masterful mastering?

I'm wondering if the following description reads like anything that could be deliberate. I started out suspecting a manufacturing error, now I'm not at all sure what the story might be.

I recorded a cassette to computer, a Carlos Montoya performance of various Flamenco pieces. It was purchased new in the mid ‘80s. There is a copyright notice dated 1981, but the original performance could be significantly older.???There are no production notes, performance notes, nor any mention of mono/stereo.

The first thing I noticed was that everything is much louder in the right headphone earpiece. Viewing the waveform confirmed this. The Statistics Average RMS number runs 7+dB higher on the right channel.

Switching to Spectral View showed something more unusual. The louder right channel has very little of anything above 2kHz while the softer left runs generally to about 10kHz, sometimes to 12 and even 14kHz.

Amplifying the left channel by 7dB did not make the two channels sound anywhere near equal loudness. The right still sounds much louder.

Frequency Analysis shows that the lower frequencies, especially from about 40 to 150Hz, are much higher on the right (frequency challenged) channel, about18dB. The two curves start converging around 100 to 150Hz and come together by 250 to 450Hz, depending upon where in the music I measure.

On the Frequency Analysis graph the louder right channel starts falling off sharply around 1700 to 2000Hz, corresponding to what I see in Spectral View, and is flat by 4000Hz, staying so until almost 21kHz. The left channel eventually drops to join the right, merging around 12 to 14kHz. I am reasonably sure there is nothing but tape hiss above 14kHz and little but hiss between 10 and 14kHz.

I tried mixing the two channels evenly and got what seems, on headphones at least, a good single channel. I took the cassette to the living room to listen on speakers and found the experience quite different than on headphones. Sound is not particularly concentrated towards one side or the other. The source seems hard to pin down, it is just there. It actually sounds pretty good.

For an interval on the first track there seems to be two instruments. One plays fast while the other plays a slower, rather sparse melodic line at seemingly higher frequencies. At this time there is somewhat more focus, the two seem to be definitely separated in space. On headphones both seem exclusive to the right channel. I don't know if it is possible for one person to be doing both parts, with the apparent separation (on speakers) a trick of EQ, or if Montoya really has accompaniment.

Anyway, back to the question. Anything could be deliberate but this is unique in my limited experience. Does it seem like any ‘normal' approach?

Can't really tell. But you have to bear in mind that this is a) a cassette, presumably mass duplicated and b) about 25 years old, so whatever approach was used may well be completely masked.

Cassette mass duplication has always been a very dodgy business - back in the 80's, you were invariably better off buying the vinyl and recording your own cassette copy if you wanted one, the commercial offerings were often that bad. And they don't really stand the test of time very well, either. Partial self-erasure isn't uncommon at all. And on top of that, recording techniques back then weren't anything like as accurate or predictable as they are now - you regularly ended up with some stuff on tape that if you look at today, seems truly awful. And that's the master tapes I'm talking about... by the time it gets to your lowest-common-denominator cassette tape, via a duplication master duped from what was probably the vinyl cutting master in an attempt to keep the sound at least similar, all bets are off, quite frankly.

It could have been stereo - it could have been mono. If it was stereo, it sounds as though the guitar(s) would have been about 8ft wide!

It sounds as though this is not the result of any deliberate technique with which you are familiar. It does seem too strange to be real but, as I said, the listening doesn't seem bad. Of course, maybe I'm just wierd that way.

I decided to make up a test CD with a few tracks both ways, the 'stereo' or whatever it is that exists as I got it off the tape, plus whatever NR or such that I decide to do, and a full mix together of the two tracks. Then I can make the final version be whatever suits my fancy from the listening tests, regardless of any original producer intentions.

The mixing together that I mentioned earlier was done with the channel mixer ‘Full Mix' preset. The two channels are completely identical after mixing; the frequency analysis graph looks like there is only one channel.

I almost always use the ‘LR to Mid-Side' preset on mono LPs, to get the sum and difference channels, but that did not seem relevant here. Well, I decided to try it, just out of curiosity.

The result is two very similar channels instead of the normal two very different channels. The Statistics measurement shows the two differing by about 2dB but there are some much larger differences on peaks. This time the right has the lower level. The Frequency Analysis graph lines are now very close together, almost on top of each other. The appearance, both in Waveform and Spectral views, is both channels very close to each other most of the time. The sound is right in the middle of my head on headphones.

This LR to Mid-Side result seems as though it could be a stereo recording that had been separated into sum and difference (or into something and something) and is now restored by what I did. Except of course that what I did normally separates, not brings together. Also, the starting point was quite unlike anything I ever get using the LR to Mid-Side preset to separate a mono LP (recorded via a stereo cartridge).

I did not notice it until I compared them directly, and I don't know how to characterize it, but the sound of the two ‘mixes' is quite different in headphones. The Full Mix version is towards the front of my head and the LR to Mid-Side version sets towards the back of my head. There are definite tonal differences too. I hope this isn't sounding too much like audiophile babble; what I hear seems very definite and stable.  I guess I will have to try three versions on my test CD

Anyway, considering what that LR to Mid-Side mix usually does, and what it did here, it seems too far fetched to believe that the stuff on the cassette was merely a manufacturing error or due to tape deterioration over the years. I guess I don't know enough about what might be going on to form any reasonable picture. Perhaps this is all nonsense.

Anyway, considering what that LR to Mid-Side mix usually does, and what it did here, it seems too far fetched to believe that the stuff on the cassette was merely a manufacturing error or due to tape deterioration over the years. I guess I don't know enough about what might be going on to form any reasonable picture. Perhaps this is all nonsense.

I wouldn't go as far as to dismiss an error at all. There have been some colossal technical cockups released quite unwittingly before now...

As for the sound moving around inside your head - well, it's not surprising. You will have altered the IID and ITD relationships significantly. At this distance, the only sensible thing you can do is to recreate the track in a manner that pleases you, and have done with it.

Having read the posts to date, it reminded me of an experience I had. When copying my old 4 track recordings into Audition, I also noticed my left track a lot lower in volume etc. I knew I recorded them in stereo so, like Andy, tried to even them up but it still annoyed me.

I then swapped the jacks around and the right side was lower. It panned out that the RCA to 1/4" jack connector was faulty. Replacing this cured the problem. Possibly useless information

Mark

All sorts of things can happen but I did a half dozen cassettes without fiddling the hardware and only one was that kind of wierd.

I decided to go with the complete mix, creating a strict mono track. I listened to a variety of selections in all three modes. It really seemed that I liked some parts best in type A, some in type B, and others in type C. That was just too much to consider. It is just an imperfect cassette and not worth all the time that would have been required to find the best individual mix for every track.