Hi, its video number 1 1 1 1. You know what that means. It's the end of 4-bit County and we're flipping over to the fifth bit. Well, we would be if we actually counted Eevblog episodes in binary, but we don't.

Mere details is video number 1 1 1 1. So I Thought we'd take a look at the world's first microcontroller, which just so happens to be a 4-bit microcontroller. None of this a bit rubbish. and to go along with it I Thought we'd take a look at the one of the world's first electronic games to use a microcontroller.

the Merlin that? check it out? Who remembers The Merlin Fantastic came out in 1978. The Merlin was released in 1978, along with the famous Simon game as well, which is a much simpler game. They basically came out at the same time, and whilst technically they weren't the first games on the market to use a microcontroller at the time, they were the two games that became massively popular because of their functionality and their low price bracket. The melon and the Simon were about 25 US dollars at the time.

some companies were even selling it and close to its cost price of around about 20 US dollars, and both of them just so happened to use the world's first micro controller, the Texas Instruments TMS 1000. So let's take a look at it. The Merlin actually was the best-selling toy in 1980. selling 2.2 million units, it was phenomenally popular.

I have one of these as a kid. It was absolutely remarkable. I Was just amazed that a computer inside like a toy. there's a computer in here because you got to remember this is 1978 and whilst like 1977, the year before was famously the year that the Commodore pet came out, the Apple 2 the Tandy Trs-80 model 1.

so you know computers are out there. the Atari 2600 video game, of course came out, but these were relatively expensive items. Um, so you know people knew what computers were, but to get a computer in a toy that cost like twenty to twenty-five dollars was mind blowing. The Merlin was actually developed by Bob and Holly Doyle who formed an independent game company a few years earlier along with Holly's brother Wendell Thomas and they were contracting to Parker Brothers at the time Parker Brothers said hey, we want one of these electronic games In fact, they had actually developed an earlier game to this called our Sector or it's our what they wanted to call it was sink the Sub and it came out slightly before the mill and I believe.

But then you know it wasn't hugely popular Because it it didn't, it really only appealed a certain niche market, whereas they wanted something that would every kid would want for Christmas And so they said, hey, give us your game ideas, what do you got and they developed at the moment and showed them a prototype and they went. Yep, let's do that And they had a lot more games than the six games in one that. came with this. and there was a bit of you know tooling of throwing over what games would be included in this, but this was more popular.

I think than the Simon because the Simon was basically a one trick pony. It just did you know a simple memory type flashing LED game. This actually had six games built in. Remarkable.

So Bob and Holly Doyle were both Harvard astrophysicists and Wendell Thomas he was actually a computer scientist from IBM So they went. What do we do? Let's make games. It's got to be money in games. And sure enough, come set 1978 there was.

So both the Simon and the Merlin used the world's first 4-bit microcontroller. Nonetheless, a bit rubbish. 4-bit TMS 1000 processor. In fact, the Merlin uses a slightly higher memory version of that, the TMS 1100.

And curiously, the microcontroller didn't actually come out in 1977 or 1978. When these made there well, the first electronic games to use, them made their appearance. it actually came out in 1974, basically four years earlier. But it took those four years for the game industry to catch on and go.

Hey, we can use these microcontrollers in electronic toys and they only cost a couple of dollars per chip. And they can sell a $20 item because as you will see in the teardown of this thing, the microcontroller is does practically everything. this. I think it really got to get their bill of materials cost down because the microcontroller had everything built in and had the RAM, the ROM and the I/o everything you wanted and that's all you had to use.

A single microcontroller could do everything and the rest of it was just some LEDs and some molding and cases and stuff like that and you could sell it for like 25 bucks. retail. Crazy. Of course we take that for granted now where you can buy a farting novelty gadget for $1 delivered on eBay but back in 1978 it was a huge deal and everyone was amazed that you could get a computer inside a toy.

So here it is and it really is a funky form factor. What I loved about this was the fact that it had these membrane touch keys which I had no idea what membrane touch keys didn't even know the name for it back when I was a kid, but I thought it was just magic that he could just touch these like they didn't feel like a real button for the time. It was absolutely magical and the fact that there was an LED behind each one of these and you could press the button and see the LED at the same time. It's like wow, how did that work anyway I'm the form factor was designed I believe to be like a familiar and at the time like a like a phone handset in apparently surveys they did show that yeah, it Was a sort of you know, kind of like a familiar form factor to people.

He did have a DC jack or a battery eliminator on the side and there was a power switch which confusingly was not labeled at all. And of course it's got a what looks like a 57-millimeter speaker up in the top here. And of course like you might think well, why does it need such a big speaker like I've got a speaker in my phone? Yeah well, that's modern, you know, electromagnetic speaker technology. Back then it was like yep, we're gonna just use a paper cone off-the-shelf 57-millimeter driver and they just designed it around that dollies Buzzard Leone Piezo buzzer.

Rubbish. No genuine speaker. So basically it was just a two-part mold in case. Of course you have to get the price point down for these games are plus the back door as well.

but apart from two part mold in very large self tappers all into plastic I Hate that snap I hate it. but well yeah, there it comes out complete with the plastic on it. Yeah, so it looks like it's got a couple of plastic clips at the end. Alright, so let's open this up.

should just star in like Flynn Check it out. Oh hello and yet classic 1970s single sided PCB Construction As you'd expect, just a phenolic base PCB Of course that's done to where I get the cost down. You don't want any Of that double sided PCB Rubbish. Not in a consumer product.

I'm And even today, you'll still find that a lot of consumer goods you know, white goods TVs and things like that use single sided. PCBs It's against the constant and you can see that there's not much in here. There's the micro controller as we'll take a look at, which was absolutely revolutionary for the day. It was so incredible that they actually got the price down and they did it with the micro controller, but also did it with various other aspects.

Actually, take a look at the rest of it here. There's our membrane keeper up our membrane. keep head under there. They've got another look at this, a complete molding like this another.

So what's that like for mouldings? They've got the upper case, the lower case, this lead, I which is also used as the LED diffuser, and the back panel. So there's four plastic mouldings like that. But the other thing about it is that you'll notice that the lids actually go into mouldings on there, not only to i defuse it, but also to hold it in place so that you don't need any screws inside this thing. There's just the two screws to hold it together.

Because getting that price point down is not about not only about putting it in the microcontroller with nothing else on there, there's just going to be some LEDs and a switch. And that's basically if the batteries are probably just powered, not even a voltage regulator in the hair down. so they're just pairing it directly from the batteries and not only to get the component bill materials cost down, but also assembly time as well because the more screws you have, if you have to put four screws in here to screw this board down, that's going to take time and time cost money. So they do away with that with the big plastic molding it makes they needed.

sort of. You know that anyway. really too wide to get the defusing on the lids. You could have used the internal defusing on the LEDs I Guess you know the domed top on them just like that and it might not being as as good but you know it's still to worked anyway.

they've got that cost down. They figured they have to do that mold in anyway so might as well make that. Now let's have a look at thee. There we go.

There's our membrane keypad ER and when I tore apart this is a kid I Thought this is absolute magic I Didn't know how it worked and never seen a membrane I keep had before. So rather than the conductive you know, like carbon of the day this I presume it's like some form of like silver or silver alloy, something like that. And they've got an initiate here which is the separator which just insulates the two like that. because there's no insulation on these traces, they're all exposed so you need that inner one.

Of course these cost practically nothing. especially Inlet the inner sheet. So the separators. and that's how they get the button you press down on that.

It doesn't require much force and that was magical. And of course you can have the cut out so the LED shines through and at the same time as the lead shining through. There you go. You can just make contact just with those keys like that.

Brilliant. Took them for revisions to get it right though. Rev D and it's the PCB A Rev D as well. I Don't know what that part number means.

Anyway, you can see that because this has to be the world's largest solder mask expansion. Look at that. I mean that's just incredible. None of this solder mask between pins rubbish that you get these days.

They just didn't need it. This is all our wave soldered. You can just tell by the the uniform nosov, the joints, and that they obviously didn't have a problem with. you know, bridging between pins or anything like that.

But you know they went to the effort to put the solder mask on there and I saw the mask expansion around the flat flex connector as well. But apart from that, I mean there's not much. It's not much on here at all. Let's flip it over.

And they didn't even bother with screws for the speaker as well. Just some double-sided sponge tape on the top and like a you know, a reasonably tight fit designed into there and just stuck on the top. Nice. But you can see now there's no form of regulation or anything like that.

The micro is powered directly from the batteries. Do they have a what's wears? Got one cap in there? it's like they wouldn't need that for decoupling and another cap here. RC Oscillator. And there's the brains of this thing.

The Texas Instruments TMS 1100. Of course, the TMS 1000 was the world's first micro controller. What defines a micro controller is that as opposed to a microprocessor is that a microprocessor needs external memory RAM and program memory as well, whereas a microcontroller has those built in. Nowadays, you think of microcontrollers as having building analog to digital converters.

you know, timers and serial, you arts and all that sort of stuff. But back then it was, you know, like to have that sort of start. They didn't really have those peripherals back then, but just having the RAM and the ROM integrated. in this case, it's a mask.

ROM None of this reprogrammable rubbish flush wasn't invented yet, a squared prom wasn't invented yet, and maybe they had a windowed version of this like in like for development that would have cost a lot of money, but this one would have cost them a dollar or two each. and but they were mask programmable, one-time programmable and there's a more modern pic microcontroller and you can see the pin pitch difference there so you know it's not your standard point one inch pitch, much finer. That was pretty advanced stuff for 1970 and we've got a diode in there as well or as I call them at the time. Went on a kid diode ease because I'd never heard anyone say the word diode before so you know that's what I called them.

Anyway, we've got a cap in there ceramic Joby what value? Hundred seven hundred n or a hundred puffs. This one looks like manufactured in Korea in the 8th week 1980. It's a modern one and that matches the day code here. Fifth week 1980 in Singapore I Don't want my viewers in Singapore Singapore Menu: two chips back then it's quite common.

So here's our hundred puffs cap on the back here. and if you follow the money from this pin here, let's go around. always follow the money right into the basement car park at midnight. It's where all you fight get all the good info.

So that's actually pin 20 which is the VSS or ground pin and that goes over to the they've switched the ground there and you can see that the two oscillator pins 18 and 19 are actually shorted out so I didn't but do believe it's actually using the internal oscillator but it's also got a pair on needs a power on reset. so maybe that's what this chip I this cap over here is doing Aa plus the diet as part of that reset circuit - And because I know you want to know what the clock rate is. Well, let's have a look 340 Kilohertz Nice Ah screaming and we'll just capture the speaker here. There we go and check this out if we have a look at the original design patent for this thing.

There's three names on there our Bob Doyle I Don't know who Arthur Venditti is, but apparently Samuel Killman was a designer at Parker Brothers and apparently he's responsible for the futuristic design of this thing and look, isn't it great? You can see that it's like it's very like Star Wars You know the Droid II who kind of look I Just love this. Anyway, it's just a design Payton It's not an actual Payton for the game from the engineering staff at Texas Instruments Incorporated and they show the dye. Love it! Copyright 1976 and we see here that the TMS 1100 is basically pin 4 be compatible if the original TMS 1000, but it's got. Even though it's a 4-bit processor, it's actually 2 K by 8 ROM and 512 bits of RAM as opposed to I Think it's half that in the original.

Now we can see just the internal block diagrams of how it works in its a microcontroller because it has the building Ram and Rama, but you'll see it's got nothing else. No newfangled ABCs no even simple you arts or anything like that. There's many variants of this R1000 family. Some of them had higher voltage capability, opened source and open drain outputs as well for driving, you know, Nixie tube displays and all sorts of other stuff.

And the standard instruction set for those playing along at home contains two two chapters of 16 pages. Each page contains 60 80 bit words, etc. etc. But it was a four bit processor even though it used an 8 bit instruction word.

It's not exactly a lot of instructions here. it's it's ok. It does the business. You know it's similar to any sort of like a modern peak process or anything like that.

similar. I Think it's 40 odd in instructions or something like that. All the instructions actually executed in six clock cycles. here.

As you can see, there's the six clock cycles and it's split up into multiple phases. so quite complicated. phase clocking internally. But yeah, I Don't believe you have to really worry about that sort of stuff, but stand to be corrected.

And here's all the different variants on there. They'll probably more than that in the end. But as you can see the TMS 1100 here nominal power supply 15 volts, but obviously operating it well below that because we operating 6 volts directly from the batteries. That has no problem doing that, but to be operating at the higher voltages means it was fully compatible with for like 4000 series.

CMOS logic and stuff like that. Very nice I'm nominal at 15 volts, hundred and five million what? So the processor wouldn't have been drawing much at all. most of them never power would have been coming from the LEDs. The LEDs weren't hugely efficient back in 1978, so that's why they're a bit piss-weak on there I Remember back when I used it, they were pretty piece weak LEDs and we weren't very bright at all, so maybe they were trading off some brightness with battery life and the output drivers are more than capable I think 10 to 15 milliamps each.

They're more than capable of driving the LEDs directly, and if you want to see their development environment, I'm here it is. They got a flow chart of how it all worked. There was a simulator apparently, and the Doyles actually developed this on an Intel development system. don't know exactly which one, but it's probably would not be too dissimilar to the one that I just found at the Computer Museum Warehouse closing thing and apparently that was worth 25 grand at the time.

That development system just to develop for your microcontroller like crazy. and this is rather interesting. the instruction programmable logic array. it's got a pla in or a p LD basically.

and they give this example of a obesity to seven-segment decoder here and there. the fuse mappings in order to map the outputs like that. To have that built into a micro? That's pretty cool. and it's a very basic micro.

Like, There's no interrupts in this thing at all, so all the logic would just be sequential. That'll be it. So can you imagine what sort of effort would be required to hand code a game like this in assembly? They did well to get these six games into the head. 2k Words Awesome.

Anyway, I hope you enjoyed that. Look at the Mill and 1978 4-bit processor technology. Oh I Was gonna like do some video actually just playing the games on this thing, but no way and included here probably include that over on the second channel link at the end. So you want to check out Eevblog too to see that anyway.

If you liked it, please give it a big thumbs up. As always, discuss down below. catch you next time you.