Jump to content

SONAR's dithering algorithms: When to use each kind?


rasputin1963

Recommended Posts

  • Replies 59
  • Created
  • Last Reply
  • Members

Well, I'm relieved to see that we all finally agree that dither is for the most part a pointless exercise!
:thu:

--Ethan

Wait. I've been gone.

 

I don't agree.

 

Dither, properly chosen and properly applied at the proper time increases statistical signal accuracy. But not only does it tend to increase accuracy, the noise from undithered alias error is more likely to be noticeable because of implicit harmonic patterns than would be simple random error, however small that likelihood of noticeability is.

 

Yes, it's not likely to produce a noticeable effect -- at least when us old coots are the auditors -- but we can verify mathematically that it tends to increase signal accuracy.

 

What's not to like?

Link to comment
Share on other sites

  • Members

Yes, it's not
likely
to produce a
noticeable
effect -- at least when us old coots are the auditors -- but we can verify mathematically that it tends to increase signal accuracy.

 

Of course.

 

What's not to like?

 

The BS I see all the time from people arguing about how important dither is, and the even sillier arguments about how this dither flavor or that sounds more open or full or whatever. The last time I compared dither versus truncated, with a golden ears type friend present, neither of us could hear it even on the fading tail of a song. He was surprised! And I was a little surprised too. I think we raised the gain 20 dB and still heard no difference. I do agree that music 40 dB down probably benefits. The simple solution is don't record music to peak at -40! :D

 

--Ethan

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...