Martin Logans and 3-Dimensionality of Sources.

[Victor]

An infallible test for determining the “rightness/wrongness” of CDs, using Martin Logans?

ML users will be familiar with the concept of 3D soundstaging and the way MLs excel at this.
What may have been LESS obvious is the DIVERGENCE between LP & CD.
LPs generate quasi-realistic 3D soundstaging i.e. consistent instrument/vocal height.
CD soundstaging is strictly 2D for *many* CDs (my ears, various equipment).

With planars and CD, it is not untypical to hear ALL voices and instruments reproduced on a 2-Dimensional plane suspended at, say, 6ft (or top of the panel).
So, if I’m listening to piano with vocal accompaniment, I must imagine that the instrument is on a stage/platform and I’m sitting in the front row of the stalls looking up!
(The voice is more problematic!)
Although not unbearable, this situation could be better...

Critically, in this comparison, what we literally have here is a “perceptional metric”, the ability to provide “3D co-ordinates” of instruments & voices for each format, rounded to the nearest foot, which is a little more direct and characterisable than general assessments of the audio properties

I should state this is not an attack on CD replay. It is possible for some CDs to match the “3D” characteristics of the corresponding LP. They appear to be the “exceptions that prove the rule”.
CD has no problem imaging horizontally and even beyond the soundstage to extreme L & R (as of course does vinyl).
Mostly, for CD replay, if you imagine a horizontal line running across the speakers and beyond, the instruments/voices will fall somewhere along that line.
There does not seem to be any rhyme or reason to this other than mastering/CD replay issues because, from experience, I know it is possible to record the “3D” behaviour of the vinyl LP and replicate it digitally as lossless 16/44 or whatever. (Hence the success even with many of You Tube’s lossy files).
Thus 16/44 digitisation is not to blame but it doesn’t stop many “classic” CDs showing “2D” behaviour.

Even though I’ve been aware of this issue for many years, recently I discovered a (possible?) explanation ventured by George Louis.

(Google: “CD Absolute Polarity”.)

The answer, if George is indeed correct, appears simple...

So, does George have a point with his “polarity reversal” explanation?
Is it possible we can “redeem” ALL CDs?
Certainly as far as his statistics are concerned my specific 3D “qualifier” convincingly supports the view that 100% of LPs are 3D whilst *some* CDs are 3D (for most systems?) but as to whether the problem may be addressed by polarity reversal depends on more user feedback using the aforementioned “height” qualifier(?)
To be fair, George does make reference to 3D presentation being one affected property, amongst others. All I’ve done here is focus on this effect.

Your opinions/experiences with your equipment will be invaluable.

 

[rhd1953]

WOW, something must be seriously wrong with your system. I've been going thru my music and the vast majority of records and cds have excellent sound staging (3d). I'm streaming digital files (mostly wav) from my pc via VLC (media player software) and a Schiit Modi 3+ (usb to rca audio) and Anthem AVM 30 pre to a classDaudio SMA power amp and then the Magnepan 0.7 speakers.
Sometimes the streaming internet radio stations (frequently mp3 format) sound flat (2d).
I've always had 3d sound from my Martin Logan CLS ii (since 1989), listening to records, reel to reel tapes (of records), later cds (both pre recorded and burned from tape), cd rips, and from digital download files (HDTracks).

 

[Victor]

Thanks for the feedback RHD. It appears you are immune (!)

I haven’t read all of George’s Blog but I did read enough to understand that if a user is fortunate enough to possess an “Absolute Polarity” player/DAC by default (as opposed to “Inverted Polarity”) then they will be mostly unaffected by this issue.
Technically that could either mean a lot of CD players have “a problem”....or...it could mean a lot of recorded CDs have “a problem”, depending on your situation.
The implication is that, for most people, the latter applies, hence the test.

I did some reading BTW and was surprised at how long the industry has been aware.
MERIDIAN (“Absolute Polarity”) CD players WITH polarity reversal switching capability have been on the market since the early 1990s(!)
Quote from an early Meridian User Manual : “...(since...the music industry has not released all music in correct polarity....in some cases....)...”inverting *phase* can make the music sound more realistic”.
WADIA were also ahead of the game in providing such a feature.
DCS (current models such as Vivaldi) also boast a polarity reversal option in their menus but the entrance fee might be too steep for some. The DCS can also swap channels for you.

Test Tip : Sit in your normal listening position and listen for the drums to appear as they tend to be a key indicator (normally I would expect them to be “drum height”). If the drums align with the top of the speakers it’s a dead giveaway that the result is probably going to be 2-dimensional. There may be rare exceptions but largely it might only take seconds per track to identify.

I tried this out on a few CDS.
The first two were “2D” and had their horizontal plane” set at panel top edge.
(Might vary slightly).
The 3rd album (a compilation) was more than halfway through, the music being consistently 2D as before when suddenly we were startled by the next track which was an excellent 3D rendition. It was a gentle track but instrument height was true “wall-of-sound” ranging from 2.5ft to over 6ft. All other tracks that followed were also 3D so the split was about 50/50 on the compilation, not 85%.
This prompted the thought that CD compilation albums may actually be of help in identifying “3D-ready versions” of albums.
If the compilation has say 20 tracks extracted from 10 albums you can use it to identify which albums pass muster (with your equipment) prior to purchase.
Yes, while it is true that some tracks are recorded in different studios and may balk this trend, there appears to be commonality throughout most albums.

Important note : Sadly, if you think this only afflicts classic CDs and that new all-digital masters might be free of “Inverse Polarity”, indications show they are equally guilty.

I know some will argue that certain instruments naturally and routinely invert polarity (even providing oscillograms as evidence) implying that finding correct polarity is “meaningless”....
My take on this (using the 3D qualifier above) is that this assumption of “meaninglessness” is untrue. IME, guitars etc. (I think you will agree) are ALWAYS fixed in placement in EITHER medium and do not drift around the Martin Logan soundstage just because the guitar player decides to pluck upwards not downwards.
If a guitar does change position it is because the Engineer did something intentionally to create that change.
As always, our ears will be the final arbiter but please keep an open mind...

 

[Len44]

This is an interesting supposition. If you would be so kind as to share the specific CD compilation album on which you noticed the differences. I would like to investigate as this is not something I have (knowingly) experienced.

 

[rhd1953]

Do you have issues with ripped cds, and then playing the audio files?
Does it occur when playing an audio cd thru a dvd player?
Have you tried "Audio Test CD - Lasertrak CD2000 (test Tones, benchmark)" from testdisc? (https://www.testdisc.com/collections/all)? I just looked and it seems to test all the individual components of a stereo system. Oops, I just ordered the last one.

 

[Gordon Gray]

Dimensionality differences are playback medium / system dependent. For the most part, it all depends on the quality of the recording and the mastering irregardless of the medium.

 

[Victor]

Dimensionality differences are playback medium / system dependent. For the most part, it all depends on the quality of the recording and the mastering irregardless of the medium.

With reference to what Gordon has said, if you feel that your source data is of poor quality for listening/evaluation then don’t use it. If you are uncomfortable about it just find another disc.

 

[Victor]

To maintain balance, yesterday I tested needledrops of LPs....
I trawled through 34 before finding one that failed the test.
This was larger than any digital sample I’ve tested so far.
Results appear to correlate with the industry(?) belief that “virtually all” of LPs are rendered as absolute polarity.

For this LP I also own a digital box set featuring CD & DVD-A plus DVD-A STUDIO NEEDLEDROP of the album(!)
(Note I have similar box sets for some of the “3D” LPs which were under test)

Did I also test the digital copies?
You bet your sweet life I did.
Results were extremely “interesting”.

With the body of evidence building up I am satisfied that George is onto something.
Identification is easy so try it for yourselves.

 

[Victor]

OK, so here’s what was tested :

1. Original UK Vinyl (my 16/44 RIP)
2. DVD-A Original UK Vinyl (Studio vinyl RIP) Pure Direct stereo.
3. CD Transfer, Pure Direct stereo
4. DVD-A Pure Direct stereo

5-6 are not same source as 1-4 so comments are only for the sake of completeness.
5. DVD-A Summit Studios DTS 5.1 (as 4.0)
6. DVD-A summit Studios (down-mixed to “Stereo”)

For obvious reasons (1) is played from a different digital pathway to (2-6)!
As a spoiler, all versions of (1) i.e. (1-4), were virtually indistinguishable SQ wise except for 3D/2D presentation although it should be said that if there is indeed correct polarity present it appears to confer other benefits such as improved bass presence.

Results :
1. My own 16/44 needledrop : After 10 mins of listening in disbelief. Result strictly 2D as mentioned in previous post.
2. DVD-A Studio vinyl RIP matches mine i.e. 2D, top of the panel.
3. Big surprise!! CD is Absolute Polarity!!?? An inversion of the previous 2 results. Result 3D.
Soundstage is decently rendered depth-wise. Drums are waist height. Guitar centred <1ft above the drum skins. Saxophone fractionally higher. LH channel high hat about same as Sax, voice centred at head height...etc.
It’s a loud, complex, messy, dynamically challenged live concert recording but it works.
If polarity is responsible (and it does look that way), the argument for getting it right is compelling, even with low-SQ source data like this.
This CD will be my reference from now on, even if I only listen in 10 yrs time.
4. DVD-A 2.0 version : Matched the CD performance exactly i.e. 3D

5. & 6. This a re-recorded set. No surprise these are the best of the lot because it’s a Studio Set mastered using something other than a domestic stereo cassette deck!
The 3D effect fills the space between the speakers. In 2.0 it matches the CD result (3D). The 5.1 mix is as you would expect, room filling at realistic height.

A quick recap :
This is strictly a “rubbish in, rubbish out” exercise.
We’re not trying to overthink it or extrapolate what reversals are occurring within the digital/analogue chain.
Each disc has an equal chance of presenting its case using the relevant player.
This why I won’t even mention what gear is being used.
Even changing turntable+codec+Digital players doesn’t alter the outcome as in (1) & (2) above.
The only thing I will say is that the players used are always solid state. No valves.

 

[Victor]

For fun, I explored the idea that CD compilation albums might prove to be a guide to “3D Ready” CD originals...
I mentioned one compilation earlier in which the first half was 2D and the 2nd half was 3D.

This new compilation CD features 17 tracks.
BTW, the various album songs weren’t clustered, they were randomly scattered throughout.
I graded each track as either 3D/2D then checked their origins.
The 3 main source albums I will refer to as 1, 2, & 3.
After sorting the results Album1 had 3 tracks, Album2 had 4 tracks and Album3 had 4 tracks in this compilation. A total of 11 tracks.
(I didn’t bother with the remaining 6 as the sources were unidentified.)
All tracks per album were found to be consistently either “2D” or “3D” except one track from Album 3.
I’ll come to that in a moment...

Conclusion was that (1) & (2) were likely to be 3D albums as their relevant tracks from the compilation were all 3D. (I don’t own these albums so cannot absolutely say.)
The 4 tracks from Album3 were 2D except for a LIVE CONCERT track - recorded back in the days when crowds were allowed!
Obviously since this was not a studio recording the results might reasonably be expected to differ from the others so no special significance was attached to it being 3D.
As a final check I referred to a CD of Album3, which i was fortunate enough to have and sure enough, it replicated those findings perfectly. No surprises. It appears to be (mostly) a 2D album.

Unlike the other compilation, this was a random scattering of 2D interspersed with 3D results but which still broke down to a near 50/50 split.
Thus, I still can’t yet verify George’s 85% figure but can affirm the effect is interesting and widespread enough to warrant action.

 

[LJSki]

Could you list the specific track-recording used on your tests (everyone, not just prime post)? I would love to be able to compare apples to apples (excepting EQ of course) and see what my perception detects. This is a highly fascinating discussion. Probably don't need all, but at least a short list of ones that were clearly detectable as to differences.

 

[Victor]

Please note several recent posts have disappeared (Restore from backup causing history loss?).
No problem.

In Martin Logan we have one of the most phase coherent and revealingly accurate loudspeakers on the market (at any price) which is exactly what we need for the “3D test”.
This test, until now, was performed using a couple of universal players as transports plus the DACs of 2x AV processors/receivers.
I have now included a dedicated CD PLAYER for further comparisons. (Wow!)

To keep the signal path/number of units simple, the CD player’s internal DAC was used and the player’s analogue outputs connected to the same dual-mono Class A/AB amp as used previously.
The usual offenders (i.e. CDs) were tested along with non-offenders.

RESULT :
The outcome for CD player presentation was IDENTICAL to the other players/DACs i.e. previously identified 2-dimensional discs/tracks still sounded 2D and vice-versa.
So far, so good/bad.

NEXT :
The next phase is to insert a switchable (polarity reversible) DAC and see what effect that will have on 3-Dimensionality. This will be split into 2 tests :
1) Can the 2-dimensional CDs be “corrected”?
2) Does the polarity switch render 3D (absolute polarity?) CDs as 2D?
I’m not presuming this to be a foregone conclusion.

A phase-switchable R2R DAC (not Schiit - overbooked) has been ordered and will arrive today.
I will report the effects. Watch this space...

 

[RCHeliGuy]

I mostly just just play ripped or streamed music. I haven't actually used the CD transport to listen for many years.

That said, I get exceptional 3D imaging from my ML speakers with digital sources. Frequently I can pick out numerous positions that are very distinct. I think it makes the music sound much more clear when the music comes from multiple sources vs all being piled up together.

That assumes excellent source material. For poorly recorded music, you usually get a properly centered singer and not much more.

Patricia Barber Companion is an excellent example of source where you can easily pick each individual instrument out and of course the singer. I purchased that CD and ripped many years back. It's a 44.1kHzx16bit FLAC file nothing special there, but it was recorded very well.

To be fair, I have to be sitting in the center of my couch to get this effect. The stereo imagery falls away rapidly if you move off to the side.


The imaging I've been getting is identical sounding between using a PC with JRiver Music Server feeding my DAC or my current Raspberry Pi Music server feeding my DAC.

 

[Victor]

LOL! I’ve managed to largely avoid playing CDs for over 30 years!
In fact I generally manage to play music without mechanisms (includes hard drives) and the not-always-reliable internet. I do own a modest collection of several hundred CDs (bought another new one this week).
It’s good to give back to the music industry in direct exchange.

I’ve been testing the invert-able DAC but while I’m formulating results - try this...

Follow the path of the solo guitar...
It starts dead centre of the LH panel then moves slowly upwards....

First, scroll to 27 mins 42 secs on the counter :

 

[RCHeliGuy]

Hocus Pocus by the Focus complete with Popeye style yodeling. LOL! Got to love it!

I'm a big fan of The Carpenters and I was recently listening to "Calling Occupants". Karen has a beautiful voice even if some songs are hopelessly dated. The twangy guitars in "Top of the World" always crack me up.