• QuadraphonicQuad welcomes you and encourages your participation! Treat all members with respect. Please keep all discussions civil, even when you have a strong opinion on a particular topic.

    Do not offer for free, offer for sale, offer for trade, or request copies or files of copyrighted material - no matter how rare or unavailable to the public they might be. We do not condone the illegal sharing of music. There are many places on the internet where you can participate in such transactions, but QuadraphonicQuad is not one of them. We are here to encourage and support new multichannel releases from those companies that still provide them and as such the distribution of illegal copies of recordings is counter-productive to that effort. Any posts of this sort will be deleted without notification.

    Please try to avoid discussions that pit one format against another. Hint for new users: make liberal use of the search facilities here at QuadraphonicQuad. Our message base is an incredibly rich resource of detailed information on virtually all topics pertaining to surround-sound. You will be surprised at what you can find with a little digging!

INVOLVE ENCODER SAMPLER

Bob Romano

Administrator
Staff member
Since 2002/2003
Joined
Apr 26, 2002
Messages
4,745
Reaction score
454
Points
1,536
Location
Viva Las Vegas
I was lucky enough to receive an Involve Encoder to test out and I now have a sampler available for download. There are 10 tracks:

01 AIX Channel Check
02 Steely Dan - Bodhisattva (Q8 4.1)
03 Deep Purple - Highway Star (SACD 4.1)
04 Spinners - I'll Be Around (QR 4.1)
05 Santana - Black Magic Woman (QR 4.1)
06 Beatles - Golden Slumbers Medley (DVDA 5.1)
07 Britney Spears - Toxic (5.1 DVDA)
08 Frank Zappa - Zeets (DVDA 5.1)
09 Billy Joel - Movin' Out (SACD 5.1)
10 Emerson, Lake & Palmer - Lucky Man (Steven Wilson version) (DVDA 5.1)

There are 2 separate sampler with the same tracks - one in Involve and one in QS - also encoded using the Involve Encoder. All sources were from discrete sources (even Golden Slumbers) and encoded to stereo. As with any matrix type of recording, it will never be 100% discrete however the results (IMO) are quite impressive. The really remarkable thing is that the mixdown of 4.1 or 5.1 sound like proper stereo with no artifacts. Burn and play in Involve 5.1 or QS 4.1.

I do not want to list the links here but if you PM me I will hit you back with the links. The zip files contain a wav and cue file that can easily be burned to a cd using ImgBurn.
 
Last edited:

LizardKing

2K Club - QQ Super Nova
Joined
Oct 13, 2009
Messages
3,992
Reaction score
102
Points
461
Location
NZ
All I can say is ...... I'm impressed... I love the Eclectic Mix too....:D

Highway star and Black Magic Woman in particular were good....actually they're all good...

I need to do comparisons with the discrete versions...
 

Bob Romano

Administrator
Staff member
Since 2002/2003
Joined
Apr 26, 2002
Messages
4,745
Reaction score
454
Points
1,536
Location
Viva Las Vegas
I think my favorite tracks are Toxic and Golden Slumbers
 
Last edited:

Disclord

900 Club - QQ All-Star
Joined
May 19, 2005
Messages
945
Reaction score
18
Points
0
Location
Plattsburg, MO (just outside Kansas City)
Your tracks sound amazing through the SM - if I hadn't known better I would have though I was listening to a discrete recording with lots of ambience and excellent localization. QS seemed to have a tad more separation than the involve but it was so minor that it didn't matter. And the Involve encoding played in stereo sounded totally normal, unlike QS. It sounded basically the same as SQ stereo playback.
 

Owen Smith

Senior Member
Joined
Apr 6, 2012
Messages
251
Reaction score
2
Points
261
Location
UK
I'm confused. It's the 21st Century, we have many multi channel lossless formats: DVD-Audio, SACD, Blu Ray (various codecs), FLAC and others. So why would we want to matrix encode discrete sources into stereo and then decode it again? It can't be for mass market acceptance, because they'll MP3 encode it and destroy any chance of matrix decoding it. Can someone explain?
 

Disclord

900 Club - QQ All-Star
Joined
May 19, 2005
Messages
945
Reaction score
18
Points
0
Location
Plattsburg, MO (just outside Kansas City)
MP3 can carry phase information - it's only the intensity coding mode that can't and major improvements have been made to MP3 in the last few years, allowing correct phase encoding at lower bitrates. An M:2:N codec like Involve allows 2 channel sources, either analog or digital, to carry multiple channels with performance that is almost indistinguishable from true discrete multichannel - and in the case of lossy codecs, it can carry multi channels in the same bitrate that a discrete signal would occupy, allowing more bits to be allocated to the two channels and thus higher audio quality with basically no penalty in sound quality in stereo or multichannel. Radio stations can broadcast 5.1 surround recordings and they will sound like normal stereo to listeners without a decoder. All the other modern matrix systems have major drawbacks in stereo compatibility and in decoding to 5.1 - none sound discrete and several, like Neural and Circle Surround, have real problems in both encoding, stereo compatibility and decoding artifacts since they are simple gain riding decoders.

there is room for both matrix formats such as Involve and discrete. One does not obsolete the other.
 

bmoura

Super Moderator
Staff member
Since 2002/2003
Joined
Mar 2, 2003
Messages
9,081
Reaction score
2,152
Points
1,396
Location
Redwood City, CA
I'm confused. It's the 21st Century, we have many multi channel lossless formats: DVD-Audio, SACD, Blu Ray (various codecs), FLAC and others. So why would we want to matrix encode discrete sources into stereo and then decode it again? It can't be for mass market acceptance, because they'll MP3 encode it and destroy any chance of matrix decoding it. Can someone explain?
I would look at this as a "proof of concept" in terms of how well the Involve encoder and decoder work. But I'm not expecting fully discrete playback.
 

chucky3042

1K Club - QQ Shooting Star
Joined
Apr 9, 2012
Messages
1,267
Reaction score
2,019
Points
356
Hello Owen

That question is really at the core of my philosophy in regards to surround sound. Disclord has already made some very valid points, but in a previous thread

https://www.quadraphonicquad.com/forums/showthread.php?16628-New-Vinyl-Based-Surround-Quad-System-(Hypothetical-Question)

I stated that in any new universal sound media (referring to records in the thread) the system should address the following issues:


1 be able to be played by all existing stereo equipment with NO audible deterioration or damage to record or stylus
2 the encode must sound identical to the stereo
3 the decode must sound identical (or better than) discrete
4 Must have no pumping, clicking, breathing, whistling sonic artifacts.
5 Be able to synthesise with a high degree of accuracy a surround from non encoded material
6 Preferably get rid of the image skewing center channel (and the wife hates it)
7 Provide a consistent image in all seats of the house with no sweet spot.
8 Be able to be transmitted in low bandwidth media such as AM radio, mp3
9 Must become the new UNIVERSAL music/ cinema recording format to eliminate market confusion and customer isolation (I still don’t have a Blueray player!)
10 Be able to produce 2, 4 and even 5.1 (I hate it) surround.


Central to my thoughts are issues of backwards compatibility and ease of use - so the user does not have to worry about some unfriendly menu system (issues of what surround format), is this disc compatible (eg the existing stupidity of Bluerays and HD DVD not being compatible with each other OR older DVD players). Remember lack of cross compatibility is one of the things that killed surround back in the 70's. Today there is only two (I claim) backwards/ forwards compatible universal systems that can play from everything from an old record player, cassette, mp3, TV, DVD, AM/ FM radio they are STEREO and INVOLVE encode (yes I am biased!!!!!!!).

In short my aim is to have recorded media encoded in Involve so if the user plays it in mono or conventional stereo they will be unaware of any deterioration in sound quality. The bonus is that when they play it back through an Involve decoder they will get a surround sound that we claim is indistinguishable from discrete for most listeners. Please note that in our own internal audience surveys we found a slight preference of Involve encode/ decode to discrete......see attached. - I can tell you why later if you wish!

Our Involve basic encoder is now available (assuming our commercial people ever get around to putting it up on our ordering page) for $695.

Regards

Chucky


I'm confused. It's the 21st Century, we have many multi channel lossless formats: DVD-Audio, SACD, Blu Ray (various codecs), FLAC and others. So why would we want to matrix encode discrete sources into stereo and then decode it again? It can't be for mass market acceptance, because they'll MP3 encode it and destroy any chance of matrix decoding it. Can someone explain?
 

Attachments

bmoura

Super Moderator
Staff member
Since 2002/2003
Joined
Mar 2, 2003
Messages
9,081
Reaction score
2,152
Points
1,396
Location
Redwood City, CA
Very impressive results in general. But I did notice that on the Channel ID Check, the QS Encode is very accurate while the Involve Encode results in the Left and Right Front channel IDs bleeding into Left and Right Center. Is that an encoder issue?
 

chucky3042

1K Club - QQ Shooting Star
Joined
Apr 9, 2012
Messages
1,267
Reaction score
2,019
Points
356
Hi Bmoura

Its an encoder issue as Involve encode works for actual music content - in static point tests it is inferior to QS. Please find attached a table comparing Intelligent Involve to QS and you will see on STATIC tone tests a 12 dB separation left front/ left back and right front/ right back when using Intelligent involve. Compared to over 35 dB in QS.

Sound pretty bad eh? Well this is where we get controversial in that we conducted other tests on panels of test "monkeys" and we injected crosstalk until they could detect a deterioration. Turned out the it was 12 db not the "essential" 100dB as our current digital brigade would claim or even the old golden rule of 20 dB. You guys need to look at the Haas precedence effect curve for more clues!

Where Intelligent Involve gets clever is that it changes the encode matrix parameters (IN 3 BANDS) to optomise for either the stereo down mix or the surround output dependent on the nature of the music. When you get complicated surround passages where all corners are working the Intelligent Involve matrix becomes QS again giving you the equivalent to the best QS encode decode. Remember we are doing this to keep the stereo down mix SOUNDING the same as stereo.

Enjoy the reading - I will ask questions later.

Regards

Chucky

Very impressive results in general. But I did notice that on the Channel ID Check, the QS Encode is very accurate while the Involve Encode results in the Left and Right Front channel IDs bleeding into Left and Right Center. Is that an encoder issue?
 

Attachments

Disclord

900 Club - QQ All-Star
Joined
May 19, 2005
Messages
945
Reaction score
18
Points
0
Location
Plattsburg, MO (just outside Kansas City)
Hi Bmoura

Its an encoder issue as Involve encode works for actual music content - in static point tests it is inferior to QS. Please find attached a table comparing Intelligent Involve to QS and you will see on STATIC tone tests a 12 dB separation left front/ left back and right front/ right back when using Intelligent involve. Compared to over 35 dB in QS.

Sound pretty bad eh? Well this is where we get controversial in that we conducted other tests on panels of test "monkeys" and we injected crosstalk until they could detect a deterioration. Turned out the it was 12 db not the "essential" 100dB as our current digital brigade would claim or even the old golden rule of 20 dB. You guys need to look at the Haas precedence effect curve for more clues!

Where Intelligent Involve gets clever is that it changes the encode matrix parameters (IN 3 BANDS) to optomise for either the stereo down mix or the surround output dependent on the nature of the music. When you get complicated surround passages where all corners are working the Intelligent Involve matrix becomes QS again giving you the equivalent to the best QS encode decode. Remember we are doing this to keep the stereo down mix SOUNDING the same as stereo.

Enjoy the reading - I will ask questions later.

Regards

Chucky
Charlie,

do you have any test results with involve or QS in the 5.1 mode, I.e center channel to L/R Front, CB, etc...? If so, could you post them? I know you hate the center channel, but I'm interested in the results - I'll let you know why in one of my typically overly verbose emails. And stop cringing right now at the thought of another email from me!

Cheers,
Ty
 

bmoura

Super Moderator
Staff member
Since 2002/2003
Joined
Mar 2, 2003
Messages
9,081
Reaction score
2,152
Points
1,396
Location
Redwood City, CA
Chucky, thanks for the response. I was wondering if the results were intended by design. Apparently so.
 

chucky3042

1K Club - QQ Shooting Star
Joined
Apr 9, 2012
Messages
1,267
Reaction score
2,019
Points
356
Hi Disclord

Hope you are feeling better.

We really had a difficult time on the center channel separation with initially unsatisfactory results and some feedback from Bob Romano at the time. After several weeks of work we finally obtained good results SONICALLY with a lack of pumping and popping. The absolute separation is quite low (from memory) but we have not produced a formal document on the test results yet, we really went by "what it sounds like".

I think (yes its my biased opinion again) people tend to forget the initial intended function of the center channel was to ensure audience members sitting off center get a stable image of central vocal material. The older simpler method of just adding L +R fails because it tends to narrow the overall image - so there is a need to separate the pure L and R components of this signal. Only problem is that it is mathematically impossible to extract 3 signals from the original 2. As a result we need to use tricks such as dominance direction clues and psychoacoustic stuff to "cobble together" an approximation of the center - we are no different.

It is really questionable if we really need great separation of the center as it is a somewhat faulty concept (now there is fighting words). Hopefully you fellas will be able to hear our Total Perspective system one day.

Regards

Chucky

Charlie,

do you have any test results with involve or QS in the 5.1 mode, I.e center channel to L/R Front, CB, etc...? If so, could you post them? I know you hate the center channel, but I'm interested in the results - I'll let you know why in one of my typically overly verbose emails. And stop cringing right now at the thought of another email from me!

Cheers,
Ty
 

atrocity

1K Club - QQ Shooting Star
Since 2002/2003
Joined
Jun 14, 2002
Messages
1,367
Reaction score
849
Points
1,566
Location
Sacramento, California
MP3 can carry phase information - it's only the intensity coding mode that can't and major improvements have been made to MP3 in the last few years, allowing correct phase encoding at lower bitrates.
Which reminds me: Rhapsody subscribers may be entertained if they play the "Chase" album. As expected, it's SQ encoded. Far less expected, the embedded artwork is the old quad cover with the can't-be-missed border!
 
Top