MQA Discussion

j2020 said:
Certainly, with an MQA dac like the 808v6, the superiority of an MQA file over a non-MQA file is indeed noticeable.
Click to expand...

But here is the deal. If MQA is supposed to be for the masses to bring in listeners and change the way the world listens to music why or how could the average person go out and spend $22,000 (Meridian 808v6) just to hear the "true" benefits of MQA. Its not going to happen but with a select few. As far as ABing well maybe the press just can't figure out how to write it up in some positive way as to not kill MQA's attempt at world domination.
 
CPP said:
But here is the deal. If MQA is supposed to be for the masses to bring in listeners and change the way the world listens to music why or how could the average person go out and spend $22,000 (Meridian 808v6) just to hear the "true" benefits of MQA. Its not going to happen but with a select few.
Click to expand...

Steven Stone heard the superiority of every MQA file through the $1995 Brooklyn too.
 
j2020 said:
Has anyone ever wondered why to date, no-one from the audio press has felt the urge to A/B an MQA dac (Meridian 808v6/Ultradac/Mytek) playing an MQA file vs the PCM/DSD version of the same file via a non-MQA dac(Berkeley, Totaldac, MSB, Nagra, Playback Designs, Chord, Esoteric, etc)?

Certainly, with an MQA dac like the 808v6, the superiority of an MQA file over a non-MQA file is indeed noticeable.
Click to expand...

I would prefer comparing an MQA file to its non-MQA counterpart in the same DAC. Seems more like an apples-to-apples comparison that way.
 
j2020 said:
Steven Stone heard the superiority of every MQA file through the $1995 Brooklyn too.
Click to expand...

Don't know Steven Stone but I do know my neighbor that actually owns the Brooklynn,(connected to a AR LS27, AR REF210, SF Olympica III's, and the 2L MQA file we compared to the 2L DSD file we just liked the DSD better. Not enough to get excited about since there is little MQA music to even enjoy that we like.
 
CPP said:
the 2L MQA file we compared to the 2L DSD file we just liked the DSD better
Click to expand...

Exactly as Dr Lesurf predicted: http://www.audiomisc.co.uk/MQA/origami/ThereAndBack.html
The anharmonic distortion from the varying aliasing will tend to "spice up" some music, but be detrimental with others...and vary with each playback!!!
John Siau independently surmised much the same https://benchmarkmedia.com/blogs/application_notes/163302855-is-mqa-doa
Some folks are going to like it sometimes...others not so much. YMMV.

cheers,

AJ

 
AJ Soundfield said:
Exactly as Dr Lesurf predicted: http://www.audiomisc.co.uk/MQA/origami/ThereAndBack.html
The anharmonic distortion from the varying aliasing will tend to "spice up" some music, but be detrimental with others...and vary with each playback!!!
John Siau independently surmised much the same https://benchmarkmedia.com/blogs/application_notes/163302855-is-mqa-doa
Some folks are going to like it sometimes...others not so much. YMMV.
cheers,
AJ
Click to expand...


The Meridian 808v6 does not do dsd and so cannot compare mqa vs dsd; we have to wait for the Meridian Ultradac to do this. But we already have the Brooklyn that can do dsd/pcm/mqa.

In a non-Meridian mqa dac like the Brookyln, dsd is preferred. In a non- Merdian non-mqa dac like the Esoteric N-05, I prefer pcm/dsd to mqa.
We are back to the issue - does an mqa file played via an mqa dac really sound superior to pcm/dsd played via a non-Meridian mqa dac or a non-Meridian non-mqa dac?

The answer so far in this thread is a "no" for a non-Meridian mqa dac like Brooklyn and another "no" for a non-Meridian non-mqa dac like the Esoteric N-05.
The audio press should be doing more of such comparisons but, but......
 
j2020 said:
The Meridian 808v6 does not do dsd and so cannot compare mqa vs dsd; we have to wait for the Meridian Ultradac to do this. But we already have the Brooklyn that can do dsd/pcm/mqa.

In a non-Meridian mqa dac like the Brookyln, dsd is preferred. In a non- Merdian non-mqa dac like the Esoteric N-05, I prefer pcm/dsd to mqa.
We are back to the issue - does an mqa file played via an mqa dac really sound superior to pcm/dsd played via a non-Meridian mqa dac or a non-Meridian non-mqa dac?

The answer so far in this thread is a "no" for a non-Meridian mqa dac like Brooklyn and another "no" for a non-Meridian non-mqa dac like the Esoteric N-05.
The audio press should be doing more of such comparisons but, but......
Click to expand...

The 808v6 does DSD via DOP.
 
still-one said:
The 808v6 does DSD via DOP.
Click to expand...

From Meridian's website under Disc Formats for the 808v6 :-

"DSD64 on S/PDIF coaxial inputs and USB (DSD over PCM) with suitable media player."

I guess it is only with the ethernet Soolos as I could not get dsd from my usb pc/server.


 
j2020 said:
From Meridian's website under Disc Formats for the 808v6 :-

"DSD64 on S/PDIF coaxial inputs and USB (DSD over PCM) with suitable media player."

I guess it is only with the ethernet Soolos as I could not get dsd from my usb pc/server.
Click to expand...

Actually the way it was explained to me, you get DOP (DSD via PCM) via USB. According to some of the Meridian 808v6 literature it says via SPDIF too. So far I haven't found anyone who has been successful at doing it that way.
 
still-one said:
Actually the way it was explained to me, you get DOP (DSD via PCM) via USB. According to some of the Meridian 808v6 literature it says via SPDIF too. So far I haven't found anyone who has been successful at doing it that way.
Click to expand...

Hmmm.......

This is interesting too in that the Soolos connects only via ethernet and not via usb.

Even if DoP does work, dsd64 is a non-starter as dsd downloads are usually 128 miminum. Dsd64 is base for sacd's.
 
j2020 said:
Hmmm.......

This is interesting too in that the Soolos connects only via ethernet and not via usb.

Even if DoP does work, dsd64 is a non-starter as dsd downloads are usually 128 miminum. Dsd64 is base for sacd's.
Click to expand...

According to the attached document DOP does handle dsd 128. I am not sure whether the Meridian gear will handle it even though the units MQA display shows tracks up to 352 which is what is required for dsd 128.

http://dsd-guide.com/dop-open-standard
 
still-one said:
According to the attached document DOP does handle dsd 128. I am not sure whether the Meridian gear will handle it even though the units MQA display shows tracks up to 352 which is what is required for dsd 128.

http://dsd-guide.com/dop-open-standard
Click to expand...

We know that DoP handles 128. I get 128 via DoP from my usb dac.
What I was referring to is Meridian's own spec for only 64 which is strange.
 
radioactive said:
Interesting comment by DCS about supporting MQA - The way this quick response reads, doing MQA in software is entirely possible, but they may go to a "hybrid" mode instead.

http://www.dcsltd.co.uk/support/are-you-planning-to-support-mqa/
Click to expand...

Hmm...I read the details on the link differently than you for some reason. I didn't get the feeling from their text that "doing MQA in software is entirely possible" as you mention above. I actually read it more as they attempted to put out a prototype MQA software solution in one of their DACs and hit issues when attempting that. Given that I am sure the men/women at dCS are expert level developers in software/firmware for their products, I am not sure unless their entire platform lacks the software extensibility (highly unlikely) or they just can't code the solution due to lack of experience/expertise (absolutely unlikely); why they wouldn't do it completely in software (at least at first). I have to believe that their statement in the link might be a way for them to get their clientele ready for a hybrid (meaning a hardware change) solution.

If my thoughts are correct, then I think it a stretch to say that "doing MQA in software is entirely possible". Or at least currently I think that statement is a little strong. or why wouldn't dCS just release the bits (code) to the public without going down the hybrid route? IMHO any software can be created to solve an issue with the requirement that the issue is solvable in software. Their "technical issues" might just mean that realistically it is not possible to do an entire software solution.

Anyway, just my thoughts as I have no like/dislike for MQA and anything to help improve digital is fine by me. NOTE: I have not heard MQA yet, so I have no idea how it sounds, nor will make any commentary until I hear it to that effect.

Thanks.
 
Interesting take on things @laynes...

Just for anyone who is unaware, all DCS DAC's use FPGA's in a configuration they refer to as their "Ring DAC". FPGA's are software-programmable. DCS' designs are certainly not the norm -- they have a lot in common with FPAA's (Field Programmable Analog Arrays). DCS knows FPGA programming, and are experts at it -- they're the pioneers of high-end use of these configurations so I don't think it's a lack of software design/programming skills. The way I read it is that they're saying they had an implementation that worked but "MQA" (the "group" formed by Bob S. and company) had a set of restrictions on how MQA is implemented and DCS didn't adhere/conform (similar to another company that had issues with MQA at CES recently -- I believe the details are in this thread a few months ago). DCS references "the MQA development team" (I underlined it below), which I take it means MQA's (the company/group formed by Bob S.) development team, not the DCS developers working on MQA.

Almost all DCS DAC's are hybrid software designs since the FPGA's run this software rather than purpose-built ASIC's which are traditionally referred to as entirely hardware and not reprogrammable or capable of running new software uploaded onto them like FPGA's.

I do know that DCS' current plan for supporting MQA is to sell a separate unit that will do "streaming" which can be connected "somehow" to existing DCS DAC's that don't have the built-in functionality. Exactly where and how the MQA decoding is done in their implementation is still unclear to me (it may be out in press, I'm just not aware of it). They may even hold off on making it until there's sufficient content to play along with demand.

Here's a snippet of the link I posted:
" ....The current situation is that our development team have prototyped an MQA solution inside a dCS DAC but this “software only” mode of enabling MQA has hit some challenges. We are working with the MQA development team to identify solutions to these technical challenges and this work may result in us changing our architecture from software only to a hybrid software and hardware solution."​


laynesaddiction said:
Hmm...I read the details on the link differently than you for some reason. I didn't get the feeling from their text that "doing MQA in software is entirely possible" as you mention above. I actually read it more as they attempted to put out a prototype MQA software solution in one of their DACs and hit issues when attempting that. Given that I am sure the men/women at dCS are expert level developers in software/firmware for their products, I am not sure unless their entire platform lacks the software extensibility (highly unlikely) or they just can't code the solution due to lack of experience/expertise (absolutely unlikely); why they wouldn't do it completely in software (at least at first). I have to believe that their statement in the link might be a way for them to get their clientele ready for a hybrid (meaning a hardware change) solution.

If my thoughts are correct, then I think it a stretch to say that "doing MQA in software is entirely possible". Or at least currently I think that statement is a little strong. or why wouldn't dCS just release the bits (code) to the public without going down the hybrid route? IMHO any software can be created to solve an issue with the requirement that the issue is solvable in software. Their "technical issues" might just mean that realistically it is not possible to do an entire software solution.

Anyway, just my thoughts as I have no like/dislike for MQA and anything to help improve digital is fine by me. NOTE: I have not heard MQA yet, so I have no idea how it sounds, nor will make any commentary until I hear it to that effect.

Thanks.
Click to expand...
 
Technology wise the D/A conversion is always done through programming. The difference is just that if a manufacturer uses a ready made DAC chip off-the-shelve, such as e.g. ESS Sabre, the functions the chip performs are always hardcoded into the silicon (ASIC). So, to include MQA, ESS Sabre would need to make a new chip with that functionality.

In case of FPGAs companies like DCS or Meitner use to make their DACs, the functionalities of the chip can be re-programmed in the field with a SW upgrade. So, as Radioactive says, for DCS, Meitner etc. it would be a SW upgrade to integrate MQA, unless they run out of space or there are functionalities in MQA that would require an entirely new chip architecture.


Sent from my iPad using Tapatalk
 
radioactive said:
Interesting take on things @laynes...

Just for anyone who is unaware, all DCS DAC's use FPGA's in a configuration they refer to as their "Ring DAC". FPGA's are software-programmable. DCS' designs are certainly not the norm -- they have a lot in common with FPAA's (Field Programmable Analog Arrays). DCS knows FPGA programming, and are experts at it -- they're the pioneers of high-end use of these configurations so I don't think it's a lack of software design/programming skills. The way I read it is that they're saying they had an implementation that worked but "MQA" (the "group" formed by Bob S. and company) had a set of restrictions on how MQA is implemented and DCS didn't adhere/conform (similar to another company that had issues with MQA at CES recently -- I believe the details are in this thread a few months ago). DCS references "the MQA development team" (I underlined it below), which I take it means MQA's (the company/group formed by Bob S.) development team, not the DCS developers working on MQA.

Almost all DCS DAC's are hybrid software designs since the FPGA's run this software rather than purpose-built ASIC's which are traditionally referred to as entirely hardware and not reprogrammable or capable of running new software uploaded onto them like FPGA's.

I do know that DCS' current plan for supporting MQA is to sell a separate unit that will do "streaming" which can be connected "somehow" to existing DCS DAC's that don't have the built-in functionality. Exactly where and how the MQA decoding is done in their implementation is still unclear to me (it may be out in press, I'm just not aware of it). They may even hold off on making it until there's sufficient content to play along with demand.

Here's a snippet of the link I posted:
" ....The current situation is that our development team have prototyped an MQA solution inside a dCS DAC but this “software only” mode of enabling MQA has hit some challenges. We are working with the MQA development team to identify solutions to these technical challenges and this work may result in us changing our architecture from software only to a hybrid software and hardware solution."​
Click to expand...

Thank you for the additional information/insight. Just to be clear (as I mentioned in my original post), I think it absolutely unlikely that the dCS folks lack the software design skills to do the job completely in software. Sorry if that wasn't as clear as I meant it to be as I have absolute regard for their products (even though I chose a different DAC for my tastes).

My reason even asking or being concerned that this is now the second DAC manufacturer (maybe more) that has had issues implementing a complete solution in software, is that I would certainly rather have this option than having to source yet another expansion card (read: hardware) for my DAC to get MQA capabilities. Just as I am sure dCS (or Meitner (from Kuoppis response above)) would much rather have a solution that can be fanned out to their customers via a download/flash update; I would rather have the same with my current DAC (which also can support "on-the-fly" updates).

Of course, this all assumes that the MQA folks have indeed created a beast that can realistically be done in a software-only solution. Thus, my original post.

Thanks again for your answers and information.
 
Kuoppis said:
Technology wise the D/A conversion is always done through programming. The difference is just that if a manufacturer uses a ready made DAC chip off-the-shelve, such as e.g. ESS Sabre, the functions the chip performs are always hardcoded into the silicon (ASIC). So, to include MQA, ESS Sabre would need to make a new chip with that functionality.

In case of FPGAs companies like DCS or Meitner use to make their DACs, the functionalities of the chip can be re-programmed in the field with a SW upgrade. So, as Radioactive says, for DCS, Meitner etc. it would be a SW upgrade to integrate MQA, unless they run out of space or there are functionalities in MQA that would require an entirely new chip architecture.


Sent from my iPad using Tapatalk
Click to expand...

Thank you the reply, and yes, I got the fact that some DACs (including my current one) allow for a field flash of firmware/bits (software). I am just concerned that the MQA folks have not made the option of a truly software-only solution realistically doable because we now have another DAC company (dCS) currently having issues doing this and contemplating a hybrid solution. As I mentioned in my reply to Radioactive, the selfish side of me would rather just get a software/firmware drop instead of having to source a companion hardware module or other hybrid solution that could be avoided.

Thanks,
 
I think what you say might well be the case, I really cannot judge for not knowing the specifics. But and as you indicated, DCS is not just some DAC maker - they develop excellent gear. And we did hear a similar story from Auralic earlier.


Sent from my iPad using Tapatalk
 
You must log in or register to reply here.
Back
Top