Jump to content

Audio Myths and Human Perception - Explored


mikebse2a3
 Share

Recommended Posts

10 minutes ago, robert_kc said:

Regarding old analog recordings that have been remastered, I have found a few that have surprising good quality. 

I was referring the 5.1 DVD-As and BDs being touted as "new"-especially from DG and some from Philips, etc. 

 

I have quite a few of the three-channel "stereo" recordings from Living Presence, etc. that are quite good, but still exhibit higher noise levels in quiet passages than more recently recorded discs.  But some of those 1960s recordings are quite enjoyable because of their historical significance (especially the Russian recordings) and their musical interpretations. 

 

10 minutes ago, robert_kc said:

This is one of the discs where I found some mastering EQ applied (slightly exaggerated highs and attenuated lows) which makes the mix somewhat strident on flat reference loudspeakers.  It's a BD which I own and was produced by Decca. 

 

If I had the ability to demaster that EQ on BD, I would be a much happier camper since the performance is quite beautiful.

 

Chris

Link to comment
Share on other sites

1 hour ago, Chris A said:

This is one of the discs where I found some mastering EQ applied (slightly exaggerated highs and attenuated lows) which makes the mix somewhat strident on flat reference loudspeakers.  It's a BD which I own and was produced by Decca. 

 

If I had the ability to demaster that EQ on BD, I would be a much happier camper since the performance is quite beautiful.

 

Chris

 

My solution - which is great for me as a hobbyist who collects tube amps - is to have different amps that pair well with different recordings.   This Swan Lake recording sounds fabulous with my Scott 299C - with the tone controls set to my taste.  Played on an MC240 (no tone controls), it can be a little bright.

 

Bottom line, I generally recommend that people get an amp with tone controls, due to the variances in recordings.

Link to comment
Share on other sites

 

 

I have an SACD that I think may have an engineering problem, but I’m not 100% certain:  Reference Recordings RR-136 Saint-Saëns Symphony No. 3 "Organ" (Kansas City Symphony) SACD.   Reference Recordings has graciously offered to accept a return and provide a refund.  (Their current opinion is that the SACD is OK, and that the problem is with my equipment.  My understanding is that they forwarded my email to their engineer, but I haven’t received any response from the engineer.)   Before I return the SACD I thought I’d check to see if anyone else owns this SACD, and can share their experience:

 


Saint-Saëns: Symphony No. 3 in C Minor "Organ Symphony"

 

My concern is that the lowest organ pedal notes on this SACD are extremely loud, in comparison with the 24bit/176kHz FLAC stereo download (HDTracks) of the same recording - particularly on track 6, 1:20 – 1:50.   And the deep bass is extremely loud in comparison with my other SACDs (5.0 and 5.1), FLAC downloads, and CDs featuring pipe organ music (including Reference Recordings 133 “Organ Polychrome” featuring the same Julia Irene Kauffman Casavant Organ in Helzberg Hall).   And I’ve played a test CD that has warble tones at a variety of frequencies, and deep bass is not exaggerated.

 

Moreover, I have full season tickets to the classical series performed by the Kansas City Symphony.  And I’ve attended several concerts in Helzberg Hall featuring the Julia Irene Kauffman Casavant Organ.  I’m familiar with the live sound in Helzberg Hall, including the pipe organ.  (Though I haven’t heard the KC Symphony perform Saint-Saens Symphony 3 live.)  IME this SACD’s deep bass is unnaturally loud compared with the live sound in the hall.

 

I’ve experimented with the software settings in my Oppo players (UDP-205, BDP-105, BDP-95).   I have the Oppo software parameters in each player set similarly – i.e., so that the subwoofer plays for SACDs.   The same settings work OK for all other recordings, and I therefore think that the problem is not with the Oppo players.   (Of course, I could be wrong …)

 

My subwoofers in my various systems (SVS Sub16-Ultra, Klipsch R-115SW, and Klipsch P-312W) are connected to an Oppo player.  The Oppo players are also connected directly to vintage tube amps - no AVR or preprocessor.   (In other words, the Oppo units provide the “pre-amp” functionality.)   The subwoofers are adjusted so that all other recordings of classical music (including pipe organ) have an appropriate amount of bass.

 

Does anyone else have this SACD, and play the SACD layer (vs. CD layer), and have their system configured to engage the sub when playing the SACD layer, and have large subwoofers?

Link to comment
Share on other sites

1 hour ago, Chris A said:

I have another solution:

 

 

Chris

 

As I said earlier, if the goal is audio nirvana – when you forget about the equipment – and lose track of time and become completely lost in the music – then I think there are many paths to the top of the mountain. 

 

 

Different people enjoy traveling different paths.  :)

 

Link to comment
Share on other sites

Do you have a picture of your setup?

 

You must own a lot of amplifiers to be able to use them as you've described.

Link to comment
Share on other sites

Here’s my current systems:

 

Living room:  Stereo speakers are Snell Type CV.   Subwoofer:  Klipsch P-312W.  The source components are Oppo BDP-95 (with USB hard drive containing high-res FLAC recordings), and Dual 1249 with Stanton 681EE.  Amps include a pair of McIntosh MC30s, Scott 296, McIntosh MX110 / McIntosh MC275, a pair of Pilot HF-56 mono receivers, an NAD pre-amp and Acurus A250 power-amp for movies, and a McIntosh 2155 driving JBL L830s in the kitchen / dining room.   A patch panel (banana plugs) allows me to connect the speakers to whichever amp I want, and a Niles AXP-1 RCA selector switch connects the Oppo to the amp.   Chromecast Audio for internet radio and Spotify Premium.

 

TV room:  Stereo speakers are Klipsch Palladium P-37F.   Subwoofer:  Klipsch P-312W.  The source is an Oppo BDP-105 (with USB hard drive containing high-res FLAC recordings).  The amps are Scott 399, McIntosh MC225, Kenwood KR-9050, Fisher 800B, Fisher X-1000, Scott 299C, McIntosh MC240, and an NAD C375BEE.   The tube amps are for music.   The solid-state amps are for movies.   A patch panel (banana plugs) allows me to connect the speakers to whichever amp I want, and Niles AXP-1 RCA selector switches connect the Oppo to the amp.   Chromecast Audio for internet radio and Spotify Premium.

 

Office: Stereo speakers are JBL L880.  Sources:   Oppo DV-980H SACD/CD/DVD, and my Windows 10 laptop with Music Streamer II DAC.  Amps: Fisher 500C, Scott 299B, Altec 353A, and an NAD D 3020 for general internet use (and summertime).   Banana jacks allow me to connect the speakers to whichever amp I want, and a Niles AXP-1 RCA selector switch connects the Oppo to the amp.  

Basement:  Front, center, and left speakers are Klipsch RF-7 II.  A single rear speaker is a Klipsch RF-7.   Subwoofers:  SVS SB16-Ultra, Klipsch R-115SW.  Source:  Oppo UDP-205
(with USB hard drive containing high-res FLAC recordings).  Amps: Scott 272, Inspire “Fire Bottle” SE Stereo Tube Amplifier HO, Scott 222C, McIntosh MX110Z tuner/preamp, Fisher KX-200, Scott 296, Pilot SA-260, Scott LK150.   A patch panel allows me to connect the speakers to whichever amp I want, and F/F RCA cables enable me to connect an amp to the Oppo, and a power amp to the MX110Z.   Chromecast Audio for internet radio and Spotify Premium.

 

Bedroom:  The speaker is a single Klipsch WF-35.  Source is an older CD player.  Fisher TA 500 (AM/FM mono receiver).  Chromecast Audio for internet radio and Spotify Premium.

 

 

Photobucket has stopped allowing linking to pics.   I'll try attaching a couple of pics here:

 

 

100_2163.jpg

 

  • Like 1
Link to comment
Share on other sites

  • Moderators
On 7/31/2017 at 9:58 PM, Chris A said:

The entire mastering culture is apparently geared to mess with your music...no matter what...and the more that they mess with it, the more credit--and therefore money--that they get.  Conversely, if they don't mess with the music, they stand to lose their livelihoods, or at least that's the belief that apparently keeps those practices recurring again and again. 

That is not only a myth, it is just simply untrue.

  • Like 1
Link to comment
Share on other sites

  • Moderators
6 hours ago, Chris A said:

Mike Bentz:

 

I again recommend setting up another thread on these specific effects...which you have in the past have been reticent to do for reasons unknown.  Then you can go in the direction that the OP (that would be you) wishes to go, instead of strong emotion-tinged responses filled with obvious angst over each response from the audience to "turn the tide in a different direction"...or perhaps other less friendly motives. 

 

I understand that you have near-and-dear thoughts on this subject--as evidenced by these type of responses.  I'd recommend channeling that energy into productive dialog and/or exposition instead.  Everyone will benefit--including you. 

 

And I certainly won't get in your way...

 

Chris

He doesn't  need to create another thread.  He is on point and on topic, maybe a little abrupt at first.  

 

Please do not tell people you disagree with to go to another thread.  If they are on point or responding to something they have as much right to do so as someone making a comment.

Link to comment
Share on other sites

  • Moderators
On 8/2/2017 at 6:52 PM, Chris A said:

I don't believe that the practice was ever widespread, in fact, never--due to the cost of trying to do things that way.  There is a reason why classical musicians can do what they do without "trying to do it over and over again".  The big name classical labels themselves have something to lose if they allowed those practices to occur on widespread basis--as is seemingly implied by those whose names on that forum.  Those CDs where the musicians are having trouble playing the music usually wind up in the clearance bins...just like anything else pretending to be espousing quality.  There is just too much talent out there to put up with that sort of thing...and the costs associated with it. The record companies only need to pick up the phone to get real talent--in any genre--that can do it in one take.

 

Chris

Splicing and editing the best takes and measures has been happening as long as tape recording, and now digital has been around.

 

Every great conductor, every great soloist. Those people are typically perfectionists and they are not afraid to get the best of multiple performances.

 

Here is just one example of 1000s.

 

 

Link to comment
Share on other sites

1 hour ago, robert_kc said:

Here’s my current systems:

 

Living room:  Stereo speakers are Snell Type CV.   Subwoofer:  Klipsch P-312W.  The source components are Oppo BDP-95 (with USB hard drive containing high-res FLAC recordings), and Dual 1249 with Stanton 681EE.  Amps include a pair of McIntosh MC30s, Scott 296, McIntosh MX110 / McIntosh MC275, a pair of Pilot HF-56 mono receivers, an NAD pre-amp and Acurus A250 power-amp for movies, and a McIntosh 2155 driving JBL L830s in the kitchen / dining room.   A patch panel (banana plugs) allows me to connect the speakers to whichever amp I want, and a Niles AXP-1 RCA selector switch connects the Oppo to the amp.   Chromecast Audio for internet radio and Spotify Premium.

 

TV room:  Stereo speakers are Klipsch Palladium P-37F.   Subwoofer:  Klipsch P-312W.  The source is an Oppo BDP-105 (with USB hard drive containing high-res FLAC recordings).  The amps are Scott 399, McIntosh MC225, Kenwood KR-9050, Fisher 800B, Fisher X-1000, Scott 299C, McIntosh MC240, and an NAD C375BEE.   The tube amps are for music.   The solid-state amps are for movies.   A patch panel (banana plugs) allows me to connect the speakers to whichever amp I want, and Niles AXP-1 RCA selector switches connect the Oppo to the amp.   Chromecast Audio for internet radio and Spotify Premium.

 

Office: Stereo speakers are JBL L880.  Sources:   Oppo DV-980H SACD/CD/DVD, and my Windows 10 laptop with Music Streamer II DAC.  Amps: Fisher 500C, Scott 299B, Altec 353A, and an NAD D 3020 for general internet use (and summertime).   Banana jacks allow me to connect the speakers to whichever amp I want, and a Niles AXP-1 RCA selector switch connects the Oppo to the amp.  

Basement:  Front, center, and left speakers are Klipsch RF-7 II.  A single rear speaker is a Klipsch RF-7.   Subwoofers:  SVS SB16-Ultra, Klipsch R-115SW.  Source:  Oppo UDP-205
(with USB hard drive containing high-res FLAC recordings).  Amps: Scott 272, Inspire “Fire Bottle” SE Stereo Tube Amplifier HO, Scott 222C, McIntosh MX110Z tuner/preamp, Fisher KX-200, Scott 296, Pilot SA-260, Scott LK150.   A patch panel allows me to connect the speakers to whichever amp I want, and F/F RCA cables enable me to connect an amp to the Oppo, and a power amp to the MX110Z.   Chromecast Audio for internet radio and Spotify Premium.

 

Bedroom:  The speaker is a single Klipsch WF-35.  Source is an older CD player.  Fisher TA 500 (AM/FM mono receiver).  Chromecast Audio for internet radio and Spotify Premium.

Which system are you using for BDs?

Link to comment
Share on other sites

  • Moderators

Someone posted a video on here awhile back  (dtel I think) of Toole talking about the studies that Harmon did on listening, consumer preferences, what golden ear audio press selected with blind testing.  It all sounded very scientific to me, I will find it.  His conclusion was with technology now, the best sounding stuff will be powered speakers with active processing such as the JBL M2.  Often wondered what you guys thought of the listening studies they did and if they are getting closer to being able to test for hi fi sound in speakers?  Or, if it is still the same ole objective v. Subjective argument with a prettier wrapper.

 

I will find it and post it.

Link to comment
Share on other sites

  • Moderators

Actually Chris posted the video of Floyd Toole I was thinking of in this thread

 

A lot of his studies are about consumer preferences, but they do include some self-described audiophiles. 

 

To me if you have speakers that meet Toole' s 5 design criteria, and you put all the science you possibly  can, there is still going to be a large amount of subjective preference that is going to vary from person to person.

 

At what point is it Hi--fi?   There are people, both in recording arts and consumers who have strong opinions that you cannot have Hi-fi below SACD quality.  I'm not talking about hearing a difference.  I'm talking about you are not getting an acceptable level of reproduction from a CD, you need 96 or better. Is that a myth?  Just personal preference?

 

I don't think we could even get agreement on what Hi-fi is.

 

@mark1101 for a time, maybe up to the present, would say that in order to get the best reproduction you need to have analog Eq and passive crossovers.  There is no doubt in my mind that to for him, to get to hi-fi you have to have that.

 

The purpose of testing, I would hope, is to try to separate myth from reality.  Nut it seems like for every step forward in psychoacoustics we.take two steps backwards. 

 

Are we any further ahead of where we were a couple of years ago?

 

 

 

 

Link to comment
Share on other sites

18 minutes ago, dwilawyer said:

Are we any further ahead of where we were a couple of years ago?

 

 

 

@dwilawyer have you read "Hearing vs Measurements" by Heyser I posted ..?

 

That article was 30 years ago..... If he had not passed away I could only imagine how much more progress would have been accomplished. .!!!

 

miketn 

Link to comment
Share on other sites

This is very insightful, and explains something I've been noticing in JRiver.

 

A lot of tracks I've been listening too lately have a measured dynamic range of 4.4dB or less, but they obviously sound much better and are very dynamic on my system.  After reading this post, I found this link: https://wiki.jriver.com/index.php/Dynamic_Range

 

"Which one is more accurate? It depends on what your criteria are. If you consider extremely narrow dynamic peaks to be an important part of the picture, then R128 may sometimes give you a misleadingly narrow impression. If you instead want to get a sense of how dynamic "most of the song" is, crest factor DR may give you a misleadingly good impression (in the example above because of a handful of clipped peaks). It's best to look at both and try to understand why when they disagree.

 

Dynamic Range (either one) is not the be all end all of musical quality. You may find you prefer some tracks that are more compressed than others. Dynamic Range isn't a perfect test, it's just more information."

 

12 hours ago, DrWho said:

The rest of us don't listen to 3 minute averages. We listen to individual instruments popping in and out as melodies and harmonies move around the sonic space, often listening through things to hear deeper inner details that an "average level" doesn't notice. That stuff is happening on timescales of tens of milliseconds (not minutes), and the cool thing is we're able to hear it because our ears are multidimensional across frequency and time. I can still hear the high frequency decay of a flute at 50dB SPL even though there's a huge 90dB low frequency drone coming from the cello section. The average SPL is still 90dB, but we're still hearing the >> 20dB dynamic range of the flute (probably closer to 50dB).

 

Link to comment
Share on other sites

Ellery, while this appears to be a new thing, in fact these "sliding integration window" dynamic range measures have been around for more than 5 years (the video below from 2013).  This is really not new and anyone that's been following the trends in industry Loudness War practices has been aware of these "new" measures which are kinder to the music undergoing dynamic range compression/limiting:

 

 

Ian Shepard makes his living doing mastering and is a major force in the "Loudness Day" and DR Database site.  He has been searching for ways to reconcile the practice of crushing of recorded track peaks (time domain) with perception of dynamic range.  This work also is apparently also related to dynamic range limiting algorithms used in radio broadcasting for many years.

 

In many ways, this new measure--which is trying to increase the understanding of brief instances in music where there is relative dynamic range--seems to me like governmental measures of unemployment: if you don't like the answers you're getting, then change the measures.  However, the new measure is an improvement to completely ignoring the effects of combined compression and limiting that have been used fairly indiscriminately in the past without acknowledgement by the practitioners what has been occurring (as Bob Katz highlighted in his now-famous video that I posted above).   Shepard uses both crest factor and these sliding integration window measures in this and other videos to give his audience a feel for how these two types of measures relate to each other in hands-on experiment fashion.

 

Shepard has been quite open with these type of measures in order to include the genre-specific dynamic range adjustments that I referred to above.  The idea is that you'd have another measure to check against--not only for radio broadcast limitations (broadcast band deconfliction of adjacent broadcast frequencies related to music dynamic range and bandpass limiting), but especially for the mastering practices used in heavily compressed/loud music tracks that he deals with for non-radio mastering, is actually is a double-edged sword.  How does one now compile the same data into databases like the DR-Database site for ease of access and intelligibility of what is being bought by the (now better informed) customer?

 

I now wonder--just like in unemployment statistics used nowadays--how does the average person integrate all these measures to get to an intuitive feel for how much loss of dynamics is occurring, and how that affects the listening experience (especially with hi-fi setups)?

 

I have a feeling that the picture has been made more granular, but at the expense of losing the patient--at least for those struggling with the technical aspects of the discussion.  This is one of the reasons I haven't discussed these new type of measures over the past few years: I believe that the casual observer is actually being left behind.  I believe that Shepard fears that situation is occurring, too.

 

YMMV.

 

Chris

  • Like 1
Link to comment
Share on other sites

I think there is less perceived loss on a system that has the best possible S/N ratios in the signal chain (e.g. a lot more dynamic range) for the reasons @DrWho describes in his two posts.

 

I know in 2013 after I bought the squashed album "Daft Punk - Random Access Memories" when it came out, I was using a much less expensive signal chain, and the album sounded very squashed (La Scala II's fed by ATI AT2007 amp and an Onkyo TX-NR905 receiver).

 

On my new setup capable of ~119dB+ of dynamic range (limited by D/A of the Xilica XD4080 when fed using AES), the album sounds noticeably better.  I think the improvement comes from tri-amping using amps with a S/N of ~132dB and having much lower IMD (from the amps themselves as well as tri-amping); also the D/A stages of the Xilica are far better than the Onkyo (plus I'm not using an AVR for volume control anymore).

 

It would appear as you increase the S/N ratio and improve IMD, it makes the hidden stuff much easier to hear on these types of tracks (e.g. background instruments and the little details still present even on the loud tracks).  I'm not saying these are the only parameters, but it would seem they are far more important than I previously thought if your goal is to get every last detail out of a so called "squashed" track.

 

As DrWho pointed out, since our ears operate in the time domain, but can differentiate frequencies at the same time, our minds can pick out quieter details even with a loud electric guitar playing in the foreground.   Further, even over the same mid-range frequencies, somehow our minds can pick out conversations in a crowded restaurant even though there could be 20 people talking at once; no SPL meter can do that.

 

5 hours ago, Chris A said:

I now wonder--just like in unemployment statistics used nowadays--how does the average person integrate all these measures to get to an intuitive feel for how much loss of dynamics is occurring, and how that affects the listening experience (especially with hi-fi setups)?

 

Link to comment
Share on other sites

If the goal is understanding of relative levels of compression/limiting...here is an R128 loudness measuring batch application that can be used on your digital tracks:  http://r128gain.sourceforge.net/

 

Perhaps I'll explore this a bit, as well as the free Melda MLoudnessAnalyzer, which is really not geared for scalar measures, but rather for interactive use on the track to zoom in on sections of the tracks.  It may bring more insights into the differences in genre and the sound of their dynamic range values.

 

Note that, at least for me, the DR-Database TT Meter plugin combined with the ReplayGain calculator plugin for foobar2000 is still pretty much my gold standard for how much compression/limiting has occurred to a track.  These are apparently the most difficult loudness measures to achieve higher values, and are most sensitive to compression/limiting processing that has been applied to the tracks.

 

ReplayGain values that are more negative than about -4 for any track will tell you how much damage there is.  I've had albums that have measured -10 dB (i.e., that's an average of 6 dB of compression over the entire album over and above the maximal loudness that can be achieved using only peak limiting in moderation.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...