Low Frequency Sine-Burst Test Signals for Room and Loudspeaker Evaluation

Status
Not open for further replies.
Hi all,

Please find attached a .ZIP archive which contains a set of programs to generate a bunch of special dynamic test signals (to be burned on CD for most convenient usage, or used otherwise) for low-frequencies (27.5Hz to 440Hz, in small steps).

This CD was originally developped many years ago to detect and possibly correct modal issues, "measuring" with no other tools than your ears at hand, plus a CD-player with remote (read: no actual measurements). All relevant info can be included on the CD when the CD-player (and compillation software) supports CD-text. With today's mobile phones etc it's nice as well, if you have a long enough cable or transmit wireless.


The signals are especially suited to identify modal problems in all its cruel details (boomyness, or "suck-outs" etc) and to test speakers and rooms for rattling/noise issues etc.

Measurements are nice and show enough for the real experts, but actual listening to this sort of signals can be very educative too, in fact you may better understand what you see in measurements compared to how you will perceive it with signals that are not as annoyingly artifical as (and better suited to real listening than) steady state sines or sweeps, or noise test signals (while those are needed and used here, too). With those static signal you may not hear the temporal effects of room modes as readily as with the modulated burst sequences provided which are much more music-like

Details can be found in the included documentation, or just ask.

As those little generator programs are really old (I did them in 2001, quick'n'dirty) they are for 16-bit MS-DOS environment. Up to WinXP this may not a problem, but for Win7 etc you may need an add-on to run them (see the info).

If anybody wants to convert the WAV outputs to MP3, upload it to a public server and post a link... they are welcome to do so, but please include the original ZIP package as well as this stuff is published under the GNU General Public License.
Even more apreciated would be the re-usage of the idea to make a real program out of it (with GUI or cmd-line options and for today's operating systems), or a least a recompile for 32-bit to settle the Win7 issue. I'm an old fart and can program only for MS-DOS and microcontrollers...

Have fun!
- Klaus
 

Attachments

I also have the Linkwitz disk.

My favorite method is an old analog sweep generator. Twittle until something resonates and go fix it. The back on my china cabinet was a big problem. Small foam block solved that. Pictures on walls, now held with velcro. One nick-knack now on a felt pad. My center channel acting like a bass trap, now gone, and a inter-glass window grill that I can't do anyting about. Not too bad for a room. May are worse.

I agree, every signal generator I ever downloaded was pretty poor. Must have used 20. I just use the built in one in my TrueRTA or SoundEasy or my old HP 406. You would think it would be simple, but no one has done a good job on one. Yes, I am begging some enterprising DIY programmer type to do something about it.
 
One feature request: could the bursts start at 10Hz?
No Problem, the next time I compile it again (about time after 11yrs) I'll add cmd-line options or a simple config file for the options the code already has but does not use at this point.

A quick fix to get the lower tones would be to force the sample rate to 1/2 or 1/4 and then resample it to get down two octaves lower (that's the way you'd do it in cooledit/audition. Probably sox can do it, too, then it could be easily automated for batch processing).
 

They seem to contain quite a lot of noise down to DC.
endofdays2.jpg


I hope it is not your goal 😉

And sorry for an OT post 😀
 
I agree, every signal generator I ever downloaded was pretty poor. Must have used 20. I just use the built in one in my TrueRTA or SoundEasy or my old HP 406. You would think it would be simple, but no one has done a good job on one. Yes, I am begging some enterprising DIY programmer type to do something about it.
Not sure what your exact complains are. I use cooledit/audition (cost some $$, though, unless you find ways around it) as my primary sound editor/software because it is really a good piece of software, unlike many other editors it was very thoughtfully programmed by a scientifically skilled person. That's probably why Earl likes it, too.

Log-Sweeps with real-time freq display (per 8k-FFT) are no problem with it. Almost as convernient as an analog generator.
 
War of the Worlds:


Lord of the Rings:


The incredible Hulk:


Hanna:


Cabin in the Woods:
 

Attachments

  • WOTW_ch5.jpg
    WOTW_ch5.jpg
    164 KB · Views: 200
  • LOTR_NazgulNoHead.JPG
    LOTR_NazgulNoHead.JPG
    156.5 KB · Views: 207
  • Hulk_II.jpg
    Hulk_II.jpg
    182 KB · Views: 213
  • Hanna_escape_lab_scene_IV_2.jpg
    Hanna_escape_lab_scene_IV_2.jpg
    145.9 KB · Views: 204
  • Cabin_Woods_ch7.jpg
    Cabin_Woods_ch7.jpg
    155.4 KB · Views: 159
Last edited by a moderator:
it is really a good piece of software, unlike many other editors it was very thoughtfully programmed by a scientifically skilled person. That's probably why Earl likes it, too.

Thanks for the code. This should compile in Visual Studio I assume.

Yes, I live by Cool Edit. And it CAN still be found on the web.

I too have been disatisfied with computer based signal generators and analog ones are just too inconvenient. So I finally wrote my own. It uses double precision in FORTRAN to generate the samples (and then to 16 bits for the wav file) and then sends this to the sound card. Its output is extremely clean and accurate.
 
Thanks for the code. This should compile in Visual Studio I assume.
You're welcome. A port to 32bit would be nice, but my code was quite dirty and there might be "specifics" that won't compile/run
I too have been disatisfied with computer based signal generators and analog ones are just too inconvenient. So I finally wrote my own. It uses double precision in FORTRAN to generate the samples (and then to 16 bits for the wav file) and then sends this to the sound card. Its output is extremely clean and accurate.
I did the same an got a clean .WAV output with zero distortion down to the 24 noise-floor of the analysis with SpectraLab.

-----:-----

So, anybody yet excersized their woofers?
 
Status
Not open for further replies.