I thought it was going to be Super Smash Brothers but I wasn’t totally surprised either way.
The linked article is a jargon festival, but that’s to be expected given the topic. I still found it pretty interesting despite not fully understanding it
I have no idea what the article was talking about except it is about some tech used by radio enthusiasts. Can someone explain the basic idea in it?
I’ve not really touched radio stuff in a long while now but here’s my attempt.
Single sideband (SSB) is a radio transmission technique for sending audio (often voice, but many data modes use SSB too) whilst being pretty efficient with the use of radio spectrum. Think like FM and AM modes on a consumer radio, except those approaches take up a bit more bandwidth compared to SSB, so you can’t pack as many stations into a radio band without interference.
And this is where I might be completely off the mark, but this novel approach is interesting compared to the more conventional approaches due to the reduction in the complexity of the components needed to do this and a reduction of waste power. As the other approaches involved essentially generating a double sideband signal (I can’t remember what the technical term is, but part of me thinks this might be standard AM) and filtering out the (typically) lower mirror band.
You’re not off the mark. Honestly not a bad overview to squeeze into a few sentences. Here’s some extra detail for those who remain more curious.
The circuit complexity reduction happens by changing the math behind the radio signal. Much like how you can describe a vector in cartesian coordinates (a point in x, y) or in polar coordinates (a point in angle and length), choosing how to represent the radio math allows for different techniques to arrive in the same answer. That’s what the author did: he picked a polar modulating scheme over a quadrature modulation scheme. (Note, there are even more mathy ways to modulate a radio signal, but those are what the author is presenting to us.)
The author’s choice avoids generating unwanted frequencies that must be filtered out before amplifying. That’s components on the board that don’t need to be designed nor exist. A solid win.
The drawback? Polar modulation is non-linear in frequency space. What that means is certain frequencies are over-represented and others are under-represented. Imagine playing notes on a piano where some keys are very loud and others you could hardly hear them. That’s the unwanted non-linearity.
Herein lies the trick: what’s bad can be turned into good. Power amplifiers typically need to be linear. Imagine a piano that works fine but the auditorium’s loud speakers make it sound terrible. Those loud speakers would be a non-linear amplifier. The trick is that it’s possible to match the modulator’s non-linear behavior with a power amplifier’s non-linear behavior to end up with a clean signal! A non-linear piano and a non-linear loud-speaker can produce beautiful music! This engineering trick unlocks all kinds of non-linear power amplifier architectures (that’s the “C/E/F” described in the article) which are drastically more energy efficient than linear ones (linear designs max out around 65% efficient).
Big thanks for providing this extra context, that made it a lot clearer for me
Now it makes more sense, thanks!
I have that exact SanDisk, rocking a rockbox firmware – glad to see its still seeing life
As for the rest of the article, yep no idea why polarising sound(?) is a difficult feat digitally but not analog
I have that exact SanDisk, rocking a rockbox firmware
Me too! Only in blue. I have no idea where mine is now, sadly.
I have that exact same SanDisk player as well rocking rockbox firmware and it’s now my portable game console at work.
