MQA is Vaporware

[mansr]

1 minute ago, Paul R said:

Are you sure? The latest firmware seems to claim to use the same filters for both MQA and non MQA. 

Maybe they changed it since I tested it. I'll have to check again.

 

1 minute ago, Paul R said:

What I really need is to capture the completely unfolded digital signal of course, but since the last unfold(s) happen in the DAC... 

I was thinking to try a capture from S/PDIF, butterflies that does not go to 384khz.

You can always tap the I2S signals going into the DAC chip. That's what I did to find out what they were really doing.

[mansr]

8 minutes ago, Paul R said:

The fact I liked a MQA version really surprised me, and means if I am honest, I need to find out why.

It may have been a remixed/remastered track.

[fung0]

4 hours ago, Paul R said:

Seriously? 

 

I'm always astonished by people who are apparently 'on the Internet' (as evidenced by their ability to post on forums like this one), yet have no access to search engines or reference sources.

 

https://en.wikipedia.org/wiki/Socialism

[4est]

7 minutes ago, Paul R said:

 

Oh, I am not Pro DSD Forrest, in fact, it would please me  greatly if both MQA ad Tidal just rolled over and went out of business.

 

The fact I liked a MQA version really surprised me, and means if I am honest, I need to find out why. I have some hearing loss at 58hz, and some issues with light tinitius in my left ear, which I think is changing how I hear a bit. It is annoying not to be able to trust what I hear, but then...  🤪

 

I expect to find pretty much exactly what Mansr pointed out today, and perhaps try Deltawave. 

Oh, OK. Sorry if I misinterpreted your stance. It just appears as if you are trying to support them. You "liking" their filters is really besides the point on some levels. Mansr has shown that we do not need MQA to get the same "effect", and it has been proven that it doesn't actually save space. Oh, sure there may be some "cool tricks", but then again, slight of hand that picks your pockets may be cool, but it is not good for you, but them.

[firedog]

2 hours ago, mansr said:

The iFi DACs (not sure about Pro iDSD) do not degrade non-MQA material.

All of them except for the Pro use MQA filters on non MQA material, once they've been engaged. Whether that "degrades" the files or not is a matter of taste. I just keep pointing this out, b/c I think when people use almost every MQA DAC to evaluate the sound of  MQA vs. non-MQA, they should be aware that the MQA filters they are hearing are a bit of a spanner in the works.

[firedog]

10 minutes ago, mansr said:

Maybe they changed it since I tested it. I'll have to check again. 

 

You can always tap the I2S signals going into the DAC chip. That's what I did to find out what they were really doing.

Acc'd to iFi themselves, their only MQA capable DAC that switches automatically between MQA and non-MQA filters is the   Pro iDSD. They wrote about how they devoted thousands of hours of programming time to arrive at this result.

[Jud]

47 minutes ago, Paul R said:

 

Somehow, the music will survive. Even if all the opposition to MQA fails, which it certainly won’t. 

 

I agree, but if MQA is widely adopted by the labels, getting new music by many artists I like in normal Redbook or hi res will become much more of a PITA. Since I (and everyone else) have enough PITAs in my life already, I'd like to avoid adding this one.  Simple as that, really, as far as avoiding support for MQA to help convince the labels this isn't going to be a financial winner.

[Jud]

5 hours ago, Paul R said:

 

It is indeed using the same set of filters, but... there does not seem to be an easy way around that. 

 

 

Later today or tomorrow I'll post a process with a "way around that."

[mansr]

33 minutes ago, firedog said:

Acc'd to iFi themselves, their only MQA capable DAC that switches automatically between MQA and non-MQA filters is the   Pro iDSD. They wrote about how they devoted thousands of hours of programming time to arrive at this result.

The 5.30 firmware on the older DACs uses the the DSD1793 built-in filters for non-MQA content. The 5.30C firmware adds their ridiculous GTO (did they accidentally drop an F?) filter. The notes on their website are rather blurry regarding when this filter is used. Guess I'll have to try it and find out for sure.

[firedog]

23 minutes ago, mansr said:

The 5.30 firmware on the older DACs uses the the DSD1793 built-in filters for non-MQA content. The 5.30C firmware adds their ridiculous GTO (did they accidentally drop an F?) filter. The notes on their website are rather blurry regarding when this filter is used. Guess I'll have to try it and find out for sure.

I'm just telling you what iFi themselves publicly stated about their 5.3-MQA firmwares. They made a very big deal (as did dCS) about how difficult it was to write software including MQA that automatically switched filtering modes between MQA and non MQA and said that b/c of the cost and complexity/resources needed they saw it as a premium feature included only in the PRO iDSD firmware.

 

As far as I know, nothing has changed since the quote below, and it applies to all their 5.3 firmware DACs, other than iDSD Pro. In other words, if you don't want MQA filters, don't use 5.3 firmware.

Quote

 

Yes, in our iDSD Pro anything non-MQA will not pass through MQA upsampling.

 

Our nano iDSD BL's MQA beta firmware currently up-samples non-MQA streams to 8 x PCM via a filter of our specification.

 

We are working with MQA to allow us to fully bypass this upsampling, until then for all iFi products our standard firmware V5.2 without MQA is available.

 

[Paul R]

1 hour ago, 4est said:

Oh, OK. Sorry if I misinterpreted your stance. It just appears as if you are trying to support them. You "liking" their filters is really besides the point on some levels. Mansr has shown that we do not need MQA to get the same "effect", and it has been proven that it doesn't actually save space. Oh, sure there may be some "cool tricks", but then again, slight of hand that picks your pockets may be cool, but it is not good for you, but them.

 

Yes, to be honest I was writing in my journal and the whole thing was astounding to me.  My first thought was the two files had been somewhat remastered,  and my second was the volume was different. Just a slightly louder volume would tend to cause a preference for the louder file. Then compression, and on down the line. It's maddening, because that wasn't at all what I had predicted - my results were exactly the opposite of what I had predicted.  Worse, this was in a blind test, so I could not even blame bias. Very frustrating. 

 

I guess the only thing I can do is void the warranty and tap into the iFi Micro, though I am going to try dropping back to an earlier version of the firmware. I do remember upgrading and thinking the sound was better, so it may also just be I like the iFi filters. Too many loose factors. 

 

I am pretty sure I understand how the MQA decoding is taking place, in pretty solid detail. That allows me to infer how the MQA Encoding is taking place, and what the results should be, and that seems to match up well too. But the "proof is in the listening."  I am hoping a few others will try the two files and see what they hear. I am more than willing to disavow my hearing if other people, like yourself come back and say the 192K or even the 384k PCM file sounds better.  Seriously, I am.

 

-Pul

 

[Paul R]

1 hour ago, mansr said:

Maybe they changed it since I tested it. I'll have to check again.

 

You can always tap the I2S signals going into the DAC chip. That's what I did to find out what they were really doing.

 

Yes, I think this is the only way. Thanks.  -Paul 

 

[Paul R]

1 hour ago, fung0 said:

 

I'm always astonished by people who are apparently 'on the Internet' (as evidenced by their ability to post on forums like this one), yet have no access to search engines or reference sources.

 

https://en.wikipedia.org/wiki/Socialism

 

I am always amazed by people who take Wikipedia as a primary source for any reference. It's good for a general read, but because people can edit it, it is far from reliable on any subject that is in the least controversial. (MQA for example.)  

 

It all depends upon what branch of socialism you are talking about, but every branch of socialism I know agrees on the "evils of capitalism" - no matter how far they depart from typical Marxist Socialism ideas or ideology. And of course, capitalism is what enabled the birth of an acceptance of socialism in the first place. As a reaction to the excesses of capitalism. 

 

The MQA issue can and should be purely capitalistic in the evaluation of the problem, and the solution to the problem. Outside of the technical issues and hard facts of course. From an individual standpoint, it is probably as simple as not buying or supporting anything with MQA on it. 

 

The economic force that the audiophile world is able to levy will be a deciding factor in whether MQA succeeds or fails, and if it does succeed, how much success it achieves. How great a factor the economic impact from audiophiles will be is the only question. 

[mansr]

18 minutes ago, firedog said:

I'm just telling you what iFi themselves publicly stated about their 5.3-MQA firmwares. They made a very big deal (as did dCS) about how difficult it was to write software including MQA that automatically switched filtering modes between MQA and non MQA and said that b/c of the cost and complexity/resources needed they saw it as a premium feature included only in the PRO iDSD firmware.

 

As far as I know, nothing has changed since the quote below, and it applies to all their 5.3 firmware DACs, other than iDSD Pro. In other words, if you don't want MQA filters, don't use 5.3 firmware.

They say a lot of things, not necessarily accurate. With the 5.30 firmware (the original MQA rendering capable one), the iFi Nano definitely does not use MQA filters for non-MQA content. I checked this by looking at the I2S signals going to the DAC chip. The newer 5.30C firmware does upsample non-MQA content using their "GTO" filter. The GTO filter is not one of the MQA set, although it shares some properties (minimum phase, piss-poor performance). I just checked, and the iFi Nano with this firmware uses the GTO filter for non-MQA input regardless of the filter switch setting. These devices are getting more and more crippled with each update.

 

The iFi website states of the 5.30C firmware that it "Adds Gibbs Transient Optimised Digital (GTO) Filter (This replaces the Minimum Phase Filter)." This suggests that it would not be used with the "standard" setting, which is evidently not actually the case.

 

As usual, don't trust a word from those people.

[mansr]

11 minutes ago, Paul R said:

I am pretty sure I understand how the MQA decoding is taking place, in pretty solid detail.

How did you come to acquire this understanding?

[rickca]

17 minutes ago, mansr said:

As usual, don't trust a word from those people.

I agree and iFi products are off my radar, period.

[Paul R]

22 minutes ago, mansr said:

They say a lot of things, not necessarily accurate. With the 5.30 firmware (the original MQA rendering capable one), the iFi Nano definitely does not use MQA filters for non-MQA content. I checked this by looking at the I2S signals going to the DAC chip. The newer 5.30C firmware does upsample non-MQA content using their "GTO" filter. The GTO filter is not one of the MQA set, although it shares some properties (minimum phase, piss-poor performance). I just checked, and the iFi Nano with this firmware uses the GTO filter for non-MQA input regardless of the filter switch setting. These devices are getting more and more crippled with each update.

 

The iFi website states of the 5.30C firmware that it "Adds Gibbs Transient Optimised Digital (GTO) Filter (This replaces the Minimum Phase Filter)." This suggests that it would not be used with the "standard" setting, which is evidently not actually the case.

 

As usual, don't trust a word from those people.

 

And it appears one can not downgrade from 5.30C to 5.30. Just refuses to do it. Can downgrade all the way to 5.2 though, so will try an incremental upgrade. How very irritating. 

 

I don't really know much of anything about the GTO filter, have you looked at it or know who Gibbs is?

 

-Paul 

 

[KeenObserver]

Minimizing or dismissing the possibility that MQA could monopolize the music distribution chain is a dangerous proposition.  That is exactly what the MQA business model calls for.

The majors control a substantial percentage of the market and have bought in to MQA.

I, for one,  do not want to go back to the time when the major studios dictated who succeeded in the music business.  I do not want to go back to the time when the studios dictated who received air play.

The thought of barriers to entry concerns me greatly.  The independents democratized the music industry and opened up opportunities for many.

If MQA becomes widely embedded,  the majors could very well decide that they will only distribute MQA. How long before it became a MQA business?

No,   the music consumer needs to completely reject MQA.

[rickca]

4 minutes ago, Paul R said:

know who Gibbs is?

Ever use Google?

https://ifi-audio.com/wp-content/uploads/2018/08/iFi-audio-Tech-Note-GTO-filter-FINAL.pdf

[mansr]

2 minutes ago, rickca said:

Ever use Google?

https://ifi-audio.com/wp-content/uploads/2018/08/iFi-audio-Tech-Note-GTO-filter-FINAL.pdf

That's not a reliable source of information about iFi products.

[Paul R]

8 minutes ago, rickca said:

 Ever use Google?

https://ifi-audio.com/wp-content/uploads/2018/08/iFi-audio-Tech-Note-GTO-filter-FINAL.pdf

 

Missed the smiley did you?   (It was a mild joke, though obviously, only a math nerd joke.)   -PR

 

What is your opinion on the filter? I have heard it on a 777, and thought then it sounded amazing. That was several years ago though, around 2012. [Edit, August of 2013 actually.]  A friend had that DAC and was so impressed with it, I actually traveled to go have a listen.  Did not buy one though. About the only DAC I would sink that kind of money into is a Wavelength. 

 

-Paul 

 

[rickca]

5 minutes ago, Paul R said:

Missed the smiley did you?

You added the smiley with a later edit.  And I am a M.Eng so yes I understand math.

Anyway, let's keep to the MQA topic.

[Paul R]

Just now, rickca said:

You added the smiley with a later edit.  And I am a M.Eng so yes I understand math.

 

No I did not, the smiley was always there. You were simply being nasty.  Man up and move on. 

 

-Paul 

 

[rickca]

1 minute ago, Paul R said:

 

No I did not, the smiley was always there. You were simply being nasty.  Man up and move on. 

 

I was actually trying to be helpful by providing a link for you to read.  Now I see why you're on my ignore list.  I won't engage further.

[MikeyFresh]

11 minutes ago, KeenObserver said:

The thought of barriers to entry concerns me greatly.

 

I'd like to think this would stop them dead in their tracks at some point, though no doubt the greedy private equity bankers and others planning and investing in such schemes have ideas as to how they might actually pull that off should it ever come to that day.