Originally Posted by jriver
Here's how we work
I see how you work by denying people's rights for support... But I am insisting on getting an answer to the valid support question instead of kicking me out of the only official support channel. It is easy to hide behind the fanboys to look good.. by making someone who points out a problem look bad and not letting him to respond. But you would look much better if the problem is not left as it is but fixed. Even if the problem is just miscommunication.
The topic remained unanswered still.
So, as I am refused to get feedback through official support channel I am re-asking it here. Please clarify what of the bellow is true about the dither used in JRiver and what are your intentions in solving it if there is a problem.
1) Is it uses triangular probability density function (TPDF) dither? (I assume this info is true)
2) Is it using correct 2LSB amplitude for TPDF dither noise and so do not have noise modulation artifacts (but also can change 'exact' values of the source signal, so it cannot be called bit-perfect (as claimed by Matt) in general understanding of the term)?
3) Is it improperly implemented by using 1LSB amplitude and so for exact values in the input there is no change, and so - the noise added depends of the input signal and it has noise modulation artifacts?
Note that 2 and 3 cannot be true at the same time. Simple direct question that needs simple direct answer that is intended to stop misinformation and solving a technical problem if it is not a communication problem. Please clarify what of the above is true in the contradictory information.
I hope you are wise person and will not take anything personally but act in the favour of doing progress and I will get an official answer by the JRiver personnel clarifying the question, so the contradictory information is stopped spreading and if there are any technical issues they are fixed. This is in my interest, in your interest and the interest of the JRiver community. Please don't behave as a scam company.