I love this art/engineering project. The whole concept of extra-large unintegrated circuits is just so amusing.
The team behind this project sells kits for two classic ICs, the 741 op-amp [1] and the 555 timer [2]. Sadly, they've said the Monster6502 is too big and complex to make a practical kit.
The Apple A8X, found in the iPad Air 2, contains about 3 billion transistors. (This is comparable to the number of transistors in modern desktop computer CPUs as well.) At the scale of the MOnSter 6502, that would take about 885,000 square feet (over 20 acres or 8 hectares) — an area about 940 ft (286 m) square.
We used to have circuits, and when integrated on one piece of silicone, we called them "integrated circuits", while the old ones became "discrete circuits".
I have been kicking around the idea of doing this to something like a NES. Then selling them preassembled. But then I remember I can solder at all and it is decently expensive to do.
Wait, so if I understand this correctly (what I don't know about hardware is a lot!), is this basically a "macroprocessor"? As in, it functions the same way as a vanilla 6502, but done at a larger scale?
Actually pretty cool that it's able to get 1/20th the speed when it's this big!
It's a discrete CPU. This is how processors were made before microprocessors. Before 1970, we had discrete component mainframe machines, in fact running at multi-MHz speeds (e.g. CDC 6600).
The main issue isn't the CPU as such, it's memory. You can reasonably make a usable CPU using relays even, but a non-trivial amount of memory takes up sooo much space.
I learned this in Minecraft. I’d wired up ttls in college for intro to ee for cs majors, which was fun. But ram was a beast. I guess I’ll build another tower and upgrade to a full 1k.
The Virtual Circuit Board game/sandbox at Steam offers an external memory, presumably for this reason.
I remember a Minecraft mod from way back that let you do "modular" redstone, going "into" blocks to build redstone inside them, which you could then reuse as a single block. I wonder if it's been updated.
Discrete logic circuits can be fast because they have a lot of area over which to dissipate heat and can guzzle current.
Moreover, most of them used either ECL or TTL, which are logic families much faster than P/N/CMOS. This being a transistor-level replica of an NMOS 6502, it can't go much faster.
> Discrete logic circuits can be fast because they have a lot of area over which to dissipate heat and can guzzle current.
... but not too fast either, because the long trace lengths are antennas so you'll end up in EMF emission regulations hell on the one side, and signal integrity issues on the other side. On top of that come losses from inductive and capacitive coupling.
Yep. The bummer is that the Apple ][ relies on the clock speed of the 6502 to handle some of its functionality (most notably video handling), so you can’t just run a cable to the CPU socket on the Apple motherboard. I wonder if any of the other classic 6502 machines (Commodore, Atari, Acorn, BBC, etc.) would work?
Again, I know basically nothing about hardware, so this is likely a dumb question: could you conceivably get an oscillator/clock chip that also operates at 1/20th the speed, wire that into the Apple ][, and then wire in this giant CPU?
It is well worth understanding how these older computers worked. All the components were doing double or triple duty, so that together all the tasks like refreshing memory, reading/writing memory, generating video, generating audio, and I/O worked. This involved interleaving bus access and similar "tricks".
There are an excellent series of talks at CCC titled "The Ultimate X talk" [0] where you are shown exactly how. I recommend The Ultimate Acorn Archimedes Talk [1] which introduced the ARM chip, but also shows how shared bus access and optimisation was the core principle of performant designs at the time.
Due to the way the video is interleaved with CPU cycles, you'd have to do something like divide the 1 MHz CPU clock by 20, and then when writing carefully only put data on the data bus for one of those 20 cycles. And the disk accesses would surely fail due to the CPU not keeping up, among other problems.
Thinking more and more about it in this thread (yes, I'm procrastinating doing something else), I'm beginning to come to the conclusion that the way with the least modifications to the computer itself is to replace the DRAM with SRAM, and to "just" slow the video circuit down as well (including color burst, horizontal scan, etc.), but to attach a custom buffered display instead of a regular monitor to the computer.
However, I wouldn't be surprised in the slightest if any other of the integrated circuits don't work at the slower speed for some reason. (Filters etc. you'd just replace/retune.)
Unfortunately, no. The CPU would work, but memory retention would be unreliable (because the DRAM is now going ~20x too long between refreshes), and the composite video output would fail to display on a monitor (because it's the wrong frequency).
Well now I'm wondering; it might not directly run an Apple ][, but what about the Apple ][ compatibles like the Laser 128 [1]? It doesn't use the same ROM as Apple, they licensed Microsoft Basic directly from Microsoft and then added the missing functionality themselves.
I guess someone would have to see how similar it is to a vanilla Apple ][.
1.023 MHz is basically the Apple ]['s fine structure constant. Change that, even by a tiny amount, and the whole universe comes unglued. Video, disk access, you name it.
There were accelerator boards, but they ran at integer multiples of the original clock speed for that reason.
Again, the answer is almost certainly no, unfortunately.
A Laser 128 isn't that fundamentally different to an actual Apple II. In fact, from a wider point of view, they are "almost identical".
DRAM needing timely refresh is universal to DRAM itself: It's because DRAM is literally made of a transistor and a capacitor holding your bit, and if you don't access it every so often, that capacitor discharges eventually. The most realistic path to "slow DRAM" would be to replace DRAM with SRAM altogether, which is effectively made of transistors only, and holds its content as long as just power is applied. It's not that hard to replace DRAM with SRAM: Apart from not needing refresh, SRAM behaves similar to DRAM. It's just more expensive (and usually faster, which is also not a drawback), and the DRAM refresh, while not doing anything useful, wouldn't hurt.
But as for video output, that's just a consequence of how Apple II's (most microcomputers of that era, actually) video output works: By literally controlling the electron gun in the CRT, with only some analogue electronics in between. The monitor however wants to move the electron beam at a certain speed (or, if you had a fancy Multisync monitor, a set or range of speeds). But even if that wasn't the case, the phosphor on the screen glows only for a short amount of time, so draw your picture too slow, and you will only see small parts of it.
The latter can be demonstrated by just pointing a camera at a CRT. In all likeliness, the camera's frequency will be slightly off from the CRT's frequency, so that you capture different parts of the beam's path at each frame, simulating what it would look like if you'd slow down the beam itself: https://pub.mdpi-res.com/sensors/sensors-22-01871/article_de...
Would it be possible to decouple the video output circuit from the microprocessor's timing? Absolutely! And this is in fact what virtually any modern computers has been doing for many decades by now: Even if you still can find analogue VGA with a CRT, that VGA card had its own clock that is completely independent of the CPU's clock. But that's just not how simple home computers of that era usually operated.
From what I can tell from documents like [1], the Laser 128 was very similar (possibly even identical) to an Apple II on a hardware level. It'd have the same issues running at a higher/lower clock speed than it was designed for.
You’ll need additional circuitry to "upscan" your 3Hz display to 60Hz.
There was a moment many years ago when I considered constructing a mechanical 6502. I quickly learned that transistors are fabricated in silicon for economic reasons.
The fun path would be to modify the display to run at 3Hz (and also implying the much slower horizontal scan rate), but due to the phosphor coating's short persistence, you'd at best see only a small sliver of the full picture at any time. Exactly like you see if you point a video camera to a CRT whose frequency does not match up exactly.
But if you'd take a long exposure photo of the CRT, it would likely work! (EDIT: Might fry your phosphor, though... as for any other components, like filters or components that may get damaged by the now only slowly changing current, I just assume you replaced or retuned that as part of the conversion process.)
The big problem is really the CRT (and possibly other ICs in the computer that don't like the extreme slowdown). As noted below, if you went through the arduous process of replacing and retuning components to be able to slow the beam down, your phosphor coating is still too fast, and probably doesn't like having the beam passing only slowly over it. (You thought regular burn-in was bad!)
However, replace the CRT with something else, like an LCD with some extra buffering circuit, and it could work. Yes, the color burst will be at a much lower frequency as well, but just demodulate color at this lower frequency, then.
The software is really not an issue. I saw the MOnSter 6502 live running BASIC, and it was MS BASIC as far as I recall. It just wasn't simply hooked up as the CPU of an Apple II or similar.
> Does it run at the full speed of an original 6502 chip?
> No; it's relatively slow. The MOnSter 6502 runs at about 1/20th the speed of the original, thanks to the much larger capacitance of the design. The maximum reliable clock rate is around 50 kHz. The primary limit to the clock speed is the gate capacitance of the MOSFETs that we are using, which is much larger than the capacitance of the MOSFETs on an original 6502 die.
Now I'm curious but in way over my head.
Could the speed be improved by using MOSFETs with better capacitance?
Can the full 1.023 Mhz be attained by throwing money at it or are there physical limitations at that scale?
I'm not an expert but a hobbyist. Take what I say with a grain of salt.
> Could the speed be improved by using MOSFETs with better capacitance?
Not using MOSFETs would make it faster, in theory. MOSFETs have a delay time and are considered slow. But it appears they want to be true to how the original was made. There definitely exist mosfets indistinguishable from dust - I made the mistake of designing with a few at one point. So probably, yes.
Another thing I noticed is that the distance between the components on the board is quite high. They might have done that to replicate the original layout, or maybe I'm just not seeing it correctly. But everything - including the traces on the board - has some capacitance.
> Can the full 1.023 Mhz be attained by throwing money at it or are there physical limitations at that scale?
Hard to answer without measuring, though I'd imagine it's possible. 1MHz isn't that fast, but if you have high capacitances or they don't switch fast enough (e.g. MOSFETs) then your signals are going to get muddy.
EDIT: to be clear, not ragging on the project. This stuff is beyond cool.
This is basically a second generation computer, whose heyday was in the first half of the 1960s – after vacuum tubes, but before the first integrated circuits.
It makes me wonder, what transistor family would give you the best performance for this (very unusual nowadays) use case? Obviously not MOSFETs.
From what I understand, germanium transistors were most common for 2nd generation computers, because (at the time) silicon transistor technology wasn't sufficiently mature to be used for that application.
The real question is why would you. This is an educational project that visually demonstrates how a CPU operates. It could clock at 0.5 Hz for that purpose.
If it ran at full speed, you could run tons of original 6502 software, which almost universally rely on cycle-accurate timing to generate video output.
I assume they use garden variety MOSFETs like 2N7002, which cost a few pennies when bought in bulk.
The gate capacitance is 20 pF. One logic gate output is typically driving several inputs, so the typical load is several-fold of this capacitance. One could check the schematics for the maximum fan-out in the real circuit. The longest, one foot long traces on the PCB add another few tens of pF. So the capacitance seen by an output of a logic gate will probably be in the range of 20-100 pF most of the time. But it is the slowest gates which determine the speed of the whole circuit, so the worst case could be much worse.
These transistors themselves are extremely fast and can switch on and off in about 3 ns. Their channel resistance is in single Ohm range even at the lowish gate voltages, so the RC time constant is also very small 100pF * 3 Ohms = 0.3 ns. Thus this is also not an issue. The slowness of the circuit comes from elsewhere.
First, from the resistors which pull the logic gate outputs to the supply voltage. There is one for each logic gate, so a total of one thousand of these, and in the first approximation half of them conduct at any given time. To keep power consumption low, these resistors must have relatively high resistance values. If we limit the current for the logic to 1 A total, that is 2 mA for each of the 500 resistors. Therefore the resistors must be 2.5 kOhm for 5 V supply voltage. Thus charging of the nodes of the circuit through these 2.5 kOhm resistors is almost a thousand times slower than discharging the same nodes through the 3 Ohm channel resistance of a fully open transistor. Still, this is not too bad -- on the order of 250 ns of delay per gate.
But the frequency for the whole circuit is determined by delay not through just one, but through many gates in series, plus there are probably parts of the circuit topology which are not optimal for implementing them with discrete transistors, where the delay does not follow from such a simple reasoning as above.
The speed can probably be easily increased by an order of magnitude if one were willing to spend 50W instead of 5W for the circuit. Beyond that, one would want a CPU circuit designed specifically for implementation in discrete transistors, not a discrete element copy of a monolithic chip.
I signed up for the mailing list when it was first on HN a few years ago, but haven't seen many signs of progress. Website still says mid 2023 for a launch.
6502 was my first CPU so I'm totally down to buy one, even though it's going to be pretty expensive. I'd be pleased if they could make them for less than $5k.
Wonder if it's feasable and how much work it would be to record a run of eg Super Mario Land and play the processor instructions back on this thing hanging on the wall.
Would be a nice conversation piece.
Dot matrix display. Progress bar is underneath the Level indicator.
Plus presence sensor. Not PIR. And RTC in/on whatever plays the instructions to always seemlessly pick up on the loop.
Beautiful work, and a great choice of processor to emulate. As I understand it, they also tried to get close to the layout of the actual chip, not just make a functional equivalent circuit.
The next time I have some time, I would love to do a discrete RISC-V, but I know how big a project this sort of thing is.
I compared the sizes of several processors[1] including these RISC-V: Glacial, SERV, PicoRV32, VexRiscv and Darkriscv. Compiled to NAND gates the results were 2063, 3595, 34463, 49214 and 50424 gates respectively. Multiply these numbers by 4 to know how many transistors a CMOS implementation would need.
I also implemented them in a bunch of different FPGAs, though that is less relevant here.
Glacial is an 8 bit processor optimized to emulate a 32 bit RISC-V and SERV is a serial implementation of a 32 bit RISC-V. They save transistors at the cost of many clocks per instruction.
There are several projects that implement RISC-V using TTLs but I have not heard of one using individual transistors.
I have my own core design that I would probably use, and you definitely don't need to use CMOS logic with your discretes. You can also get away with a lot more things than verilog lets you do, like you can use shared buses with internal 3-state drivers. That stuff cuts down your transistor count a lot.
Gigatron is a retro 8-bit computer with no microprocessor. Instead, its processing is done through thoughtful application of “TTL” logic chips. It’s another project that helps you “see” the processor’s internals. Kits aren’t for sale anymore through the official website, but you can find unpopulated PCBs and the components shouldn’t be hard to find.
And had much larger "nodes", given that those tubes are much, much larger and farther apart than the SMT transistors on this 6502 replica. It's not even close.
Those little square "chips" are not actually integrated circuits, but effectively small PCBs with surface mounted discrete components (including transistors).
The early PDPs used discrete CPUs, so they'd probably be the closest examples, the 12-bit PDP-5 is the smallest those CPUs went though, which isn't hideously more complex than a 8-bit CPU.
I thought about CPUs with either discrete transistors, or with e.g. 74xx ICs. But the former intuitively seemed too big to me (because I was assuming they use large transistor packages instead of SMT), and the latter too small (because a single 74xx logic chip can pack a lot of transistors).
So this “hybrid” of having SMT components in a can by IBM seemed close.
Complete working transistor-scale replica of the classic MOS6502 microprocessor - https://news.ycombinator.com/item?id=33841901 - Dec 2022 (38 comments)
MOnSter 6502 - https://news.ycombinator.com/item?id=26507525 - March 2021 (31 comments)
The MOnSter 6502: transistor-scale replica of classic MOS 6502 microprocessor - https://news.ycombinator.com/item?id=17969472 - Sept 2018 (81 comments)
A working, transistor-scale replica of the MOS 6502 microprocessor - https://news.ycombinator.com/item?id=14386413 - May 2017 (44 comments)
The MOnSter 6502 - https://news.ycombinator.com/item?id=11703596 - May 2016 (74 comments)