|
Post by PaulBe on Apr 15, 2020 9:56:37 GMT -5
|
|
|
Post by PaulBe on May 23, 2020 15:10:49 GMT -5
Subjective evaluation of high resolution recordings in PCM and DSD audio formats www.researchgate.net/publication/289595500_Subjective_evaluation_of_high_resolution_recordings_in_PCM_and_DSD_audio_formatsHigh-resolution audio production and consumption are increasing attraction supported by releases of the relatively affordable audio recorders from multiple manufacturers and broader bandwidth of the Internet. However, differences in audio quality between high-resolution audio formats are still not well known, especially between the different formats available for the audio recorders. In order to evaluate the differences between subjective impression of the sounds recorded using high resolution audio formats, three audio formats - PCM (192 kHz/24 bits), DSD (2.8 MHz), and DSD (5.6 MHz)- recorded with multiple studio-quality audio recorders were evaluated in a double-blind A-B comparison listening test. Six sound programs evaluated by forty-six participants on eight attributes revealed statistically significant differences between PCM and DSD but not between the two sampling frequencies (2.8 MHz and 5.6 MHz) of DSD. Article (PDF Available) at link. May be viewed online too.
|
|
KeithL
Administrator
Posts: 9,970
|
Post by KeithL on May 24, 2020 4:22:19 GMT -5
Interesting.....
And they were quite specific that they were testing "subjective impressions"..... And, from a market perspective, it is certainly useful in describing the subjective preferences of listeners....
However, the one thing which I might stress, which they did not, was that most of the metrics they chose were also non-specific in direction.
The metrics they used were: image width, • image depth, • image definition, • timbral brightness, • timbral richness,• temporal separability, • overall quality, and • overall preference.
However, what they apparently didn't do was to compare any of the recordings to the original performance. Therefore, what they actually did was to ask their subjects "which recording they liked better - in the context of several specific attributes".
For example, the fact that one format was perceived by the subjects as "conveying more of a sense of image width than the other" is interesting.... And so is the fact that the subjects preferred the greater sense of image width they felt was conveyed by that format.... But that does not tell us which format conveyed the sound of the original performance more accurately.... It could be that their test subjects preferred the format that was most accurately conveying the image width of the original performance...
Or it could be their test subjects preferred a less accurate reproduction of the original with an exaggerated sense of image width... (We don't know... but we cannot safely assume that "the version with the greater sense of image width" was the more accurate one.)
And, overall, we don't know if the format that was preferred overall was the most accurate one... Or if it simply added some subtle colorations that many subjects found pleasing... (Which is something that most audio formats have been accused of at one time or another.)
This is not at all unlikely... since it is well known that many of the attributes they chose to evaluate can in fact be easily altered. For example a recording may sound "timbrally rich" because it accurately reproduces all of the harmonics in an original performance well... But that can also be exaggerated or faked by artificially boosting or adding harmonics that weren't there in the original (as tube amps often do). And, if done well, the result can be a reproduction that "sounds more real than the original" (whatever exactly that means).
Or, to use an analogy to something that is commonly done in video and imaging.... A video or still image can be made to appear sharper by adding tiny exaggerated contrast halos on either side of high contrast boundaries. (This is a time-honored method called "unsharp masking"... and it looks very good... but technically it is added distortion that looks pleasing.) (It is also an effective method for causing an image to "be perceived as being sharper than it really is".)
It would have been very interesting if the subjects had also heard the original live performance...
And then been asked BOTH: - which recording they liked better? - which recording they thought was closer to the original? Make no mistake that I think this was an excellent study... with the methodology... and the results... fairly and accurately described and presented.
Subjective evaluation of high resolution recordings in PCM and DSD audio formats www.researchgate.net/publication/289595500_Subjective_evaluation_of_high_resolution_recordings_in_PCM_and_DSD_audio_formatsHigh-resolution audio production and consumption are increasing attraction supported by releases of the relatively affordable audio recorders from multiple manufacturers and broader bandwidth of the Internet. However, differences in audio quality between high-resolution audio formats are still not well known, especially between the different formats available for the audio recorders. In order to evaluate the differences between subjective impression of the sounds recorded using high resolution audio formats, three audio formats - PCM (192 kHz/24 bits), DSD (2.8 MHz), and DSD (5.6 MHz)- recorded with multiple studio-quality audio recorders were evaluated in a double-blind A-B comparison listening test. Six sound programs evaluated by forty-six participants on eight attributes revealed statistically significant differences between PCM and DSD but not between the two sampling frequencies (2.8 MHz and 5.6 MHz) of DSD. Article (PDF Available) at link. May be viewed online too.
|
|
|
Post by PaulBe on May 24, 2020 9:57:15 GMT -5
Interesting.....
And they were quite specific that they were testing "subjective impressions"..... And, from a market perspective, it is certainly useful in describing the subjective preferences of listeners....
However, the one thing which I might stress, which they did not, was that most of the metrics they chose were also non-specific in direction.
The metrics they used were: image width, • image depth, • image definition, • timbral brightness, • timbral richness,• temporal separability, • overall quality, and • overall preference.
However, what they apparently didn't do was to compare any of the recordings to the original performance. Therefore, what they actually did was to ask their subjects "which recording they liked better - in the context of several specific attributes".
For example, the fact that one format was perceived by the subjects as "conveying more of a sense of image width than the other" is interesting.... And so is the fact that the subjects preferred the greater sense of image width they felt was conveyed by that format.... But that does not tell us which format conveyed the sound of the original performance more accurately.... It could be that their test subjects preferred the format that was most accurately conveying the image width of the original performance...
Or it could be their test subjects preferred a less accurate reproduction of the original with an exaggerated sense of image width... (We don't know... but we cannot safely assume that "the version with the greater sense of image width" was the more accurate one.)
And, overall, we don't know if the format that was preferred overall was the most accurate one... Or if it simply added some subtle colorations that many subjects found pleasing... (Which is something that most audio formats have been accused of at one time or another.)
This is not at all unlikely... since it is well known that many of the attributes they chose to evaluate can in fact be easily altered. For example a recording may sound "timbrally rich" because it accurately reproduces all of the harmonics in an original performance well... But that can also be exaggerated or faked by artificially boosting or adding harmonics that weren't there in the original (as tube amps often do). And, if done well, the result can be a reproduction that "sounds more real than the original" (whatever exactly that means).
Or, to use an analogy to something that is commonly done in video and imaging.... A video or still image can be made to appear sharper by adding tiny exaggerated contrast halos on either side of high contrast boundaries. (This is a time-honored method called "unsharp masking"... and it looks very good... but technically it is added distortion that looks pleasing.) (It is also an effective method for causing an image to "be perceived as being sharper than it really is".)
It would have been very interesting if the subjects had also heard the original live performance...
And then been asked BOTH: - which recording they liked better? - which recording they thought was closer to the original? Make no mistake that I think this was an excellent study... with the methodology... and the results... fairly and accurately described and presented.
Subjective evaluation of high resolution recordings in PCM and DSD audio formats www.researchgate.net/publication/289595500_Subjective_evaluation_of_high_resolution_recordings_in_PCM_and_DSD_audio_formatsHigh-resolution audio production and consumption are increasing attraction supported by releases of the relatively affordable audio recorders from multiple manufacturers and broader bandwidth of the Internet. However, differences in audio quality between high-resolution audio formats are still not well known, especially between the different formats available for the audio recorders. In order to evaluate the differences between subjective impression of the sounds recorded using high resolution audio formats, three audio formats - PCM (192 kHz/24 bits), DSD (2.8 MHz), and DSD (5.6 MHz)- recorded with multiple studio-quality audio recorders were evaluated in a double-blind A-B comparison listening test. Six sound programs evaluated by forty-six participants on eight attributes revealed statistically significant differences between PCM and DSD but not between the two sampling frequencies (2.8 MHz and 5.6 MHz) of DSD. Article (PDF Available) at link. May be viewed online too. As you know, I don’t have a dog in the hunt to make SACD test or appear better than PCM. That being said, here goes – Subjective testing in reproduced sound is a large component of arriving at objective conclusions. A lot of good basic research in speaker development has been and is done by asking listeners what they prefer. Some fine scientists have done great work in this area. Floyd Toole is one of my favorite. Some so called ‘objective’ measurements have become relatively meaningless. Some objective measurements are now standards for determining sound quality. These standards are derived from the results of good controlled double blind subjective studies. Those studies used some of the metrics in this study. Recordings are an illusion. No comparison with a live event will tell us if a recording is accurate. Microphones are not ears; they don’t respond the same as ears (even to each other or different types); they are not placed in the position to record as our ears are placed to listen. Speakers are not original sources. I have compared recordings to original performances – my own recordings. Some with me performing. There was no time where reality was suspended. A recording is never the original performance. The best and most accurate recordings are 1st generation. I use the word “accurate” with some reservation. The production and reproduction system either supports the illusion or it doesn’t. We use controlled subjective tests and experience to determine what objectively works. I used microphone types and positioning to create the illusion. I used my ears to determine what worked and what didn’t work. The proof was in smiles and gaping mouths upon listening. Thankfully, it mostly worked. Occasionally it didn’t, and I’m still learning. Recordings are an illusion. I think we still have the most trouble trying to record and reproduce what happens between 300Hz to 3KHz… telephone bandwidth and frequency…hardly dog whistle territory. Back to PCM vs SACD The authors write about bit equivalency between DSD and PCM but also note that the analog to digital conversion process is different. Does either PCM or DSD resemble the code transmitted between our ears and brain? Processed by our brain? Does it matter? The standard deviation in a normal hearing test is about 10dB. The authors state “…differences in audio quality between high-resolution audio formats are still not well known…”. Perhaps that is a kind way of saying “We can’t tell what is going on up there because few of us have the hearing of bats and dogs”. The article was presented to the 136th AES Convention in 2014. Since we are still talking about potential differences, I assume the differences are still not well known, and still potential. Or, it’s monetarily convenient to beat this ever diminishing dead horse until all the money is extracted. I think it is some of both. I think we will arrive at real conclusions when the diminishing dead horse has turned to dust, (published till perished), discussing potential tomato tomatoe differences between SACD and PCM, and All the gold has become fool’s gold. Sony and Phillips will ride off into the sunset with nary a discouraging word to be heard. Nor will the subject be part of collective memory… just an oddity in the dustbin of audio history. Rhetorical and comical question - Where is the clamor to turn all the LPCM, Dolby, and DTS sound traks for movies and music into DSD? Or, to never insult our senses with anything besides DSD ever again? “Make no mistake that I think this was an excellent study... with the methodology... and the results... fairly and accurately described and presented.” Agreed.
|
|
Chris
Emo VIPs
Posts: 424
|
Post by Chris on Jul 5, 2020 18:34:39 GMT -5
And, yes, in the future, the USB inputs on the RMC-1 and the XMC-2 will support DoP (but we're still working on that).
Are you ripping your SACDs into DSF files and running them through a DAC? That's how I listen to DSD. My DAC does DoP (DSD over PCM). Today I tried to play the new Octave Records Don Grusin album that comes with a data disc with the DSD ."DSF" files I copied these to my server and then played (Roon) without issue via DoP through my Benchmark DAC2. When I tried to play them with my XMC-2 via the HDMI connection on my Roon Core NUC this did not work? I changed Roon to "Convert to PCM" and I can now listen via HDMI. Will the XMC-2 be updated to support DoP or "Initial DCS) over HDMI at some point?
|
|
|
Post by Casey Leedom on Jul 5, 2020 21:45:42 GMT -5
The XMC-2/RMC-1L/RMC-1 DSD support is thin at best. What probably happened on your HDMI connection is that it came through at a DSD rate that your XMC-2 couldn't handle so it wasn't recognized. Or it just simply wasn't recognized as DSD. I just checked the RMC-1 manual and I can't find the DSD rate limit over HDMI. (Over USB it says single (DSD64) and double (DSD128) via DSD over PCM.) The AK4490 supports up to DSD256 (sometimes called "quad rate").
Casey
|
|
|
Post by PaulBe on Sept 6, 2020 9:21:09 GMT -5
From Mark Waldrep, Ph.D. To Hi-Res or Not to Hig-Res? Dr. AIX, It's been just short of a month since my last post. Even in times when I remain pretty close to home, I'm still pretty busy. I managed to get my 15 year old Euphonix System 5 digital console up and running for a new studio tenant. About 4 years ago, I disassembled it and put all of the modules up in the attic area of my building after a different tenant failed to learn how to use it. They didn't want to boot up a recording console. And I've been studying a lot - getting ready for my final pilot oral exam and check ride. I'm pretty close to getting my Private Pilot Glider Certification. I passed the written exam and have been flying once a week all summer. Soaring in active thermals at 10,000 feet plus in the mountains north of Los Angeles is a guaranteed way to forget about all of the challenges that exist back on the ground. Big fun. Finally, I've spent the last week or so writing up my 2020 AES Convention presentation/paper. And it's finally finished and submitted. more... secure.campaigner.com/csb/Public/show/566i-2a4yxp--recl1-dvrqt1a3
|
|
novisnick
EmoPhile
CEO Secret Monoblock Society
Posts: 27,230
|
Post by novisnick on Sept 6, 2020 9:56:29 GMT -5
From Mark Waldrep, Ph.D. To Hi-Res or Not to Hig-Res? Dr. AIX, It's been just short of a month since my last post. Even in times when I remain pretty close to home, I'm still pretty busy. I managed to get my 15 year old Euphonix System 5 digital console up and running for a new studio tenant. About 4 years ago, I disassembled it and put all of the modules up in the attic area of my building after a different tenant failed to learn how to use it. They didn't want to boot up a recording console. And I've been studying a lot - getting ready for my final pilot oral exam and check ride. I'm pretty close to getting my Private Pilot Glider Certification. I passed the written exam and have been flying once a week all summer. Soaring in active thermals at 10,000 feet plus in the mountains north of Los Angeles is a guaranteed way to forget about all of the challenges that exist back on the ground. Big fun. Finally, I've spent the last week or so writing up my 2020 AES Convention presentation/paper. And it's finally finished and submitted. more... secure.campaigner.com/csb/Public/show/566i-2a4yxp--recl1-dvrqt1a3Sounds like you’ve done an amazing job! Congratulations! Soar Like an Eagle!
|
|
|
Post by PaulBe on Sept 6, 2020 10:32:48 GMT -5
From Mark Waldrep, Ph.D. To Hi-Res or Not to Hig-Res? Dr. AIX, It's been just short of a month since my last post. Even in times when I remain pretty close to home, I'm still pretty busy. I managed to get my 15 year old Euphonix System 5 digital console up and running for a new studio tenant. About 4 years ago, I disassembled it and put all of the modules up in the attic area of my building after a different tenant failed to learn how to use it. They didn't want to boot up a recording console. And I've been studying a lot - getting ready for my final pilot oral exam and check ride. I'm pretty close to getting my Private Pilot Glider Certification. I passed the written exam and have been flying once a week all summer. Soaring in active thermals at 10,000 feet plus in the mountains north of Los Angeles is a guaranteed way to forget about all of the challenges that exist back on the ground. Big fun. Finally, I've spent the last week or so writing up my 2020 AES Convention presentation/paper. And it's finally finished and submitted. more... secure.campaigner.com/csb/Public/show/566i-2a4yxp--recl1-dvrqt1a3Sounds like you’ve done an amazing job! Congratulations! Soar Like an Eagle! I'm a bit confused. Just to be clear, I am Not Mark Waldrep/Dr. AIX. The letter is his. I posted only the first paragraph. Go to the link for more, including the first page of his AES Paper - Native High-Resolution versus Red Book Standard Audio: A Perceptual Discrimination Survey.
|
|
novisnick
EmoPhile
CEO Secret Monoblock Society
Posts: 27,230
|
Post by novisnick on Sept 6, 2020 11:27:03 GMT -5
Sounds like you’ve done an amazing job! Congratulations! Soar Like an Eagle! I'm a bit confused. Just to be clear, I am Not Mark Waldrep/Dr. AIX. The letter is his. I posted only the first paragraph. Go to the link for more, including the first page of his AES Paper - Native High-Resolution versus Red Book Standard Audio: A Perceptual Discrimination Survey. LOL Still sleep deprived from 10 days of chaos! I woke up several times saying generator! Gas! LOL Every 7-9 hours refilling them to have ac and some lights for mom! Shes 86 now, Blessed! I should have gone back to the beginning og the thread but I’m not all with it yet. Need more sleep. Thanks for pointing out my mistake. 😳😁
|
|
|
Post by PaulBe on Sept 6, 2020 12:40:49 GMT -5
I'm a bit confused. Just to be clear, I am Not Mark Waldrep/Dr. AIX. The letter is his. I posted only the first paragraph. Go to the link for more, including the first page of his AES Paper - Native High-Resolution versus Red Book Standard Audio: A Perceptual Discrimination Survey. LOL Still sleep deprived from 10 days of chaos! I woke up several times saying generator! Gas! LOL Every 7-9 hours refilling them to have ac and some lights for mom! Shes 86 now, Blessed! I should have gone back to the beginning og the thread but I’m not all with it yet. Need more sleep. Thanks for pointing out my mistake. 😳😁 I hope Mom is well and that you are able to get some rest soon. God Bless You.
|
|
|
Post by PaulBe on Feb 27, 2021 10:51:40 GMT -5
CLAUDE CELLIER Merging Technologies, Switzerland Trust your ears! The quest for the optimum (digital) audio representation Distinguished Lecture, October 24, 2019 In this talk Claude Cellier, CEO of Merging Technologies, known for his invention of the DXD (Digital eXtreme Definition) range of high-end audio products, reviews some of the formidable changes that the audio industry underwent over the last 60 years. He presents a variety of audio examples (some of which are produced at McGill) supported with intuitive graphs that will convince us of the difference between traditional encoding standards such as PCM (Pulse Code Modulation) and the newer standards such as DSD (Direct Stream Digital) and DXD. If you think it is sufficient to stick to current audio representation methods, watch this video to understand what important information we are missing out on because of age-old audio capture and reproduction techniques. idmil.gitlab.io/CIRMMT_visualizations/?/Instruments,%20Devices%20and%20Systems%20(RA1)/Audio%20Processing/Claude%20Cellier www.youtube.com/watch?v=dKoZmxP7rSA
|
|
|
Post by PaulBe on Mar 1, 2021 15:14:17 GMT -5
|
|
|
Post by PaulBe on Mar 1, 2021 15:16:40 GMT -5
|
|
|
Post by PaulBe on Mar 1, 2021 15:32:48 GMT -5
MQA: Sonic Degradation and Huge Losses www.realhd-audio.com/?p=7218MQA is slightly off topic. But, the bottom line of the article is very on topic, and always timely. Audiophile arguments are "...largely motivated by money not fidelity enhancement." SACD vs PCM is right in the thick of it.
|
|
NicS
Sensei
Will the G4 upgrade help quell my RMC1-L frustrations...?
Posts: 210
|
Post by NicS on Nov 28, 2022 18:17:23 GMT -5
This is hardly the most active corner of the forum, but... here goes...
When using the XMC or RMC line of processors, is it possible to set basic level controls for the five channels of a standard SACD 5.1 mix?
It seems when I play SACDs I have to manually wind down the level of the surrounds every time by about 5dB to get a reasonable level match.
Q1 - can I do this so the level controls persist with SACD/DSD playback, without bringing in any other controls? Q2 - is having this basic DSD multichannel control been on the radar for Emotiva as something to include in future firmware upgrades? It doesn't have to be output control of the surround channels; input level would also suffice to balance the surrounds and perhaps subs (though I do this at the sub itself for basic balance)
|
|
geebo
Emo VIPs
"Too bad that all the people who know how to run the country are driving taxicabs and cutting hair"
Posts: 24,204
|
Post by geebo on Nov 28, 2022 19:40:10 GMT -5
This is hardly the most active corner of the forum, but... here goes... When using the XMC or RMC line of processors, is it possible to set basic level controls for the five channels of a standard SACD 5.1 mix? It seems when I play SACDs I have to manually wind down the level of the surrounds every time by about 5dB to get a reasonable level match. Q1 - can I do this so the level controls persist with SACD/DSD playback, without bringing in any other controls? Q2 - is having this basic DSD multichannel control been on the radar for Emotiva as something to include in future firmware upgrades? It doesn't have to be output control of the surround channels; input level would also suffice to balance the surrounds and perhaps subs (though I do this at the sub itself for basic balance) There are no separate Levels adjustment for DSD. You could adjust the Levels in one of your User Presets to use just for DSD.
|
|
NicS
Sensei
Will the G4 upgrade help quell my RMC1-L frustrations...?
Posts: 210
|
Post by NicS on Nov 28, 2022 20:45:13 GMT -5
This is hardly the most active corner of the forum, but... here goes... When using the XMC or RMC line of processors, is it possible to set basic level controls for the five channels of a standard SACD 5.1 mix? It seems when I play SACDs I have to manually wind down the level of the surrounds every time by about 5dB to get a reasonable level match. Q1 - can I do this so the level controls persist with SACD/DSD playback, without bringing in any other controls? Q2 - is having this basic DSD multichannel control been on the radar for Emotiva as something to include in future firmware upgrades? It doesn't have to be output control of the surround channels; input level would also suffice to balance the surrounds and perhaps subs (though I do this at the sub itself for basic balance) There are no separate Levels adjustment for DSD. You could adjust the Levels in one of your User Presets to use just for DSD. I did not think this was possible. DSD from SACD bypasses all level settings as far as I can tell.
|
|
geebo
Emo VIPs
"Too bad that all the people who know how to run the country are driving taxicabs and cutting hair"
Posts: 24,204
|
Post by geebo on Nov 28, 2022 20:59:57 GMT -5
There are no separate Levels adjustment for DSD. You could adjust the Levels in one of your User Presets to use just for DSD. I did not think this was possible. DSD from SACD bypasses all level settings as far as I can tell. Works on my RMC-1L. Levels work in the analog domain and don't use the DSP.
|
|
NicS
Sensei
Will the G4 upgrade help quell my RMC1-L frustrations...?
Posts: 210
|
Post by NicS on Nov 28, 2022 21:06:32 GMT -5
I did not think this was possible. DSD from SACD bypasses all level settings as far as I can tell. Works on my RMC-1L. Levels work in the analog domain and don't use the DSP. Hmmmm. I’ve tried changing settings whilst in DSD mode, which is automatically engaged. Doesn’t allow me to do much.
|
|