Welcome to another episode of the 8-Bit Guy.
Today I'm going to show you two very interesting technologies from the 1980s.
The first is the light pen, which is basically a pointing device that was used for drawing
and just general operating system usage.
Then the other thing I'm going to show you is the zapper gun which was a light gun that
was used with some games on the Nintendo and most of the other 8-bit systems had a similar
gun like this available.
Now, both of these technologies work by sensing light.
But that's probably about all they have in common.
So I'm going to show you how each one of these works and how they're different.
The light pen is actually an amazingly simple device, usually just containing a lens and
a photo sensor.
You can place it up to the screen and the computer will immediately know where on the
screen it is pointing.
It works just as well on large screens, and small screens, color, or even monochrome screens.
I used to be mystified as to how the computer knew where the pen was positioned.
Well, as it turns out that the computer doesn't as much know where it is on the screen, but
rather WHEN it is.
So as many of you know, when you look at an old CRT television or monitor like this, you
see a full, static, sharp image, at least to your eyes.
But that's only because of persistence of vision.
Basically, our eyes are too slow to see the moving pixel or electron beam that is essentially
lighting up the screen one pixel at a time.
This is similar to how at night you can use a sparkler or flashlight and make circles
that look coherent even though you know the light only exists in one place at a time.
As a result of this, it's actually pretty difficult to capture stable images on a camera.
So, as you're watching this, you're probably seeing some sort of scrolling line or something
moving across the screen, which I am not seeing when I look at this here in person.
And that's an unfortunate result of the timing between the camera and the CRT.
So in reality, at any given microsecond, there's really only one pixel lit up on this screen.
Now, the computer knows exactly which pixel it has lit up and exactly what time.
So, I've created this simulation in order to try to show you what it would look like
if you were able to slow down the beam of the CRT display to the point you could actually
see it.
Now, I also enlarged the pixels in order to make them a little bit easier to see as well.
Now, if you were to slow it down even more, it would look like this.
And eventually, if you slow it down enough you would literally only be able to see one
pixel at a time.
And that's what it would really look like if you could slow it down that far.
Now, eventually, the pixel underneath the light pen's lens will light up.
The light pen will see this, and that information will be transmitted to the computer and the
computer will know exactly which pixel it was drawing at the time that the signal comes
in, thus locking in the coordinates of where the light pen is on the screen.
Amazingly I don't even have to touch the light pen to the screen.
In fact, it works almost as well several inches away from the screen.
Now, some light pens have buttons on them.
This particular one is on the front so by pressing the light pen harder against the
screen, activates the button.
Some light pens had buttons on the side, and some like this Trojan model have no button
at all.
So, one annoyance when using a light pen on the Commodore 64 is that not all software
is compatible with all light pens.
Because, apparently the button is not wired the same from one light pen to the next.
If you look at the joystick port on the Commodore 64, you'll notice there is a special line
for light pen.
This line is for the photosensor and goes straight to the video chip inside the computer.
Now the button could be connected to any of these other input lines.
As an example of this problem, I have a copy of Koala Painter here.
Yes, this is the same software I showed in a previous episode being used with the KoalaPad.
Only this version is modified to work with a light pen.
But the button on this light pen does not work with the software, presumably because
it is wired to the wrong pin.
However, they were generous enough to design the software to also accept the 1 key on the
keyboard as a substitute for the button.
So I can at least partially use the pen in this software.
I have another piece of software that came with the light pen, and this one's actually
pretty cool because it actually allows you to select several different input devices.
But, the button works just fine on this piece of software.
Now this might not have been a problem on other platforms.
I'm not really sure, but I do know the Apple II, IBM PC, and Atari 8-Bit systems could
also use light pens.
OK, so here is another downside to the light pen.
Notice it has trouble when I try to position the pointer over a black area of the screen.
And, the reason is simple.
When there are no pixels being lit up, the pen can't see anything, so the computer
doesn't know where or when the pen is.
They try to compensate for this by using like a flashing pointer that follows the pen.
So if you move slowly enough, you can get the pointer to follow you into the black area,
and with this it can actually see the flashing pointer.
Another downside has to do with accuracy.
The pen does tend to see some nearby pixels sometimes, and send inaccurate results.
So drawing by hand often results in pretty jagged edges.
In fact, on the C64 the official spec actually shows that the Light pen can only really distinguish
about 160 pixels horizontally, but the C64 actually has 320 pixels.
So that right there should give you an idea of accuracy.
Now, this Flexidraw software actually has a special mode called filter.
And what that does is uses a special algorithm to smooth out the movement of the pen to help
out with this problem.
But ultimately, the light pen works best for clicking menus and working on zoomed in graphics
where the accuracy is not nearly as important.
And I suppose the last thing I can mention about these is that they only work on a CRTs.
Due to the fact that LCD monitors deal with drawing the screen in a totally different
way, light pens are totally useless on LCD screens.
So now that I've shown how the light pens work, I thought we might tackle the question
of why were they never popular even back during the 1980s and why did they die out.
I decided to invite my friend Robin to come explain this.
Robin is a professional developer at PIXL games and he also has a band where he sings
about vintage computer hardware.
Robin was also one of the original developers of the software for the Commodore all-in-one
DTV joystick.
And, I featured this in a previous episode before, and in fact there's an easter egg
on here where I'll show you, that you can actually find an original picture of the development
team and you'll see Robin there before he had his beard.
Thanks David.
I think there's a few reasons the light pen didn't catch on.
You already mentioned the accuracy problem.
At least on the Commodore 64, you can't do pixel perfect drawings with the light pen.
So, you need to zoom in.
But if you are working zoomed in like this, you might as well use keyboard or joystick
controls.
And that's exactly what most artists did back in the day.
You also mentioned compatibility.
Even if you found some software worth using, there's no guarantee your light pen would
work with that software, since there's no standard for how the button on the pen, if
any, would be wired into the computer.
Also, it seems that add-on input devices rarely catch on, unless there's killer apps or
games or strong operating system support.
Even modern game consoles have this problem.
The Kinect and especially Playstation Move did poorly, despite being first party devices
with huge companies behind them.
On top of all that, the light pen was just plain uncomfortable to use.
Having to stretch your arm out to the TV or monitor might be fine for a couple minutes,
but if you're going to try to get actual work done, your arm is quickly going to tire.
Remember that back then, a lot of people used their TV set with their computer, so it wasn't
going to be in an ideal position.
Even if you had a dedicated computer monitor, it would typically be sitting in a place that's
good for looking at while typing or playing games, not for writing on.
The idea of using a pen on a screen was a good one, and finally found mainstream success
in the late '90s with devices such as the Palm Pilot.
It was a lot nicer to use because the small, lightweight screen could be held in a comfortable
position for using the stylus.
So, next I'm going to talk about the Zapper gun on the NES.
Most people should be familiar with this since the gun was usually included as part of the
original system and Duck Hunt was often bundled on the same cartridge as Super Mario Brothers.
OK, so first of all, I've gotta point out a little piece of trivia.
I think a lot of people are actually totally unaware that when playing Duck Hunt, you can
hand somebody the second controller and they can actually control the ducks.
So, that aside, how does the zapper gun know where on the screen it is pointed.
Well, you might think it works similar to the light pen, and thus works by tracking
the beam of the CRT.
But it doesn't.
It actually works on a much simpler principle.
If you notice, when you pull the trigger on the gun, even if it isn't pointed at the
screen, you'll see these white flashes of light around the ducks.
In fact, let's slow this down and look at it frame by frame.
So when you pull the trigger, the screen goes completely black for a single frame.
This allows the gun to calibrate, then the next frame is a white box.
And then the following frame goes back to the duck and everything else.
So, essentially the zapper gun is looking for a white flash of light during that one
frame.
So, if you have it aimed correctly it is going to see the light, and if you don't, then
well, it won't.
So it does work similar to a light pen, in that it is detecting light from a CRT.
But, it really has no idea what part of the screen it is looking at.
All it really knows is whether it saw a flash of light or not.
So, you might ask the question, if it doesn't know where it is looking at on the screen,
how does it handle two ducks?
Well, let's watch that in slow motion.
Again, we get a black screen, followed by one flash of light, then in the next frame,
a different flash of light.
So depending on when it receives the flash of light, it will know which of the two ducks
you shot, assuming you managed to hit either of them.
So let's take the zapper gun apart and see what secrets it holds inside.
Disassembly is actually pretty simple with just a few screws to remove.
It appears this part is just a decorative bezel.
OK, so the real action happens inside here.
So at first I was mystified what this metal cylinder was for.
I eventually decided its only purpose must be to add weight to the the otherwise plastic
toy.
Ok, so on the front we have a pretty basic lens.
This one is surprisingly scratched up, but probably doesn't affect the function any.
This piece also had my mystified.
I eventually decided this is just another weight to make the gun heavier.
I just thought it was weird how it was shaped to allow the light to pass through it, rather
than just putting the weight somewhere else.
And of course, here's where all the real action happens.
This is the photosensor.
This is not a camera, it is just a really simple device that detects the presence of
light, and that's about it.
So the next question to be asked is why was the zapper gun not used in more game titles.
So, I've decided to ask my friend The Obsolete Geek, who is far more versed in Nintendo history
than I am, to see if he can explain this for us.
Thanks, David.
Yeah, you know it is hard to say with any certainty why more Nintendo games did not
support the light gun.
Actually, a little more than a dozen officially licensed games actually supported it.
But, the NES was not the only system to use a zapper.
You have the Master-System by Sega, or more obscure platforms like the Atari XEGS, which
is one of the systems that was actually bundled with a light gun.
And I think even fewer games than the NES supported the light gun on this system.
Then, of course, you've got some of the early light guns that you could actually rob
a bank with.
They look like real guns, such as this one for the Magnavox Odyssey.
If I had to speculate as to why these just didn't maintain their popularity is, they're
fun to play but not for a very long period of time.
I think the replay value is pretty low.
You also have to remember too that, for most systems, these were additional accessories,
so that means the game developer had to make sure that their game had to support the light
gun, which means a game developer is going to sell even fewer games, because even fewer
households would actually have this accessory.
So, both of these technologies have been made obsolete by the proliferation of LCD monitors
and televisions.
The light pen will no longer work because there is no longer a single pixel, electron
beam scanning across to pick up to know where the heck it is looking at on the screen.
The zapper gun won't work for a slightly different reason, because you see these modern
LCD televisions, they process the image, and there's usually a millisecond delay between
when the signal goes to the TV and when the TV actually displays it, thus totally screwing
up the timing of the zapper gun and what the Nintendo is looking for in regards to the
flashes of light.
But, still, they are both very interesting pieces of computer and gaming history, and
I hope you found this episode interesting.
And stick around, because as always, I have more stuff coming up!
Không có nhận xét nào:
Đăng nhận xét