- Mar 11, 2004
- 23,076
- 5,557
- 146
MQA?
Really interested to hear your thoughts on this one. The initial stuff I was reading made this sound like it was a power grab by the recording industry, essentially forcing a sorta DRM (only without it explicitly being DRM in the traditional sense, meaning it would require you to pony up for MQA certified hardware, which of course would be licensed out), as a means to try and force people to pay up for the quality version of recordings. It'd force a sub-CD quality version if you didn't have that (but you'd at least still have access to a version of it, hence not typical DRM). So it'd be sorta like SACD all over again (which oddly enough DSD is seeing a surge in popularity for some reason?), only focused on like streaming or something? Even after further reading that most of the alleged sound quality improvements is due to some new shaping. And it's apparently lossy (but somehow that not only isn't a problem but won't prevent it from curb stomping Redbook and of course MP3 in fidelity).
Trying to sort through this for instance:
https://www.computeraudiophile.com/...emy/A-Comprehensive-Q-A-With-MQA-s-Bob-Stuart
DSD and newer PCM (they're pushing 32bit 352.8/384/even 768kHz and I forget what DSD rates they're at these days)? Isn't this going way overboard? I could almost even understand it in the production space, wanting to eliminate technical roadblocks, but even there, is there any real benefit of this (so say you have a whole chain that can do all of that without any converting, is there any realistic benefit as isn't that level of sampling mostly sampling frequencies far out of human hearing, and even typically get filtered? I saw some about pushing the noise into inaudible frequencies, helping to lower the audible noisefloor?), or is it marketing spec numbers to sell newer and better products? DSD samples different from PCM, but a lot of DACs (even older ones) can do both without much issue it seems. There's some debate about which is better, but DSD seems to be gaining steam (especially now that it's not locked to SACD format).
Object based audio (Dolby Atmos and whatever DTS are calling theirs)? Granted, despite their claims to the contrary, this seems more about just adding some extra channels (either higher up or in the ceiling) to give some extra positional queues/altering the sound field to add "height".
The idea behind it though actually seems the most interesting to me, as I've thought for some time now that basically changing things so that there's spatial data that could then be processed depending on your setup (or even preferences) and so in theory they could place things as desired, and your system, be it headphones, stereo speakers, surround, etc, it would sound "right" (and then you could tweak it if you wanted, like if you want a certain instrument to be more pronounced, then you could, you could widen the sound field, change the "venue" etc, similar to the type of processing a lot of stuff already has, only this way you could do an "accurate" mode and so they'd push a calibrated reference aspect where you could get what the artist intended).
A lot of my interest in this (object based) is multi-part. I feel like it would help alleviate issues in mastering/mixing (poor mixes could be adjusted) and maybe it would help make the initial recording get retained but in a tidier way (so recordings would be better archived for the future, meaning remasters would have better source to go on, and it would be neatly contained, where they have the data all contained). This would also be able to deliver on the "studio source" that we keep getting promised but never actually receive (and that MQA seems to be more of; same with Neil Young and Pono). The next is that I'm a headphone listener, and while there are a lot of crossfeed designs and things that try to help adjust it to sound more like speakers, most of those have noticeable flaws and rarely live up to the potential, so I feel like it could be better (without hurting speaker setups, plus it would scale well with 2 channel speakers or whatever new surround they try to do). Next, I think it would help set standards for gaming audio (that again, would be more forward facing, so that it wouldn't depend on junk like EAX/etc). And, regardless of the environment being "real" or imagined, good VR is highly dependent on audio for immersion, and so it will help with that. It would also, although certainly more wishful thinking, put a stop to Dolby or Creative or whomever dictating stuff.
I was going to try and find the discussion about DACs and looking for a newer way that could be objectively and subjectively better, but I can't remember where I found it, seems to have gotten lost in the noise of discussion about all this other stuff. It got into various things like what is the important parts of sound that aren't being accounted for in objective measurements.
Really interested to hear your thoughts on this one. The initial stuff I was reading made this sound like it was a power grab by the recording industry, essentially forcing a sorta DRM (only without it explicitly being DRM in the traditional sense, meaning it would require you to pony up for MQA certified hardware, which of course would be licensed out), as a means to try and force people to pay up for the quality version of recordings. It'd force a sub-CD quality version if you didn't have that (but you'd at least still have access to a version of it, hence not typical DRM). So it'd be sorta like SACD all over again (which oddly enough DSD is seeing a surge in popularity for some reason?), only focused on like streaming or something? Even after further reading that most of the alleged sound quality improvements is due to some new shaping. And it's apparently lossy (but somehow that not only isn't a problem but won't prevent it from curb stomping Redbook and of course MP3 in fidelity).
Trying to sort through this for instance:
https://www.computeraudiophile.com/...emy/A-Comprehensive-Q-A-With-MQA-s-Bob-Stuart
DSD and newer PCM (they're pushing 32bit 352.8/384/even 768kHz and I forget what DSD rates they're at these days)? Isn't this going way overboard? I could almost even understand it in the production space, wanting to eliminate technical roadblocks, but even there, is there any real benefit of this (so say you have a whole chain that can do all of that without any converting, is there any realistic benefit as isn't that level of sampling mostly sampling frequencies far out of human hearing, and even typically get filtered? I saw some about pushing the noise into inaudible frequencies, helping to lower the audible noisefloor?), or is it marketing spec numbers to sell newer and better products? DSD samples different from PCM, but a lot of DACs (even older ones) can do both without much issue it seems. There's some debate about which is better, but DSD seems to be gaining steam (especially now that it's not locked to SACD format).
Object based audio (Dolby Atmos and whatever DTS are calling theirs)? Granted, despite their claims to the contrary, this seems more about just adding some extra channels (either higher up or in the ceiling) to give some extra positional queues/altering the sound field to add "height".
The idea behind it though actually seems the most interesting to me, as I've thought for some time now that basically changing things so that there's spatial data that could then be processed depending on your setup (or even preferences) and so in theory they could place things as desired, and your system, be it headphones, stereo speakers, surround, etc, it would sound "right" (and then you could tweak it if you wanted, like if you want a certain instrument to be more pronounced, then you could, you could widen the sound field, change the "venue" etc, similar to the type of processing a lot of stuff already has, only this way you could do an "accurate" mode and so they'd push a calibrated reference aspect where you could get what the artist intended).
A lot of my interest in this (object based) is multi-part. I feel like it would help alleviate issues in mastering/mixing (poor mixes could be adjusted) and maybe it would help make the initial recording get retained but in a tidier way (so recordings would be better archived for the future, meaning remasters would have better source to go on, and it would be neatly contained, where they have the data all contained). This would also be able to deliver on the "studio source" that we keep getting promised but never actually receive (and that MQA seems to be more of; same with Neil Young and Pono). The next is that I'm a headphone listener, and while there are a lot of crossfeed designs and things that try to help adjust it to sound more like speakers, most of those have noticeable flaws and rarely live up to the potential, so I feel like it could be better (without hurting speaker setups, plus it would scale well with 2 channel speakers or whatever new surround they try to do). Next, I think it would help set standards for gaming audio (that again, would be more forward facing, so that it wouldn't depend on junk like EAX/etc). And, regardless of the environment being "real" or imagined, good VR is highly dependent on audio for immersion, and so it will help with that. It would also, although certainly more wishful thinking, put a stop to Dolby or Creative or whomever dictating stuff.
I was going to try and find the discussion about DACs and looking for a newer way that could be objectively and subjectively better, but I can't remember where I found it, seems to have gotten lost in the noise of discussion about all this other stuff. It got into various things like what is the important parts of sound that aren't being accounted for in objective measurements.