DSD genuine improvement or marketing tool?

[Mivera]

Well I expected more. I can see I need to lower my expectations by at least a factor of 10.

So can you explain what the LG G5 does so much better than my DAC? Oh, I am sorry, there I was expecting too much again. So let me dial it down another 10x.

I guess you can also suggest Amir go get an LG G5. I believe he uses a Mark Levinson that is about as old or older than my Tact.

I'm just talking about measurement wise. If straight measurements were compared, the phone would probably win.

What measurements do you look for to tell you which DAC is the best?

 

[JoeWhip]

I think I have had enough!

 

[Mivera]

Mike, I think folks here would like to give you the benefit of the doubt. I'll wait for the measurements, myself. But you need to understand that in the meantime, you sound just like every subjectivist on every audio forum, and all you really have is what you "hear." The measurements you do have come from the manufacturer and don't really show anything we should expect to be audible. And your answer to those who doubt, is that their equipment isn't good enough to reveal the "obvious" superiority of what you're promoting. And you are promoting. While you aren't currently trying to sell anyone a DAC, we all know you're in the business of developing DSD solutions and have the most powerful psychological motivation to hear what you want to hear and convince others to hear it with you.

It is a perfect storm of subjectivist argument, gift wrapped in tech speak. Very high levels of skepticism, even straight-out disbelief, are to be expected. A more appropriate approach, for this forum anyway, would be something like "I think I hear a significant advantage in DSD, even when it is just PCM converted to DSD. I know that's counter intuitive and the only measurements I have come from a manufacturer and don't really show anything that should be clearly audible. Yet I think I hear it. I've just bought the Audio Tester of the Gods 4.1, and I intend to test every parameter available to me, searching for solid evidence. Stay tuned."

This is, of course, very different from the path you've taken, which is the same argument used by every audiophool to justify every empty tweak ever created. Humility and patience would work much better for you here.

Tim

I'm actually more focused on PCM performance. As I mentioned earlier, as well as Sonny, 99% of people listen to PCM. The future of high end audio is streaming services. Currently, the best streaming services only offer redbook PCM. Soon they will also offer MQA. So my focus is on best performance from those 2 formats.

 

[TBone]

Anecdotal story.

A few years ago I attended a demo at my local high end audio shop in Berkeley. In attendance, amongst others, were Yuki Sugiura of Sony, a genuinely nice man with a wealth of knowledge, because the demo included the then relatively new Sony top of the line flagship speakers, and Cookie Marenco of Blue Coast Records, a vocal proponent of DSD. The seminar was held in the main showroom. Electronics used included the then top of the line dCS stack.

Cookie stood in front of the group in attendance, which included about 20-30 folks, and shared what she believed about the differences in the formats, including things we all have heard like longer tails with DSD or redbook is incapable of capturing the sound of a violin.

Cookie brought with her a recording of some violin music which she had recorded in DSD and then put out on her label on both SACD and CD. She then played both discs. Three people in the group in attendance were violin players. All 3 stated, rather forcefully, that the redbook sounded truer to what they knew to be the violin sound. Cookie's jaw dropped and her eyes glazed over. She recovered quickly and responded with words to the effect of: "Well, that's why we offer the music in different formats."

another tale ...

Not all that long ago, I was invited to a high-priced audio demo in which many of the participants were invited to bring along their mega-buck digital components for comparisons. Nothing was hidden from view, this was not intended to be a blind test. When I ask if I could bring along my CDP, I was told in rather plain english that few attendees would be interested in hearing a circa 1998 16/44 player; that it simply wouldn't be competitive in such elite current-day company. My reply was, at the very least, this player could act as a reference, an indicator of how far digital has improved ...

Anyway, on arrival, its addition was received with much skepticism, so they decided to audition it first, get it out the way, so-to-speak. We all got to choose our specific music, so I brought along a CDR copy of one of my many Pink Floyd's WYWH pressings. Anyway, the cut was played in its entirety without the heckling I fully expected. Although it appeared that people were enjoying this rendition, nobody said anything positive or negative after. So the demo proceeded, replaying "superior" digital equipment after "superior" digital equipment, vast majority being DSD/hi-rez based, which by default, this audiophile crowd thinks superior.

In the end, while packing up, I was approached by more than a few attendees who "whispered" they thought my 1998 CDP was surprisingly the "most enjoyable" of the bunch. Of course, nothing to that affect was stated during the actual demo, and not one person inquired about the provenance of my CDR.

 

[amirm]

Mike, do you have any way of performing a blind test with your setup? Can you get help from someone to make an AB switch and run a few trials and report?

 

[Mivera]

Mike, do you have any way of performing a blind test with your setup? Can you get help from someone to make an AB switch and run a few trials and report?

I probably could. What are you referring to comparing?

 

[amirm]

HQPlayer conversion of PCM to DSD versus playing native PCM.

 

[Phelonious Ponk]

I'm actually more focused on PCM performance. As I mentioned earlier, as well as Sonny, 99% of people listen to PCM. The future of high end audio is streaming services. Currently, the best streaming services only offer redbook PCM. Soon they will also offer MQA. So my focus is on best performance from those 2 formats.

Good to hear. So PCM converted to DSD? Regarding that, I will remain skeptical unless and until there are really good numbers or a blind listening study at least as buttoned-up as the much maligned Meyer and Moran. I wouldn't have much trouble believing that the conversion could be made transparently. And improvement seems very unlikely, but I look forward to your measurements.

Tim

 

[Mivera]

Good to hear. So PCM converted to DSD? Regarding that, I will remain skeptical unless and until there are really good numbers or a blind listening study at least as buttoned-up as the much maligned Meyer and Moran. I wouldn't have much trouble believing that the conversion could be made transparently. And improvement seems very unlikely, but I look forward to your measurements.

Tim

How the audio is processed in the gear doesn't matter. What matters is how the audio you feed the gear sounds when it comes out of the analog outputs of the DAC. 98% of all modern DAC's use upsampling technology to improve performance. How they do it isn't as important as the end result.

 

[Phelonious Ponk]

How the audio is processed in the gear doesn't matter. What matters is how the audio you feed the gear sounds when it comes out of the analog outputs of the DAC. 98% of all modern DAC's use upsampling technology to improve performance. How they do it isn't as important as the end result.

I couldn't agree more.

Tim

 

[Mivera]

HQPlayer conversion of PCM to DSD versus playing native PCM.

Well I already did the ultimate blind test of that last year. I had Jussi offline convert 3 redbook tracks to DSD 128. I copied the tracks onto an SD card along with 3 tracks I converted to DSD 128 with a different offline conversion software:

http://audiventory.com

And the 3 original tracks in their native redbook format. The DAC I had at the time was the Resonessence Mirus which has an SD card transport on it. So I had the owner of a local audio dealership come over and asked him for his impression of which he liked better. He had no idea what he was listening for, or even that it was possible to convert redbook to DSD. He also had no idea which resolution each format was in. With all 3 tracks, he chose the Hqplayer converted one as his favourite.

I repeated the same experiment with several since and so far 100% prefer the Hqplayer converted tracks. Even my wife who is far from being an audiophile.

 

[amirm]

Thanks for that. It is a step in the right direction . The things that need to be done further are:

1. Checking for levels. Resampling can easily create differing levels.

2. Not setting expectations for listeners that there are differences and making sure the test reflects this. That is sometimes you want to play the same file against itself. Sometimes not.

3. At least 8 to 10 trials. We cannot establish statistical confidence with lower numbers.

 

[Mivera]

Thanks for that. It is a step in the right direction . The things that need to be done further are:

1. Checking for levels. Resampling can easily create differing levels.

2. Not setting expectations for listeners that there are differences and making sure the test reflects this. That is sometimes you want to play the same file against itself. Sometimes not.

3. At least 8 to 10 trials. We cannot establish statistical confidence with lower numbers.

1: I did match the levels perfectly for the testing.

2: I never set any expectations. Nobody even had a clue what they were listening for, or what any of the format resolutions were

3: I've tried the test with more than 8-10 people. I've did this demo to almost everyone who's listened to my system since. But the difference is so profound, and blind test results were so conclusive, that I'm beyond that phase now. That was last years news.

Do you really think so many people rave about Hqplayer because of the clunky GUI?

 

[Purité Audio]

I don't think Iwouldget too excited about Meridians MQA . I cannot see another proprietary format taking off.
Keith.

 

[Mivera]

I don't think Iwouldget too excited about Meridians MQA . I cannot see another proprietary format taking off.
Keith.

Well if they offer the option to stream it for free from Tidal, than what's the harm? If it's better than redbook then perfect. I'll listen to it. Jussi even developed a filter to optimize MQA.

 

[Purité Audio]

Meridian say they go back to the original master, is that correct?
Archimago's take is here,
http://archimago.blogspot.co.uk/2016/01/measurements-mqa-master-quality.html#more

Keith.

 

[Mivera]

Meridian say they go back to the original master, is that correct?
Keith.

Based on what Dustin Forman from ESS told me, it's no better than 24/176 PCM. The only real advantage is the compression that shrinks the size down for low bandwidth streaming. But there's no free lunch. It does add artifacts, which is why Jussi developed a special filter to remove them.

Here's some measurement results from Jussi on the matter:

http://www.computeraudiophile.com/b...oded-flac-vs-normal-optimized-hires-flac-674/

 

[amirm]

2: I never set any expectations. Nobody even had a clue what they were listening for, or what any of the format resolutions were

It doesn't work that simply unfortunately. You must have controls that determine that. I have had countless examples of "having no expectation," hearing a difference, then being able to prove there was no difference. Our hearing is elastic and needs no expectation bias to imagine hearing differences that may not be there.

Do you really think so many people rave about Hqplayer because of the clunky GUI?

I can put a coat hanger in a box and have people put it inline with speaker wires and rave about it sounding better with it. Heck even I "heard" differences when Synergistic Research guy put little thumbtacks on the front of the speaker. So did the entire room full of audiophiles. None of that means the little thumbtacks changed the sound that was arriving at our ears. These are all faults of not performing controlled testing.

When we go through the work of making changes, it is natural to listen carefully and hear things we did not hear. Once there, we then apply the reverse to the "before" configuration making it sound bad. With discipline, you can play two identical files and have them sound different on demand! I can do that with ease. I can make such differences be there such as air, resolution, etc. And just as easily imagine them gone. It is all how I think about what I heard, not what I actually heard.

To get reliable data that is translatable to everyone we need to guard against such things.

 

[Mivera]

It doesn't work that simply unfortunately. You must have controls that determine that. I have had countless examples of "having no expectation," hearing a difference, then being able to prove there was no difference. Our hearing is elastic and needs no expectation bias to imagine hearing differences that may not be there.


I can put a coat hanger in a box and have people put it inline with speaker wires and rave about it sounding better with it. Heck even I "heard" differences when Synergistic Research guy put little thumbtacks on the front of the speaker. So did the entire room full of audiophiles. None of that means the little thumbtacks changed the sound that was arriving at our ears. These are all faults of not performing controlled testing.

When we go through the work of making changes, it is natural to listen carefully and hear things we did not hear. Once there, we then apply the reverse to the "before" configuration making it sound bad. With discipline, you can play two identical files and have them sound different on demand! I can do that with ease. I can make such differences be there such as air, resolution, etc. And just as easily imagine them gone. It is all how I think about what I heard, not what I actually heard.

To get reliable data that is translatable to everyone we need to guard against such things.

Everyone could have easily picked any of the 3 formats with all 3 songs. But for some reason they pick the Hqplayer one every time. It's not often that you roll a dice that many times and land on the same number every time. But as long as the odds are working in the favour of Hqplayer, it's what I will be going with. After all the customer is always right. I'm not going to argue with my clients and tell them they shouldn't like the sound of my system because there hasn't been any properly conducted testing performed on the upsampling algorithms.

 

[NorthSky]

I am reading attentively the discussion, and I find it very informative and instructive and constructive and very very rewarding.

I've read quite a bit in the past about TacT. @ Emotiva site, from Amir, from several owners of their products in Europe and here in America.

I have also read from Mike (Kelowna), Richard (Opus111), Norman (Europe), and @ diyAudio, Gearslutz, EmotivaLounge, AVSForum, and the other sites of the internet like Audiogon, ComputerAudiophile, great DAC designers, best implementation, parts use, best DACs, etc.

This, is a science audio forum, with great knowledgeable people with instrumentation to measure the performance of audio products.
This is of the utmost interest to me. Mike and Amir are two great individuals with great brain inside the head reposing on their shoulders.

DSD is the new wave/rage right now, and there's no better place to learn about the science behind it than right here.
The members ask the right questions, and they bring intelligent points; that's how we explore and learn.
I'm not here to judge, criticize, promote, ...nothing else than learn about good music reproduction devices/components, and for down-to-earth prices without over exaggeration or extrapolation or exploitation or overcharging the music lovers, us.

I have nothing against customers who paid $30,000 and $85,000 for a DAC or CD player, or SACD player; their money is theirs and they are free to spend it the way they see it fit.

I am interested in the best sound reproducers (PCM, DSD, hi-res audio, ...), from the best coffee beans. ...Way of speech.
Reality prices... Hi-end DACs (DSD included, 2x, 4x, 8x, etc.) they can be had from $200 to $2,000
And from $2,000 to $4,000 for 10% better performance (assumption, guesstimate).

From $4,000 we can go all the way up to $100,000 for another 5% better overall performance (maybe, depending of the parts used and implementation).

I watched a movie on Blu-ray last night; Carol, → Cinematographer Edward Lachman shot Carol on Super 16 mm film.
The way we record and reproduce music and the pleasure we get from it is not much different to movies.
The best rewarding films are not the most expensive ones shot with this camera or that one, but with that eye and this one.
And replay it @ home does not require a $100,000 screen and projector to fully enjoy. We can get the full benefit from a $1,000 70" TV.

_________________

Anyway, there are many ways we can contribute in the music and movies we love to listen to and watch and from the recordings using the mics and cameras chosen by the artists recording maestros, composers, directors, orchestration directors, ...and to the reproduction systems @ home.

• DSD - a genuine sound improvement or just a marketing tool? ...I have great confidence in some of my peers...

P.S. I can see a bright future, from this new audiosciencereview website forum right here...
The best is not the most ultra expensive, it's the smart inexpensive well designed product by a well calibrated audio designer of pure genuine ingenuity.
...DSD or not, but more than nut.