The Calculator that spawned the Microprocessor (2009) | Hacker News Reader

The Calculator that spawned the Microprocessor (2009)

vintagecalculators.com

125 pointsfortran7746 comments

Just in case anyone wondered Mostek (designers of the MK6010 "calculator on a chip") is a completely different company from MOS Technology (designers of the 6502 processor). I used to find this confusing.

Mostek was also one of the producers of the z80 (and as a Mostek z80 was in my Nascom I was always aware of the company as different from MOS)

Interesting that using microprocessors in (simple) calculators was very short-lived, they were very quickly replaced by custom-made controllers: "While the 4000 series microprocessor chip-set was being developed, the "calculator on a chip" MK6010 was developed for Busicom by Mostek.".

But the TI84 series uses a Z80 so it seems that the microprocessors won in the end.

If the TI84 were the only calculator, that would indeed be the conclusion. It's not, though.

Perhaps we could say that the microprocessors didn't totally lose.

Or is it more generally true that all high-end calculators use microprocessors?

I'd argue that the most common calculators are the four-function calculators (plus GT, MR etc.) used by sellers worldwide in the non-Western world that don't want to use full-function POS systems and still use paper ledgers – the ones with solar cells and battery lifetimes in the order of months.

While they used to have special-purpose controllers, sadly, they seem to be shoving Android in them now... perhaps for a good upgrade path, though. [1]

[1] https://www.youtube.com/watch?v=ui2UVUXiIzY

Keep in mind that Busicom was pretty much asking for a custom made controller in the beginning (more accurately, a set of chios), but Intel suggested a simpler solution. In the end, calculators ended up being sold in such large volumes that it ended up being cheaper to have an application specific controller. (That said, more sophisticated calculators still use microcontrollers.)

Counter example: People now use their smartphone's calculator app, and that is powered by a general purpose CPU core.

The irony of course being that the GUI requires thousands of more calculations than the actual math problem you're calculating.

it's the future, we made our calculators so big and complicated that we have to calculate a smaller calculator for our calculations

You just need to ask Chat GPT then ask Gemini and CoPilot if the answer is right.

Interesting that Japanese companies sometimes help competitors, such as Sharp helping Busicom in the article. I also think Yamaha helped Korg in their early years [1].

[1] https://www.soundonsound.com/series/history-korg

Canon and Nikon. Sometimes it’s due to personal relationships and owed favors that would be dishonorable to ignore.

Can you think of an example where they helped each other? It's not an easy topic to google.

The closest I can think of is Nikon making the lenses for the first Canon cameras. I don’t really have a source for that offhand, though.

https://www.mir.com.my/rb/photography/companies/nikon/htmls/... claims that was because they weren’t competitors yet:

“Unfortunately, Nikon had not yet perfected a camera body. In fact, there was no Japanese camera that could compliment the NIKKOR lens. Then one day, Canon found its way to Nikon's door. The courtship was short and in 1934, the marriage was struck. The result you are seeing here:

The first Japanese 35mm camera, which was made in 1935, and its Nikkor 50mm F3.5 lens. NlKKOR lenses continued to be used in Canon cameras until 1948, when Nikon created its first camera, the Nikon 1.”

I'm also not sure if it qualifies as help but Nikon F80 was used by Kodak and Fuji to make digital cameras on the Nikon F mount system for many years. We know what happened to Kodak, but Fuji went on to make their own successful cameras and lenses for the same market

I love how the traces are smooth and rounded rather than square and thin like modern pcbs.

They look like they were hand-drawn, and they probably were.

Does anybody knows if modern pcb cad software can do that? Like, KiCAD or others ?

I don't know, but I remember reading that curved is better but harder to do in software (at least in the past). Reddit agrees: <https://old.reddit.com/r/AskElectronics/comments/iejapa/why_...>. I wonder if the machines used to print boards play a role, too.

There are several plugins for making curved traces in KiCAD.

These boards were basically laid out using masking tape on transparencies. The rounding comes naturally from turning a corner with the narrow tape. The teardrops around pads and other more complex shapes (like footprints) came from pre-fabricated stickers. The layouts were produced in 4:1 or similar scale and were optically reduced during photolithography of the resist on the actual PCB (just like semiconductor masks).

Way back then you also had some concerns about manufacturability because sharp corners tend to be undercut during etching when your process is poor. That's probably why they usually added little fillets to T and similar connections.

Yup. On a smaller production scale, we did prototype PCBs with photoresist directly on copper clad boards.

There's a not-OSS, not cheap autorouter called TopoR that will do this. There are plugins of varying quality for other tools. And you can usually do something manually if you have spare time.

https://www.youtube.com/watch?v=euJgtLcWWyo

Huh. It never occurred to me that the word chipset was the descendent of a set of chips. That's nifty

Even until recently (a few generations ago), motherboards still had northbirdges and southbridges.

Desktop Zen3 has a chip you can point to and call Northridge, and a chip you can point to and call southbridge. They just moved the Northridge next to the 1-2 CPU chips on the CPU side of the socket. Note that laptop (APU) Zen3 uses a single chip containing CPU, GPU, and Northridge functionality.

I always wonder where we’re going to see the next kind of world changing innovation like this.

Immunotherapy? Maybe.

Recently all I’ve gotten is a phone that listens to my conversations and serves me targeted ads.

Feels like a lot of people who should know have been saying biotech for a while, but I’m unclear what big breaks are even being worked on. Not that targeted cancer therapies isn’t super cool and worthwhile work, but it does feel a successful result still basically fits the existing industry shape; it’s not the kind of thing that would be spawning entire new industries the way computers and the internet did.

I think a big part of that comes down to regulation when it comes to biotech, which, frankly, I think is a good thing. After all, skimping on software (usually) won’t kill anyone, but imagine how much worse things could have been with something like Theranos absent regulation.¹

⸻

1. For those rushing to point out that regulation didn’t stop Theranos, it might not have kept the fraud from starting, but it did keep it from continuing.

Pointing out Theranos would be like pointing out the existence of double agents despite rigorous counter-measures. Things will happen; one wants them to happen less.

The groundwork isn't there. Computers were the result of blending Maxwell, Boolean algebra, and some fairly obscure speculations about computability in the most abstract sense.

There's nothing equivalent waiting for someone to industrialise it. AI is being hammered together out of bits and pieces, but there's no Grand Theory. Robotics is the same, but further behind. Bio is similar but maybe a little ahead.

Physics and math just haven't primed the pump for another revolution.

How do you know? The pump could be primed with all sorts of obscure, non-obvious things the importance of which only becomes evident after the fact when someone combines them.

I think the importance of Boolean algebra and computability theory to early computers is vastly overstated. I see no influence of computability theory on ENIAC, for instance. As for Boolean algebra, it seems more descriptive than useful for early computers. But I'm very interested if people have counterexamples.

Millions of people, at all stages of the age curve, not dying or losing their marbles to diseases that can be treated by immunotherapy? I foresee many opportunities.

Time will tell…

I think that we’re definitely seeing some world-changing things happening in the vaccine world with mRNA vaccines. Covid accelerated development, but there was a lot of movement before that already.

LLMs no good for you?

Not OP, but what game changing uses are you finding for LLMs? I mostly find they're good for brainstorming ideas of activities to do, asking "is there a thing like this and what is it called", and writing trite correspondence to acquaintances. It's useful for me but hardly life changing.

Putting it like that, and I like putting it like that, sounds like LLMs are to Wikipedia what shoes are to people who want to go from A to B.

That computers can reasonably handle natural language instructions is pretty amazing.

As an aside, immunotherapy is pretty neat stuff. My fathers bladder cancer is in remission thanks to it.

Yes, there are still some issues being worked out (possibility to cause diabetes) but it is just neat to see in action. The medical field is one that still seems to be making these kinds of leaps. It is wild seeing some diseases that only a decade ago were a death sentence or had extreme solutions now becoming treatable.

The ancient design of the page is very topical.

Simple, fast-loading and unobtrusive, definitely better than a lot of "modern" web design.

I dig it. As a vintage tech enthusiast, the last thing I want is a bunch of javascript or framework-of-the-moment hoohah pummeling my browser. I want pictures and info. HTML with a bit of CSS is perfectly fine for that. There's a time and place for the girthy websites of the modern era, but sometimes my viewport just needs a break.

I used to browse this website on my grandparents' dialup in like 2004 and I swear it looks the same as it did back then. It still brings the good feelings.

Perusing the web source: `<META NAME="Generator" CONTENT="NetObjects Fusion 11 for Windows"/>`. A tool developed in 1996...

Reply on news.ycombinator.com