Testing audiophile claims and myths

[bigshot]

I did a very careful line level matched direct A/B comparison test with this recording...

http://www.amazon.com/Stravinsky-Histoire-Dumbarton-Concerto-Orchestra/dp/B0000XKB9W

The redbook layer was audibly identical to the SACD on multiple headphones and three different really good sound systems, including the one in the lab of my sound engineer friend.

 

[analogsurviver]

  What are the audiable differences between pcm and dsd?, the only difference I see is dsd recordings can store and playback even more inaudiable frequences than 192khz pcm.

To be blunt - time domain. DSD has MUCH better pulse response - even in real world, where filtering is still needed ( thus taking away some of its superiority in time domain, a necessary compromise in order to keep the ultrasonoc noise at an acceptable level ), than anything PCM - DXD included.
 
I have promised photos from the analog oscilloscope for the squre wave response of various analog and digital devices. I am anything but an accomplished photographer - but this post will come sooner or later. 

 

[bigshot]

There's no impulse in recorded music that even comes close to not being covered by 44.1. A redbook sample represents a very very VERY small sliver of time. Aside from that, an impulse faster than redbook can capture would need to occur at a frequency beyond the range of human hearing by definition. (see Nyquist)

 

[JRG1990]

Ive never heard audiable timing errors with pcm.

 

[bigshot]

I love it when people talk about "fast" timing and PRaT and point at the attack and decay of a snare drum hit as an illustration of what they're hearing. They don't have any idea how many samples go to make up that snare drum hit from attack to decay!

 

[analogsurviver]

  Ive never heard audiable timing errors with pcm.

It is both timing AND amplitude problem.  If listening to PCM recording ONLY, without frequent exposure to the real thing, one can "adjust" to that sound and - not miss anything. 
 
And then be "shocked" how much different live sound is. 
 
This is audible in all but the "slowest" instruments ( organ...it can not change volume fast - no matter what, it has slow start and yet slower decay, particularly in large venues ) - but easiest is to listen for this to a drum solo live WITHOUT ANY MICROPHONES/SPEAKERS - and then to the CD by the same drummer.  
 
Believe me, this laaaarge caaaaaanyon of the difference experienced above can be bridged - but most definitely not with redbook CD. 

 

[bigshot]

Redbook is more than capable of reproducing both amplitude and timing correctly. The redbook standard was created to exceed the range of a human's ability to hear. The differences between a live feed and redbook are going to be inaudible.

The area that standard redbook can't cover is in dimensional sound fields. SACD does sound better than two channel redbook for that. But multichannel sound is a lot more work to do right, so people focus on things that don't matter that are easy instead.

 

[analogsurviver]

Redbook is more than capable of reproducing both amplitude and timing correctly. The redbook standard was created to exceed the range of a human's ability to hear. The differences between a live feed and redbook are going to be inaudible.

The area that standard redbook can't cover is in dimensional sound fields. SACD does sound better than two channel redbook for that. But multichannel sound is a lot more work to do right, so people focus on things that don't matter that are easy instead.

Now - PLEASE - go to a concert hall, set up ANY semi decent microphone(s) setup ultimately giving two channel output - and adjust the direct feed from those mics AND output from any redbook CD standard recorder within 0.1 dB or less, ABX switching done by quality ANALOG switch , sighted or multitple blind -  and listen with any decent headphones. 
 
And then ask yourself, honestly - if you still can claim the above statement .
 
( I certainly can hear the difference between the live feed and DSD128 - perceptible as slight loss of "air" around the instruments and voices - let alone the "blanket over everything" imparted by the redbook CD ) . 
 
One area we might eventually agree upon is dimension in audio. Regardless how one might call it - soundstage, depth, width, height, dimensional fields, etc. It is an area where redbook CD is at the poorest.
No, to do this right, no more than two channels are required - it has been around for ages, it is called binaural ( or artificial head, or Kunskopf, etc ). To be listened on headphones - this IS head-fi after all. Only with the advent of DSD, and DSD128 in particular, there is a recording medium available that does binaural finally justice . There IS a decisive advantage of DSD128 vs DSD64 (SACD) in binaural - redbook CD with binaural is a crude joke in comparison. Higher rez PCMs up to 192/24 are increasingly better than redbook, but simply can not provide the same degree of realism as attained by DSD128.
 
A good DSD128 binaural recording is capable of fooling a person hearing it for the first time into "conversation" with a person that was there a couple of seconds/minutes ago - producing astonished and at the same time amazed faces upon discovery that the "conversation partner" has left the room and the recording was so realistic to fool even - spouses .
 
Not gonna happen with redbook CD...

 

[cjl]

  Try http://www.discogs.com/Omnibus-Wind-Ensemble-Music-By-Frank-Zappa/release/1820934
 
If you can not hear the difference between the CD and SACD layers of this disc -  then nothing can help you. Even on the $ 120 Sony Blu-ray player.
 
Earlier CD only release is MUCH worse - and all policarbonates are eclipsed by the original vinyl LP release, now next to unobtainium.
 
Recording and mastering of this calibre is scarce, so more "mainstream" repertoire worth of DSD ( SACD is only an encrypted form of it ) is hard to recommend.
Omnibus Wind Ensemble does have a Mozart etc disc, should Zappa prove too much to stomach. 

It is entirely possible that there is an audible difference on that disk, since some SACDs use different mastering for the CD and SACD layers (so the difference is clearly audible). However, CDs are perfectly capable of replicating the same audio as SACDs, down to every audible detail.

 

[OddE]

However, CDs are perfectly capable of replicating the same audio as SACDs, down to every audible detail.

 
-With the (obvious; just thought I'd mention it) exception of multi-channel SACDs. (Where multi>2...)
 
Then again, that may already be covered by your comment if one reads it to claim that the contents of a channel on a SACD is indistinguishable from a channel on a CD (provided the master is the same, of course) - in which case I agree 100%.

 

[cjl]

   
-With the (obvious; just thought I'd mention it) exception of multi-channel SACDs. (Where multi>2...)
 
Then again, that may already be covered by your comment if one reads it to claim that the contents of a channel on a SACD is indistinguishable from a channel on a CD (provided the master is the same, of course) - in which case I agree 100%.

Fair enough, and that is a worthwhile distinction between CDs and DVD-audio/multi-channel SACDs.

 

[bigshot]

Now - PLEASE - go to a concert hall, set up ANY semi decent microphone(s) setup ultimately giving two channel output - and adjust the direct feed from those mics AND output from any redbook CD standard recorder within 0.1 dB or less, ABX switching done by quality ANALOG switch , sighted or multitple blind -  and listen with any decent headphones. 

And then ask yourself, honestly - if you still can claim the above statement .

Been there, done that in the recording studio.

 

[bigshot]

Multichannel sound is the next frontier for home audio. If standards for recording and playback can be firmly established, and some way of dummyproofing installation and room acoustics can get worked out, it will be as much of a leap over normal 2 channel as stereo was over mono. Unfortunately, 5:1 straight out of a box doesn't even come close to touching its potential. It takes a experience to tweak it properly- perhaps installation services that make house calls and set the whole system up... like a cable man. I had a professional install my sound system, but the only sort of calibration service he offered was THX certification, which in my opinion is overpriced and not at all what I was looking for. I had to do all of my tuning myself.

By the way, binaural recording is tremendously limiting and only appropriate for certain types of recordings. 5:1 is much more flexible and has the potential to improve every kind of recorded sound, from television broadcasts to recorded music of all kinds.

 

[esldude]

http://www.aes.org/sections/pnw/ppt/ivantash/loudspeakerarraysforfocusinganddiffusingsound.pdf
 
Nice AES presentation about beam forming arrays and some possibilities that exist to exploit for multi-channel sound and other purposes.  Long arrays of small speakers with the properly filtered and delayed input are cable of sending controlled beams of sound for mono, stereo or any number of other channels from a single up front position.  I gather the result would improve even more with multiple arrays in more positions.

 

[analogsurviver]

Been there, done that in the recording studio.

I will answer your unedited reply ( copy-pasted from my email subscription to this thread ).
 
It certainly is possible to compare live feed to output from the recorder of any kind - analog and digital.  Analog has the advantage of including only  the defects that creep in during recording ( wow & flutter only of the recording part, not the combined w & f of recording and plaback , for example ), meaning that what you hear in monitor at the time of recording is the best that will ever come out of an analog machine.
 
With digital, it depends how data is stored. If it is optical ( CD-R), there is a fair chance that ultimately finished CD will sound better than the monitor off the CD-R being recorded. If the data is being recorded directly to hard disk or solid state device, it is the best you can ever have.
 
Korg DSD recorders record to a hard disk ( solid state is , even under best/most expensive scenario, barely capable of 100 MB/s requirement  for glitch-free DSD recording - besides, it apperently does not like numerous deleting/overwriting , leading to premature destruction, all of which is totally unacceptable for the requirements of DSD ) - and you can only hear its analog output AFTER it has been recorded to the HDD. It is the very same thing listening to the monitor or playback - so DSD recorders can be said to  "burn their audio on the fly - in real time ".
 
DSD can be burned to disc http://www.ps3sacd.com/dsddiscguide.html I never did it - because it can only be DSD64 (same resolution as SACD) and after hearing DSD128 it just no longer makes any sense. So I can not comment how DSD disc sounds compared to DSD off HDD.