So, 11 of these filters in series:
sounds like a telephone on those settings
NT
On a sunny day (Mon, 9 Feb 2015 06:28:28 -0800 (PST)) it happened snipped-for-privacy@care2.com wrote in :
Yes sort of, I use it for radio communicatons, gets rid of some noise, I only use ONE though. It is specificaly for speech, almost becomes a brick wall filter with 11.
It does show how nice stdin to stdout is in Unix. I wrote several other audio procesing programs that that way can be combined. I do the same with video, using the mjpeg tools streaming format.
What idiot even thought that even two in series would be "better" than one?
Do you have a plot of the entire thing... are all the filters the same? Are the units on the gains dB?
George H.
I did wonder about the in-band f response.
NT
On a sunny day (Mon, 09 Feb 2015 16:03:33 -0800) it happened Robert Baer wrote in :
You are the idiot who does not understand filters. It IS better.
Talk bout something you actually have experience with, like a Chinese LC meter LOL. Cannot even use that!
On a sunny day (Mon, 9 Feb 2015 16:25:42 -0800 (PST)) it happened George Herold wrote in :
You want me to cook you dinner too? You can downkoad and plot all you want.
On a sunny day (Mon, 9 Feb 2015 16:52:03 -0800 (PST)) it happened snipped-for-privacy@care2.com wrote in :
And are aliens among us.
How do you know they are in series?
What Virtual Audio Patchbay software are you using to line them up in series?
I am guessing they are actually acting in parallel.
Plus, the spectra shown for each EQ window is all wrong. There should be practically nothing showing for the bottom 2-3 octaves, based on the slider settings as shown.
I bet one EQ window's spectra is all wrong by itself. And they just put some random jaggy stuff in there to look real.
:)
Hell of an experiment in Audio Processing Latency, though.
On a sunny day (Tue, 10 Feb 2015 02:37:26 -0800 (PST)) it happened snipped-for-privacy@gmail.com wrote in :
OK, this about Unix (Linux): PROGAM (whatever) reads date from the input, proceses that data, and writes processed data to the output. In Unix input is called stdin, and output is called stdout(for data) or stderr for messages Many prgrams allow you to write a '-' for input and output '-'
For example my_program - - my_program now reads from stdin, and writes to stdout.
Now look at the command line in the picture: | xpequ -i - -z 24000 -c 1 -b 16 -o - -s | xpequ -i - -z 24000 -c 1 -b 16 -o - -s | etc etc ever the same The '|' symbol stands for connect output (preceding) to input(next). this is the power of Unix over things like MS crap, and as old as computahs. So that is SERIES. Just like 2 resisisters in series.
On a sunny day (Tue, 10 Feb 2015 02:41:45 -0800 (PST)) it happened snipped-for-privacy@gmail.com wrote in :
They is me, and yes the spectrum is correct.
Not thar bad really, still very usable on a single core, try one of those quads with hyper threading..
For processing or recording audio it makes no difference whatsoever if there is any latency, not even for transmission. You should, as experiment, if you can (have digital) listen to two staions that transmit the same digital content, seconds delay can sometimes be heard. Via satellite 2 x 40,000 km, even more, I am sure you can dodo ze maz.
In a DVB-T country, sometimes 5-6 s delay between FM audio and DVB-T can be observed. The GOP (Group of Pictures) parameter itself is responsible for 0.5 s delay, but from where does the rest come from ?
Telephone conversation from Europe to say Hawaii was once a pain due to the two way delay through two satellites.
If I understand correctly, telephone conversations are no longer routed through two satellites, at most though a single satellite. Anyway in the days of ENG, the reporter at some remote location will have to wait 1-2 seconds for the question to arrive into his/her ear,
er, uh, if you say so, boss...
On a sunny day (Tue, 10 Feb 2015 05:59:01 -0800 (PST)) it happened snipped-for-privacy@gmail.com wrote in :
It correct because I tested it. You can download and stop polluting s.ed. with nonsense and check for yerselves.
>On a sunny day (Tue, 10 Feb 2015 15:13:53 +0200) it happened snipped-for-privacy@downunder.com wrote in :
Probably routed via NSA center in 'merrica ;-)
But more seriously, it is likely that these DVB-T stations do their own signal processing, they have to encode again (transport stream bitrate) so the 'bouqet' fits in the transponder bandwidth
It would not surprize me a bit if for example cable picked it up from satellite to begin with, they had a bunch of dishes here, but was long time ago I did anything in broadcast. So cable viewers would be serviced even later, or maybe DBT-T used cable feed?
It all does not matter, woke up this morning and the world had not ended yet. Its OK to be notified 2 seconds late. ;-)
This is till with multiple contributers in news programs. I watch China TV (both satellite and internet),
I watch China as its reporting is more neutral towards the Ukrain shit US is doing (like Vietnam). Most of Europe TV sings songs from the 0bombma teleprompter.
To thwart the obvious *nix vs. Windows troll:
I maintain your spectra analysis/display code has 'issues'; reviewing xpequ source code, will report back.
PS. If you adjust one EQ band slider, does it adjust all of them?
What does a software EQ have to do with Electronics Design in the first place, sir and/or ma'am. :)
You posted this bug in xwequ long ago. It is shown here in your article.
How much xpequ fft and eq code comes from xine? Quick gloss of the source code involved shows the xine fft code is designed for a fancy audio 'visualization', not necessarily 'scientific' or 'clinical'.
I rest my case, jan. :)
It's not a bug after all.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.