ???????????????? How come latency/delay is irrelevant ? Ronan is in the real of live music, so the audience will hear both direct from the stage sound and PA one - the interference between the two directly depends on latency/delay - nobody cares what cause the phase shift - digital (+ FIR) latency or IIR/analog filter group delay - the latter two are the same in their physical/mathematical nature. > So, can he replace his analog filter board with a digital one. I would > worry about this at high frequencies when the latency-- the time between > reading a word from the sound card input to that byte or anything affected > by that byte being delivered to the output of the soundcard-- becomes a > worry. The argument that the ear cannot tell if something comes in 1/100 > of a sec late is a reasonable one to say one should not have to worry about > the latency. I am not sure it is right, but it could be. > > But latency at 10-100 Hz is completely irrelevant. See above about latency and interference. > > (Note if you know the signal you can tell its frequency in far less than > one cycle. If I hand you a sine wave with infinite precision you can tell > its frequency in an arbitrarily short time.) If you have an unkown signal > shape, then its period is both hard to define and to measure in less than a > period. > > But sound comes with no tag attached, saying it's a 1Hz signal. And that's why any DFT has Fsample/Number_of_samples spectral resolution. Regards, Sergei. -- Visit my http://appsfromscratch.berlios.de/ open source project. ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ Alsa-user mailing list Alsa-user@xxxxxxxxxxxxxxxxxxxxx https://lists.sourceforge.net/lists/listinfo/alsa-user
