An Interesting Experiment

[Overture]

Ahoy there,

After being asked if two encoded sources would decode properly at the same time if mixed together (think crossfading or using dual cdj or vinyl decks through a dj mixer and into the surround master) I thought I'd do a test. I took two tracks that have very stark surround sound introductions - The Lady Don't Mind (extended edition) that David Byrne remixed himself in '99 and no-one listened to, but it has a lot of surround content, and our own encoded version of the Alan Parson's quad mix of Money from DSOTM.

I've attached a zip file with a short clip of each track separately for you to decode for fun, as well as the mixed version to test out. (less than 30 seconds of each, under fair use)

It's startling how well we're decoding two entirely different tracks and they're remaining stable, coherent and also just by coincidence kind of fit well together.

Anyway, something interesting for your mid to late week.

~Captain Dave (Mrs)

 

[Quadwreck]

I used to do that with a QS decoder and SQ decoder, directing one to the front channels and the other to the rear channels. Not great results, but interesting with some sources. Sgt Pepper was one.

 

[ummagumma]

That is pretty wild!

Out of curiosity: what happens if you feed two completely unrelated signals into the L/R inputs? One signal per input

 

[kfbkfb]

I think both matrix encoded signals would have to use the same matrix - both QS or both SQ - although SQ decoding of QS or QS decoding of SQ would still provide a surround sound effect, before, during and after the crossfade.


Kirk Bayne

 

[Overture]

I think both matrix encoded signals would have to use the same matrix - both QS or both SQ - although SQ decoding of QS or QS decoding of SQ would still provide a surround sound effect, before, during and after the crossfade.

Kirk Bayne

I suspect, well if you've ever listened to QS through an SQ decoder, the fronts are extremely similar but the rears become a dog's breakfast. You'd get good QS decoding and a weird nebulous SQ rear. Maybe.

That is pretty wild!

Out of curiosity: what happens if you feed two completely unrelated signals into the L/R inputs? One signal per input

That's a fascinating idea. I guess you'd get a front to rear "audio graph" of sorts where any common material directly in phase or out of phase was spread across the sound stage. I might try it with some EDM, since they have so many complex phase relationships, there are bound to be some interactions.

 

[kfbkfb]

I suspect, well if you've ever listened to QS through an SQ decoder, the fronts are extremely similar but the rears become a dog's breakfast.

Maybe recommend the QS decoding setting only for DJing/crossfading - that way nearly any input (QS, stereo [even electronically rechanneled stereo], SQ, UHJ and other matrix encoded content) would have some sort of surround sound output, in a large listening room it might sound OK even if it's the wrong decoder for the content.


Kirk Bayne

 

[par4ken]

A related idea is to listen to two or more decoders decoding that same material simultaneously. My homebuilt system uses a mixer input stage instead of switching the quad inputs, so the odd time I've unknowingly had more than one decoder running at a time! It didn't even sound that bad!

 

[barfle]

Out of curiosity: what happens if you feed two completely unrelated signals into the L/R inputs? One signal per input

I have a few stereo records of “wind in the trees” recordings that are primarily pink noise. Through a Tate, quite a bit of similar sounds come out if the back speakers. I also have a test record with a 300 Hz sine wave on one channel and 303 Hz on the other. As you might expect, there’s a major swirl effect from that!

 

[MidiMagic]

Actually an SQ decoder with the 10/40 blend WILL decode a QS record and place all the sounds in approximately the correct directions. It also works with Dolby Surround.