Jump to content
IGNORED

Was Meyer and Moran debunked by Robert Stuart?


esldude

Recommended Posts

Here is what is left after converting one of Barry Diament's 192 files to 44/24 and back to 192/24.

 

Interesting question; which one sounds better in your system, the 44.1/24 or the one converted back to 192/24?

 

I assume you don't hear any difference between the original 192/24 and the converted-back-to-192/24?

 

How about original converted to 384/24 or 768/24?

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment

I just tried an informal "shuffle play" test with a 24/192 download and a copy downsampled with sox to 16/44. Unfortunately, with my Squeezebox Touch as a USB source, my Auralic Vega makes a loud "bloop" noise when switching between the different files, an obvious tell. So I think I need to downsample then upsample for a valid test.

 

The file is track 2 from:

 

eClassical - Holmboe: Chamber Symphonies

 

which was recorded in DXD.

Link to comment
I just tried an informal "shuffle play" test with a 24/192 download and a copy downsampled with sox to 16/44. Unfortunately, with my Squeezebox Touch as a USB source, my Auralic Vega makes a loud "bloop" noise when switching between the different files, an obvious tell. So I think I need to downsample then upsample for a valid test.

 

The file is track 2 from:

 

eClassical - Holmboe: Chamber Symphonies

 

which was recorded in DXD.

 

Yes, this is pretty typical. Upsample back to 192 and the effects of lost resolution and bandwidth by having been 44/16 will still be there for comparison.

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
Did you brickwall filter this or something?

 

 

 

Oh, how sneaky of them.

 

I don't know they were being sneaky (though there is a cable company who did such thing in some downloads you could do). Most resampling software will drop level a fraction of a db. The reason is to prevent intersample peaks that create momentary clipping that was not in the original file.

 

Now I didn't brickwall the file. When you resample to 44 the software has to brickwall filter it. When you sample back up to 192 you can't put back what was filtered out in the down conversion. So everything below 20 khz cancels pretty well and things above 22 khz or so don't cancel at all.

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
Funny part in this test was that a multi-format DVD-A/SACD player was used, based on Mediatek chipset doing (known bad) conversion to PCM for all SACD sources...

 

So the comparison tells mostly that A/D/A conversion to RedBook PCM didn't sound worse than original conversion to RedBook PCM. In this case.

 

Well actually they used 3 different players. Pioneer, Sony, and Yamaha. The Pioneer at least when using the analog out didn't drop to redbook. I do believe it switched SACD to high rate PCM. If you used the digital connection, which they didn't, it converted to 48khz/24 bits. The Sony was an SACD player reportedly it sampled CDs to DSD for playback. You can find tests of it from Stereophile which would show if it dropped to PCM, which it didn't.

 

Strange how all these ideas circulate when you can look at the report and a follow up on the equipment to see it isn't so.

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
Interesting question; which one sounds better in your system, the 44.1/24 or the one converted back to 192/24?

 

I assume you don't hear any difference between the original 192/24 and the converted-back-to-192/24?

 

How about original converted to 384/24 or 768/24?

 

I don't have a DAC to play back at more than 192 so can't answer that question.

 

Using ABX protocols they all test out the same.

 

When listening sighted, I thought 44/16 was a bit different while if I switched to 48/24 instead it seemed equal to all the higher rates. Don't believe I have bothered with 48/16 to any great extent. Don't know if I should trust myself when I know I am listening to the 44/16 file.

 

Am I remembering right Miska that you have a Focusrite Forte? I have one though haven't gone back and A/B'd such things extensively with my main rig. Is there a preferred rate for the Forte in your experience?

And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. 

Link to comment
I just tried an informal "shuffle play" test with a 24/192 download and a copy downsampled with sox to 16/44. Unfortunately, with my Squeezebox Touch as a USB source, my Auralic Vega makes a loud "bloop" noise when switching between the different files, an obvious tell. So I think I need to downsample then upsample for a valid test.

 

The file is track 2 from:

 

eClassical - Holmboe: Chamber Symphonies

 

which was recorded in DXD.

 

I did the down and up conversions using sox:

 

sox foo.flac -b 16 -V foo.1644.flac rate 44100 dither -s

sox foo.1644.flac -b 24 -V foo.1644up.flac rate -v 192000

 

Hardly definitive -- I'd probably need more trials with different clips -- but I think I managed to convince myself that I could not hear a difference in my system. I suppose that someone could counter that my system is not "highly resolving", and it's probably true because I don't select gear for "resolution". I may try later with my solid-state integrated and headphones.

Link to comment

Do we really need a new paper to be sure higher resolutions are better than Redbook? It looks to me that the importance of accurate reproduction of fast transients and the deleterious effect over the time/phase domains of filters have been known for years, no?

 

Frequency reproduction is a solved problem, has been for decades.

 

The importance of the paper is not to debunk the older one - anyone who knows how to listen properly with proper gear knows the old one is flawed - but it can bring more information about MQA, the process and format coding.

 

This to me is where the interest lies.

 

Just too bad for those who can't hear better sound. Some people can't hear differences between MP3s and uncompressed. A bunch of papers won't fix their hearing.

Dedicated Line DSD/DXD | Audirvana+ | iFi iDSD Nano | SET Tube Amp | Totem Mites

Surround: VLC | M-Audio FastTrack Pro | Mac Opt | Panasonic SA-HE100 | Logitech Z623

DIY: SET Tube Amp | Low-Noise Linear Regulated Power Supply | USB, Power, Speaker Cables | Speaker Stands | Acoustic Panels

Link to comment
Okay so it is a strawman argument. You have just confirmed it.

 

http://www.drewdaniels.com/audible.pdf

 

You talk of all these multiple conversions when that is not what happened. There is a description and block diagram in the paper in the link above. A "professional CD recorder" took the analog output of the SACD/DVDA player doing an AD conversion to 44/16. It then did the DA conversion outputting the result . Level matching was done on the analog signal, and that was it. No multiple conversions you describe. Though often described as an AD/DA loop it actually was a once through AD/DA pass through.

 

As for discussing the sound of USB cables, no it doesn't sound logical or rational.

 

 

OK, it has been a while since I read the paper, and I would have sworn that there was more than one A/D/A loop, but it looks like I misremembered. Still, the results were that there was no statistical difference found between the original and the output of the A/D/A loop, and I still say that this result is highly unlikely. You can insert a single transistor unity-gain buffer stage between the output of your pre-amp and the input of your power amp and even that one transistor will significantly alter the sound on a direct DBT. So Meyer/Moran's results say that an entire A/D/A chain (at a lower resolution than original) will result in no change in sound? I've been involved in audio too long to buy that.

George

Link to comment

The first curious thing is the digital low pass filter was said to be like that used in mastering and playback equipment. ... What is curious is they used merely 500 hz wide transition bands between the cutoff and the flat response. Typically you would see 2050 hz and 4000 hz transition bands in consumer playback gear.

 

While 500Hz is narrow for playback and recording hardware (i.e. DAC and ADC chips), it is wider than what many of today's software resamplers do. Much wider.

 

But all of this is a bit apples and oranges, as many resampling solutions (HW and SW), are half-band, meaning that they will introduce some aliasing/imaging, something entirely absent in the Meridian study.

Link to comment
Still, the results were that there was no statistical difference found between the original and the output of the A/D/A loop, and I still say that this result is highly unlikely. You can insert a single transistor unity-gain buffer stage between the output of your pre-amp and the input of your power amp and even that one transistor will significantly alter the sound on a direct DBT. So Meyer/Moran's results say that an entire A/D/A chain (at a lower resolution than original) will result in no change in sound? I've been involved in audio too long to buy that.

 

AD-DA conversion is very good these days, and the transparency via nulling and listening tests has been shown over and over again.

 

I would not be surprised if one transistor changed the sound, since the feedback gain is not very high, and thus it introduces distortions.

Link to comment
Am I remembering right Miska that you have a Focusrite Forte? I have one though haven't gone back and A/B'd such things extensively with my main rig. Is there a preferred rate for the Forte in your experience?

 

Yes I have it for room measurements. I don't really ever listen anything through it...

 

I did some of my own percussion recordings with it for testing digital filters (@192/24). It is pretty decent device and I think it performs the best at 192/24 as many other modern devices do, but I haven't done my usual measurement set for it's DAC side yet.

 

There's some of the gear I have listed in "recommended hardware" category on my web page here:

Signalyst

Plus of course bunch of my own DACs not listed there.

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment
AD-DA conversion is very good these days, and the transparency via nulling and listening tests has been shown over and over again.

 

I would not be surprised if one transistor changed the sound, since the feedback gain is not very high, and thus it introduces distortions.

 

Except that the Meyer/Moran test wasn't done "these days" IIRC it was done in 2008 or maybe even earlier.

George

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...