analog everything

Curse the digital world. Curse it all to hell.

I was re-researching SCA broadcasts, and finally read up on all of this IBOC HD Radio digital broadcast stuff. And I read back on an older Motorola technology called Symphony. It really looked like Motorola’s effort was all we needed. Digital-quality broadcasts in analog, with multipath resolved and dynamics concerns virtually eradicated. Symphony looked like the way to go. But digital broadcasts are winning out anyway, and I can’t figure out why.

And yet all we end up with is digital everything, presumably because people like zeros and ones better. And, they prefer telling you whether you’re allowed to make a copy of a specific sequence of zeroes and ones or not. Grrr. There’s so many interesting things that can be done with analogue signals in general. And our technology has clearly caught up without having to redo everything as digital. The days of just rigging up a simple AM radio out of parts you have lying around your basement are going fast…

I’m angry again. There is a better way, I’m sure of it. We can fix this. Even if I have to have classes and classes of Grade 3 kids building AM transmitters and receivers, just to prove it can be done.

8 thoughts on “analog everything

  1. I’ve just contacted my local ARRL group to see about getting my amateur license….I wonder how much longer such analog tech. will be around.

  2. The problem is that analog electronics have a much harder learning curve, and as the device grows you still have to create all of it. I don’t think it’s analog vs. digital per se, but the fact that digital could be modularized to IC’s that made it adventageous over analog circuitry.

  3. I disagree a bit, toaster. Analogue actually has a pretty straighforward learning curve, compared with the Fourier transforms that must exist in all A/D/A conversions. Not only that, there’s plenty of highly integrated analogue technology, at VLSI levels to be sure. Much of it is used in radio transmission equipment.

  4. With all due respect, you’re on crack. Obviously, one can do anything in “analog” that one can do in “digital” and vice-versa, and it’s possible to define some interesting “analog” modulation schemes, however:
    1. Digital codes allow one to almost trivially choose any possible tradeoff between robustness of transmission and channel bandwidth used for a given rate of information. Obviously you have to obey the channel coding theorem whether your system is “digital” or “analog” but tuning system behavior is vastly easier if the system is digital.
    2. Even if you did have some really nice “analog” coding scheme that you were going to use, practical implementations would almost certainly do signal processing digitally because it’s vastly easier than making sure your “all analog” design is going to behave nicely over the entire range of temperature and supply voltage even after all the components have spent several years installed in the dashboard of a car.
    3. Limiting oneself to doing things “digitally” isn’t actually a limitation. The building blocks required for doing stuff “digitally” are perhaps more complex but less varied and numerous than the building blocks for doing stuff “analog.” I don’t know about you, but I certainly would rather write software than build circuits – each iteration of the design cycle is a lot shorter that way. “Digital” systems let us design the hardware robustly just once and then use it for a multitude of purposes by changing software – that is the beauty of programmable machines. Yes, there are analog programmable machines, but, as programmable machines are really an exercise in arithmetic and as arithmetic is discrete, it makes more sense to build digital ones.

  5. Problem (in the USA at least) is the FCC is auctioning off spectrums for other uses. We see this ongoing push of an all digital broadcast scenario in the USA, but so far it keeps getting pushed back (thank god). I remember talking to some guys at NBC about this — the attitude was “its so distant and technology is progressing so quickly, everything will be fine”. Frankly I disagree with the push of all digital due to many points (i.e. leaving people behind, upgrading to a technology no one may want due to no improvements, etc). Then again, look at the legacy of black and white television has and how, due to compatibility concerns, we’re kinda stuck with a ‘satisfactory’ colour standard.

    Personally I would like to have both. Digital is great for many purposes, but the simplicity of analog (not to mention compatibility, etc) should make a case for itself. It’s easier to piece together or fixi an analog radio or than a digital radio. Emergency situations, being in remote locations and being used as a learning aid are good examples of how analog is still important. And I mean, for TV transmission, granted a 70% quality analog signal may have snow, but it’s more watchable than a digital signal with macroblock artifacts and total distortion and popping audio (as an example).

  6. It does look like Symphony is the way to go for analog signal radio. Radio amateurs have been using DSP on very weak moonbounce signals for a long time now. Analog or digital, software-defined radio is the way to go if one wants the nec plus ultra of capability.

    I skimmed through the IBOC HD whitepaper, and much of the complexity seems to stem from having to gracefully support both analog and digital systems at both the transmitter and the receiver. Something like that is bound to make for complex engineering. But that’s market forces for you. :P

    AM receivers and transmitters are easy to build because the scheme is simple, but it’s also less efficient power and spectrum-wise than more complex methods like FM. Conversely, a spark-gap transmitter with a coherer receiver would be even simpler, but no one in their right mind would use such a beast except as a last resort.

    It’s the usual “appropriate technology” argument:
    (And yes, I agree that IBOC HD might not be appropriate)

    Kids can build an AM radio from a simple diode, but they cannot make the diode itself. Same for digital building blocks, except that the technology is not stable and established at all. If there was one, or few, stable forms of digital radio out there, it would be easy to make kits for it. I’m all for that.

    I also think there has to be a better way, but I’m not sure what it is either.

  7. Actually, an amendment: s/last resort/last resort or for educational value/

    I think that might be “the better way”: show what’s simple first. Have kids build simple radios, and go up from there, regardless of market forces.

  8. On the contrary – hunk of galena and a cat’s whisker placed just-so… and wallah! A diode.

    (That said, doing analog work sucks, IMVHO. I hates it. It’s tons harder to debug.)

Leave a Reply

Your email address will not be published. Required fields are marked *