A Matrix Decoder curiosity item on Japanese Auction site

QuadraphonicQuad

Help Support QuadraphonicQuad:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

bigbillquad

1K Club - QQ Shooting Star
QQ Supporter
Joined
Dec 6, 2010
Messages
1,579
Location
New Zealand . Hobsonville Rd. West Harbour
A nice looking Decoder no real info on it …...link to site
I wonder if this was a kit set project

https://www.jauce.com/auction/d360381157
39919
39920
39921
 
Last edited:
Seems to be a Diatone (never heard of them myself but apparently a Mitsubishi brand) DA-Q100 "4 Ch Decoder - Separation Enhancer". Given the crude Dymo labels, the two model numbers (Q-2000 & DA_Q100), the lack of a manufactures name and absence the of a serial no. / mains voltage rating plate the one offered for sale in BBQ's link looks very much like a prototype. The production ones looked like this:

39924


39926


That strange function gets variously labelled QM-4 or MQ-4, one of those many obscure Japanese formats limited to some small record companies there.

More info here, including some amusing stuff about their logic circuit:
https://audio-database.com/MITSUBISHI-DIATONE/etc/da-q100-e.html

They appear to abound on Japanese auction sites but not sure they ever made it to a worldwide market back in the day. There was also this variant (the DA-Q200) which had a rear power amp in from the looks of it:

39925
 
Last edited:
Seems to be a Diatone (never heard of them myself but apparently a Mitsubishi brand) DA-Q100 "4 Ch Decoder - Separation Enhancer". Given the crude Dymo labels, the two model numbers (Q-2000 & DA_Q100), the lack of a manufactures name and absence the of a serial no. / mains voltage rating plate the one offered for sale in BBQ's link looks very much like a prototype. The production ones looked like this:

View attachment 39924

View attachment 39926

That strange function gets variously labelled QM-4 or MQ-4, one of those many obscure Japanese formats limited to some small record companies there.

More info here, including some amusing stuff about their logic circuit:
https://audio-database.com/MITSUBISHI-DIATONE/etc/da-q100-e.html

They appear to abound on Japanese auction sites but not sure they ever made it to a worldwide market back in the day. There was also this variant (the DA-Q200) which had a rear power amp in from the looks of it:

View attachment 39925

Great research that's very interesting. It's easy to forget that manufacturer's use different names for different markets such as Trio/Kenwood Akai/Roberts, etc.
 
First thing that came to mind is that it would be way cool to adapt a Surround Master V2 to this enclosure.

HAHA we think alike! When the SMv2 gets here I will certainly use it as is. But I really want more than 1 input, a L/R balance control and a phase or panorama control. That could be an external box. But I would also like to have a 5.1 discrete input and another set of buffered outputs to my Panny quadscope. That would require pulling the SM v2 board out & reinstalling in another enclosure. That Diatone has just the right amount of knobs & switched for that.

Black panel faces are much easier to re-purpose than brushed aluminum. One of my more successful projects was gutting a Sansui QS-500 & reusing it as quad multi-source input control center. Doing the same thing with the earlier (but very very classy looking) QS-1 would have been impossible.
 
HAHA we think alike! When the SMv2 gets here I will certainly use it as is. But I really want more than 1 input, a L/R balance control and a phase or panorama control. That could be an external box. But I would also like to have a 5.1 discrete input and another set of buffered outputs to my Panny quadscope. That would require pulling the SM v2 board out & reinstalling in another enclosure. That Diatone has just the right amount of knobs & switched for that.

Black panel faces are much easier to re-purpose than brushed aluminum. One of my more successful projects was gutting a Sansui QS-500 & reusing it as quad multi-source input control center. Doing the same thing with the earlier (but very very classy looking) QS-1 would have been impossible.
Great minds think alike ;)

I'm sure there would be plenty of space inside to put/build/assemble whatever you want in the enclosure, say, a phono preamp and a proper preamp to begin with.
 
It is just me or... these Diatone units have a feeling of JVC?
 
Great minds think alike ;)

I'm sure there would be plenty of space inside to put/build/assemble whatever you want in the enclosure, say, a phono preamp and a proper preamp to begin with.

For those that might be interested in this the current cost is about $55 plus another 10 to ship to the JAUCE warehouse and then who knows how much to ship to your home. As I write 21 hours to go. Has anybody used this service? Any info? They also have a QSD-1 considerably more expensive.....

On the website they say:
CAUTION! The word "JUNK" included in this page refers to an item with unknown working condition or unknown origin and other details.
In other words sold for parts only. If re-purposing that doesn't matter so much. The meters light up! But UGH! Look at the dirty dirty RCA jacks!. Hard to replace since they are held in by rivets.
 
The Japanese like to use the word junk to absolve themselves of any possible claims. I tend to read that then look carefully to decide if I will bid or not. I prefer the Buyee service over Jauce or Goody Japan. If one uses Chrome they can translate the original Yahoo auction pages.
 
The Japanese like to use the word junk to absolve themselves of any possible claims. I tend to read that then look carefully to decide if I will bid or not. I prefer the Buyee service over Jauce or Goody Japan. If one uses Chrome they can translate the original Yahoo auction pages.

Thanks for the reply & info. I use CD Japan & Yes Asia for J/K music I can't get domestically. I have never done business with a proxy buyer or shipping service. My anime friends speak highly of Buyee also. I might have to look into that.

Ultimately I'll let this one pass buy as Jauce rules & regs seem pretty restrictive. First you open a Jauce acct & transfer $X from your verified PayPal acct to your Jauce acct. If you win good you get it but also stuck with numerous surcharges & fees that would surly exceed the bid on this decoder. If you lose the bid you don't get your money back it stays in your Jauce acct for the next time. So either was, win or not, your $$'s are gone.

Buried somewhere in a closet I have a patent assigned to Mitsubishi for a high seperation quad decoder. I remember it being very complex & interesting design. I might have to dig it out. Yes, I consider studying patents to be fun reading.
 
Ultimately I'll let this one pass buy as Jauce rules & regs seem pretty restrictive. First you open a Jauce acct & transfer $X from your verified PayPal acct to your Jauce acct. If you win good you get it but also stuck with numerous surcharges & fees that would surly exceed the bid on this decoder. If you lose the bid you don't get your money back it stays in your Jauce acct for the next time. So either was, win or not, your $$'s are gone.
I sympathise with your caution - the uncertainties and complexities of that process have always put me off attempting to get anything from them.
Buried somewhere in a closet I have a patent assigned to Mitsubishi for a high seperation quad decoder. I remember it being very complex & interesting design. I might have to dig it out. Yes, I consider studying patents to be fun reading.
Yes that would be interesting, hope you can find it. I wonder if it has anything to do with the supposed improved separation of this thing? The page linked to by https://audio-database.com/MITSUBISHI-DIATONE/etc/da-q100-e.html which rather boldly claims "has realized about the same separation rate as discrete", although the quoted specification gives a rather more modest -
Separation rate Between each channel: A maximum of 15dB
 
I sympathise with your caution - the uncertainties and complexities of that process have always put me off attempting to get anything from them.

Yes that would be interesting, hope you can find it. I wonder if it has anything to do with the supposed improved separation of this thing? The page linked to by https://audio-database.com/MITSUBISHI-DIATONE/etc/da-q100-e.html which rather boldly claims "has realized about the same separation rate as discrete", although the quoted specification gives a rather more modest -
Separation rateBetween each channel: A maximum of 15dB

I didn't think that on line patents would go that far back but it is: Mitsubishi Quatenary Decoder

What the heck is quaternary?
1.fourth in order or rank; belonging to the fourth order.
Well that explains everything.

The Diatone unit has SQ but a quick study shows this patent relies in RM/QS matrix encoding. Of course I'm not sure that this patent design was used in the Diatone or ever made it into production. Me thinks it's as good a chance like so much else of that vintage that it was basic RM and SQ some kinda logic.

The decoder circuit can best be described as each output ch consists of 1 fixed coefficient and 2 variable coefficients. Unlike the Sansui use of FET's as a shunt to ground for variable mixing this uses it as pass through. Instead of measuring direction between F/B & L/R this does it on the diagnol. If a left front output is desired than the cross talk is between right front & left back. It senses the high degree of seperation between left front & right back & applies crosstalk cancellation appropriately.

It's this approach that inspired me several years ago to do QS decoding on the PC differently than all the other approaches on the QQ forum. As briefly as possible I make two copies of the stereo original. In AA3 I use channel mixer to create the following:
Left = left + .414 Right
Right = right + (-.414 left)
You now have respectively left front & right back together. I save as Lf_Rb. With the next copy you do the opposite to produce right front & left back. With AA3 Center Channel Extractor you apply to each pair & it cancels all the crosstalk on a diagnol basis. It is proportional;set the cancellation to 10dB or 25dByou get 10dB or 25dB crosstalk reduction. This as well as any pre-synthisis enhancement can be done with scripts. The down side is you must now open up a couple of instances of AA 3 & copy paste these files into the correct left front/right front pairs & same for the rears. Time consuming but great results.
 
I didn't think that on line patents would go that far back but it is: Mitsubishi Quatenary Decoder

What the heck is quaternary?

Well that explains everything.

Thanks for that - but it wasn't an easy read! I think fewer words would have conveyed more information (although I realise that is not the purpose of patents at all)!

It's this approach that inspired me several years ago to do QS decoding on the PC differently than all the other approaches on the QQ forum. As briefly as possible I make two copies of the stereo original. In AA3 I use channel mixer to create the following:
Left = left + .414 Right
Right = right + (-.414 left)
You now have respectively left front & right back together. I save as Lf_Rb. With the next copy you do the opposite to produce right front & left back. With AA3 Center Channel Extractor you apply to each pair & it cancels all the crosstalk on a diagnol basis. It is proportional;set the cancellation to 10dB or 25dByou get 10dB or 25dB crosstalk reduction. This as well as any pre-synthisis enhancement can be done with scripts. The down side is you must now open up a couple of instances of AA 3 & copy paste these files into the correct left front/right front pairs & same for the rears. Time consuming but great results.

I'm impressed - you clearly have more patience than me!
 
Somewhere in my digging in the past few months, I think I remember a quick blurb on the QM-4 format. I wanted to say it was yet another variant of Regular Matrix in Japan,
but of course I can't put my fingers on that right now. It's curious how QM-4 is it's own setting and not included as a sub-function under "matrix" (along with SQ and RM). It
might just be an ambient extractor along the lines of a Hafler circuit.
 
Back
Top