WEBVTT

00:00:00.240 --> 00:00:05.839
Everyone knows that in 1995, Toy Story

00:00:03.280 --> 00:00:10.960
made history as the first fully computer animated film. But what most people

00:00:07.919 --> 00:00:12.960
don't know is that a full year earlier,

00:00:10.960 --> 00:00:18.640
a little Canadian studio called Mainframe Entertainment released the

00:00:15.040 --> 00:00:20.880
first fully computer animated TV show

00:00:18.640 --> 00:00:24.640
reboot. It was quirky. It was heartwarming. And it was full of

00:00:22.640 --> 00:00:29.679
adventure with each episode exploring exciting new worlds. So then why hasn't

00:00:27.680 --> 00:00:34.320
anybody heard of it? Well, being Canadian, it wasn't as broadly

00:00:31.679 --> 00:00:38.640
distributed. And making matters worse, due to poor media preservation practices

00:00:36.559 --> 00:00:44.399
at the time, the original quality masters were thought to be lost for many

00:00:41.520 --> 00:00:50.079
years. So, in spite of an official release existing, for decades, the best

00:00:47.680 --> 00:00:57.120
way to enjoy the show has been a pirated Russian DVD. Reboot and its fans

00:00:53.840 --> 00:00:59.199
deserved so much better, and they're

00:00:57.120 --> 00:01:04.159
finally getting it. partly thanks to the sponsor of this video, Kioxia. Data

00:01:01.760 --> 00:01:08.960
integrity is their passion, and this project was only possible thanks to

00:01:06.400 --> 00:01:13.920
their support. But what data? Reboot only found a limited audience during its

00:01:10.960 --> 00:01:17.759
TV run and the only means that mainframe had of reading those original master

00:01:16.159 --> 00:01:23.840
tapes with the full quality digital files was discarded years ago.

00:01:20.799 --> 00:01:25.680
Thankfully, in 2023, Jacob and Raquel,

00:01:23.840 --> 00:01:30.960
the creators of the Reboot Rewind documentary, got a brilliant idea of

00:01:28.640 --> 00:01:35.280
asking Mainframe, pretty please, very nicely, if they could try and do

00:01:33.360 --> 00:01:40.960
something with those tapes, cuz you know how companies normally do that, right?

00:01:38.079 --> 00:01:47.920
Except Mainframe said yes, lending them all of the master tapes for reboot. All

00:01:44.479 --> 00:01:51.600
of it. So, huzzah, problem solved,

00:01:47.920 --> 00:01:54.000
right? Well, no. See, it turns out that

00:01:51.600 --> 00:01:58.159
getting any files off of these thick boys is pretty much like pulling the

00:01:56.560 --> 00:02:04.719
data logs from a crashed alien spaceship. So, the Reboot Rewind project

00:02:01.360 --> 00:02:07.119
was stuck with no plan, no budget, and

00:02:04.719 --> 00:02:11.760
no idea where to even look for a machine that could read these tapes. They threw

00:02:09.280 --> 00:02:16.720
up a desperate Facebook post. Does anyone have a D1 deck? As luck would

00:02:14.800 --> 00:02:23.120
have it, someone over in Germany had not one, not two, but three up for auction.

00:02:21.520 --> 00:02:29.440
Huzzah. Yeah. If you thought they were going to

00:02:25.680 --> 00:02:31.440
be cheap and not broken, that's where we

00:02:29.440 --> 00:02:35.519
came in. When our community flagged the incredible preservation work that Reboot

00:02:33.440 --> 00:02:39.920
Rewind was doing, we partnered up with Kyioia to cover the cost of both the

00:02:37.599 --> 00:02:44.160
machines and some of the rare parts and supplies that would be needed to get

00:02:41.519 --> 00:02:49.680
them up and running. And now one of them is and you can actually see the results

00:02:46.480 --> 00:02:52.720
for yourself on YouTube. Reboot is a

00:02:49.680 --> 00:02:54.560
little older and it does show. But hey,

00:02:52.720 --> 00:02:59.519
at least it's no longer held back by mid90s encoding quality and it looks

00:02:57.200 --> 00:03:04.239
beautiful in its own way. As for the other machines, well, they're kind of

00:03:02.080 --> 00:03:09.519
nice to haves now that the first one is working, but because the footage needs

00:03:06.640 --> 00:03:13.760
to be captured rather than just dragged and dropped, having more machines will

00:03:11.680 --> 00:03:19.680
help the process go faster. And with the tapes continuing to deteriorate by the

00:03:16.480 --> 00:03:23.120
hour the clock is ticking. So, our goal

00:03:19.680 --> 00:03:25.200
for today is to pop open this absolute

00:03:23.120 --> 00:03:29.440
treasure of cuttingedge technology from a bygone era and maybe get it working.

00:03:28.159 --> 00:03:34.640
If you want to learn more about the Reboot Rewind team, by the way, they've done a series of documentaries that you

00:03:32.640 --> 00:03:37.920
can check out at the link down below. We, on the other hand, are going to get

00:03:36.400 --> 00:03:43.599
right into the tech tips and talk about a tape format that I'd never even heard

00:03:40.319 --> 00:03:46.799
of before this project, D1. Imagine

00:03:43.599 --> 00:03:49.360
this. It's 1986 and the executives at

00:03:46.799 --> 00:03:55.360
Sony, riding a bubble economy, have told their worldclass engineers that money is

00:03:52.319 --> 00:03:58.080
no object. Make the absolute best video

00:03:55.360 --> 00:04:05.599
format in the world. That's what this is. It's like the McLaren F1 of8s video

00:04:01.760 --> 00:04:08.480
technology. D1 physically is the largest

00:04:05.599 --> 00:04:14.640
video cassette tape ever made. And like DVHS, which would arrive in home cinemas

00:04:11.280 --> 00:04:17.840
over a decade later, it is purely

00:04:14.640 --> 00:04:20.479
digital. But unlike DVHS, it was

00:04:17.840 --> 00:04:26.160
intended for professional use. So it was completely uncompressed. No codec, just

00:04:23.919 --> 00:04:32.240
bitmapped images written directly to tape. That is why Kyokia's help was so

00:04:29.840 --> 00:04:37.280
critical. Remember guys, I said we needed to capture this uncompressed

00:04:34.560 --> 00:04:42.639
video. So, their CM7R drives were an essential part of the entire workflow

00:04:39.120 --> 00:04:44.800
because they can easily handle the 40 GB

00:04:42.639 --> 00:04:50.240
of raw video from each of these 30-year-old tapes safely and without

00:04:48.240 --> 00:04:54.639
bottlenecks. And that's essential because with the state of these tapes,

00:04:51.919 --> 00:04:58.880
they might literally only get one shot at a good transfer. By the way, if you

00:04:56.560 --> 00:05:05.040
thought the tapes were big, the decks are on a whole other level. These

00:05:01.280 --> 00:05:08.720
behemoths are 250 lb of metal and glass,

00:05:05.040 --> 00:05:12.479
draw 650 W of power, and inside each

00:05:08.720 --> 00:05:15.199
one, you will find six or seven Intel

00:05:12.479 --> 00:05:21.120
80186 processors communicating on an internal Ethernet network. This is no

00:05:18.000 --> 00:05:23.840
compromise design at its absolute best,

00:05:21.120 --> 00:05:28.880
but almost no one has ever seen one because to buy one new adjusted for

00:05:26.160 --> 00:05:34.479
inflation would be a cool third of a million. So, I guess the 15 grand we

00:05:32.560 --> 00:05:38.000
paid was a deal even though they were all broken. I should probably

00:05:35.919 --> 00:05:42.720
acknowledge, by the way, that we are not on our normal set. Welcome to

00:05:40.400 --> 00:05:47.199
Disappearing Inc., a local company that's dedicated to not only rescuing

00:05:45.039 --> 00:05:52.560
classic computers, but also repairing and restoring them to their former

00:05:48.880 --> 00:05:54.479
glory. And the place really is glorious,

00:05:52.560 --> 00:06:01.280
isn't it? Almost as glorious as the skill set of the man who runs it, Mark.

00:05:58.080 --> 00:06:04.720
Hello. At the peak of the D1's run,

00:06:01.280 --> 00:06:06.800
there were maybe 100 people who were

00:06:04.720 --> 00:06:11.199
qualified to work on these machines. 30 years later, he might be the only one

00:06:09.199 --> 00:06:15.759
left who can fix them and definitely the only one under the age of 40. How'd you

00:06:13.039 --> 00:06:19.039
um how'd you learn that? Uh local community college course.

00:06:16.960 --> 00:06:22.880
>> No, no, no, no. I was accidentally training myself for this job from a

00:06:21.039 --> 00:06:25.440
young age because I grew up with a fascination with video equipment, but I

00:06:24.319 --> 00:06:31.360
couldn't afford any of this stuff. So, everything I got was broken. It was all castoffs and old computers generally no

00:06:29.840 --> 00:06:36.960
one wants to talk about them at HP and Dell and things like that. So, I just kind of had to figure it all out. So

00:06:34.400 --> 00:06:40.720
sourcing some test sacrificial tapes and recapping the shoe box size power

00:06:38.560 --> 00:06:44.400
supplies back here was no problem for Mark. But what was a problem was the

00:06:43.280 --> 00:06:49.919
brakes. >> Yeah. And it didn't look like it originally. We got these errors from the

00:06:47.520 --> 00:06:54.479
machines about the clock signal being somehow too fast and spent way too long

00:06:52.880 --> 00:07:00.720
trying to figure that out. And finally at the end of it discovered that the same board that sets up the clocks is

00:06:58.400 --> 00:07:02.639
also the one that runs the brakes. So when it saw something was wrong with

00:07:02.240 --> 00:07:07.039
him, >> it just stopped working. >> Yeah. It just gave up. >> So if you'd known the brake was the

00:07:06.319 --> 00:07:15.039
problem, you could >> have saved months. >> And that was exactly in time for another

00:07:11.759 --> 00:07:17.840
huge problem to raise its head or heads

00:07:15.039 --> 00:07:20.960
as it were. The tape heads to be precise.

00:07:18.960 --> 00:07:24.880
>> For you younglings out there, the tape head is the part of the machine that

00:07:22.880 --> 00:07:30.639
actually reads the zeros and the ones from the tapes. And the D1's heads are

00:07:27.360 --> 00:07:34.800
remarkably precise. So much so that they

00:07:30.639 --> 00:07:37.039
are only rated for 500 hours of use. 500

00:07:34.800 --> 00:07:42.720
hours. I thought this is supposed to be a professional standard. After that

00:07:39.360 --> 00:07:45.199
time, they're intended to be replaced.

00:07:42.720 --> 00:07:53.120
And the most workingest of the three machines we bought. Yeah, that tape head

00:07:48.080 --> 00:07:54.560
was already at 650 hours. So then what

00:07:53.120 --> 00:08:00.240
did we do at that point then? >> Well, uh, we panicked and we despared.

00:07:58.560 --> 00:08:05.280
And then a miracle happened. Herbert Han, a man in Germany who has been

00:08:03.039 --> 00:08:08.879
incredibly helpful to us, said he knew a technician from back in the day. He was

00:08:07.120 --> 00:08:15.120
an elderly gentleman now, long retired, but he might still have a couple of spare parts for us. And boy, did he. In

00:08:12.319 --> 00:08:19.039
fact, he had not one but two brand new, never used BTS refurbished heads still

00:08:17.520 --> 00:08:27.080
in his garage. >> Wow. >> It was just a cool $8,000 to buy them.

00:08:22.240 --> 00:08:27.080
>> Right. but wasn't our money.

00:08:29.680 --> 00:08:36.479
>> That was actually one of the points in this project where it could have all

00:08:33.760 --> 00:08:41.279
gone completely south because those parts, my understanding is ended up all

00:08:39.360 --> 00:08:46.800
getting kind of parted out and shipped off shortly after we made our offer. So,

00:08:43.919 --> 00:08:50.880
if we hadn't gone in and taken them, >> yeah, within weeks those were sold off

00:08:48.880 --> 00:08:53.839
to other companies who still have to maintain these crazy things. And we've

00:08:52.640 --> 00:08:57.519
spoken to a couple of those companies and they're not fond of sharing parts.

00:08:56.399 --> 00:09:01.200
>> Makes sense. >> So, we had one shot and we took it and

00:08:59.680 --> 00:09:06.240
it's all thanks to you guys. >> Well, we we do our best. As far as I'm

00:09:03.760 --> 00:09:10.880
concerned, uh if we can get the show restored, then mission accomplished.

00:09:08.959 --> 00:09:14.399
>> Well, I am so happy to say that it's because of these opportunities. We got

00:09:13.279 --> 00:09:16.959
some big news to share. >> Oh, is this Are we dropping something on

00:09:16.800 --> 00:09:21.440
you? >> No, wait for later in the video. >> What? >> Yep.

00:09:20.320 --> 00:09:26.880
>> Wait, you're not going to tell me? >> No. No. >> Okay. I guess we're getting to that

00:09:23.920 --> 00:09:29.120
later. For now, we're going to try and repair the brake on this one.

00:09:28.160 --> 00:09:34.240
>> We are indeed. >> Okay, let's do it. >> So, are we doing this for real?

00:09:32.000 --> 00:09:38.560
>> Yeah. Yeah. So, we'll just put a super cut down version in the YouTube video,

00:09:36.240 --> 00:09:41.839
but at LMG.gg/flatplane, people can get the whole video of us

00:09:40.160 --> 00:09:46.880
going through and giving it our best shot. >> Well, in that case, let's do it. These

00:09:44.000 --> 00:09:52.120
machines need a lot of service, and you can tell that the Germans put some

00:09:48.320 --> 00:09:52.120
thought into it. So,

00:09:53.440 --> 00:09:58.480
It goes like that. >> So, the Germans can make a thing

00:09:56.640 --> 00:10:01.200
serviceable when they want to. >> Someone should tell BMW. The work I'm

00:10:00.160 --> 00:10:06.399
going to be doing is mostly from the top. Okay. >> So, we're not actually working under here. I do need you to warn you, there's

00:10:04.880 --> 00:10:10.320
a chance the power supply explodes while we're working on it. >> Sick. >> Cuz um haven't rebuilt this one yet.

00:10:09.839 --> 00:10:15.920
>> Sick. >> So, if smoke starts pouring out the back, that's good content.

00:10:13.440 --> 00:10:20.800
>> This guy knows YouTube. Right. >> Big thing on these decks is that because

00:10:18.240 --> 00:10:24.959
the tape is moving so fast, there has to be a really good, reliable way to speed

00:10:23.279 --> 00:10:29.279
it up and slow it down and stop it on a moment's notice. What if something goes wrong with the machine? What if the

00:10:27.440 --> 00:10:32.000
power goes out? What if you just get to the end of your scene and want to stop

00:10:30.560 --> 00:10:36.560
it? So, they've got these brakes like a car that like clamp down on the spools

00:10:34.240 --> 00:10:38.959
and stop the tape and the machine checks every time it turns on to make sure

00:10:37.680 --> 00:10:42.560
those are working. Like, it spins it up and it clamps it on

00:10:40.720 --> 00:10:45.600
>> and they just wear out like anything else, >> right? The brakes do.

00:10:44.000 --> 00:10:49.760
>> They do. They're mechanical. So, every time you use them, plus just when you let them sit for 30 years, there's these

00:10:48.320 --> 00:10:54.079
big springs inside, they kind of get loose. So, we're going to go in there and do a very non-German repair to these

00:10:52.640 --> 00:10:58.800
things. It involves grabbing some springs with a pair of pliers and like kind of twisting them a bit and, you

00:10:57.360 --> 00:11:02.320
know, causing some people to roll in their graves and putting a little fresh

00:11:00.480 --> 00:11:04.800
grease in there and crossing our fingers and with a little bit of luck, it'll get

00:11:04.079 --> 00:11:09.120
that error cleared. >> What are you doing right now? >> We're borrowing the control panel that

00:11:07.440 --> 00:11:12.640
works, which is to say, we're returning it to the machine we borrowed it from before.

00:11:10.800 --> 00:11:15.680
>> Sick. So, we have two out of three. Unfortunately, as you saw inside,

00:11:14.399 --> 00:11:21.040
there's a lot of custom chips in these things. And one of the custom chips, and one of the control panels has gone dead.

00:11:19.120 --> 00:11:24.320
So, we have three machines, but only two of them that we can actually talk to at

00:11:22.399 --> 00:11:29.680
the moment. So, we simply put it back on over here. And this machine has what

00:11:27.360 --> 00:11:33.200
looks like a serial port, but is in fact something incredibly German and

00:11:30.959 --> 00:11:35.920
proprietary. Uh, and the control panel will go, "Oh, hey, you're a 500."

00:11:35.120 --> 00:11:40.000
>> Okay. >> Some of these people did go on to work at BMW then.

00:11:38.720 --> 00:11:45.440
>> Yeah. Yeah. I suspect >> this I guess goes here. >> If you can get it in there for me,

00:11:43.839 --> 00:11:50.320
that'd be great. >> Small hands power activate form of small

00:11:49.600 --> 00:11:54.399
dainty hands. >> That's perfect. I got these big man hands. I can't get that thing in there

00:11:52.800 --> 00:11:57.920
without 5 minutes of swearing. Fantastic. Look at that. So, that's

00:11:56.160 --> 00:12:03.440
enough that we can control this machine. >> I helped. >> Yeah. Don't let anyone tell you Lionus

00:12:01.360 --> 00:12:07.680
was instrumental in this project. So, the brakes on this thing are deep inside

00:12:05.440 --> 00:12:09.600
here. They're underneath these sort of brown spindle looking things,

00:12:09.200 --> 00:12:14.240
>> right? >> Because they're the spindles, you see? Right. >> And we need to get in there. And the

00:12:12.639 --> 00:12:17.279
only way we can do it is if we take the entire part you put the tape into out.

00:12:16.880 --> 00:12:23.120
>> Brilliant. >> So, the good news is it's surprisingly serviceable.

00:12:20.639 --> 00:12:25.040
>> We're going to once more raise this thing up

00:12:23.760 --> 00:12:31.920
>> and then we're going to start unscrewing things. We need to loosen a couple of

00:12:27.760 --> 00:12:33.680
screws that hold the mechanism together.

00:12:31.920 --> 00:12:40.399
Oh, this is the real pain in the ass one cuz when this raises up, it kind of

00:12:35.760 --> 00:12:41.279
blocks this. So, I need to kind of brace

00:12:40.399 --> 00:12:48.639
the mechanism a bit. >> Oh, well, I can hold it. >> Um, it's if you do that, it might break.

00:12:45.600 --> 00:12:49.839
>> I need you to hold on to this as tight

00:12:48.639 --> 00:12:53.279
as you can and try and prevent it from turning. And I'll do the same over here.

00:12:51.839 --> 00:12:55.839
And then we get to do the opposite. There it goes. When we go to put this

00:12:54.880 --> 00:13:02.120
thing back together. >> And I helped. I think if we're very

00:12:59.120 --> 00:13:02.120
careful,

00:13:04.000 --> 00:13:08.480
>> it feels like it should run in a clean

00:13:07.760 --> 00:13:13.920
room, >> right? >> Like it looks like something that should

00:13:11.120 --> 00:13:18.240
only work in a completely sterile, perfect environment, but like you had it

00:13:16.160 --> 00:13:21.600
just open there and it was just working. I don't get it. >> Well, the good news is that once the

00:13:20.160 --> 00:13:24.639
thing gets spinning fast enough, it kind of flings the dust out of the way.

00:13:23.279 --> 00:13:28.720
>> Oh, genius. >> Yeah. So, in a way, it's not quite as

00:13:27.440 --> 00:13:33.200
critical as you think. The other half, though, is you just clean the thing constantly as you're using it because

00:13:31.440 --> 00:13:37.120
more than the dust that comes in from the outside is the bits of tape that get

00:13:35.680 --> 00:13:40.880
flung off every time you play the thing. So, I spend more time cleaning brown

00:13:38.880 --> 00:13:43.519
junk out from the inside than I do about the dust in the room.

00:13:42.160 --> 00:13:47.920
>> All right. So, brakes. >> Yes, they are under here. And you can

00:13:46.399 --> 00:13:52.399
feel if you try and turn that, it turns easier one way than the other. >> Oh, yeah. Yeah. Yeah. It's super easy

00:13:50.720 --> 00:13:55.519
clock counterclockwise and then it's really tight clockwise.

00:13:53.760 --> 00:14:00.160
>> Yeah. And it should be even tighter than that. So, it should really clamp it

00:13:58.639 --> 00:14:03.440
hard. But it's not doing that anymore. So, we need to go in there and make some

00:14:02.240 --> 00:14:08.320
adjustments. >> Okay. That's the spring we're adjusting.

00:14:06.639 --> 00:14:13.120
>> That underneath this little piece of foam or felt, I should say, that is the

00:14:11.760 --> 00:14:17.680
brake. >> Seriously? >> Yep. >> Well, how the hell are we supposed to

00:14:16.079 --> 00:14:21.760
make adjustments to that with any degree of precision? >> We're not. We're going to use pliers.

00:14:20.079 --> 00:14:26.000
So, this here is a brake. Doesn't look like much, but this is the entirety of

00:14:24.560 --> 00:14:28.560
the core mechanism. This is the part that actually grabs it together. Pulls

00:14:28.000 --> 00:14:32.399
it apart. >> Yeah. So, we go grab. >> Yep. >> Brakes are on.

00:14:31.199 --> 00:14:36.800
>> We go apart. >> Breaks her off. >> Breaks her off. Easy peasy.

00:14:34.720 --> 00:14:39.920
>> Right. Unfortunately, it's not squeezing it quite hard enough. And when it does,

00:14:38.480 --> 00:14:45.040
sometimes it gets stuck. And oh yeah, there was originally some sort of grease in here. I have no idea what, but it's

00:14:42.880 --> 00:14:50.880
kind of sticky now. So, we're going to replace that. So, we just tighten the

00:14:48.079 --> 00:14:55.360
whole thing like that just a little bit. We don't need much. Just a little

00:14:52.480 --> 00:14:57.760
dabble, do you? >> Thank you.

00:14:56.000 --> 00:15:02.240
>> I knew I was here for a reason. >> And there's a little piece of felt whose

00:15:00.639 --> 00:15:04.480
job is basically to stop that grease from getting thrown out into the machine.

00:15:04.079 --> 00:15:10.000
>> Mhm. >> You know, if it starts flinging around at high speeds, it'll go in here instead

00:15:07.600 --> 00:15:13.360
of onto a tape. Then we need our magic piece of brass that holds everything

00:15:11.920 --> 00:15:17.600
together. And I look at the other one to figure out how I have to reassemble it.

00:15:15.199 --> 00:15:20.800
See, real technicians still have to look at the instructions sometimes, too. But

00:15:19.040 --> 00:15:24.079
in lie of that, we leave one of them assembled so we can look at it and see

00:15:22.160 --> 00:15:27.920
how it goes back together. That's a real important tip. >> A tech tip even you might say.

00:15:25.920 --> 00:15:29.839
>> You might say you would see how it engages sooner. It pulls a little

00:15:29.360 --> 00:15:34.560
farther. >> All right. >> And so Oh, yeah.

00:15:32.399 --> 00:15:36.880
>> Brakes on, brakes off here. Here we go. >> So brakes on.

00:15:35.680 --> 00:15:40.399
>> See how tight that is now? >> Yeah. So that's pull that >> here. You can actually really Oh, no.

00:15:39.360 --> 00:15:46.160
No. Here. Don't pull it. Don't pull it yet. This is really great cuz you can

00:15:42.480 --> 00:15:48.560
really see that that is not moving. Get

00:15:46.160 --> 00:15:54.320
the rod. Now she spins. Oh, and she stops.

00:15:52.240 --> 00:15:57.040
>> We fixed the thing. Lus, can we get a vending machine to work?

00:15:55.440 --> 00:16:02.320
>> You want a vending machine? >> Yeah. Yeah. >> Oh, just like for the fun of it kind of

00:15:59.360 --> 00:16:07.279
thing. >> I mean, they're not expensive.

00:16:04.240 --> 00:16:09.680
>> Yo, root beer.

00:16:07.279 --> 00:16:14.399
>> This is a good surprise. Now that you've seen some of it for yourself, you may

00:16:11.440 --> 00:16:18.000
have a concept of an idea of how hard Mark has been working not only to get

00:16:16.000 --> 00:16:22.880
these machines up and running, but to keep them that way throughout the

00:16:19.680 --> 00:16:24.399
capture process. And the most fun part,

00:16:22.880 --> 00:16:30.079
remember how I said the condition of the tapes ain't too great? Yeah, that was an

00:16:27.839 --> 00:16:34.959
understatement. These are suffering from something called sticky shed syndrome,

00:16:32.959 --> 00:16:39.920
which basically means that the glue holding the magnetic particles to the

00:16:36.880 --> 00:16:41.839
tape is literally turning into gunk with

00:16:39.920 --> 00:16:45.519
every passing day. So, if you guys want to support this project, make sure you

00:16:43.600 --> 00:16:49.040
donate to their cause at the link in the description. Are you guys selling these

00:16:47.279 --> 00:16:54.079
shirts, by the way? Cuz they're super cool. I wish. Oh, okay. Apparently,

00:16:51.600 --> 00:16:58.399
there was some, you know, IP complications. I guess you'll just have

00:16:55.680 --> 00:17:02.959
to buy our shirts then at ltstore.com. Now, our focus today has been on the D1

00:17:00.560 --> 00:17:07.120
decks and on the tapes, but that's just part of the puzzle and it would be a

00:17:04.799 --> 00:17:12.480
crime to not also acknowledge the work that this man has done. Thanks to Brian,

00:17:10.160 --> 00:17:17.120
the team has a solid capture workflow that includes custom software to manage

00:17:14.720 --> 00:17:21.760
the decks, track errors, and make sure that the restoration process is moving

00:17:19.520 --> 00:17:27.520
efficiently. They can currently transfer up to a dozen tapes on a good day with

00:17:24.160 --> 00:17:30.480
the end goal being capturing all 260

00:17:27.520 --> 00:17:35.039
tapes, which is the 47 episodes plus multiple edits, unfinished copies, and

00:17:32.720 --> 00:17:39.440
bonus content like the cutscenes from the PS1 game.

00:17:36.320 --> 00:17:41.360
>> How's that progress going, Brian?

00:17:39.440 --> 00:17:45.679
>> It's a long process, but you know, this is the problem with this technology. You

00:17:43.039 --> 00:17:50.720
know, they put so much effort into editing every episode possible. At the

00:17:48.400 --> 00:17:55.120
same time, this was done on CRT TVs. And for all those youngans, it hurts to say

00:17:52.720 --> 00:17:58.559
that that didn't grow up with CRTs, they're not familiar with, you know,

00:17:56.559 --> 00:18:02.240
video interlacing where persistence of vision makes it look like a full

00:18:00.240 --> 00:18:05.760
picture. But when you're digital now, you get scan lines or, you know,

00:18:04.000 --> 00:18:09.120
>> and it looks nasty. >> Exactly. Right. So, just like VSync,

00:18:07.760 --> 00:18:12.880
right? You know, nobody likes to have VSync off and then see tearing. Well,

00:18:11.200 --> 00:18:18.240
that's exactly what we're seeing now that we're all digital. So, we're actually going back and basically quasi

00:18:16.160 --> 00:18:21.440
re-editing the show while keeping it true to its source. Oh, you know,

00:18:20.080 --> 00:18:26.799
>> good gravy. >> Yeah, you know, 43,000 frames per

00:18:24.640 --> 00:18:31.840
episode and and you know, we're dealing with multiple tapes per episode.

00:18:28.960 --> 00:18:35.360
>> So, I'm sure Kio would love for me to ask how their technology contributes to

00:18:34.320 --> 00:18:38.720
that part of the workflow. >> Absolutely. Yeah, Kioxia has been a huge

00:18:37.600 --> 00:18:42.320
help on this. You know, we've been dealing with spinning hard drives as

00:18:40.640 --> 00:18:46.720
cheap as we can get. It's a lot of data to work through. So Kioxia's drives have

00:18:44.559 --> 00:18:51.200
allowed us to basically work through the files at a rate that is much more real

00:18:50.160 --> 00:18:55.520
time than previous. >> Can you give me like FPS comparison?

00:18:53.840 --> 00:19:01.679
>> Yeah. So when we're doing a pipeline direct from our RAWs over to uh ProRes,

00:18:58.640 --> 00:19:03.360
we're we were getting about 7 FPS on a

00:19:01.679 --> 00:19:07.840
good day if you're splitting across spinning hard drives >> and that's going to wear out your read

00:19:05.840 --> 00:19:12.720
head. That's going to be just bad all around. >> Absolutely. and you know data redundancy

00:19:10.480 --> 00:19:16.480
and the whole project here is focusing on the need to retain that data longterm

00:19:14.960 --> 00:19:20.960
and Kyok says drives are actually offering us the ability to trust that we

00:19:18.799 --> 00:19:24.400
can rely on that data basically being there when we power the systems back up

00:19:22.640 --> 00:19:28.160
>> and what kind of FPS can you do with a high performance SSD by comparison

00:19:26.240 --> 00:19:33.520
>> essentially we are at almost real-time playback so 25 35 frames per second on a

00:19:31.200 --> 00:19:37.600
good time so we can actually go and see all of our edits and play them back in

00:19:35.440 --> 00:19:42.240
real time and not have to focus and see individual frame issues that you really

00:19:39.360 --> 00:19:46.240
wouldn't see frame by frame. And Jacob and I both, we have spent tons of nights

00:19:44.559 --> 00:19:50.480
uh rendering these. And when you spend 2 or 3 hours on an episode, render it,

00:19:48.240 --> 00:19:54.720
submit it over, find a frame issue, go back, do it again, and it's just

00:19:52.559 --> 00:20:00.080
crunching hours on hours, right? So, we've got 260 tapes here. Do the math of

00:19:58.080 --> 00:20:02.400
that offset. I mean, Kioxia definitely saved us massive amounts of time.

00:20:01.440 --> 00:20:07.840
>> Hundreds of hours. >> Absolutely. >> How many episodes are left to capture?

00:20:04.880 --> 00:20:08.960
Well, in fact, as of today, there's only one.

00:20:08.559 --> 00:20:15.600
>> Really? >> There is one reboot episode tape left to capture, and we're going to start it up.

00:20:14.080 --> 00:20:19.600
>> Oh, is that the special surprise? >> That's the surprise. >> Oh, sick.

00:20:17.520 --> 00:20:23.440
>> You get to watch the last reboot episode tape come across.

00:20:21.200 --> 00:20:26.160
>> Let's go. Hopefully, I'm not a bad luck charm. Okay. Well, let's do it.

00:20:25.600 --> 00:20:30.720
>> Let's do it. >> We've got a couple other pretty special people here with us for this final

00:20:29.280 --> 00:20:37.280
capture. Jacob, do you want to tell the people what you're about? >> Hi. Yeah, I co-directed and edited the

00:20:34.480 --> 00:20:42.000
Reboot Rewind documentary. Uh, and I was one of the people that kind of found the

00:20:39.360 --> 00:20:45.280
these tapes in Mainframes archives sort of with the original intention of just

00:20:43.760 --> 00:20:48.880
finding better quality footage of the show for the documentary, but now it's

00:20:47.039 --> 00:20:52.159
turned into a much bigger project. >> Very cool work. And then we've also got

00:20:50.559 --> 00:20:56.480
someone like really special that we hit around in the back here. This is Gavin

00:20:53.600 --> 00:21:00.640
Blair, who was one of the co-creators of Reboot. Uh, wrote, did some voice

00:20:58.880 --> 00:21:06.000
acting. Uh, >> how cool is this? This is very cool. I

00:21:02.960 --> 00:21:08.559
mean, we raised a generation of nerds

00:21:06.000 --> 00:21:12.880
and now they're doing remarkable work, deserving history, which, you know, as

00:21:11.280 --> 00:21:15.440
an old guy who's not long for this world, it's great to see this history

00:21:14.799 --> 00:21:21.039
being preserved. >> Thanks for being here, creating the show and uh just kind of showing showing

00:21:19.440 --> 00:21:25.360
support for what these guys are doing. I I think we can all agree that it is

00:21:23.039 --> 00:21:30.799
tragic when these sorts of things are just lost forever. I mean, not lost, but

00:21:29.360 --> 00:21:36.080
not preserved with the kind of quality that they could be. >> Absolutely. Yeah. We made history with

00:21:33.520 --> 00:21:39.520
the quality of the show we made, and you're seeing it now.

00:21:37.280 --> 00:21:43.919
>> It didn't do justice for so many years. >> Now, finally, the way we saw it when we

00:21:42.320 --> 00:21:48.400
made it, which is remarkable. >> That's a great final line. It's time to

00:21:46.159 --> 00:21:51.440
press capture. Are we ready, Mark? >> Yeah. Let me pull out the tape. All

00:21:50.080 --> 00:21:59.120
right. Are we ready to do this? >> Yes. >> So, we're going to go ahead insert our

00:21:55.919 --> 00:22:01.200
tape. Watch the magic happen.

00:21:59.120 --> 00:22:06.799
>> That's terrifying the way it just like manhandles that fragile, delicate

00:22:04.720 --> 00:22:09.760
>> tape that's coming apart at the glue. >> This has a little auto adjust menu here

00:22:09.039 --> 00:22:15.200
right on screen. >> Clean, isn't it? >> And it's about to get better. So, here on the screen, we can actually see the

00:22:13.520 --> 00:22:20.240
four different video heads, their signal strengths. This is the uncorrectable

00:22:17.679 --> 00:22:22.640
errors. We're go to the corrected errors. We're going to do an auto

00:22:21.440 --> 00:22:27.440
adjustment. This is looking at the levels of the signals coming off. It'll glitch momentarily on the video. And

00:22:25.600 --> 00:22:31.760
then if there's still an outstanding one, it's got a tiny smidge here. What

00:22:29.840 --> 00:22:34.960
we do is go for the Allen keys and we're gonna just go slightly adjust.

00:22:34.159 --> 00:22:40.080
>> That's insane. >> While it's running, >> that's insane what you're doing right now, sir.

00:22:37.760 --> 00:22:44.400
>> Mhm. Yeah, that's too far. Back it off. >> She But that was that was up to like

00:22:42.640 --> 00:22:49.039
here at the beginning and now it's gone. Now we're going to do another auto adjust to make sure it's settled.

00:22:47.360 --> 00:22:53.440
>> And again, it'll glitch momentarily. That's normal. But there we go. Now

00:22:51.039 --> 00:22:56.480
these are uh correct or detected errors and they're almost gone. we switch over

00:22:54.720 --> 00:23:00.320
to the uncorrectables and there's basically zero. So that's the difference

00:22:58.400 --> 00:23:04.240
between just naively shoving the tape into the machine and hoping for the best

00:23:02.000 --> 00:23:07.440
and tuning it to each one. There's some things where we simply cannot go and get

00:23:05.679 --> 00:23:11.919
another copy of. And what turns out when this thing's in frame by frame slow-mo,

00:23:09.280 --> 00:23:15.840
the tracking has time to catch up to wild changes. We've seen a couple tapes

00:23:14.480 --> 00:23:20.960
where the deck that recorded it must have had a broken roller or something cuz the tape was just doing this the

00:23:18.799 --> 00:23:24.159
whole way through. And so I could tune it and it would work and then it go

00:23:22.159 --> 00:23:29.919
right back out again. All right, let's see it. >> All four channels of audio. We resume

00:23:26.960 --> 00:23:36.159
our data log over here. Uh, we spin you up and from now this computer is in

00:23:32.960 --> 00:23:38.400
control. I click clip and it instructs

00:23:36.159 --> 00:23:41.280
the machine to start moving and it will instruct the machine to stop when it

00:23:39.679 --> 00:23:44.880
gets to the end. This is the real-time error log that Brian developed and that's where I was adjusting it. You see

00:23:43.919 --> 00:23:47.600
those spikes? Yeah, >> that was me tuning it.

00:23:46.960 --> 00:23:53.120
>> Right. >> And now you can see how just flat it is. That's what we want.

00:23:51.600 --> 00:23:56.480
>> And that's it. There it is. >> That's the last tape of reboot episodes.

00:23:55.600 --> 00:24:00.880
We've done it. >> That is so cool. Well, thanks for

00:23:58.960 --> 00:24:05.520
letting me be here for it. This is >> making it possible. >> Absolutely incredible.

00:24:03.840 --> 00:24:09.360
>> Now, this has been a dream come true, I think, for all of us. I never imagined

00:24:07.520 --> 00:24:13.600
little six-year-old me looking at that tiny CRT in my living room that 30 years

00:24:11.840 --> 00:24:19.200
later I was going to be a part of rescuing a show just so incredible. and

00:24:17.279 --> 00:24:23.440
to have gone through this, let's be honest, incredible pain in the ass

00:24:21.120 --> 00:24:27.279
journey. I would absolutely do it again if I had to. This has been a pleasure.

00:24:25.600 --> 00:24:32.640
>> And that's almost the most important point. I think as cool as the tech was,

00:24:30.320 --> 00:24:37.679
the best part of this story is the human side. The incredible outpouring of fan

00:24:35.600 --> 00:24:40.799
support has helped to show Mainframe, the studio that created Reboot, that

00:24:39.200 --> 00:24:44.960
there is still a massive amount of appreciation for both their art and for

00:24:43.279 --> 00:24:50.080
their innovation. Again, I can't emphasize this enough. It was the first

00:24:47.440 --> 00:24:54.080
real 30inute TV show that was rendered on a computer, pioneering a new

00:24:51.679 --> 00:24:58.880
animation standard that is still used by kids shows today. It was so far ahead of

00:24:56.640 --> 00:25:03.279
its time and I'm so impressed by the work that's been done and will still be

00:25:01.440 --> 00:25:06.720
done and I'm so proud to have had an opportunity to be part of it. Massive

00:25:04.960 --> 00:25:10.480
shout out to Kyioia. They have been an incredible partner for us. When there's

00:25:08.799 --> 00:25:15.120
something that we just look at and we go, "Wow, wouldn't that be so cool to

00:25:12.799 --> 00:25:19.679
do?" whether it's getting a pi calculation world record or or doing

00:25:17.200 --> 00:25:24.159
this. If it involves storage, they basically go, "All right, cool. We're

00:25:22.080 --> 00:25:27.360
in." And it makes incredible things like this possible. Check them out in the

00:25:25.760 --> 00:25:31.679
video description. Genuine thank you to them. And a huge thank you to you guys

00:25:29.360 --> 00:25:34.960
for watching. Uh, and hey, if you're looking for something else to watch, why

00:25:33.039 --> 00:25:38.080
not check out Reboot Rewind? If you like videos about older video formats, why

00:25:36.720 --> 00:25:43.919
don't you check out the time that PLF and I watched the same movie on 10

00:25:41.440 --> 00:25:48.559
different mostly obsolete video formats. There were some really wacky ones over