Note: I was editing live and added "lossy" - just mentioning it in case you think it was intentional.
So, if they are trying to develop a better perceptual lossy codec and charge for it, isn't it reasonable to expect the burden of proof to be on them as far as the benefits are concerned?
I see two ways in which they could do that
- some fully objective way as in providing lossless master samples, their encoded files and maybe some open competitor files and using a metric showing which files are closest to the original signal. In some fields I am familiar with some variation of MARD metrics would do the trick, but I am ready to accept other perceptually based metric. Unfortunately they don't seem to do that, quite the opposite in fact.
- some softer but still scientific approach, such as
independent testing on users, not unlike what Harman did with speakers. We'd have a softer result, such as 78% of users prefer MQA or something similar. Unfortunately, they don't seem to do that.
The consequence of not doing this (and not allowing users to do this in an open but patent protected way à la MP3) is that the only material we have is their "unfortunate" marketing...