Support

If you have a problem or need to report a bug please email : support@dsprobotics.com

There are 3 sections to this support area:

DOWNLOADS: access to product manuals, support files and drivers

HELP & INFORMATION: tutorials and example files for learning or finding pre-made modules for your projects

USER FORUMS: meet with other users and exchange ideas, you can also get help and assistance here

NEW REGISTRATIONS - please contact us if you wish to register on the forum

sound quality

DSP related issues, mathematics, processing and techniques

sound quality

Postby tester » Mon Feb 11, 2013 8:40 pm

I have a small quality problem here. Sound quality at low volume levels, but audible enough to be painful.

In zip is one cycle modulated sine wave attached, for the reference what I talk about.
One channel is original -50dB, other one is normalized so that you can hear.

The same problem is in both, FS and SM (tested on exe, soundcard output is 16bit).
The same problem is with static tones too; modulation emphasizes difference at various levels.
The same problem is with built-in sine and with wavetable with loaded sine.
The problem is less noticable at higher levels (but is audible), which suggests that this is not the oscillator problem.
At higher frequencies, and at such type of modulation, cross-zero glitchess seem to appear.
Problem becomes not greater after scaling the volume down and up again in FS.
You may need better headphones (used on studio monitors)

The design is simple. 1) oscillator, 2) multiplier for gain, 3) out.

And here is my question.
Is this quality issue related to 16-bit resolution of playback/files?
Or is it related to 32-bit procesing in SM/FS?
What and how can be done about that in FS?
How to approach it, to improve the sound quality?
Attachments
design.fsm
(613 Bytes) Downloaded 1282 times
reference-L-norm-R-origin-50dB.zip
(30.11 KiB) Downloaded 1294 times
Need to take a break? I have something right for you.
Feel free to donate. Thank you for your contribution.
tester
 
Posts: 1786
Joined: Wed Jan 18, 2012 10:52 pm
Location: Poland, internet

Re: sound quality

Postby tester » Mon Feb 11, 2013 9:59 pm

My guess is following.

Soundcard outputs 16-bit, which equals 96dB dynamics. So at volume = - 48dB (in relation to 0 as maksimum), sound quality is there at 48/6 => 2^8 => 8-bit. Sine is a very specific type of sound due to it's clarity, so in these remaining 8-bit could sound so dirty? Considering the fact how sine is calculated - maybe yes? I don't know. I need to check it on another computer, with 24-bit output (on exe playback, because I guess FS app follows soundcard).
Need to take a break? I have something right for you.
Feel free to donate. Thank you for your contribution.
tester
 
Posts: 1786
Joined: Wed Jan 18, 2012 10:52 pm
Location: Poland, internet

Re: sound quality

Postby MyCo » Mon Feb 11, 2013 10:00 pm

is the recorded wave the output from your soundcard with this schematic? Damn... you need a better one.

I don't hear that much noise with my old Prodigy HD2. At low volume I can hear the sine modulation and some ground noise (the noise that is constant, because of cables and interfering electronics). It has nothing to do with the digital domain... I've attached a demonstration, as you can see, the difference between original and down-and-upscale signal is by 10 ^-6 worst case. That's -120dB !!
Attachments
mathtest.fsm
(1 KiB) Downloaded 1241 times
User avatar
MyCo
 
Posts: 718
Joined: Tue Jul 13, 2010 12:33 pm
Location: Germany

Re: sound quality

Postby tester » Tue Feb 12, 2013 12:04 am

Well it's my 12 years old sound blaster live! :-)
Im using it for reference, but not that one!

I start checking on 24-bit playback resolution here, we see then what happens.
I need to include that in manual :-)

Plus I just dicovered, that wavetables can be greater than 256 pts, so another thing to check out.
I don't know why I memorized, that higher ones require some mix up.

Thanks for help. I let you know where I ended up with my tests.
Need to take a break? I have something right for you.
Feel free to donate. Thank you for your contribution.
tester
 
Posts: 1786
Joined: Wed Jan 18, 2012 10:52 pm
Location: Poland, internet

Re: sound quality

Postby Urnsoft » Tue Feb 12, 2013 10:42 am

Hi
I ran a test and it seems the problem is with the DS Out but ASIO out is working fine .
I think its soundcard related.
cheers
Attachments
bug-01.fsm
(42.97 KiB) Downloaded 1268 times
Urnsoft
 
Posts: 15
Joined: Wed Aug 08, 2012 7:44 pm

Re: sound quality

Postby tester » Wed Feb 13, 2013 1:19 am

I did listening experiment on SB Audigy for notebook, which is set by default to 24bit resolution playback (...but is there a way to deremine whether audio streams going out of FS exe app are 16 bit or 24?...) and noticed few interesting things.

First - there is no such distortion at the same volume levels like I had on computer with regular old SB Live. Audible distortion starts much deeper (around -85dB), so indeed it is a matter of soundcard. Direct sound mode.

Second - audible sound ends exactly at "-90.3dB" which is "3.05492e-005" in terms of float primitive non-zero value. Direct sound mode. Guess this answers my question about bit depth (direct sound plays 16bit).

Third - there are no issues with ASIO, which confirms what @Urnsoft said.

So if you encounter similar problems - just make sure that this is not due the soundcard type/quality and bit depth resolution.

It can be really confusing and nasty issue.
Need to take a break? I have something right for you.
Feel free to donate. Thank you for your contribution.
tester
 
Posts: 1786
Joined: Wed Jan 18, 2012 10:52 pm
Location: Poland, internet


Return to DSP

Who is online

Users browsing this forum: No registered users and 23 guests