Ask questions to Audio insider Sir Terrence

[Sir Terrence]

Originally Posted by Stereo24192

So, for movies, they never make a 2 channel master of any kind in any stage of the recording/mastering process?

This question I can answer. Never.

I wonder if the way 5.1 is downmixed in the player is any more sophisticated with the new formats on Blu-ray. I read somewhere that with Dolby Digital there is a fixed standard where they simply lower the level of the surrounds 3 dB and mix it in with the fronts (and maybe the same with the center).
That's what I call a compromise!

Actually as compromised as it sounds to you, its really quite elegant of a processes when you only have a pipeline for two channels. Yes the surrounds are lowered 3db, and mixed "out of phase" into the front L/R channels. The center is raised 3db, and mixed "in phase" into the fron L/R mains. All you have to do is do the reverse at the decoding stage, and you have your matrix surround mix. You can never acheive perfect seperation as you had before encoding, but the steering logic of the decoder can help with that.

Since I don't like compromises I almost feel forced to get a surround setup -and that would be too expensive right now since I'd want the same quality as I have now in my stereo setup (NAD Masters M3, B&W 703).And I don't want two separate systems.

You really should upgrade, because the future of movies and music is in surround, not stereo. Now I am sure that B&W has a high quality center and surround speakers to go with your 703, and there is a way to use a high quality two channel pre-amp along with a processor for 5.1. It takes some creative wiring, but I have heard of folks who are doing it. Surround is not a compromise, its a progression beyond the constraints of two channel stereo. And let me tell you, there are alot of constraints in mixing for two channel. When folks fully understand how much signal processing is used to get two channels to sound clear and good, they would not be so gung ho about it.

[Woody]

Originally Posted by Sir Terrence

This question I can answer. Never.



Actually as compromised as it sounds to you, its really quite elegant of a processes when you only have a pipeline for two channels. Yes the surrounds are lowered 3db, and mixed "out of phase" into the front L/R channels. The center is raised 3db, and mixed "in phase" into the fron L/R mains. All you have to do is do the reverse at the decoding stage, and you have your matrix surround mix. You can never acheive perfect seperation as you had before encoding, but the steering logic of the decoder can help with that.



You really should upgrade, because the future of movies and music is in surround, not stereo. Now I am sure that B&W has a high quality center and surround speakers to go with your 703, and there is a way to use a high quality two channel pre-amp along with a processor for 5.1. It takes some creative wiring, but I have heard of folks who are doing it. Surround is not a compromise, its a progression beyond the constraints of two channel stereo. And let me tell you, there are alot of constraints in mixing for two channel. When folks fully understand how much signal processing is used to get two channels to sound clear and good, they would not be so gung ho about it.

Are you from Cleveland, or is the cavaliers avatar for something else? I don't recognize the logo if it actually a Cleveland Cavaliers logo.

Just curious.

[jdc115]

Originally Posted by PeterTHX

Yes, but you prove my point and cite music videos. Which usually have dedicated stereo mixes on BD anyway. The way films & concerts are recorded and mixed is very different.

Plus you may be hearing a difference between compressed lossy and uncompressed there in your example.

I'm curious to see what people would think when the playing field is level and they hear something like Shakira: Oral Fixation Tour both presented in either 5.1 or 2 channel 24-bit PCM. Personally, I prefer the 5.1 mix, I feel like I'm in the audience (I also have a pretty high-end system).

I am waiting to be able to for that time. I have the Dave Matthew, David Gilmour, Shakira, NIN, and Legends of Jazz. I just do not have a HDMI processor yet but hope to change that in about a month and I will then probably go for 5.1 music.

[jdc115]

Sorry, I forgot I was on an insider thread before when I responded. Back to the insider

[DaViD Boulet]

You really should upgrade, because the future of movies and music is in surround, not stereo.

Agreed.

I'm a serious audiophile. The kind of guy who likes tube amps, LPs, and can hear the difference in jitter signatures between different transports and digital cables.

And I loved the 5.1 24/96 TrueHD track on the David Matthew's Band Blu-ray Disc. One of the most realistic audio reproductions ever to take place in my listening room.

Two-channel stereo is an artifact from two sides of a record groove. God never decreed that thou-shalt-listen-to-two-channel-stereo. 3 channel, 4 channel, five channel... it gets better and better the more channels you have to represent the authentic soundfield as long as the fidelity of each of those channels is maintained.

That's the audiophile's responsibility: put together a multi channel system that has proper timbre and fidelity for all channels.

And my 5.1 system still performs as a fantastic two-channel system when I listen to left-right stereo recordings (I don't apply surround DSP to my high-end 2-channel recordings), but it's also there for native 5.1 audio as well.

My REAL gripe is with 5.1 "music" mixes that don't use the center channel!?!? There seems to be a "better mix this phantom since the listener probably has a crappy center speaker" mentality that needs to stop...

[Stereo24192]

Originally Posted by Sir Terrence

This question I can answer. Never.



Actually as compromised as it sounds to you, its really quite elegant of a processes when you only have a pipeline for two channels. Yes the surrounds are lowered 3db, and mixed "out of phase" into the front L/R channels. The center is raised 3db, and mixed "in phase" into the fron L/R mains. All you have to do is do the reverse at the decoding stage, and you have your matrix surround mix. You can never acheive perfect seperation as you had before encoding, but the steering logic of the decoder can help with that.



You really should upgrade, because the future of movies and music is in surround, not stereo. Now I am sure that B&W has a high quality center and surround speakers to go with your 703, and there is a way to use a high quality two channel pre-amp along with a processor for 5.1. It takes some creative wiring, but I have heard of folks who are doing it. Surround is not a compromise, its a progression beyond the constraints of two channel stereo. And let me tell you, there are alot of constraints in mixing for two channel. When folks fully understand how much signal processing is used to get two channels to sound clear and good, they would not be so gung ho about it.

First, thanks for taking the time with us. I like this idea with "insider discussion"

I agree that 5.1 done right has a bigger potential than 2.0 if both systems are of the same caliber.Yes B&W certainly has options if I were to go 5.1, and when NAD introduces the lossless formats to their Master series, I'll surely be tempted

A comment on a topic discussed earlier:
I don't agree with people who say that 48kHz is all you need if you want the best possible sound.I personally would even hesitate to call it "High Resolution audio" like it says on the back of many Blu-rays.(even if it's 24 bit)
I feel that higher sampling frequency's gives you a better sense of realism, 3-D and detail.
"The movie people" are so eager to ditch the old and inferior 2.0 format but at the same time they stubbornly refuse to even try recording in 96 or 192kHz
Aren't 96kHz almost standard in music studios nowadays?

[Sir Terrence]

Originally Posted by DaViD Boulet

Agreed.

I'm a serious audiophile. The kind of guy who likes tube amps, LPs, and can hear the difference in jitter signatures between different transports and digital cables.

And I loved the 5.1 24/96 TrueHD track on the David Matthew's Band Blu-ray Disc. One of the most realistic audio reproductions ever to take place in my listening room.

Two-channel stereo is an artifact from two sides of a record groove. God never decreed that thou-shalt-listen-to-two-channel-stereo. 3 channel, 4 channel, five channel... it gets better and better the more channels you have to represent the authentic soundfield as long as the fidelity of each of those channels is maintained.

That's the audiophile's responsibility: put together a multi channel system that has proper timbre and fidelity for all channels.

And my 5.1 system still performs as a fantastic two-channel system when I listen to left-right stereo recordings (I don't apply surround DSP to my high-end 2-channel recordings), but it's also there for native 5.1 audio as well.

My REAL gripe is with 5.1 "music" mixes that don't use the center channel!?!? There seems to be a "better mix this phantom since the listener probably has a crappy center speaker" mentality that needs to stop...

Not using the center channel is an issue of the formerly two channel guys jumping over to mixing in multichannel. Alot of two channel guys just do not like the sound of the center channel, because they think it actually contracts the frontal soundstage(They are stuck in the phantom image mode). That is because rather then mixing things are unique to the center channel, these guys were pulling stuff from the L/R channels "inward", and that WILL make the center channel contract the overall frontal soundstage. Once they tried it and it didn't work right, they never use the channel again. This is a theme that has played out time and time again when you talk to two channels guys jumping into multichannel. Those of us who came from mixing film in 5.1, and jumped into 5.1 music have no problem with the center channel, but we are the minority out there unfortunately.

David, I agree with you about having a good HT system, that happens to be a very good music system as well. Two channel recordings played through my Left/Right mains has all of the same qualities as any dedicated high end two channel setup. I believe if you have a desire to create a very good music and movie system from the very start, it should be able to do both equally well.

[DaViD Boulet]

Sir Terrence,

thanks for the feedback about the center channel issue. I had never realized the context behind so many music-5.1 mixes not utiliizing the center properly. Glad you're on the side of authentic 5.1 mixing!

A comment on a topic discussed earlier:
I don't agree with people who say that 48kHz is all you need if you want the best possible sound.I personally would even hesitate to call it "High Resolution audio" like it says on the back of many Blu-rays.(even if it's 24 bit)
I feel that higher sampling frequency's gives you a better sense of realism, 3-D and detail.

Agreed. Switching to the 96 kHz soundtrack on the Dave Matthews BD produces an experience I've never had listening to 44.1/48 kHz signals... the speakers literally disappeared and the audio landscape just "appeared" as if it there wasn't really a hi-fi system in the room. Very, very impressive. I'd love to see 96 kHz offered more and more when available.

[Sir Terrence]

Originally Posted by Stereo24192

First, thanks for taking the time with us. I like this idea with "insider discussion"

No problem, I actually enjoy doing this here at bluray.com more than any other place I have setup a online home at.

I agree that 5.1 done right has a bigger potential than 2.0 if both systems are of the same caliber.Yes B&W certainly has options if I were to go 5.1, and when NAD introduces the lossless formats to their Master series, I'll surely be tempted

I would advise you to upgrade to multichannel when you can. As you can see, year after year high quality two channel music is getting harder and harder to find. At least I find that is the case for me. Either its hard to find, or expensive as hell.

A comment on a topic discussed earlier:
I don't agree with people who say that 48kHz is all you need if you want the best possible sound.I personally would even hesitate to call it "High Resolution audio" like it says on the back of many Blu-rays.(even if it's 24 bit)
I feel that higher sampling frequency's gives you a better sense of realism, 3-D and detail.
"The movie people" are so eager to ditch the old and inferior 2.0 format but at the same time they stubbornly refuse to even try recording in 96 or 192kHz
Aren't 96kHz almost standard in music studios nowadays?

Well, the best possible sound means different things to different people, that is for sure.

First I want to address this 192khz thing. It is unnecessary, wasteful, is not faithful to the original source, and has little to no benefit for the amount of space it takes to archive it, and the amount of processing to work with it.

I personally would like to see movie soundtracks mixed in 24/96khz, but I am afraid that the amount of space it would take to archive an entire movie for storage would strain the servers at any studio. Secondly you have to balance that with the fact that most sound effects are archived at either 16/48khz or 24/48khz, and they would not benefit much from an upconversion to 96khz, because lets face it, once you record and archive something at 48khz, you have already cut your frequency limit at 24khz. You will get nothing but air from 24khz(48khz sample rate) to 48khz(96khz sample rate). That is a waste of disc space IMO. Now if you start from scratch recording dialog, music, and use freshly recorded effects at 96khz, then maybe you have something there.

Next you have to decide if anyone would benefit from the higher sample rate. And that invites the question, "how many folks have speakers with a response to 48khz?". When you answer that question "not many"(which is true), then 96khz sampling rate as a playback rate is a waste. Who wants a bunch of frequencies that get lost in your tweeters?

24/96khz is a great bit and sample rate for recording music, as this insures that you capture every note and harmonic intact. But as a playback rate, it is a waste of bandwidth, and this opinion is supported by Bob Stuart of Meridian digital audio. Now there is a flip side to this. I believe that less processing is more, so if I recorded in 24/96khz, I would want that to be reproduced in 24/96khz as it does not require any downsampling. There is nothing wrong with downsampling from 96khz to 48khz, this can be done without any digital side effects. I just do not like extra steps within the workflow personally.

If we can hear more detail in a 96khz signal, where would that extra audible detail be located? It wouldn't be at frequencies over 20khz, because we cannot hear anything that high. That detail would be within the range of our hearing which is below 20khz which suggest that 44.1khz is fine. However we have to filter out any signals above 20khz, and that requires very steep filters that have audible side effects below 20khz. So its not that 24/96khz reveals more detail, it is 24/96khz does not have brickwall filters to contend with. A signal using a 44.1khz sample rate, and oversampled 4 times would also avoid the brickwall filters audible degredation. Most dedicated CD players oversampled either 4 times, or 8 times. Now that we have gone to using DVD or bluray players that do not oversample, then a push to higher sample rates for playback becomes necessary if only to eleminate the need for brickwall filters.

I would suggest to you that alot of what you hear from 24/96khz does not come from the sample rate directly, but from a well mixed, minimal processed recording that is void of the effects of any anti-aliasing filters. I say this because there are plenty of 24/44.1khz recording that have loads of detail, sound 3D, and have that you are there realism. Sample rates deal with bandwidth, not detail. Bit depth has more influence on the amount of detail we hear.

Man am I out of breath!

[Stereo24192]

We can't hear a sine wave over 15-20kHz, you're right about that. But didn't they do some test that showed changes in brain activity when they cut off/switched on the above 20kHz content.

I've also read that we, to detect direction of sound can perceive time differences of 10 microseconds. (wouldn't you need 200kHz sampling for that?)
http://en.wikipedia.org/wiki/Interaural_time_difference

From what I've read/learned there are different opinions among the experts about the importance of high sample rates.Both in terms of theory/technology and practical listening experiences.

[DaViD Boulet]

If we can hear more detail in a 96khz signal, where would that extra audible detail be located? It wouldn't be at frequencies over 20khz, because we cannot hear anything that high. That detail would be within the range of our hearing which is below 20khz which suggest that 44.1khz is fine. However we have to filter out any signals above 20khz, and that requires very steep filters that have audible side effects below 20khz. So its not that 24/96khz reveals more detail, it is 24/96khz does not have brickwall filters to contend with. A signal using a 44.1khz sample rate, and oversampled 4 times would also avoid the brickwall filters audible degredation. Most dedicated CD players oversampled either 4 times, or 8 times. Now that we have gone to using DVD or bluray players that do not oversample, then a push to higher sample rates for playback becomes necessary if only to eleminate the need for brickwall filters.

I would suggest to you that alot of what you hear from 24/96khz does not come from the sample rate directly, but from a well mixed, minimal processed recording that is void of the effects of any anti-aliasing filters. I say this because there are plenty of 24/44.1khz recording that have loads of detail, sound 3D, and have that you are there realism. Sample rates deal with bandwidth, not detail. Bit depth has more influence on the amount of detail we hear.

Fantastic post.

I would never argue that humans need to hear above 20K or that stereo systems need to produce 48 kHz response in order for 96 kHz recordings to sound better.

I just know that they *do* sound better... in general... for whatever reason.

And I think that the possibilities you suggest are probably the biggest reasons why... moving the brick-wall filter so far away from audible frequencies that it has very little signature introduced into the recording where the ear is sensitive (and that's true at both ends... a/d and d/a).

I just know it sounds better... so bring on the 96 kHz!!!

[Clark Kent]

Don't you think any serious audiophile should have completely separate 2 channel and 5.1/7.1 systems(preferably different rooms)? Jack of all trades, master of none I say.

[Sir Terrence]

Originally Posted by Stereo24192

We can't hear a sine wave over 15-20kHz, you're right about that. But didn't they do some test that showed changes in brain activity when they cut off/switched on the above 20kHz content.

It was Oohashi paper, and there have been papers submitted to AES that dispute his finding as well. So at this point this issue is still unresolved and inconclusive.

I've also read that we, to detect direction of sound can perceive time differences of 10 microseconds. (wouldn't you need 200kHz sampling for that?)
http://en.wikipedia.org/wiki/Interaural_time_difference

If you carefully look at how the experiment was conducted(a single source emitting a signal) you can plainly see that we do not listen that way. We listen binaurally. My question to you is can you determine location and direction with a 48khz sample rate?

This effect is called a HRTF, or head related tranfer function. Our ears and brain are very effective at dealing with this, as it has been around since the beginning of man. We do not need to increase the sample rate to deal with this at all. Our ear/brain uses time and intensity to determine direction. And it is very effective at doing that.

From what I've read/learned there are different opinions among the experts about the importance of high sample rates.Both in terms of theory/technology and practical listening experiences.

And there will be forever I am afraid.

[Sir Terrence]

Originally Posted by Clark Kent

Don't you think any serious audiophile should have completely separate 2 channel and 5.1/7.1 systems(preferably different rooms)? Jack of all trades, master of none I say.

Absolutely not. That's old school, and things have progressed far beyond this limited type of thinking. If you set out to put together a system that can do both well, there is no reason to seperate the two systems.

The very idea that a high quality well designed hometheater system cannot do music well smacks of pure snobbery.

[Sir Terrence]

Originally Posted by Woody

Are you from Cleveland, or is the cavaliers avatar for something else? I don't recognize the logo if it actually a Cleveland Cavaliers logo.

Just curious.

Actually this Cavaliers just happens to be a Drum and Bugle corps from Rosemont Illinois. Just to give you an idea of what Drum Corps is, visit this site www.dci.org.

[dialog_gvf]

Originally Posted by DaViD Boulet

I would never argue that humans need to hear above 20K or that stereo systems need to produce 48 kHz response in order for 96 kHz recordings to sound better.

I just know that they *do* sound better... in general... for whatever reason.

I don't believe Nyquist/Shannon ever talked about quality. At the exact 1/2 sample rate all you can have are the extremes of the amplitude represented. Does that produce a QUALITY result?

Let's talk about a frequency many people can hear, say 12Khz. What does having eight samples per cycle (96Khz), versus four samples per cycle (48Khz) do to the quality of the cycle?

A car that drives well at 200mph will probably drive like a dream at 100mph. That's why people buy the ones that can do what they can't legally try. The performance at the speeds they can do is better.

Gary

[Simplayer]

If you only need two points to completely define the wave, taking any number more won't define it "more completely".

[Sir Terrence]

Originally Posted by Simplayer

If you only need two points to completely define the wave, taking any number more won't define it "more completely".

Bingo!

[dialog_gvf]

But, it isn't ONE wave, it's a mix of N different simultaneous sine waves.

Consider this:

1 (48Khz): 10000 0 10000 0 10000 0 10000 0 ... (pure 24Khz tone)

2: (98 Khz): 10000 1000 0 2000 10000 3000 0 4000 10000 5000 0 6000 10000 7000 0 ...

Contrived, of course, but it illustrates that information can be lost.

Gary

[Woody]

Originally Posted by Sir Terrence

Actually this Cavaliers just happens to be a Drum and Bugle corps from Rosemont Illinois. Just to give you an idea of what Drum Corps is, visit this site www.dci.org.

Cool

Being as knowledgeable about audio as you are, I can't help but notice that you use horn loaded speakers for HT and music.

What is it in particular that endears you to horns?

I personally have never been too keen on horns for music.

Even listening to JBL synthesis 1 systems (I sell them) with music is just not as accurate as I would like.

I must admit that I have no idea what your custom speakers consist of though.