Too Many Pixels

Sometimes simpler is more impressive than complicated, and part of this is certainly due to Arthur C. Clarke’s third law: “Any sufficiently advanced technology is indistinguishable from magic.”. It’s counter-intuitive, though, that a high-tech project would seem any less amazing than a simpler one, but hear me out.

I first noticed this ages ago, when we were ripping out the blue laser diodes from Casio XJ-A130 laser projectors back when this was the only way to get a powerful blue laser diode. Casio had bought up the world’s supply of the 1.5 W Nichias, and was putting 24 of them in each projector, making them worth more dead than alive, if you know what I mean. Anyway, we were putting on a laser show, and the bright blue diode laser was just what we needed.

RGB Laser show
A sweeter setup than mine, but you get the idea. 

Color laser setups take three or more different lasers, combine the beams, and then bounce them off of mirrors attached to galvos. Steer the mirrors around, and you can project vector images. It’s pretty cool tech, and involves some serious fine-tuning, but the irony here is that we were tearing apart a device with 788,736 microscopic DLP mirrors to point the lasers through just two. And yet, a DIY laser show is significantly cooler than just putting up your powerpoint on the office wall.

The same thing goes for 2D plotting machines like the AxiDraw. The astonishing tech behind any old laser printer is mind-numbing. Possibly literally. Why else would we think that art drawn out by a pen in the hands of a stepper-powered robot is cooler than the output of a 1600 DPI unit coming from HP’s stable? I mean, instead of running an hours-long job to put ink on paper with a pen, my Laserjet puts out an image in ten seconds. But it’s just not as much fun.

So here we are, in an age where there’s so darn much magic all around us, in the form of sufficiently advanced technology, that comprehensible devices are actually more impressive. And my guess is that it’s partly because it’s not surprising when a device that’s already magic does something magical. I mean, that’s just what it’s supposed to do. Duh!

But when something beautiful emerges from a pair of mirrors epoxied to shafts on springs turned by copper coils, that’s real magic.

39 thoughts on “Too Many Pixels

      1. You’re mistaken. Two years ago in lab where I work we did some tests on both original HP toner and that refill stuff from China. Turns out, while laser printers are generally bad for your health, the original HP toner was mostly carbon and some kind of polymer-based binder. Not something you should inhale but also not very toxic. Composition of the Chinese stuff on the other hand was comparable to sooty residue from burning used motor oil – pretty much concentrated cancer.

  1. Great thought! Have you noticed how for years they would use ancient mainframe style computers to depict advanced futuristic units, because the old tech was much more visually impressive. A steam engine locomotive is also 100 times more impressive in the flesh than a far more sophisticated electro-diesel. Same applies to tube amplifiers, except they are actually better. A naked …. OK – let’s stop there. ;-0. The bottom line is humans like to see how things work.

    1. I live close a car museum that has load of stuff from the first part of the 20th century, including Ettori Bugattis personal car. That thing is way more impressive to look at than any Tesla.

    2. Or washing machines. Most of them are front loading and have a glass door, which serves no other purpose than to watch your undies spinning around. From a technical viewpoint this is an inferior design, top loaders can be built smaller and cheaper, are less noisy and last longer. But people prefer to watch moving things, or at least the marketing guys think so.

        1. Who prefers and who doesn’t.

          With less water, you have to use less detergent or you’ll get itchy residue on your clothing, but with less detergent you don’t necessarily get clean clothes. Sure, some special expensive detergents might work, but when you’re buying the cheap stuff you need to use more detergent and more water, so with any modern machine I always end up using the extra rinse option anyways.

          It’s just another case of too much optimization – it’s good on paper but doesn’t meet real world conditions and the effect is either no saving or negative savings.

          1. Most clothes don’t get that dirty. If you work in an office, the biggest thing to worry about is sweat.

            Low water and little detergent work just fine for those conditions. Heck, you can often times skip the detergent entirely – there’s enough residue in the machine to take care of the little bit of real dirt that such clothes have.

            If your clothes get really grubby (yard work, grease and oil from machines, or people who work in the great outdoors, etc.) then you need more detergent. You can run an extra rinse cycle if needed.

            Not that a little detergent residue matters much to many people. They use those dryer thingies to “reduce static cling” and give their clothes a “spring fresh” smell. Translation: Fill my undies with wax and perfume. That junk is itchy and scratchy. A little extra detergent doesn’t make a difference.

            Most of our clothes are washed with (far) less detergent than the detergent manufacturer and the washer manufacturer recommend. They come out clean. We dry most clothes on a line – either in the house (winter) or in the back yard (rest of the year.

            Never use detergent pods. Those are just a way to force you to buy and use more detergent.

          2. Detergent pods also have plastic that “dissolves” into the water and adds to plastic pollution in our water systems. Even if it’s maid of some kind of starch or “biodegradable” plastic, it’s still an unnecessary waste.

          3. > A little extra detergent doesn’t make a difference.

            It does to me. I don’t use scented detergents or dryer sheets.

            Another thing is the energy saving features which reduce temperature to 30 C instead of 40 to 60 C, which actually make some of the detergent not dissolve properly. There may be visible flecks left on the clothing, which is another reason I use a double rinse. I refuse to buy detergent that is 3-4 times more expensive to save a bucket of water and a penny on electricity – it’s just false economy.

      1. Our clothes come out much dryer from our front loader than they ever did from our older top loaded machines. Then they take less time to dry. Plus we stacked them so they take up less space.

    3. “A steam engine locomotive is also 100 times more impressive in the flesh than a far more sophisticated electro-diesel.”

      Well, yes and no. Some may find the steam engine to be boring and dirty, while same person may find an early electric tram (ie, a vintage Berlin tram) to be elegant and fascinatic. Let’s just think about Jule Verne’s novels. It all depends.

      Electron tubes are also fascinating, because someone can see and feel electricity at work. The phenomenon behind it may also be more interesting than that of the bipolar transistor (similar to the working of the field effect transistor).

      But rose-coloured glasses aside, electron tube technology is much more sophisticated/ingenious than solid-state. Strictly speaking, solid-state dates back to the crystal detector, the predecessor of the glass diode. The transistor is nothing more than an evolution of the diode. Unlike the thermionic valve, which is a very unconventional invention.

      Another advantage of the tube is its electrical robustness. Except for the heating, it is pretty much invulnerable. It can handle electrical mismatches, lile shorts, that would destroy transistors quickly. It can handle strong inputs and EMPs without breaking. When overloaded, it goes softly into saturation.

      If todays world just would be more open minded and give the tube a second chance. With today’s manufacturing processes, highly efficient tubes or nuvistors could be made. They’d ge ideal in RF front-ends, for example. A thunderstorm which would normally possibly damage any receiver wouldn’t harm such a tube.

      Another advantage is that tubes have a high impedance output. They can work very well in tandem to high impedance inputs of certain transistor circuits (base as ground/emitter as input). Long story short, they’d make a good team. 💙

      1. If the base is grounded then the emitter input would need to be negitive. Electrically it is equivalent to a base input, other than inverted and the collector output can drop to .8 v or whatever. But why woud you want to drive a transistor from a vacuum tube? Would the transistor not get fried?

  2. Love this thought provoking article. Elegant simplicity is truly sublime. I think there comes a point when you know something about a topic where you can truly understand and respect when something simple functions way better than a complex solution, and understanding that compared to an unimpressed layperson is very satisfying.

    I’d say some surgical instruments, with their organic curves that have been refined over 100s of years to do exactly what they are supposed to do, epitomize this. Sure, we have million dollar DaVinci robots but they aren’t as elegant as a simple CastroViejo needle holder, IMO. Also, another related thought- I used to live on the route for an annual marathon, and the leaders looked like they weren’t working very hard but, of course, they were international elite athletes. They were just beautiful and smooth to watch in motion.

  3. It’s only magic if you can’t even begin to imagine how to do it. That’s where the “sufficiently advanced” part comes into play.

    It doesn’t have to be sublime, but that helps. Case in point: fusion power. A giant tokamak certainly is advanced technology, but if you look at all those coils, you can kind of see what’s going on there, and in that regard, there’s no magic. But if you could do the job in a pickle jar, *that* would be magic. And that’s probably why that little diversion got so much attention; because we were drawn to the magic of it

  4. Great thought. My theory is humans love to visually see technology working. Our whole technological evolution is based on a “visualize – build – see – upgrade” loop. This is why sci fi movies use giant ancient mainframes to depict scary AI vs beige boxes. It’s why a steam locomotive knocks the socks off an electro-diesel locomotive in the looks department. And we love motorbikes with no clothes on. We consider tube amps to be stunning compared to black bricks, except they are also actually better. Mechanical watches even made a come back until the Apple Watch just got too useful to resist. I guess we are all steam punks under the skin (or copper embossed leather cladding), if you are a victorian robot.

    1. This causes some practical problems, because it means the smartest people (Who tend to be the most interested in, and capability of, understanding devices like engines and tube amps) have little interest in the most advanced and most useful tech, which is basically defined by the fact it’s function doesn’t follow from it’s materials in any obviously way.

      We’ve advanced to where the most practical and efficient option is almost always just the same exact computer chip programmed in the same language with slightly different IO capabilities, and it’s hard to do anything about that, what makes modern tech special, the fact that we can arbitrary program it to do whatever we want, the low use of real resources, and the lack of moving parts, is exactly what makes it boring, and seem almost disconnected from reality.

      1. You nailed it. We love when function follows form. That can apply to coding as well, but you need to look under the hood. Lots of form and function in an architecture or algorithm.

  5. >And yet, a DIY laser show is significantly cooler than just putting up your powerpoint on the office wall.

    Apples to oranges. Suppose you were to use couple pixels off the DLP projector to perform a light show. It would put out a million times less power than the laser diode and it wouldn’t draw that cool beam in the fog because it’s meant to be projecting entire images.

    Likewise for the laser printer: it’s putting out a dithered raster image made out of molten plastic that looks like a xerox copy (because it is), whereas the plotter produces a continuous line of ink that could be mistaken for hand-drawn. When you hold it in your hand, you can tell it’s made by a different process.

    It’s just not doing the same thing. The error of the premise is assuming that projecting light is just projecting light, and printing is just printing.

      1. That’s another thing. A laser printer is very picky about what it will print on, what size, etc. whereas a plotter will draw on any flat thing that you can place under the pen.

  6. But it’s not always and only “lights” and images that make the magic, it’s technology. Starting in 1988, I used a 16 bit Intel 80C196 processor, originally made for car engine controllers (forced by lower pollution and higher mileage requirements) into a controller and monitor for an ILS (Instrument Landing System) ground system. Initial development took a bit over a year, but evolved over nearly two decades of upgrades including newer software safety requirement (DO-178, etc.). Part of the initial effort was to design a simple (<500 byte) but nimble multitasking OS. The system was eventually was upgraded from CAT I to CAT II and III (the latter is for zero visibility). Variations are installed in thousands of airports all over the world — all with less than 64KB of code and SRAM space (later models used a technical "trick" to make 32KB of this both code and SRAM, for a total of 96KB). This "invisibly" produced magic that few can fully appreciate other than it lets planes land safely in zero visibility weather. There are so many examples of invisible engineering "magic" out there. Pretty much ANY microcontroller is magic and can be used to make magic. It's a journey that can take your breath away when you try to imagine "what's next?"

  7. I built some Blinkenlights objects on walls, in trees or small examples for the bookshelf. If you place some hundreds of LEDs you get just another screen. I did experiments on the numbers of LED that will give the most impressive effect for me and was quite happy with around twelve.

    Instead of putting my effort in resolution I spent a lot of time optimizing dimming resolution to get better experience at low brightness. I experimented with gamma correction and rhythm of lights and patterns with a grain of randomness, spatial distribution and color.

    1. Thanks! I think you’ve hit on something else — the quantity / quality distinction.

      But maybe it’s also the same, because you _can_ focus so much more on what’s being displayed on the LEDs. Imagine having to pay that much attention to 16 megapixels. Each one of them…

      Anyway, twelve sounds about right to me! :)

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.