 Okay, welcome. Wow, got a pretty good crowd here. I'm gonna repeat something I said a little earlier. We're gonna be doing some actual live demos with different pieces of equipment on stage and we intended to have a webcam that was gonna show it up there, but it turns out the webcam doesn't work on the dual headed display. So if you're sitting in the back, you might actually get a little bit more from this talk if you come forward. Just a recommendation, not a requirement. Anyway, hello. Thank you for coming. I know it's early. How many of you were at parties last night and barely made it here? Yeah, that's, yeah, same here. Thank you for coming. Awesome. I'm Casey, also known as Freak Monkey. This is, yeah. I'm Dodger Chaos and I am Ms. Dodger Holligan. And we're gonna show you a really interesting old piece of equipment. How many of you saw my YouTube video already, actually? Oh good, I was afraid it was gonna be everybody. So, this is about a modem that I have and I'll get into what it is and where I got it in a moment. But first I wanna explain what modems are for anybody who's never used one. Has anybody here ever never used a modem? It was worth asking. I mean, it is almost getting to that point in history. A modem, as most of you probably know, is it's a device that converts, that represents digital signals in an analog format and vice versa. It stands for modulator, demodulator. The modulator side is the side that takes a digital stream and represents it in an analog form. And the demodulator takes an analog stream and then transmits it digitally over an electrical line. I say represents it because it is a common misnomer. People say that modems are analog to digital converters. They don't convert anything because what you have on the analog side is still a representation of the digital stream. It's not a video file or an audio file. It's actually digital on both sides, just one in an electrical form and the other in represented in analog form. So, as you know, everybody refers to modems and their baud rate. There's an interesting point about that. There's two sides to a modem, right? There's the analog side and there's the digital side. The digital side is represented in its speed in bits per second or kilobits per second or more recently, megabits per second or gigabits per second. The analog side is represented in its baud rate and one baud is actually an analog, what's called a symbol change. So 300 baud means there's 300 symbol changes, 300 shifts of some sort on the analog side every second. Turns out 300 baud was the last time we used that term correctly. 300 baud is 300 bits per second on the digital side and 300 symbol changes per second on the analog side. What we always called 1200 baud is actually 1200 bits per second on the digital side and 600 symbol changes per second on the analog side. The reason for that is what they figured out was they could do a symbol change that represented more than one bit. Like, they were shifting between two frequencies at 300 baud, well they figured out for the next speed they could shift between more than two frequencies to represent more than one bit. And so at doing that 600 times a second, they got 1200 bits per second of digital data out of it. So what everybody called a 1200 baud modem is actually 600 baud. They then improved the modulation better and what everybody calls a 2400 baud modem is also actually 600 baud. So we've been calling modems by the wrong thing ever since the 300 baud standard went to 1200 baud. Just worth mentioning because I'm gonna get into how the modulation works in a minute. So here's a quick timeline of modem history. In 1959, somewhere it's really hard finding history of the modems anywhere, even on the internet and Wikipedia and even looking through old bell documentation. There hasn't been much written about their history. We keep making them faster and upgrading and upgrading and then just forgetting our past, which is kind of where this talk gets interesting. But somewhere in the 1950s, a company called Anderson Jacobson introduced modems for least line, dedicated least line use for the US military, primarily NORAD, missile defense stations, things like that. It was how they sent data between the computer systems at the different military stations. These were, they ran over private lines. They did not run over public POTS telephone lines. So they weren't modems the way we think of modems. They were more like what we think of as the old multiplexers and stuff that you had hooked to dedicated lines. In 1962, AT&T, well Bell at the time, they weren't AT&T yet, introduced the first consumer plain old telephone service line modem and it was called the Bell 103. It ran at 300 BOD, 110 BOD did exist. That was the one used on the private lines, but the first consumer modem available for actual telephone line use ran at 300 BOD. So despite the fact that there were 75 BOD teletype machines prior to that, there were 110 BOD Anderson Jacobson modems prior to that. The first modems that you could buy to use on a phone line were 300 BOD, that is the original oldest speed. The modem that I have that I'm gonna show you and tell the story about was made around 1963. I had 1964 in the beginning of the presentation and in my old presentation on YouTube and one of the two things happened. One of the original VPs of Livermore Data Systems contacted me and said, you might be off by about a year because there was an interim model between that and the Model B that you're using to date it. So yours is actually probably about a year older. So somewhere around 1963, which means this modem was made within a year of the invention of the first telephone line modems. Interestingly, I mean you all have seen, how many of you have seen the movie War Games? Right, the acoustic coupler in there, that's where most people remember acoustic couplers from. Those of you who haven't have homework for this evening. You know, the reason, it's interesting, everybody says oh yeah, modems were so old back then we had to actually sit the handset down on them to make it work. If you think about that, that's kind of ridiculous. It's not that we didn't have the technology to have signals traverse wires, right? Everything else was connected with wires. The problem was that Ma Bell that owned the Bell system, all the wires also owned the telephones in the houses and in the offices. And it was illegal to connect anything to those wires that was not made by and owned by Ma Bell. So you couldn't go out and buy a piece of equipment and plug it into the phone wires. And to go along with that, they didn't even have standardized modular jacks like we do now because they didn't want you plugging things in. A lot of phones were actually hardwired into the wall. Yeah, some people nod in their heads, there's some people out there who remember that. I don't, it's a little before my time. A company called Carterphone wanted to make competing products and in typical American style, they kept appealing to the Congress and filing lawsuits and antitrust lawsuits and eventually won one in 1968 in what was called the Carterphone Decision where it was stated that, okay, third party devices must be allowed to connect to this network because it's just important for interconnectivity and for the expansion of communications and what have you. So until 1968, nobody else could make a device to be connected directly to the telephone network. That's why we had acoustic coupled modems. So it didn't advance things as quickly as we'd like because what the court said was that anybody else can make a device that can be connected to the bell system as long as the bell system approves of the connection. Right, once again, the concern was preventing damage to the telephone system. The case that Bell made was, if we don't know about these devices, if we don't know anything about them, someone could connect something up that would mess up the network and then nobody could use the phone to bring the whole thing down. They were scared. So it was still several years, almost a decade actually, before directly connected modems became common. Speaking of which, and I think this is one of the interesting points that Casey and I both found rather amazing. And that's the duration of 300 Baud as a standard. If you look at this timeline, what's remarkable is, as Casey pointed out, we're talking 1962, the introduction of the 300 Baud standard. It's not until 1976. Until we see the creation of 1200 Baud? Yeah, 1972, a company called Vatic made a 1200 Baud, a 1200 bits per second. They were actually 600 Baud. In 1972, Vatic made a 1200 bits per second modem, but it was proprietary and they refused to release how it worked so nobody else could build anything and talk to it. And so it didn't become the standard. And then a few years later, Bell, AT&T Bell, looked at that modem, reconstructed what they were doing, made the same changes to their modulation and introduced what became the 212A, or the V.22, the V.22 modulation, which we all know is 1200 Baud. So yeah, 1962 to 1976, all modems were 300 Baud. That's probably the longest span that there were no bandwidth improvements in telecommunications since the beginning, which was 1962. So, you know, we were gonna put some other, we didn't have room on the slide, we were gonna put some other interesting historical events on here. I was gonna put the moon landing on here until I learned from Joe Grand that it was a 1940. But no, the moon landing was, you know, the year after the cargo phone decision. So this modem I have was several years old when we landed on the moon. Bulletin boards came out in about 1982. The AT command set from Hayes was invented in 1981. So this modem really does represent the beginning of the era of telecommunications over phone lines and two different offices without having to run hard, dedicated lines for communication. So anyway, with that, I know y'all are dying to see this thing, so I'm gonna get it out right now so we can see it. I'm his Vanna White. I give you the only wooden computer peripheral I have ever owned. This is, I said circa is the right word. It's a circa in 1964. There's no way to know its exact date. It's, let's see in the next slide. Okay. That doesn't matter. I just forgot to look at the list. So it's early, I know. So anyway, yeah, everybody says, well, where did you get that? Yeah, it, I'm circa in 1974. So it outdates me. When I was a teenager, my dad had a company building PCs inspired by my interest in PCs and the difficulty in buying PCs and getting support for them in the early 1980s. And so he started a company assembling PC clones. I worked right along with them. And one day, this guy, this really old guy comes into our shop. I don't know, I was 15, so really old might mean 50. But anyway, now this old guy comes into the shop. He was retired, he was in his 70s. And he wanted help with his PCs at home, but he didn't want to buy anything, which my father wasn't interested in, but I was. I was like, you know, hey, for a little bit of money, you know, pick me up and I'll come out there and help you fix your old IBM PCs and install Debase and all that fun stuff. Did that for a couple of years. The guy was a retired engineer. He's, I remember he was a mainframe guy and he didn't know anything about PCs, which is why he wanted our help. And he passed away. I didn't hear from him, obviously, for a while. And his widow called me and explained that he had passed and said, you know, I'd like you to come out and do one more thing. And that is, I have all this junk that was his that was all stored in the garage and in the attic and it's all just electronic mumbo jumbo. I don't know what any of it is. Could you come help me sort through it and figure out if any of it's worth anything and what have you. And so we were going through this stuff. We figured out what stuff was like. There was a computer that was fairly recent that, you know, they gave to their daughter and I opened a box and I found this and I opened it and I realized what it was and I just was slack jawed. I was like, that is amazing. And she was like, do you want that? I was like, well, it's an antique. It's actually very, very unique. She goes, well then you should have it because Bob would, I don't know what to do with it. Bob would have wanted you to have it if you appreciate it because nobody else knows what it is. Hence the comment. I told this story in my YouTube video and then this was the first comment posted on gizmodo.com. So anyway, that was, I was probably 16 or 17 when that happened. I had, there was no way to figure out this thing's history at that point. I took it, I looked it up, looked up the company at the library. I mean, it was the late 80s. I just, so I just kept it as a conversation piece, frankly, because it's in a handcrafted wood box and you don't see many computer parts in wooden boxes. So. For those that can't see this in the back, you'll have an opportunity at the end to talk. This is a pretty amazing piece of artistry in addition to being a 45 year old modem. It's a pretty fantastic piece of art here. So, I carted it around with me. I kept it on my desk at the various jobs I had and just as a conversation piece. And then when I moved sometime in the late 90s to a house in downtown Atlanta, I packed it up in a Tupperware bin with some other stuff and I put it in my garage and I forgot about it. And it sat there until four or five months ago. So a little over 10 years. Completely forgot about it. I'm unpacking some stuff in my garage, cleaning up and I find it again. I'm like, oh my goodness. So, I'm like, well, you know what? Now the internet exists. I can Google it. So I'm just curious. I'm like, I wanna know how old is it and who is it made by and all this stuff. And so I start Googling around and I find one, the only thing I can find similar to it is this is a Model A. I find one almost identical to it that's a Model B and it's at the UC Davis Computer History Museum and it's dated 1965. I was like, oh my, that's when it occurred to me what I might actually have a fairly unique piece of equipment here. So, talking about its physical characteristics a little bit, I was gonna have a webcam up here to show it, but the webcam doesn't work on that screen. You all will get to see it up close in person after the talk if you wanna come up. But yeah, it's in a wooden box. I'm gonna sit down here so I can hold it while I'm talking. As you see, it's in a wooden box that has handcrafted dovetail joints down the sides. There's no nails or screws. It's got brass hardware on it, a leather handle on the top of it. Yeah, it's got a long piano hinge on the back that is actually sunk at 45 degree angles into the wood instead of being screwed on. You can see it's got a DB25 RS232 port that actually caused some interesting. There were a lot of people that commented on my YouTube video that said, that's a parallel port, stupid, because they had never seen a 25-pin RS232 port. I never answered any of those comments, I just silently chuckled. One guy actually emailed me and said he could tell it was fake because this handle is clearly off of a 1960s Stratocaster amp. The only place those handles were ever used. Yeah, clearly. It's not like this company went to a computer supply place to buy parts for this thing. This was early 1960s. It's got a label on the bottom that says Livermore Data Systems, Livermore Data Systems model a serial number 279 and a big gold sticker that says, this unit is not serviced by IBM Field Engineering if repair is required pack suitably and sent to the repair contractor, Livermore Data Systems. And the main part of the modem, which is the two big transducers that you actually sit the phone hand set down onto so it can talk over the bell phone system which you couldn't legally connect anything electrically to. So that's the physical characteristics. Oh, there's a neat little notch in the side here so that when you sit the phone down on it you can close this so that ambient noise doesn't interfere with the conversation. As you'll shortly see, I don't know if in case you didn't quite point out, it's kind of hard to see, but there's also kind of a custom little power adapter here on the side and that was kind of part of his hardware hack. A lot of what he had to do was to determine how to get this device working. So he pulls this box out of a Tupperware bin in his garage and of course the key is not only does he not know if it works but he doesn't know if he's got the parts to make it work. So it's got a special little custom power adapter but again, lacking the webcam, we can't quite show you at this distance what he had to do in order to make power work but again, those that are interested, you're more than welcome to come see his excellent electro hack. I took a PC power cord, the really big fat PC power supply power cord and I noticed it was just a little bit bigger than that. So I just took a pocket knife and I whittled it until all the electrodes were showing and then I'd hollowed them out so that they would fit over the pins. So yeah, this is a personal fire hazard. When I was, I told some of these guys earlier, I did the run through of this demo at home last week and I set everything up and I was pre setting up everything and just not thinking, I plugged the power cord in and laid it down and then when I was moving stuff around, I grabbed it. So thankfully I'm still here to give the demo and the good news is I know better now than to do that on stage. It would have been much more fun demo though. It actually would have been pretty funny. So technically, technically this modem is actually just as interesting to me as it is physically. It's Bell 103 Modulation which I explained a minute ago. It's 300 bot and 300 bits per second. It's Bell 103 Modulation runs over two different sets of carrier tones. One's called originate mode and the other's called answer mode. And the reason they're on different frequencies is so that they can both talk at the same time. It's duplex and the two won't interfere with each other and so the receive side has a band pass filter for the receive side on the originate mode modem has a band pass filter to listen only to the carrier tones for the answer mode modem and vice versa so that each one can only hear the other and not itself. Cause on a phone line you hear yourself when you talk. It uses what's called frequency shift keying which I'll get into in a second. There's no error correction at all. In fact, the first standard that had error correction was V dot 22 BIS I think that's 2,400 bits per second. We called it 2,400 bot and there was a MNP5 error correction add-on for that. 22 is 12, V at 23 is 24. Maybe it was V 23. You know what, I didn't put that in the slide so. Don't quote me on that but yeah, it was late 80s with 2,400 bit per second modems before there was any line error correction. This device is actually pretty much from a circuit point of view, this device is a purely analog device. It doesn't have any air internal clock. It has no idea what speed it's speaking at. It doesn't know it's talking at 300 bot. What it knows is when it sees 12 volts in the serial port, which is idle, it puts out one tone and when it sees the bits, it puts out a different tone. It doesn't know what speed it's doing that at. It's two oscillators and a switch. It requires the full plus or minus 12 volt RS232 level, RS232 levels to work and the reason that's significant is somebody said this some years ago and I cannot find on the net who it was. Somebody said in another presentation that RS232 is really more of a rumor than a standard. I wish I could cite the person who said that because it was a brilliant quote. It's true. If you look in the IEEE standard for RS232, first of all, there's no connectors specified. The pinouts were invented by each vendor who needed them which is why we went from 25 pin to nine pin to RJ45 and why sometimes those pinouts aren't always the same. That was never specified as a standard. The other thing that the IEEE standard says is when you transmit RS232, you do it at plus 12 volts when you're idle and then negative 12 volts for bits across grounds. It's a 24 volt swing between idle and bits. And they said the receive side needs to be able to discriminate and decode plus three volts to negative three volts or greater. So you've got this great discrepancy in what is allowable in there. The reason they did that is because in the 1950s when this was proposed as a standard, its purpose was to be able to put peripherals like terminals to mainframes around say a big campus or office environment. We didn't have, this was the only networking we did at the time other than like thick twin acts on 5250s and what have you. So that discrepancy is to deal with the resistance of just straight copper wire over several thousand feet. So you could put a terminal two buildings away from a mainframe and it would still work. The signal's 12 volts when it leaves the mainframe and it's only four and a half volts when it gets to the other end of this ridiculous aluminum wire and it will still work. The problem with that is over the years, manufacturers figured out if you're only gonna have a three foot cable, you don't need that plus minus 12 volts coming out of the port. If this can detect three volts, you could do it at five volts which is what TTL chips use. So all the modern USB driven serial ports and USB serial adapters actually put out plus or minus five volts for the transmit line. This modem, the switch in it that switches the oscillators needs more than five volts. So if you plug it into a more modern laptop or a USB adapter, it won't put out any tones. The only reason I discovered this is because I'm a bit of a hardware hacker and a fan of Joe Grant and thank you all for blowing off his talk, come to mind. They didn't know he was talking, Jesus. Joe Grant's great and he's inspired me to do a lot of this stuff. And I learned a while back that programming pick microcontrollers directly from a serial port that works on hardwired UART serial ports and doesn't work on USB serial ports and that's what inspired me to learn that, thankfully so that when I started testing this thing I didn't give up on it after the first thing I connected it to didn't work. So anyway, Bell 103 Modulation, as I was just explaining, it's frequency shift keyed. There's really very little to it. When the RSC32 line is idle at plus 12 volts, it puts out, it's what's called mark idle in telecommunication terms. There's two frequencies, mark and space. Mark is what you think of as the carrier tone. It's a solid tone you hear when it's not talking and every time the RSC32 line drops just in and out a bit this thing just switches the oscillator and puts out the space tone instead of the mark tone. This is an originate mode only modem which is actually a little disappointing when I discovered that because it meant if I found somebody else with one we couldn't make them talk to each other because they speak both in originate mode and one has to speak in originate mode and the other has to speak in answer mode. The neat thing about Bell 103 Modulation while I'm on it is that there's no handshaking. There's no in band signaling. There's no negotiation goes on. It's literally two tones that represent two voltage levels. When the voltage changes, the tone changes. That's it, which is why this modem could be so simple. It's got no microcontrollers in it. It's got no digital circuitry in it. There's no command set. You don't talk to it, you talk through it. You send voltages, it turns them into tones. When it hears tones, it turns them into voltages. That's it. I'm saving demos for the end of that one. Yeah. This is my slide, yes. All right, so the question then is what use is 300 bot? And this is pretty important because it doesn't serve all of the same purposes that later higher speed modems did effectively serve. For the most part, this was for remote communications and remote access to mainframe devices. Casey's got some interesting comments here with regards to program submission. And it's kind of an interesting, I guess, method by which the 300 bot modems were used for communications and for submission of programs to early data sets. Yep, absolutely. BBSs didn't come about. The use of modems for entertainment communications didn't come about until the 80s, which struck me as odd at first, but then it occurred to me this because the IBM PC came out in the 80s. Until then, nobody had things at home to hook to modems. So they were only used by engineers, by remote data centers, college students occasionally that had access to terminals and wanted to submit their programs at night to the IT lab. I'm curiousity. How many here saw HD's talk on Warbox and or know of Warbox by HD more? All right, so the reason I mention this is for those that don't know, you'd probably be amazed at the number of 300 bot modems still live in major organizations, universities, et cetera. There are a lot of 300 bot modems still out there answering. Now we go to the trouble of trying to find them all. Fortunately, KC had the ability to put together some equipment to essentially provide our other end, but suffice it to say, there's a lot more systems answering on 300 bot today than you might expect. So this guy, Bob from Australia, when all was his email address, I didn't want to put his full email address. Is Bob here? There's no way, yeah. Before I talk about him, he sent me this email. I asked for anybody to email me with any stories they had of using modems like this. He sent me this email that he used the same model in about 1964. He talked about dirty lines. There was no error correction, like I said. So a lot of times when you're seeing data over it, if noise is introduced, it'll go only one way. You won't see it. If you're uploading a program to a mainframe, that could be catastrophic, as he explains here. He used it for transfer data for computation between computers. He didn't know that dirty line existed. The worst case was when we, these are his words, somehow got an infinity loop happening in the mainframe and all the terminals froze because they were uploading data to the mainframe through old wooden modems that didn't have any error correction. So I thought that was an interesting story. It kind of gives an account of how these things were used back then. I got a bunch of other emails. I'll run through these fairly quick and then we'll get to the demos. This was the first one that I found when I was trying to figure out what my modem was. Immediately, I was like, Eureka, it's the same one. But there's actually two important differences. The first important difference is that it has, one thing I noticed is that the dovetail joints on this one are square. They're not dovetail joints, they're just square joints. And the joints on this one are dovetail. So the craftsman who was building the wooden boxes started to get lazy when he had to build a lot more of them in the second model. Oh yeah, you can see right there, these are the joints I'm talking about. And on this, they're dovetail joints. And on here, they're just square joints, which is interesting. The other thing is that this one has two lights on it. And those aren't LEDs, just by their size. It was 1965, they're little 12 volt lamps. And one of them, according to an email I got, like I said, I was contacted by a guy from Livermore Data Systems who used to work on these. By the way, he said he almost fainted when he saw the video. But one of them is carrier detect, and the other one shows data. So here's the carrier from the other modem, one lights up, and when there's data, the other one lights up. This modem actually has two little tiny circles on the box in that same spot. So they designed it for that feature, they just didn't put them in the first model. You got the cheap one. Yeah, I got the prototype. This is a Model C, which is the next, the third model that they made. It's dated 1968 by the Stanford Computer History Museum. One thing you'll see on it is this ugly green stuff. That's a foam pad for sound deadening. This one actually originally had the foam pad in it too, but by the time I got the modem, it was already crumbling and falling apart, as you see it is on there too. I just ended up pulling it out, because I was tired of it leaving schmutz all over the inside of the modem. That one has, it has a switch. The switch switches it between originate and answer mode. So that one can do both. One of the other knobs is a full and half duplex switch, so it'll run in half duplex mode as well. And I think the other one is an attenuation to a volume control knob to deal with different sound levels out of telephones. So they really started getting fancy in 1968. Livermore Data Systems went on from there to build one in a metal, finally in a metal in a plastic case called the Series 71, which became their main product throughout the 70s. I got some emails from people that had ones very similar to this sitting on their desks, just like I did. This was from a guy named Rob. As you see, that's an office cubicle environment. It was sitting on the desk of his coworker in the next cube over when he saw my video. And he jumped up and ran over there and took a camera phone picture of it. It's a Model B also. The other thing I find interesting is it has an IBM, they had IBM added on a part number tag. So IBM apparently provided these to their customers at some point. This is one that, this is what, the first one that made me realize this one might be 1963 and not 1964. It's an interim model between the Model A and Model B. It's like a Model A, it doesn't have the lights on it. But unlike a Model A, it has this switch right here that switches between half and full duplex mode. And on the tag, which is actually on the inside of that mode instead of on the bottom, it says Model AH. So somebody needed half duplex mode, so they just took the Model A, they added a switch to the side of it and made a Model AH in between. And then I just thought this was damn cool. This guy in Norway had an acoustic coupler that was two separate pieces. It was made, it was designed for use on like pay phones and such. And he was driving down the road in Norway and he had a little TRS-80 portable computer and he dialed in, I know there's no way you can read this, but what that is is a login screen for a Phytonet BBS. That, how many people remember Phytonet? Awesome. I ran a Phytonet node in Charlotte, North Carolina when I was like 17. But you can tell it's Norway too. Ice all over the roads. But yeah, he sent me that and I just thought that was way too cool to not show. I love the fact that in Norway, on icy roads, this guy's driving while taking a picture. Accessing Phytonet via his Phytonet, fantastic. He was way ahead of his time. Anyway, so yeah, it's been 30 minutes and I've been up here babbling. You all wanna see this thing function, don't you? Demo time. All right, yes, we're gonna do a live demo, especially since you probably didn't ever give me the videos, did you? We made videos of all our demos in case something went horribly wrong and I think he didn't bring them. I know I have them, but actually it's on the video camera. Oh, okay, so I guess if we can see. So the very first thing I'm gonna do is I'm gonna connect the modem. Connect the power cord to the modem first. This is much more fun this way. Do you have the right cable for it next door? Of course you have the right cable for it. This power cord's probably pretty common in the 1960s, aren't they? Oh, I'm glad you, yeah, you're the guy running the PDP-11 next door. Thank, I'm glad you took a break to come over here. Awesome. If any of you haven't been in the retro room next door, you obviously find this stuff interesting. Go over there after this talk and check it out. It is quite a collection. The fire hazard's much more fun for live demo purposes. Yeah, you know, it might end up being a power technique show too. Exactly. No, this thing's fine. The nice thing about this power, this little molar adapter is that it is grounded and the cable's grounded. So the worst case scenario is I'll blow the circuit breaker on the power strip if something screws up, but I've been using this cable like this for a couple of months, it's been fine. He says before he plugs it in, yep, yep. So, okay, you wanna do that? Let's see, it's this one. So the first thing we do is we put power to it and there's nothing connected to it. So it's not getting the plus 12 volts and we'll turn the modem on. Y'all hear that? So that's, it's putting out a tone. It's not waiting for any commands or anything. Like I said, it's just an analog oscillator device. And what's interesting is, since it's not receiving the plus 12 volts, it's actually putting out the space tone. It's not putting out the mark tone, which makes it mark idle. When it receives 12 volts from the serial port, it puts out the space tone. The other thing you see happening here is the demodulator side, which is running all the audio here through a couple of band pass filters and looking for frequency changes. Here's noise. And so it's turning them into bits. DC-404, we are the line noise. Yes, that's right. Yeah, everybody make a lot of noise. Yeah, you are the line noise. Congratulations. Who was whistling? Yeah, okay. If you can print your name up here, you win a prize. All right, so now that we've now shown you the modulation, we're gonna actually dial in to something with this. Now, you know, as you all saw, the way this modem works is you actually pick up a analog handset. How many of you guys have seen one of these before? Excellent, excellent, great start. There were some people who didn't raise their hand. I love that. Yeah, just go ahead and dial the number. It is a touch tone because my Vonage adapter doesn't support pole tiling. All right, so we've put it on there. So we're still getting some line noise, I'll close this. And I hit enter too many times, so we get to watch it scroll by a couple of times. This is for everybody who asked, because a lot of people in the YouTube video were like, you idiot, you logged in live on a video, we all saw your password. This is a Linux box I built for this purpose. It's, uh-oh, there it goes. I think I've got a, yeah, I mean, we got a, let's just leave it, just leave it. If I re-run Minicom, it'll probably fix the, no. Man, it didn't like that, it all did it. Let's go to seven data bits so that we don't have the high characters, because we're only doing ASCII anyway. That didn't help. All right, let's call it back. Just do this. So yeah, like I said, there's no, there's absolutely no error correction, and there's a lot of noise in here, plus I'm over a Vonage VoIP line that's going out over the DEF CON internet connection. I don't know if this is going to work. It must be. All right, let's reset it and start over. All right, we're going to try the dial-up portion one more time. I don't want to waste all the time. I have a couple of other demos to do, so it's good news. This is white light demo's rule. Yes. Chad here says he has a lot of problems dialing in with his shoeshine kit as well. Turning the modem on helps. Man, we're getting a lot of noise, but. Well, we have dial-up. Yeah, that's not going to work. It's close in Validho's name, yeah, okay. Anyway, you see the result. This is, ooh, it really doesn't work there like that. What's nice is I registered Google Foofle right before this, and so I'm actually getting click-through revenues every time you pull up this site. Awesome, thanks. It's just not going to work. I get too much line noise. All right, so yeah, so there you go. Connected to the internet, albeit sort of flaky, at 300 bot through a 46-year-old modem. So. Demo number two. Right, well, yeah, that worked, that worked well enough. That worked well enough, so. You see that the modem works, that's the main point of that. And because the Bell 103 standard became the standard, it can talk, the modem at the other end is a US Robotics Sportster 14-4 modem. It actually is just what I happen to have laying around. So that's what we were dialing into. For what it's worth, do you guys remember that back in 1990, a US Robotics HST dual-currier modem cost $1,300 retail? Rock! So, one of the interesting comments, we're talking about dialing into the internet real quick, one of the interesting comments I kept getting, both on slash.dot, it's got featured on slash.dot, on slash.dot and on YouTube, and I guess Modo was, you weren't really on the internet because you weren't doing IP across the modem. You should have done PPP to put the modem on the internet. I understand the argument, but these are obviously people who are not very good at math. 300 bits per second is 30 characters per second. An IP header is 20 characters? And options, options is 20 bytes. I mean, yeah, yeah, 20 bytes. An IP header is 20 bytes. A ping packet is, you know, without anything else added to it is another 56 bytes. So, a single ping packet takes a couple of seconds just to transmit. A DNS lookup is a series of several of those types of packets, and they're just UDP, so there's no handshaking going on. But a single DNS lookup, and the reason I can give you these quote times is because I actually did, as a result, I decided to try it. It took me several days to get PPP options relaxed enough that the PPP handshake wouldn't time out before the modem was done sending the PPP options to the other end. Once it did finally negotiate and come up and work, when you tried to reach anything, the DNS queries alone saturated the line and took several seconds. So I turned DNS off and I put static entries in the host file. Then ping works, and simple UDP things work. One TCP connection at a time will work because the three-way handshake takes 15 or 20 seconds to complete. If you tried two TCP connections at the same time, the second one times out while the first one saturating the line with just the send-send-ac packets. So, did God help you if you run a browser? So I shouldn't have turned on pipelining then? Is that what you're saying? Yeah, right, exactly. So, yeah, I did get a browser to launch. I had to turn off pipelining, so we'd only do sequential connections, not concurrent connections. I had to put the entries in the host file. Then I had to go in and turn off auto updates, turn off plug-in support, turn off, because it flooded the line with all this stuff. It just killed the line. And, yeah, so you can't do, there's a reason we didn't start doing IP over modems until 9,600 bot, 9,600 bits per second, excuse me. Yeah, we gotta start using the terms right. Well, now we don't, because the terms are now obsolete. But we used them wrong throughout the entire generation of the terms, it just cracks me up. So anyway. Yeah, but then I also have to say like, kibba bite, and I will never, ever say kibba bite. So that's why we didn't do PPP over it. We would have been here for another hour. All right, what's up next? Yeah, what's up next is the, so I was at one point in time, hi, my name is Casey Budd, and I was once a fan of TechnoPop. Once. And there was this band called Information Society. You all might remember when they had a song called, they had a song called I Wanna Know What You're Thinking or something like that. It had that spot quote, Pure Energy. Yeah, and I bought the successor to that album, which was, well, anyway. Because the final track on it is labeled, it has like 11 tracks, and track 12 is labeled 300 BPS in eight one. Like really? So an interesting thing about Bell 103 Modulation, there's no handshaking. You don't have to get a carrier and do any kind of handshaking with an old modem. With a new modem, the new modem's implemented all those features for all the body rates. But the old original modems, they just, like I said, turn bits into signals and what have you. So, let's see if my, see if the CD player works first. This is an audio CD on a crappy little 1990s, 1990s audio CD player. And so we advanced, we advanced to that last track and we played it into the old modem. According to Wikipedia, and this CD player is really, really loud, interestingly, so we're not getting the same line noise problem even when the top open. In fact, you can hear it. So according to Wikipedia, that makes this CD, which came out in 1991, three or four years before CD ROMs on computers and five, six years before the CDR extended format. That makes this CD the first consumer audio CD that had digital data on the end of it, which is now a common thing, right? The story actually gets really stupid and keeps going on and on and on, so we're not gonna go all the way to the end of it. Sony actually tried to have them put a root kit on these. Yeah, yeah, Sony, that's right. It's the best they could do at 300 bot, so. Yeah, that's right. Yeah, yeah, Sony couldn't get the root kit on there yet, but okay, so that's that. We don't wanna take up all the time with that. And that's who we loved it so much though, we figured we need to do our own. Exactly. So what we did was, this is a cheap $29 thumb drive MP3 player that has a voice recorder on it, little mic voice recorder. What we did is we just sent data to the modem and set this down on the earpiece to see if it record the data and what would happen if we had it do that. And so we'll see if that works. I hate cheap MP3 players. Yeah, well, I can do that. I just need to get this thing to do the right track again. There we go. This is the reason that 300 bot was never really appropriate for the purpose you'll see it used here. Yeah, so in the 1960s, if you wanted to get data at 300 bot, yeah, I don't know if y'all can tell what that is, but that is. That's right. So there you go. ASCII Art Porn at 300 bot through. That was how they had to do it in the 1960s and 70s. All right, we've actually only got three minutes left. We want to even attempt, you know what, I don't think that that will work because we couldn't even get to work with the handset directly on the, there's too much noise. Yeah, one thing we could do, let's try to do it with the CD. CD's loud. So we had one other idea, but it was going to require that the phone demo worked perfectly and it didn't work perfectly, but. Can I get the context from your phone real quick? The idea was, oh yeah. So we were inspired, if it comes up immediately, otherwise I'll just say it. Ah, you know what, I'm not gonna bother. We were inspired by Senator Ted Stevens. Some of you remember him? Yes, this just turns data into audio and so anything that transmits audio, anything that transmits audio, right, can transmit the data. So our inspiration by Ted Stevens was to send the audio through a series of tubes. We couldn't, we were gonna do it to the telephone so that we could actually get on the internet through a series of tubes, but since we can't get on the internet, we'll approximate it and use the CD player. Put the tube over this end and we'll see if it actually decodes it. Yeah, it doesn't like the line noise as much with the hose pipe as we say in the south in the way, but there you go. So we're transmitting data through a series of tubes. This is for you, Ted Stevens. So, that's our demo. I hope you enjoyed it. If there's room, I'm thinking I'd like to get this set up in the retro, I'm not ahead yet there, so I'd like to get this set up in the retro room next door if anybody wants to walk up and look at it up close and personal, especially since I couldn't get it up on the screen. Holy crap. It was almost, yeah, it's working there. It looks for pretty good distance. Yeah, we almost got it to do data over my iPhone at a demo I did at DC-404. It turns out that AT&T, okay, that's getting annoying. This is a test of the emergency broadcast system. So, it turns out that AT&T, cellular network, if you try to transmit a solid tone across it, it'll carry the solid tone for about three seconds and then it attenuates it down. It goes, well, that can't be right. That doesn't sound like voice and the compression kills it. So, it stops being a solid tone so this doesn't work over a cellular phone. That was gonna be my other attempt, but we almost got it to work. We did a partial success. We draped the earbuds into the modem. We did get receive-only for iPhone earbuds. No, we didn't actually, but I'm getting the wrap-up signal so I gotta close up here. This would be the, we're 16 seconds.