Re: [digitalradio] Understanding soundcard basics ?
Hello Andy, >Is calibration really an issue of concern IF an application can enable a >re-calibration process ? If an application enables re->calibration, does >that only "hold" for that application or can it correct the soundcard for >other applications. Yes that holds the application. The process is just to measure the "real" sound card sampling speed (the "standard" being the PC clock which has a precision better than 0.02%) and to consider this measured speed in your application. There is no way to calibrate the sound card itself . You simply take it as it is... For standard narrow digital modes (as PSK31), if your AF level is good (let's say around 50 %, but not critical), there is no important need to have a very good sound card. For wide digital mode (Packet, ALE, MT63-2000 Hz, 110A), it would be a problem if the amplitude vs AF frequency would be not flat at all (the sound card is not supposed to be a filter inside the telephone bandwith (300-3000 Hz)). For SdR the problem is completly different because you need a real good dynamic. With a basic sound card, having the 10th bit noisy is not important for digimodes, but it would be very bad for a SdR if the input signal is very low (your real dynamic being bad, even if it is supposed to sample on 16 bits). 73 Patrick - Original Message - From: "obrienaj" To: Sent: Wednesday, September 23, 2009 1:01 AM Subject: [digitalradio] Understanding soundcard basics ? > From what I have read in the past, there is a difference between > inexpensive sound cards and the high quality ones. I recall past articles > that suggest the high quality ones can result in some very weak signals > being detectable in a waterfall, whereas cheap cards may not reproduce > the signal. However, as most of us know, even the cheap sound cards > effectively render the average ham signals, even quite weak ones. > > So, aside from the higher end ones rendering weak signals on a waterfall > better, what are measurable difference between a poor cheap one and a > really good top-of-the-line one ? Can someone explain this is plain > English? > > I am aware of the "calibration/timing" issue. Although that too does not > seem to make a huge difference with many digital modes. Of the numerous > digital modes I have tried over the years, PC-ALE and JT65A in WSJT have > been the most impacted by calibration issues. I have seen WSJT not decode > at all when timing of the soundcard is not correct. Do higher end sound > card have less problems with timing/calibration than cheap ones? > > Is calibration really an issue of concern IF an application can enable a > re-calibration process ? If an application enables re-calibration, does > that only "hold" for that application or can it correct the soundcard for > other applications. > > I raise these questions out of general interest, but also because of > recent WINMOR test where the poor performance has been blamed , in part, > on cheap sound cards or sound cards not dedicated to the application. I > don't know enough to argue the point, but my suspicion is that it is > really not that sound card related. > > Andy K3UK > > > > > > Announce your digital presence via our Interactive Sked Pages at > http://www.obriensweb.com/sked > > Recommended digital mode software: Winwarbler, FLDIGI, DM780, or Multipsk > Logging Software: DXKeeper or Ham Radio Deluxe. > > > > Yahoo! Groups Links > > > > >
Re: [digitalradio] Understanding soundcard basics ?
This is a topic I am also interested in. I have a box full of sound cards (and several systems with internal sound), and I have tested several of them using the 'Sound Card Sample Rate Checker' that is included with MixW (aka CheckSR.exe). This program checks the sound card against the computer clock, which its self is only accurate to about 100 PPM from what I have read - so this is a somewhat crude method. Anyway, it does give you a ballpark figure. For example, an old SoundBlaster Live card had an error of input: 204ppm, output: 189ppm. My best card is a Soundblaster X-fi with input: -8ppm, output: -9ppm. The internal sound on my Dell Optiplex GX280 was -33ppm and -79ppm. Here is a message I found in a MixW forum, that gives a better method of finding out how much error your card has: http://forums.qrz.com/archive/index.php/t-126869.html Stability of your radio can also be an issue, so the method they describe (using WWV and a freq counter) will let you know if that is a problem. I really don't know how much this error will affect WINMOR's ability to do its thing... I have passed messages with a friend using the worst of my cards (internal audio on a Dell laptop), and then been unable to decode connect requests from another friend nearby who is using a SignaLink USB (with the Jumper set, and all levels appear good). His signals sound perfect to me, and vice versa... but still I can't decode his connect requests, and he can't decode mine. Running CheckSR.exe might not be a bad idea, just to get some idea of how much error you are dealing with. Another issue is noise floor / spurs. My internal cards tend to have some pretty nasty spurs. My X-Fi is much better in this area. I'm sure there are many other factors besides these to consider. -Josh KD7PAJ > From what I have read in the past, there is a difference between inexpensive > sound cards and the high quality ones. I recall past articles that suggest > the high quality ones can result in some very weak signals being detectable > in a waterfall, whereas cheap cards may not reproduce the signal. However, as > most of us know, even the cheap sound cards effectively render the average > ham signals, even quite weak ones.
RE: [digitalradio] Understanding soundcard basics ?
Hello Andy, There was a very good article in QST a few years ago about sound cards. They ran five different cards through quite a battery of tests in the ARRL Lab and yes indeed, you really do get better performance out of some cards. But dollar for dollar, the performance was not linear. As you know, you can do quite well on a cheap card but do marginally better on a 60 dollar card and the lab reports showed as much. I am sure that most of the cards they reviewed are no longer being manufactured or have changed at least a little so it is difficult (like anything else electronic these days) to keep up with what is good. I am sure that QST article is available in the back issues if somebody wanted to dig for it but I learned a lot. I have no trouble with my card but I am not fighting the WINMOR Battle with it yet. Contentment and enlightenment await those who are not early adopters of technology . . . I have not yet reached the 24th stage of WINMOR awareness. :-) And I am prepared to wait on a few more beta releases - Yes I am a member of the WINMOR Yahoo group. I suspect that any calibration done in one application is only good for the use of that card in that application and nowhere else in that computer. But hey, I could be wrong on this. Rick - KH2DF _ From: digitalradio@yahoogroups.com [mailto:digitalra...@yahoogroups.com] On Behalf Of obrienaj Sent: Tuesday, September 22, 2009 6:01 PM To: digitalradio@yahoogroups.com Subject: [digitalradio] Understanding soundcard basics ? >From what I have read in the past, there is a difference between inexpensive sound cards and the high quality ones. I recall past articles that suggest the high quality ones can result in some very weak signals being detectable in a waterfall, whereas cheap cards may not reproduce the signal. However, as most of us know, even the cheap sound cards effectively render the average ham signals, even quite weak ones. So, aside from the higher end ones rendering weak signals on a waterfall better, what are measurable difference between a poor cheap one and a really good top-of-the-line one ? Can someone explain this is plain English? I am aware of the "calibration/timing" issue. Although that too does not seem to make a huge difference with many digital modes. Of the numerous digital modes I have tried over the years, PC-ALE and JT65A in WSJT have been the most impacted by calibration issues. I have seen WSJT not decode at all when timing of the soundcard is not correct. Do higher end sound card have less problems with timing/calibration than cheap ones? Is calibration really an issue of concern IF an application can enable a re-calibration process ? If an application enables re-calibration, does that only "hold" for that application or can it correct the soundcard for other applications. I raise these questions out of general interest, but also because of recent WINMOR test where the poor performance has been blamed , in part, on cheap sound cards or sound cards not dedicated to the application. I don't know enough to argue the point, but my suspicion is that it is really not that sound card related. Andy K3UK