WEBVTT
Kind: captions
Language: en

00:00:00.000 --> 00:00:03.060
[ The comments being read snarkily ]

00:00:11.200 --> 00:00:12.720
...the rest of that comment was really insightful,

00:00:12.720 --> 00:00:13.620
thanks for sharing.

00:00:13.620 --> 00:00:15.380
OK, could I have been a little more clear?

00:00:15.500 --> 00:00:18.679
“I’ll show you Sony’s system later on,
but if you…”

00:00:19.120 --> 00:00:19.720
Yes.

00:00:19.730 --> 00:00:20.730
Yes I could have.

00:00:20.730 --> 00:00:26.449
But, this is in fact “later on”, so perhaps
maybe we shouldn’t jump to conclusions so quickly.

00:00:26.449 --> 00:00:30.359
Tell you what, I’ll work on being a little
more precise in my choice of words.

00:00:30.359 --> 00:00:35.650
And perhaps you, dear viewer, could refrain
from jumping to conclusions quite so quickly.

00:00:35.650 --> 00:00:40.000
If we put our minds to it, together, we can
enhance our understanding through complex

00:00:40.000 --> 00:00:43.960
thought both on the part of the speaker, and
on the part of the listener.

00:00:43.960 --> 00:00:48.020
♫ And that’s how the world gets better ♫

00:00:48.020 --> 00:00:50.440
Alright, this video is coming at you in three
parts.

00:00:50.450 --> 00:00:51.680
As we discussed in the last video--

00:00:51.680 --> 00:00:53.200
NO WAIT,
wait wait wait.

00:00:53.200 --> 00:00:55.470
The ding was way too loud, let’s tone it
down.

00:00:55.470 --> 00:00:56.340
[ Ding sound from Jeopardy! ]

00:00:56.340 --> 00:00:57.380
Much better.

00:00:57.510 --> 00:01:00.760
So as we discussed in the last video, we’re
gonna take a look at some of the waveforms

00:01:00.770 --> 00:01:05.339
coming from the components of this Magnavox
(which is really a rebadged Philips because

00:01:05.339 --> 00:01:08.960
Philips couldn’t use that name in the US
due to it’s passing similarity to Philco,

00:01:08.960 --> 00:01:11.970
I guess, but that’s a stretch) CD player.

00:01:11.970 --> 00:01:13.530
That will be part one.

00:01:13.530 --> 00:01:17.540
In part 2, we’ll take a look at a Sony laser
pickup and discuss the differences.

00:01:17.540 --> 00:01:21.700
Plus I will tear this one apart to expose
the laser diode and photodiodes.

00:01:21.700 --> 00:01:25.950
And in Part 3, we’ll discuss why Sony’s
pickup solution would become the default standard,

00:01:25.950 --> 00:01:27.740
used to this day in optical drives.

00:01:27.740 --> 00:01:29.670
We’ll begin with Part 1.

00:01:29.670 --> 00:01:32.600
And Part 1 will be quite a bit different from
most other videos.

00:01:32.600 --> 00:01:35.370
That’s because much of it is unscripted.

00:01:35.370 --> 00:01:36.430
Off the cuff.

00:01:36.430 --> 00:01:37.580
Just gonna wing it.

00:01:37.580 --> 00:01:38.180
Yeah.

00:01:38.180 --> 00:01:39.440
But first, a few corrections.

00:01:39.640 --> 00:01:43.620
As usual, the Internet has provided answers
that I did not find, and has revealed some

00:01:43.620 --> 00:01:46.570
mistakes and false assumption from the last
video.

00:01:46.570 --> 00:01:47.570
Of course.

00:01:47.570 --> 00:01:51.210
But that’s OK, we all make mistakes, and
now I’m gonna tell you what we learned thanks

00:01:51.210 --> 00:01:54.600
to the diligent research of more knowledgeable
people than me.

00:01:54.600 --> 00:01:56.420
Remember it’s OK to not know everything.

00:01:56.420 --> 00:01:59.340
And it’s OK to ask for advice from professionals.

00:01:59.340 --> 00:02:00.220
♫ And that’s h--♫

00:02:00.220 --> 00:02:04.119
Let me reveal the most dumb mistake, and it
was right on screen.

00:02:04.119 --> 00:02:06.690
The best kind of mistake.

00:02:06.690 --> 00:02:11.090
I had been assuming that pin 10 was the output
signal from the focus chip, and that it would

00:02:11.090 --> 00:02:13.260
be a binary stream from the pits and lands.

00:02:13.260 --> 00:02:15.440
Boy was that wrong on many counts.

00:02:15.440 --> 00:02:18.769
First, it’s not pin 10 that does that, it’s
pin 3.

00:02:18.769 --> 00:02:21.560
And it’s labeled right here on the datasheet.

00:02:21.560 --> 00:02:26.330
Pin 3 is also labeled to decoder, which I
missed, and most importantly it’s a sum

00:02:26.330 --> 00:02:30.610
of the outputs from all 4 photodiodes as illustrated
right here.

00:02:30.610 --> 00:02:35.980
Yes pin 10 also gets the sum of the 4 diodes,
but unlike pin 10, pin three has an amplifier,

00:02:35.980 --> 00:02:38.879
equalizer, and another amplifier to boost
the output.

00:02:38.879 --> 00:02:39.980
Hm hmmm.

00:02:39.980 --> 00:02:45.029
I want to give a big thanks and shoutout to
Zim0256 who not only created the helpful diagram

00:02:45.029 --> 00:02:50.230
that I used on a few occasions but who also
gave more complete info on what is going on.

00:02:50.230 --> 00:02:55.569
The Motorola processor is a custom chip based
around an HC08 microcontroller, and it is

00:02:55.569 --> 00:02:58.769
in fact the decoder that is doing most of
the work.

00:02:58.769 --> 00:03:03.019
I couldn’t find its datasheet because the
P on the end was limiting my search results.

00:03:03.019 --> 00:03:06.180
And to think, we just learned about minding
our p’s and q’s.

00:03:06.180 --> 00:03:10.379
Anyway, in fact the decoder is handling all
the raw data; undoing the eight-to-fourteen

00:03:10.379 --> 00:03:12.970
modulation, isolating the subcode, and all
that.

00:03:12.970 --> 00:03:17.069
It even digitizes the output from pin 3, which
at this point is still analog which we’ll

00:03:17.069 --> 00:03:18.180
see in a moment.

00:03:18.180 --> 00:03:22.670
It sends the extracted subcode to the processor
so it knows how to handle the disc and respond

00:03:22.670 --> 00:03:23.829
to user requests.

00:03:23.829 --> 00:03:28.040
Really, the processor is just managing the
living and breathing tasks of the machine,

00:03:28.040 --> 00:03:31.859
such as the user interface, display, and other
goodies like popping the disc tray in and

00:03:31.859 --> 00:03:34.059
out and getting the whole process going.

00:03:34.059 --> 00:03:37.939
Speaking of the disc tray, a few people had
asked how the machine can tell that the disc

00:03:37.939 --> 00:03:42.959
tray has been pushed in manually as opposed
to someone hitting the open/close button.

00:03:42.959 --> 00:03:47.120
Some other people noticed this strange button
switch near the decoder.

00:03:47.120 --> 00:03:49.969
These two groups of people are discussing
the same thing.

00:03:49.969 --> 00:03:53.879
To the right of the disc tray is a little
series of levers that rest on top of this

00:03:53.879 --> 00:03:56.329
button when it’s assembled.

00:03:56.329 --> 00:04:00.529
Little nubs on the bottom of the tray will
cause this to push down on the button in either

00:04:00.529 --> 00:04:02.870
the open or closed position.

00:04:02.870 --> 00:04:06.459
This not only tells the processor that the
tray has reached the end of its travel in

00:04:06.459 --> 00:04:10.560
either direction, but it also means that it
can detect if the tray has been pushed in,

00:04:10.560 --> 00:04:14.730
as the button will be released (and thus the
switch opened) as soon as the tray is moved

00:04:14.730 --> 00:04:15.829
out of its open position.

00:04:15.829 --> 00:04:18.329
OK, it’s time for the script to be put to
the side.

00:04:18.329 --> 00:04:22.310
I’ve got a sort of odd setup here, because
I need to see both the oscilloscope screen

00:04:22.310 --> 00:04:26.930
and what I’m poking on the board, so we’re
gonna do some picture-in-picture editing magic

00:04:26.930 --> 00:04:29.069
and throw the scope screen up high.

00:04:29.069 --> 00:04:30.069
Excellent!

00:04:30.069 --> 00:04:35.100
And you know what, maybe we’ll throw another
camera up here so you can see what I’m doing.

00:04:35.100 --> 00:04:35.820
Even better.

00:04:35.820 --> 00:04:36.640
All the cameras!

00:04:36.640 --> 00:04:37.820
All the tripods!

00:04:38.260 --> 00:04:40.620
Well this is… awkward.

00:04:40.620 --> 00:04:42.680
It’ll have to do.

00:04:42.680 --> 00:04:45.169
As promised, we’re gonna poke around on
some of the things here!

00:04:45.169 --> 00:04:49.639
Now I brought my laptop here so I can actually
go and look and see what I’m looking at,

00:04:49.639 --> 00:04:54.110
and you’ll notice these, uh, wires that
I’ve soldered onto chips underneath.

00:04:54.110 --> 00:04:56.180
I’ll bring up the image of that.

00:04:56.180 --> 00:05:00.430
Part of the issue with this player is that
it has to be--the ribbon cable between the

00:05:00.430 --> 00:05:05.560
drive and the board is very very short, and
there’s pretty much no way to operate this

00:05:05.560 --> 00:05:06.580
unless it’s assembled.

00:05:06.580 --> 00:05:10.990
I really don’t know how people would service
these things, ‘cause it, ya know… a lot

00:05:10.990 --> 00:05:16.830
of the… a lot of the stuff that I wanna
see is underneath the CD reader like the,

00:05:16.830 --> 00:05:20.069
uh, the tracking and focus chips are basically
directly under here.

00:05:20.069 --> 00:05:22.050
That’s not helpful.

00:05:22.050 --> 00:05:25.870
But we can see what else we can take a look
at.

00:05:25.870 --> 00:05:32.600
I did find, just from poking--at one point
I found the raw datastream.

00:05:32.600 --> 00:05:35.250
I think it was one of these guys.

00:05:35.250 --> 00:05:38.380
Unfortun… what is happening here?

00:05:39.620 --> 00:05:40.200
Oh.

00:05:44.960 --> 00:05:48.259
Well this is kind of interesting, what’s…
uh I wonder what’s on this line, let’s

00:05:48.260 --> 00:05:49.900
see if I can figure it out.

00:05:49.900 --> 00:05:54.380
Notice that we’re on a really long time
scale.

00:05:54.389 --> 00:05:55.919
I think this is the subcode.

00:05:55.919 --> 00:05:59.990
If I can hopefully… get the... trigger…

00:05:59.990 --> 00:06:00.990
I’m...eurgh…

00:06:00.990 --> 00:06:04.129
I’m really...have no experience with a scope.

00:06:04.129 --> 00:06:05.139
So, just.

00:06:05.140 --> 00:06:06.460
Just so ya know.

00:06:07.460 --> 00:06:09.360
Let’s try…

00:06:09.360 --> 00:06:10.379
Oops that’s what I meant.

00:06:10.379 --> 00:06:11.639
That’s what I meant!

00:06:11.639 --> 00:06:12.639
The other way…

00:06:12.640 --> 00:06:13.140
Oh!

00:06:13.140 --> 00:06:14.140
It’s probably you, OK.

00:06:15.760 --> 00:06:16.600
Yep, OK.

00:06:16.600 --> 00:06:17.520
Ha ha!

00:06:17.520 --> 00:06:18.520
I found it!

00:06:18.530 --> 00:06:25.530
So this here is the waveform that’s coming
from the amplifier chip from the au, from

00:06:25.530 --> 00:06:28.800
uh the TDA8808T augh that one, that one.

00:06:28.800 --> 00:06:35.229
So this, this is the amplified output of the…
oops, of the pits and lands.

00:06:35.229 --> 00:06:39.949
So this is the raw analog signal coming from
the disc.

00:06:39.949 --> 00:06:48.940
And if we look a little closer… you can
see if I touch the disc… see how that’s…

00:06:49.500 --> 00:06:50.680
Oh. Well I’ve killed it.

00:06:51.120 --> 00:06:52.120
Great.

00:06:52.280 --> 00:06:53.840
Great job!

00:06:53.840 --> 00:06:57.940
So if I kinda just touch it you can see the--it
slows down a bit.

00:06:58.600 --> 00:06:59.980
Let me zoom out a bit.

00:06:59.980 --> 00:07:02.900
Oop--zoom out a bit.

00:07:04.660 --> 00:07:06.259
See that?

00:07:07.560 --> 00:07:10.120
It just completely--it gave up there.

00:07:10.120 --> 00:07:16.200
That click sound you heard I believe is the
laser kicking up all the way.

00:07:19.260 --> 00:07:23.060
But yeah so this--this isn’t the raw output
from the photodiodes but this is the amplified

00:07:23.060 --> 00:07:26.979
output coming from pin 3, going to the decoder.

00:07:26.979 --> 00:07:34.280
The output from the photodiodes is very very
small, this is… if we take a look and compare

00:07:34.280 --> 00:07:36.520
the voltage level there’s almost nothing
there.

00:07:37.140 --> 00:07:39.900
And it’s so noisy I can’t see anything.

00:07:39.900 --> 00:07:45.319
These all should be the same.

00:07:46.680 --> 00:07:49.840
There’s nothing there.

00:07:49.849 --> 00:07:52.789
Which is a shame because I really wanted to
show you that!

00:07:52.789 --> 00:07:54.680
Oh well.

00:07:56.020 --> 00:08:00.080
Wanted to show you some interesting stuff,
and I think that we saw some interesting things

00:08:00.080 --> 00:08:03.550
but not as many as I was hoping.

00:08:03.550 --> 00:08:06.969
Well, with that disappointing segment out
of the way, let’s move on to the part that

00:08:06.969 --> 00:08:08.009
I didn’t forget about.

00:08:08.009 --> 00:08:12.180
To see how Sony handled the optical system
differently, we need to get our hands on a

00:08:12.180 --> 00:08:15.949
Sony CD player, hopefully of similar vintage
to this Philips one.

00:08:15.949 --> 00:08:20.129
Oh perfect, a Sony CD changer from 1992!

00:08:20.129 --> 00:08:23.819
Now it doesn’t really matter that it’s
a changer, there’s still a standard CD mechanism

00:08:23.819 --> 00:08:28.219
in here, it’s just on a pivot and will happily
pop up and fall down to greet or say farewell

00:08:28.219 --> 00:08:30.060
to the disc it plays.

00:08:30.060 --> 00:08:34.570
It works in tandem with the disc tray, which
rather than just being an inny-outty operation,

00:08:34.570 --> 00:08:37.400
it’s an inny-outty with rotational flair.

00:08:37.400 --> 00:08:42.450
Now I embarked on the frightening task of
getting this mechanism out of this player.

00:08:42.450 --> 00:08:47.480
Armed only with an ordinary screwdriver, I
went hamfisted and started unscrewing stuff,

00:08:47.480 --> 00:08:52.840
before I realized I hadn’t filmed the close
up segments of the laser tracking a disc.

00:08:52.840 --> 00:08:55.700
After undoing what I had begun, I began once
more.

00:08:55.710 --> 00:08:57.040
The screws are the enemy!

00:08:57.040 --> 00:08:58.040
Be gone!

00:08:58.040 --> 00:09:02.620
Or, be twisted counterclockwise sufficiently
until such time that they can be gone!

00:09:02.620 --> 00:09:07.500
Having gained unrestricted access to our prize,
I loosened the four screws holding it in place.

00:09:07.500 --> 00:09:09.120
But it wouldn’t budge.

00:09:09.120 --> 00:09:13.680
Yet more screws needed liberation, until finally
I had successfully removed the crown jewel

00:09:13.680 --> 00:09:15.960
from the temple of forbidden technology.

00:09:15.960 --> 00:09:17.600
And, uh, here it is.

00:09:17.600 --> 00:09:21.520
One of the big differences you’ll see between
this and the Philips mechanism is everything

00:09:21.520 --> 00:09:23.360
is much more integrated.

00:09:23.360 --> 00:09:27.780
In fact this one chip here, if the datasheet
I found is at all correct (link below) has

00:09:27.780 --> 00:09:33.120
32 k of RAM built in, does the EFM demodulation
as well as the CIRC error correction, handles

00:09:33.120 --> 00:09:37.870
the focus, tracking, and spindle motor (using
this chip as a driver to handle the power),

00:09:37.870 --> 00:09:41.870
extracts the subcode, provides a data output
for the DAC, and can even do such exotic things

00:09:41.870 --> 00:09:46.120
as double-speed playback and variable pitch
playback (assuming the player it’s inside

00:09:46.120 --> 00:09:48.020
has those functions enabled).

00:09:48.020 --> 00:09:49.450
Did I mention it’s bilingual?

00:09:49.450 --> 00:09:53.470
It’s interesting how just 4 years time managed
to integrate functions that are spread out

00:09:53.470 --> 00:09:58.030
via multiple chips on this Philips machine
into just this one chip on the board of the

00:09:58.030 --> 00:09:59.440
laser pickup.

00:09:59.440 --> 00:10:00.870
But that makes a lot of sense.

00:10:00.870 --> 00:10:04.480
See, this Philips board only works with
one mechanism.

00:10:04.480 --> 00:10:08.560
Because its driver components are part of
the main board design, anything but a Philips

00:10:08.560 --> 00:10:13.000
CDM4 pickup won’t work without at least
some modification.

00:10:13.000 --> 00:10:17.280
But for the Sony machine, this ribbon cable
is essentially just proving a power, ground,

00:10:17.280 --> 00:10:21.180
a communication bus to the CPU, and a datastream
for the DAC.

00:10:21.180 --> 00:10:25.530
This component can be completely redesigned,
and so long as it accepts the same inputs

00:10:25.530 --> 00:10:29.710
and produce the same outputs, the main board
really won’t care.

00:10:29.710 --> 00:10:33.730
This allows Sony to make production improvements
and even radical changes to this mechanism

00:10:33.730 --> 00:10:36.220
and how it works without disturbing the main
board.

00:10:36.220 --> 00:10:40.010
You can see how Sony played a modular approach
via the markings on the board.

00:10:40.010 --> 00:10:43.620
This one board was used in many different
players, with different components marked

00:10:43.620 --> 00:10:45.600
“FOR X player only”.

00:10:45.600 --> 00:10:48.290
By my count it was used in at least 5 players.

00:10:48.290 --> 00:10:51.650
Really, Sony’s markings are just a lot more
helpful.

00:10:51.650 --> 00:10:53.580
Wondering which one of these is the DAC?

00:10:53.580 --> 00:10:55.660
Hey look, it’s labeled right here.

00:10:55.660 --> 00:10:58.010
And you’ll find a PULSE chip on the bottom.

00:10:58.010 --> 00:11:01.430
I’m pretty sure these are 1-bit DACs, based
on some vague research.

00:11:01.430 --> 00:11:04.720
I’m sure someone will correct me if I’m
wrong, and I’m counting on it.

00:11:04.720 --> 00:11:07.940
And the fact that they’ve labeled these
jumper connections on the top of the board

00:11:07.940 --> 00:11:10.690
for what they are sure is helpful.

00:11:10.690 --> 00:11:11.640
That would’ve been nice,

00:11:11.640 --> 00:11:12.320
PHILIPS.

00:11:12.320 --> 00:11:15.120
Anyway, back to the laser because that’s
the real difference.

00:11:15.120 --> 00:11:17.030
First, gone is the swing-arm.

00:11:17.030 --> 00:11:21.880
Instead, the laser is mounted on a sled, which
moves linearly with the help of this rack

00:11:21.880 --> 00:11:23.270
and pinion drive.

00:11:23.270 --> 00:11:28.270
Now, you might suspect that this drive arrangement
does not have the precision to track the disc.

00:11:28.270 --> 00:11:32.090
The floating arm of the Philips system means
it can move subtly and precisely to track

00:11:32.090 --> 00:11:33.400
an off-center disc.

00:11:33.400 --> 00:11:37.920
But this brutal, inelegant plastic gear train
can’t possibly be used for the tracking.

00:11:37.920 --> 00:11:38.820
And it isn’t.

00:11:38.820 --> 00:11:42.640
Just like the Philips system, this lens is
floating and can move up and down.

00:11:42.650 --> 00:11:47.170
But unlike the philips system, this lens can
also move left and right.

00:11:47.170 --> 00:11:50.720
With the help of a diode I have laying around,
you can see that the laser can move quite

00:11:50.720 --> 00:11:52.230
a bit to either side.

00:11:52.230 --> 00:11:55.150
Let’s take off this plastic shroud to get
a better look.

00:11:55.150 --> 00:11:57.600
Ah this one is perfect for showing this.

00:11:57.600 --> 00:11:58.600
Yes.

00:11:58.600 --> 00:12:03.250
OK, I’m sure you can see these rectangular
coils surrounding a vertical structure.

00:12:03.250 --> 00:12:06.690
These coils will push the lens upward if a
voltage is passed through them.

00:12:06.690 --> 00:12:10.900
But if you look very carefully, you’ll also
see a pair of circular coils on the outer edges.

00:12:11.340 --> 00:12:15.100
When a voltage is applied to these, depending
on the polarity it will attract itself to

00:12:15.110 --> 00:12:19.680
these handy magnets placed on the sides, or
it will push itself away.

00:12:19.680 --> 00:12:23.800
This causes the lens to pivot either left
or right, depending on which way you slice it.

00:12:23.800 --> 00:12:28.320
The best part of this setup is that the lens
is completely independent of the sled that

00:12:28.330 --> 00:12:29.460
it’s riding on.

00:12:29.460 --> 00:12:32.960
If the sled moves a bit, it doesn’t care--it
will just react to that movement as though

00:12:32.960 --> 00:12:36.910
the track of pits it’s... tracking has moved
slightly.

00:12:36.910 --> 00:12:40.940
That means that the sled only has to move
in coarse steps, and the lens will take care

00:12:40.940 --> 00:12:41.940
of the rest.

00:12:41.940 --> 00:12:43.670
But that’s just one significant difference.

00:12:43.670 --> 00:12:46.160
The other is in the photodiode arrangement.

00:12:46.160 --> 00:12:47.260
You may have seen the term

00:12:47.260 --> 00:12:49.893
THREE-BEAM LASER
TRACKING

00:12:49.893 --> 00:12:52.080
before.  Or something like that.

00:12:52.080 --> 00:12:56.621
In these systems, rather than using only four
photodiodes like in the Philips system, we

00:12:56.621 --> 00:12:57.880
use six.

00:12:57.880 --> 00:12:59.580
In the middle are four cells arranged…

00:13:00.200 --> 00:13:02.500
actually
exactly like the Philips system.

00:13:02.510 --> 00:13:06.150
These are used in an identical fashion for
focus--an elliptical reflection indicates

00:13:06.150 --> 00:13:10.890
an out-of-focus beam, and its orientation
indicates if focus is near or far.

00:13:10.890 --> 00:13:15.300
But to correct tracking errors, two additional
diodes to either side are monitoring for an

00:13:15.300 --> 00:13:16.400
off-center beam.

00:13:16.400 --> 00:13:20.240
In these systems, a diffraction grating splits
the laser beam into three.

00:13:20.240 --> 00:13:23.790
In reality it splits it into infinite beams
of increasingly weak intensity as they deviate

00:13:23.790 --> 00:13:27.300
from the center but we only care about the
center beam and the two immediately surrounding it.

00:13:27.300 --> 00:13:31.760
Now, when tracked correctly, the outside beams
shouldn’t land on the stream of pits.

00:13:31.770 --> 00:13:36.020
They should instead land just outside of them,
which will thus cause them to reflect the

00:13:36.020 --> 00:13:37.670
featureless boundaries.

00:13:37.670 --> 00:13:42.030
The reason this works to track the stream
is that if the track deviates, suddenly one

00:13:42.030 --> 00:13:44.630
of the tracking diodes will start seeing the
datastream.

00:13:44.630 --> 00:13:48.810
That’s not supposed to happen, so the player
will react by nudging the lens in the direction

00:13:48.810 --> 00:13:50.890
of the activated tracking diode.

00:13:50.890 --> 00:13:55.190
This system is certainly more complex on the
hardware side, but given how simple the trigger

00:13:55.190 --> 00:14:00.230
is--no need to compare ratios, simply see
if a single thing is happening--it’s probably

00:14:00.230 --> 00:14:02.300
a lot easier to implement.

00:14:02.300 --> 00:14:06.820
However, this make me wonder if this system
is fundamentally worse at tracking discs.

00:14:06.820 --> 00:14:11.010
See, I’ve heard anecdotal evidence that
the original Philips single-beam tracking

00:14:11.010 --> 00:14:14.700
is superior at tracking scratched or damaged
discs.

00:14:14.700 --> 00:14:19.290
And with the knowledge of how the three-beam
tracking works, that kind of makes sense.

00:14:19.290 --> 00:14:22.630
Imagine a scratch appears just next to the
datastream.

00:14:22.630 --> 00:14:26.950
If that scratch is bad enough, it might bend
one of the tracking beams towards the raw

00:14:26.950 --> 00:14:31.570
datastream, causing the lens to deflect for
a defect that isn’t really there.

00:14:31.570 --> 00:14:36.180
With the Philips system, it wouldn’t see
this scratch because it only has the one scanning beam.

00:14:36.180 --> 00:14:38.200
It would ignore it and thus be unaffected.

00:14:38.200 --> 00:14:41.900
Now of course, these conventional systems
aren’t remarkably inferior.

00:14:41.910 --> 00:14:43.730
They can tolerate scratches pretty well, too.

00:14:43.730 --> 00:14:48.890
But it makes me wonder if there is truth to
the anecdote that the swing-arm servo, Philips

00:14:48.890 --> 00:14:51.730
single beam tracking whatever is indeed superior.

00:14:51.730 --> 00:14:54.460
One last thing before we answer why this became
the standard.

00:14:54.460 --> 00:14:57.880
I’m not afraid to tear this apart and find
what inside.

00:14:57.880 --> 00:15:02.020
So first, you can see the actual laser diode
here as opposed to the Philips machine where

00:15:02.020 --> 00:15:03.500
its housed in this plastic.

00:15:03.500 --> 00:15:08.140
This, like in the Philips setup, projects
the light sideways where it hits a prism,

00:15:08.140 --> 00:15:12.700
get shot up at the disc surface (which the
floating lens helps to keep tracked and focused)

00:15:12.700 --> 00:15:16.130
and it gets reflected back, goes straight
through the prism, and lands on this little

00:15:16.130 --> 00:15:17.580
chip here.

00:15:17.580 --> 00:15:21.860
Now this is interesting, because on this machine
we can actually see the backside of the photodiode

00:15:21.860 --> 00:15:25.840
array, unlike on the Philips machine where
we could only see the board.

00:15:25.840 --> 00:15:30.440
If I keep going to remove the lens and get
everything apart, you can see the prism in here.

00:15:30.440 --> 00:15:34.420
Take a look, it does a great job of reflecting
light at a 90 degree angle.

00:15:34.420 --> 00:15:36.880
And now, let’s see what we can see of the
photodiodes.

00:15:37.340 --> 00:15:40.680
Well, don’t get your hopes up--they are
pretty much microscopic.

00:15:40.680 --> 00:15:44.940
To be fair, we are dealing with laser light
focused on microscopic pits, so its

00:15:44.940 --> 00:15:47.440
stands to reason that these are probably pretty
small.

00:15:47.440 --> 00:15:49.340
I don’t have a microscope handy

00:15:49.340 --> 00:15:50.140
--yet--

00:15:50.140 --> 00:15:52.920
but I do have my ridiculous macro lens setup so

00:15:52.930 --> 00:15:55.380
I could get these extreme closeups.

00:15:55.380 --> 00:15:59.840
You can definitely see that these are three
clusters, with one used for data capture and

00:15:59.840 --> 00:16:03.330
focus control, and the two to either side
for tracking.

00:16:03.330 --> 00:16:07.230
To show how small this is, I’ve placed a
dime next to it for scale.

00:16:07.230 --> 00:16:10.750
For a more International perspective, how
about a micro SD card?

00:16:10.750 --> 00:16:13.720
And just for fun, here’s the point of a
safety pin.

00:16:13.720 --> 00:16:15.260
These really are tiny.

00:16:15.270 --> 00:16:19.510
So now, why did Sony’s floating lens and
stepping platform become the de facto method

00:16:19.510 --> 00:16:21.430
of reading optical discs?

00:16:21.430 --> 00:16:26.650
With more photodiodes and greater mechanical
complexity, you might think it to be inefficient.

00:16:26.650 --> 00:16:31.720
Well it may be inefficient in some ways, but
in most ways it’s in fact much more efficient.

00:16:31.720 --> 00:16:34.170
First, let’s take size.

00:16:34.170 --> 00:16:38.020
If we look at the Philips system, you’ll
notice that the entire laser pickup is essentially

00:16:38.020 --> 00:16:39.930
behind the disc.

00:16:39.930 --> 00:16:44.400
The pivot point for the swing arm has to be
farther to the rear of the player, otherwise

00:16:44.400 --> 00:16:49.100
it would simply rotate around the circumference
of the disc, and that wouldn’t be very helpful.

00:16:49.100 --> 00:16:52.420
This off the bat limited its potential to
miniaturize.

00:16:52.420 --> 00:16:55.020
And boy, did this get smaller quickly.

00:16:55.020 --> 00:16:57.720
Portable CD players burst on the scene in
1984.

00:16:57.730 --> 00:17:01.880
And they would only continue to get smaller,
with one CD player featured by Techmoan actually

00:17:01.880 --> 00:17:04.240
being smaller than a CD.

00:17:04.240 --> 00:17:08.209
And that was in 1988, the same year as this
Philips machine was made!

00:17:08.209 --> 00:17:11.470
In fact, this very mechanism is larger than
it needs to be.

00:17:11.470 --> 00:17:16.370
If these parts were shifted downward, it could
be no longer than the distance here.

00:17:16.370 --> 00:17:18.440
And this laser platform could get smaller.

00:17:18.540 --> 00:17:19.160
And smaller.

00:17:19.340 --> 00:17:19.840
And smaller.

00:17:20.260 --> 00:17:25.380
Until eventually you’re into modern slim
drives for PCs which have the tiniest of lenses

00:17:25.380 --> 00:17:27.220
and thinnest of assemblies.

00:17:27.220 --> 00:17:32.150
And the PC market would only continue to make
this Philips system harder to justify.

00:17:32.150 --> 00:17:37.130
In fact, one fatal flaw of the Philips reader
would make sure it fizzled into obscurity:

00:17:37.130 --> 00:17:38.250
its mass.

00:17:38.250 --> 00:17:42.070
Imagine in a portable player, which can be
set in any position, or even moved around

00:17:42.070 --> 00:17:45.440
while playing, that it had this swinging laser
pickup.

00:17:45.440 --> 00:17:50.130
It’s so loosey-goosey that it’s doubtful
it will track very well.

00:17:50.130 --> 00:17:53.270
And its mass also limits how quickly it can
move.

00:17:53.270 --> 00:17:58.290
When CD-ROM drives appeared, and getting faster
and faster and faster, the laser would have

00:17:58.290 --> 00:18:02.910
to be able to wiggle itself back and forth
at frequencies in excess of 10 kilohertz.

00:18:02.910 --> 00:18:06.780
That’s easy if you’re just wiggling at
small piece of plastic back and forth, but

00:18:06.780 --> 00:18:10.320
a lot harder if you’re wiggling this big
swinging thing.

00:18:10.320 --> 00:18:14.940
And so, Sony’s three-beam laser tracking
would be miniaturized and improved over the years.

00:18:14.940 --> 00:18:19.260
Developments in anti-skip functionality meant
that CD walkmen could be tossed around without

00:18:19.260 --> 00:18:23.530
fear of skipping or glitches, or getting stuck
repeating a bit of the track.

00:18:23.530 --> 00:18:27.780
That was accomplished simply by reading the
disc at faster than normal speeds, creating

00:18:27.780 --> 00:18:31.860
an intermediary data buffer between the disc
and the processor.

00:18:31.860 --> 00:18:35.970
The DAC could have upwards of 40 seconds at
its disposal for the laser to get back on

00:18:35.970 --> 00:18:37.820
track and resume the datastream.

00:18:37.820 --> 00:18:42.620
Which, thanks to the timecode, is easy to
piece together if a problem does arise, thus

00:18:42.620 --> 00:18:44.740
eliminating the effects of skipping.

00:18:44.740 --> 00:18:48.700
And of course, later on the wavelength of
light would change to red, packing the pits

00:18:48.700 --> 00:18:52.680
closer together and creating the MUSE high
definition laserdisc.

00:18:52.680 --> 00:18:56.680
And then later the DVD, an obscure digital
video format you might have heard of.

00:18:56.680 --> 00:19:01.240
Later we’d say, enough with the red, in
with the blue and we’d be packing upwards

00:19:01.250 --> 00:19:06.650
of 25 gigabytes onto a slightly different
silver plastic thing, after we briefly reenacted

00:19:06.650 --> 00:19:09.180
the videotape format wars of the 1980’s.

00:19:09.180 --> 00:19:10.180
Because why not.

00:19:10.380 --> 00:19:11.860
But we’re getting ahead of ourselves.

00:19:11.860 --> 00:19:13.780
There’s lots more to talk about.

00:19:13.780 --> 00:19:18.270
When we next check in on optical disc technology,
we’ll discuss the Yellow Book, the standard

00:19:18.270 --> 00:19:22.210
published in 1988 that defines the specifications
of the CD-ROM.

00:19:22.210 --> 00:19:27.820
Then we’ll talk about some writable disc
technologies like CD-R and CD-RW, and then,

00:19:27.820 --> 00:19:31.990
if interests persists, we might dig a little
deeper into the DVD and some of the follies

00:19:31.990 --> 00:19:33.450
of its development.

00:19:33.450 --> 00:19:39.430
Such as how this logo mysteriously features
a disc for a solid state memory format.

00:19:39.430 --> 00:19:40.430
Hmm.

00:19:40.430 --> 00:19:43.010
Thanks for watching, I hope you enjoyed the
video!

00:19:43.010 --> 00:19:46.250
If you missed the first two videos on the
Compact Disc, you can check them out in the

00:19:46.250 --> 00:19:49.740
playlist on Digital Sound that will quietly
pop up above me.

00:19:50.020 --> 00:19:51.760
So serene.

00:19:52.060 --> 00:19:55.900
As always, a great big thank you to every
who supports this channel on Patreon, especially

00:19:55.910 --> 00:19:58.720
the fine folks that are scrolling up your
screen.

00:19:58.720 --> 00:20:02.400
If you’re interested in pledging some support
to the channel to help it grow, please check

00:20:02.400 --> 00:20:03.670
out my Patreon page.

00:20:03.670 --> 00:20:06.160
Thanks for your consideration, and I’ll
see you next time!

00:20:06.340 --> 00:20:08.060
[ everyone’s favorite, gloriously tacky
music plays ]

00:20:08.640 --> 00:20:13.540
Hey one thing, my most viral video to date
was about how Sony used the tracking servos

00:20:13.550 --> 00:20:18.230
in a CD player to create the copy protection
scheme in the original Playstation.

00:20:18.230 --> 00:20:22.270
You might want to check that video out, but
be warned that there are a number of annoying

00:20:22.270 --> 00:20:25.470
things about the flow of information over
there.

00:20:25.470 --> 00:20:29.440
It goes into a lot of detail about the PlayStation
and some of its history, and the comments

00:20:29.440 --> 00:20:35.480
on there (if you dare to look at them) indicate
that many people were… not happy about it.

00:20:35.480 --> 00:20:38.720
So if you haven’t seen it, check it out,
and apologies in advance.

00:20:38.720 --> 00:20:40.860
Ha, and another thing!

00:20:40.860 --> 00:20:44.950
So if you didn’t know, there’s a second
channel - Technology Connections 2 - where

00:20:44.950 --> 00:20:47.750
I upload weird stuff from time to time.

00:20:47.750 --> 00:20:52.559
And I just uploaded a video of a couple of
CD players that are very similar but have

00:20:52.559 --> 00:20:53.559
some unique differences.

00:20:53.560 --> 00:20:57.860
So if you want to check if out, please feel
free to click the link down below or up above.