Lets see if I can help make it a bit clearer. Adding another individual track of the same size is not doubling the track width. The benefits of doubling track width apply to unbroken width increase without a guardband in between. For example, the width of an individual track on ½” 8-track is about .040 inch. Even with ¼ half-track you roughly double the individual track width (more or less depending on manufacturer).
In other words, a track must consist of a wider single element and not two tracks separated by a guardband for the track width rules to apply. A wider track covers more square inches of tape per second, involving more magnetic particles on the tape, manipulated by a wider magnetic element (pole piece). With two or more separate narrower tracks for each left/right feed, the multitrack rules apply, not the track width rules.
That being said it is correct in theory that if two like signals of equal amplitude are added together, the total amplitude is doubled, which is a 6 dB increase. However, they must indeed be of equal amplitude. That means in the example from the OP the four tracks would have to be precisely balanced so that each contributes the same level to the total.
The noise/hiss generated by multiple tape tracks is cumulative and will be apparent as the dynamics of the music drop closer to the noise floor of the tape, such as in quite passages. Multiple tracks don’t help at the other end of dynamic range either. That is, you do not realize an increase in level before distortion by using two or more tracks instead of a single wider one. This is because the tracks are still individual elements covering the same area of tape. You can’t hit two individual tracks with any more level than you can one track of the same width.
For anyone interested in the math the equation is
L = 10 log
n, where
L = noise in dB and
n = number of tracks. For each doubling of tracks there is a 3 dB increase in tape noise. So combining four tracks increase noise by 6 dB.
In addition, the fact that you add electronic noise is not only relevant, it is half the issue. Each track has its own op-amp, resistors and other components that generate noise. Add to this the additional mixer channels needed to record and sum the tracks, and you have more hiss than you would using just two mixer channels (left/right). The hiss wouldn’t make the result unusable, but it does demonstrate that combining like tracks is of no benefit.
In summary, the multitrack rules apply here even when several tracks record/reproduce the same material. The track width benefit does not apply because you are adding tracks, not adding width.
Considering the phase issues, which Mr. Gillett mentioned, there are no good reasons to combine multiple tracks in a mastering role. If an 8-track is all you have you will get the best results using two tracks, preferably 2 & 7. Track 2 and track 7 are chosen because they are far apart and are normally used to set azimuth on a ½” 8-track. Yeah tracks 1 and 8 are the farthest apart but edge tracks are more susceptible to high frequency dropouts.
Hope this helps or at least makes it more confusing, which is the first step to figuring it out.
References:
1. Sound Recording Handbook - John M. Woram 1989
2. Practical Recording Techniques - Bruce Bartlett 1997
) - to Four Identical 1/4" four-track (two track (stereo) one way) master-recorders sinchronized together and then four of left channels outputs from each recorder mixed together to one "master left channel" and four of right channels outputs from each recorder mixed to one "master right channel" - and there you have your master.
). Anyway, doesn't work that way (unfortunately), as the guys above already mentioned....
