ALL TESTS ON THE SURROUND MASTER

QuadraphonicQuad Home Audio Forum

Help Support QuadraphonicQuad Home Audio Forum:

chucky3042

2K Club - QQ Super Nova
Joined
Apr 9, 2012
Messages
2,309
Dear All

Good ol OxFORD Dickie (bless his cotton socks) has inspired me to create this place to post all known tests on the Surround Master.

For shits n giggles I post OD's (AKA John Ford......"the Matrix") scientific report on the SM - Apparently its a pile of useless poo and not able to decode QS or QS . Wish my brain was cleverer and I could actually understand his test criterion, methodology or anything......folks are dumb where I come from.

 

chucky3042

2K Club - QQ Super Nova
Joined
Apr 9, 2012
Messages
2,309
Here is the earliest internal review we did by Dave the Bitch in 2009:

Appraisal of listening test of Involve audio.

Firstly, when tested with surround material, the decoder not only pulled all the details out and put them where they should be around the room, but I have never heard it at this level of detail and clarity before. Specifically the surround gunshot track; the track made perfect sense to me for the first time, and more detail and precision was brought out than I’ve heard before.

Likewise listening to material recorded in stereo, the surround image was consistent and natural, the frontal image was not compromised in any way, there was no degradation of the frequency response and no distortion was audible. The quality and detail of the audio that was being extracted to the rear was startling, and I was hearing new things yet again in recordings that I thought I’d all but exhausted in terms of new audio discoveries.

Testing audio that was run through the Involve encoder, everything that went in to one of the four channels into the stereo mix came out of the same channel when decoded from the 2-channel back into four. The numbers show crosstalk attenuation of 30-40db from any channel to any other channel, and the effect of this is quite apparent.

Additionally, just listening to the two-channel down-mix output, the 2 channel signal sounded like pure, precise and pleasant stereo, with perhaps a widening effect. It was in no way detrimental, and could definitely be used as a professional recording format.

Using the Dark Side of the Moon recordings for testing, the 2 channel down-mix decoded to a reasonable surround version of the recording. The 4 channel discreet reference mix shows no difference between listening in discreet, or Involve encode to Involve decode.

This is a remarkable result.

In all of the tests, whenever audio passed from the front to the back, or left to the right it moved smoothly and precisely; there was no sense of shifting or instant change when drifting from the front to the back.

Something else I noticed through the whole listening test is that it didn’t sound unnatural in any way. I wasn’t entertaining the notion of leaving, I didn’t feel uncomfortable, I just enjoyed the natural extension of stereo, and I now have a preference towards it.
There is no breathing, pumping, or drift, and every instrument and vocal is precisely positioned and whole rather than smeared between channels.

The most startling of all was when I finally realized that I had seated myself off-center without thinking about it, and didn’t even realize it. That’s right, the entire time I was listening to total perspective as well and I never once actually noticed. This suits me just fine, the surround image sounded perfect. This was ideal, considering Total Perspective was not intended to be a subject of the listening test.


Appraisal by David Alexandrou – conducted Friday 29/05/09 on digital implementation of Involve Audio.
 

chucky3042

2K Club - QQ Super Nova
Joined
Apr 9, 2012
Messages
2,309
Involve vs Discreet

Room setting.



Author: David Alexandrou

Internal use only.​






Quick intro:​



Involve audio is a 4/5 to 2 encode/decode matrix that can also decode surround from existing 2 channel sources.



System comparison:​



Discreet audio vs. Involve Audio

Set-up:​


Clip used: Money – Dark Side of the Moon 4 channel DVD-audio

Speakers: RTA Electrostatic Total Perspective speakers

Presentation: Audio only

Speaker width to listener ratio 1:1

Note: Central Imaging is traditionally considered difficult in this configuration.

Test Subjects: 11


Format:​

Subject was played the audio, switching between the discreet and Involve versions of the audio. The systems were only identified as either No. 1 or No. 2.


The subject was then asked to comment on various qualities of the sound, paying attention to

Any discernable differences between the two systems

Audio Quality

Surround quality.


The subject was then asked if they had a preference to any of the systems in particular, and asked to give reasons for their answer.






Preferences

Subjects were asked if they had a preference for either system. The results were as follows:

Involve: 5

No Preference: 4

Discreet: 2


Observable qualitative results:

In each case where Involve was picked as the preference, the results point to either an increase in the sound richness or fullness, and better distinction of surround sound elements.

The subjects who preferred discreet had a different reason to each other.


Conclusions

The test data as it stands indicates strongly that Involve matrix decoding is as good as or better than discreet surround sound.
 

chucky3042

2K Club - QQ Super Nova
Joined
Apr 9, 2012
Messages
2,309
Here is an early 2009 performance test.......oh we do detail test conditions etc!!
 

Attachments

  • DSP implementation test results V2-converted.pdf
    151.7 KB · Views: 25

chucky3042

2K Club - QQ Super Nova
Joined
Apr 9, 2012
Messages
2,309
1652533953990.png
 

chucky3042

2K Club - QQ Super Nova
Joined
Apr 9, 2012
Messages
2,309
Here is a test report done by little Max our physicist / Mechanical engineer. Really tough test of simultaneous 4 tones vs the QSD1
 

Attachments

  • Surround Master Involve Proper.pdf
    612 KB · Views: 26

chucky3042

2K Club - QQ Super Nova
Joined
Apr 9, 2012
Messages
2,309
And here is my personal favorite done again by little Max (soon to be Mexico Max - setting up our office there) , the really interesting part is at the end on the X shaped linearity tests simulating cross imaging.,

Here are the 4 killer graphs that I think contradict OD's waffle:

1652535045178.png


1652535083136.png


1652535127730.png


1652535172586.png


All detail in attached report. Please note our only instruction to little Max was "to find a problem". He tried to dig!
 

Attachments

  • QS tests.pdf
    842.7 KB · Views: 19

ar surround

2K Club - QQ Super Nova
QQ Supporter
Joined
Apr 3, 2010
Messages
3,455
Location
New Joisey
You forgot to post 'The Mother Of All Tests':

 

gene_stl

800 Club - QQ All-Star
QQ Supporter
Joined
Nov 3, 2019
Messages
824
Location
St.Louis
This is the best test. You have a forum full of committed , knowledgeable quadies who love it. There is a thread called "Stuff that sounds amazing with the Surround Master" ! End of story as far as I am concerned.

For a lot of electronics where you are listening for frequency response, linearity, HD, IMD and noise level, the ear is a terrible, terrible, almost useless, measurement instrument.

But for multi channel it is almost the ONLY measurement instrument.
 

Scott65

1K Club - QQ Shooting Star
Joined
Nov 2, 2015
Messages
1,162
Location
Tasmania, Australia
You forgot to post 'The Mother Of All Tests':

The real world test (QQ) is the best. I'm also into photography and I get sick of seeing posts in photography forums about which cameras are best (especially when referenced to photos of test charts). I much prefer getting out there and taking photographs in the real world.
 

kfbkfb

1K Club - QQ Shooting Star
QQ Supporter
Since 2002/2003
Joined
Feb 20, 2003
Messages
1,296
Location
Midwest USA
The SM was designed from the ground up to be a variable matrix QS and SQ matrix decoder, IIRC, script based
"logic" assisted decoders use an option in the script process that reduces volume of content common to L and R, that process may or may not be optimum for mimicking variable (QS/SQ) matrix decoding.


Kirk Bayne
 

chucky3042

2K Club - QQ Super Nova
Joined
Apr 9, 2012
Messages
2,309
The SM was designed from the ground up to be a variable matrix QS and SQ matrix decoder, IIRC, script based
"logic" assisted decoders use an option in the script process that reduces volume of content common to L and R, that process may or may not be optimum for mimicking variable (QS/SQ) matrix decoding.


Kirk Bayne
And prone to pumping
 

chucky3042

2K Club - QQ Super Nova
Joined
Apr 9, 2012
Messages
2,309
The SM was designed from the ground up to be a variable matrix QS and SQ matrix decoder, IIRC, script based
"logic" assisted decoders use an option in the script process that reduces volume of content common to L and R, that process may or may not be optimum for mimicking variable (QS/SQ) matrix decoding.


Kirk Bayne
The variomatrix approach is very much like the opposite of a mixer....... A de-mixer. Or it can be thought of as a see saw, when one is up, the other is down but the average level remains constant.

This inherently reduces pumping
 
Last edited:

chucky3042

2K Club - QQ Super Nova
Joined
Apr 9, 2012
Messages
2,309
Impressive stats.
Stats are good, QQ ears even better.
I confess I don't read every post on QQ but all I've read about the SM sounds very good.
I need a cash infusion!
If you are tight on $$ at the moment, try our development board at $150. It wont do SQ but does the rest.....without a box and knobs
 
Top