03:37:59
##forth
<lf94>
tabemann: i return
03:38:10
##forth
<tabemann>
hey
03:38:11
##forth
<lf94>
I spent my downtime reading the RP2040 datasheet today lol.
03:38:27
##forth
<lf94>
You had already filled a lot of information
03:38:34
##forth
<lf94>
One thing I think I havent done is funcsel
03:38:36
##forth
<lf94>
GPIO_CTRL_FUNCSEL!
03:38:57
##forth
<lf94>
PIO is also... insanely complex.
03:39:02
##forth
<tabemann>
that doesn't hurt
03:39:10
##forth
<lf94>
I think it's part of this not working though
03:39:12
##forth
<tabemann>
I just finished a 'matrix' demo for the PicoCalc
03:39:22
##forth
<lf94>
Daaamn nice!
03:40:46
##forth
<tabemann>
it's actually pretty slow (I didn't need to add any delays or like) because it does all its coloring with ANSI escapes, which the PicoCalc terminal emulator has to parse
03:41:18
##forth
<tabemann>
it sure looks pretty though
12:53:49
##forth
<cleobuline>
forthBot: LOAD ini.fth
12:53:50
##forth
<forthBot>
File ini.fth with moon loaded
12:53:55
##forth
<cleobuline>
forthBot: MOON
12:53:56
##forth
<forthBot>
Phase de la lune pour Mon June 16 2025
12:53:56
##forth
<forthBot>
🌖 Gibbeuse decroissante La lune decroit, une nuit douce vous attend ! Illumination 74%
14:53:55
##forth
<forthBot>
Environment for cleobuline inactive, freeing...
16:58:26
##forth
<tabemann>
back
17:00:34
##forth
<cleobuline>
forthBot: LOAD ini.fth
17:00:35
##forth
<forthBot>
File ini.fth with moon loaded
17:00:44
##forth
<cleobuline>
forthBot: QUOTE
17:00:44
##forth
<forthBot>
When we are dead we do not know that we are dead. It is for others that it is difficult. When we are stupid it is the same - Jean Claude Vandamme -
19:00:44
##forth
<forthBot>
Environment for cleobuline inactive, freeing...
20:07:48
##forth
<xentrac>
lf94: that's awesome! what do you think of the RP2040? have you thought about upgrading to the RP2350?
20:09:07
##forth
<xentrac>
tabemann: I didn't know about zeptoforth! thanks to lf94 for mentioning it
20:09:30
##forth
<xentrac>
I like Cortex-M a lot, although I think Thumb2 is a lot more bearable than Thumb
20:11:10
##forth
<xentrac>
lf94: the main reason for the weird RST instruction system on the Gameboy is that the 8080 didn't have an interrupt pin, so peripherals had to do interrupts by jamming an RST instruction on the bus when the processor went to fetch an instruction
20:11:20
##forth
<xentrac>
as I understand it
20:11:59
##forth
<xentrac>
02:56 < MrMobius> not that you have to do it like everyone else but if you want your bot to be more or less standard, you'll need to do it how most other forths do
20:12:29
##forth
<xentrac>
it's already case-sensitive, and capitals-only, which is a pretty unpleasant departure from the standard; you can't program the bot without yelling at it
20:14:33
##forth
<xentrac>
I think invert was a pretty reasonable choice; when you have a name clash between existing implementations (like figForth not vs. Forth-83 not), it's better to define a new name that is unambiguous. I don't like that invert is so long though
20:17:39
##forth
<xentrac>
04:18 < tabemann> oh, characters can't be specified with 'x'
20:18:41
##forth
<xentrac>
the 'x' syntax isn't in dpANS94 FORTH but does seem to be in the forth-standard.org Forth, which seems like a major syntactic departure; if you're going to have sigils then you could expand them far beyond character literals
20:45:45
##forth
<lispmacs[work]>
hi, is there anyone here much interesting in analog computing? was interested in starting a chat community about that but having trouble finding others interested in the subject. started a channel at #analogcomputing
20:46:23
##forth
<lispmacs[work]>
I built my own analog computer and also am saving up to buy one
21:49:38
##forth
<xentrac>
I wish people wouldn't call them "analog computers". they're not instances of what we call "compuers'
21:53:27
##forth
<xentrac>
it's like calling me a "computer" when I extract a square root with pencil and paper. Historically accurate, but anachronistic, and almost certain to be misunderstood
21:54:15
##forth
<xentrac>
the standard term nowadays for what used to be called "analog computers" is probably "analog electronics", although perhaps you mean more specifically "analog differential analyzers"
21:55:52
##forth
<tpbsd>
xentrac, the term is certainly not 'analog electronics', I specialise in that being instrumentation, psu's, motor control etc
21:56:17
##forth
<xentrac>
instrumentation and motor control are very much within the scope of what used to be called "analog computers"
21:56:24
##forth
<tpbsd>
'analog computation' is probably accurate
21:56:39
##forth
<xentrac>
I think it's still misleading
21:56:49
##forth
<tpbsd>
and so it's not a long jump to 'analog computers'
21:57:02
##forth
<xentrac>
digital computers are called that because the publicly known ones were initially funded to replace women who computed tables of logarithms or artillery aiming etc. but they turned out to be something completely different
21:57:21
##forth
<xentrac>
you can't encrypt a message, compile a program, or run an IRC client on an "analog computer"
21:57:46
##forth
<tpbsd>
Ive never heard anyone, ever describe instrumentation and motor control as 'analog computers' and Im 71 years old
21:58:33
##forth
<xentrac>
71 year ago the term was already falling out of use
21:58:34
##forth
<tpbsd>
motor control is mainly closed loop feedback and instrumentation is mainly accurate amplifiers
21:59:07
##forth
<xentrac>
if you showed an instrumentation circuit with a few op-amps in it to various electrical engineers from 80 years ago, the people who recognized it would be the people who were then building things like electronic fire control "computers"
21:59:46
##forth
<tpbsd>
ok, good point, but very arcane
22:00:03
##forth
<xentrac>
it's not arcane. that's precisely when people talked about building "analog computers" seriously
22:00:09
##forth
<xentrac>
because that's really where op-amps and the related techniques came from, although early versions came out of the Bell System
22:01:00
##forth
<tpbsd>
80 years ago people only had analog, and what they had was was never going to lead to GPU's and AI tho
22:01:32
##forth
<xentrac>
no, 80 years ago was 01942. there were already analog-to-digital converters
22:02:09
##forth
<tpbsd>
gtg, bbs
22:02:58
##forth
<xentrac>
op-amps are called "operational amplifiers" because they can carry out operations like differentiation, integration, and addition. That's what the paper that introduced the term was about: electronic analog differential analyzers, to supplant the mechanical analog differential analyzers that were already in sue
22:03:03
##forth
<xentrac>
*use
22:06:07
##forth
<xentrac>
he thanks Vannevar Bush, who is the guy that made mechanical analog differential analyzers practical, for "helpful encouragement and criticism"
22:11:09
##forth
<xentrac>
toward the end of the thesis he shows how to do binary arithmetic with relays, and ENIAC was proposed in 01942 (though using one-hot base-10!)
22:13:15
##forth
<xentrac>
of course people had been using non-electronic digital algorithms for numerical computation since at least ancient Sumeria
22:14:22
##forth
<xentrac>
anyway, so, lispmacs[work], if you want to talk to people about building circuits that do analog calculations electronically such as aerodynamic simulation, maybe try ##electronics
22:19:50
##forth
<xentrac>
also, though, since now you can buy four op-amps on a tiny chip that runs on 5 volts instead of an op-amp being a two-vacuum-tube device the size of a coffee cup that runs on 300 volts, the kinds of op-amp circuits that were considered "analog computers" in the 01940s and 01950s are now ubiquitous in things like... instrumentation, motor control, and power supplies
22:20:26
##forth
<xentrac>
as tpbsd says, nobody calls them "analog computers" any more, they just say "analog circuits"
22:54:13
##forth
<tpbsd>
I didn't say that at all. People have been calling analog computional devices 'analog computers all my life, and I'm 71 years old
22:54:57
##forth
<tpbsd>
xentrac, are you a troll ? you sound like one when you quote things I didnt say
22:56:08
##forth
<tpbsd>
lispmacs[work], you go right ahead and call your analog computer a 'analog computer' because people always have
23:01:17
##forth
<xentrac>
no, I'm serious. you said that all your life you've "never heard anyone, ever describe instrumentation and motor control as 'analog computers'", even though you acknowledge they contain precisely the kinds of circuit designs that people called "analog computers" 80 years ago
23:02:44
##forth
<xentrac>
even a simple analog PID controller already has most of the parts of the devices that people sold as "analog computers" 70 years ago, even including user-settable gains; the only thing missing was the plugboard
23:02:58
##forth
<tpbsd>
no, youre making a massive jump of logic, youve jumped right into troll land
23:03:02
##forth
<xentrac>
which, if you're building the circuit on a solderless breadboard, you also have
23:03:46
##forth
<xentrac>
well, if I'm making an error of logic, I'd be happy to find out about it and correct it.
23:04:01
##forth
<tpbsd>
Im not arguing with you any longer, you're a hopeless debater, try someone else
23:04:29
##forth
<xentrac>
it sounds like you're the one who's engaging non-seriously with the issue, not me.
23:05:23
##forth
<tpbsd>
and I thought this channel was free of the usual Usenet Forth channel crazies
23:06:16
##forth
<xentrac>
I've outlined the historical development of the field, including the dates of specific key developments, and the economic and social reasons for the changes in terminology. Everything I've said is objectively verifiable.
23:06:57
##forth
<xentrac>
And I'm open to being corrected if I've misunderstood something or drawn an invalid conclusion.
23:07:05
##forth
<tpbsd>
your arguments are specious, anyone can look up wikepedia and they will find "An analog computer or analogue computer is a type of computation machine (computer) that uses physical phenomena such as electrical, mechanical, or hydraulic quantities behaving according to the mathematical principles in question … Wikipedia"
23:07:13
##forth
<lispmacs[work]>
just got back from lunch
23:07:44
##forth
<lispmacs[work]>
there is a bit of a confusion unfortunately, because the term "analog" can be used in two different ways, both of which can be applicable here
23:07:57
##forth
<lispmacs[work]>
"analog" can mean "not digital" of course
23:08:07
##forth
<xentrac>
Yes, I agree that people do use the term in that way. I'm arguing that it's a misleading term, much like "adult industry", that should be avoided for the reasons I described above.
23:08:38
##forth
<lispmacs[work]>
but another meaning is "analog" in the sense of a computer that works by being itself an analog (parallel) to the computation being performed
23:08:59
##forth
<xentrac>
lispmacs[work]: I don't think those are actually two different meanings; I think those are the same meaning.
23:09:11
##forth
<tpbsd>
there is no real difference between a analog and digital computer in that their essense is they compute. Sure digital is the modern method in most cases (but not all) because of miniaturization advantages that analog doesn't have at present
23:09:24
##forth
<lispmacs[work]>
the main idea of analog computing is that you have components that operate in parallel, each component being some kind of analog to that part of the computation
23:09:39
##forth
<tpbsd>
lispmacs[work], I agree with you
23:09:51
##forth
<lispmacs[work]>
there was a few years in which digital analog computers were popular, known as digital differential analyzers
23:10:45
##forth
<xentrac>
I don't agree that DDAs are "analog computers"; all the discourse around them at the time described them in contrast to analog differential analyzers
23:10:54
##forth
<tpbsd>
xentrac, so you are now seeking to redefine the term 'computer' as a last resort, that's just sad imho
23:11:39
##forth
<xentrac>
Well, I do think it's true that the term "computer" has acquired a different, ontologically coherent meaning, but I'm not the one who did it; it happened 60 years ago
23:11:49
##forth
<lispmacs[work]>
DDAs are definitely analog computers, in the sense that I described, though not to be confused with differential analyzer simulators, in which a general purpose CPU steps through performing the different operations
23:13:07
##forth
<lispmacs[work]>
analog computers can be based on mechanical principles, hydraulics, analog (non-digital) electronics, or digital modules
23:13:31
##forth
<xentrac>
And I think that the historical use of the term "analog computers" to describe analog calculating devices confuses people a lot, because in fact there is an enormous difference between analog and digital computers in their essence. There's nothing similar to Turing-completeness in the analog domain, and it's fundamental to almost everything we do with computers nowadays.
23:13:45
##forth
<tpbsd>
anyone can build a OP-AMP multiplier with two analog gauges and two calibrated knobs and multiply numbers with it. Bingo, "analog computer".
23:14:31
##forth
<lispmacs[work]>
tpbsd: yes, that is an analog computer. Though obviously it is not general purpose
23:14:38
##forth
<tpbsd>
there are even areas where it's unsafe to use digital computers, and only analog ones will do. Rare, but they have existed for decades.
23:15:16
##forth
<xentrac>
The essence of digital computers is *not* that they *compute*, in the sense of doing numerical calculations. It's that they can do Turing-complete computations; any one of them can emulate any other.
23:15:39
##forth
<lispmacs[work]>
for several decades, electronic analog computers were preferred to digital computers for many simulation and modeling applications, because, due to their parallel nature, they could perform calculations and modeling much faster than the digital computers available
23:15:45
##forth
<lispmacs[work]>
with better interactivity
23:15:50
##forth
<xentrac>
Analog calculating devices, by contrast, are *never* general-purpose, being at best modestly reconfigurable.
23:16:13
##forth
<tpbsd>
lispmacs[work], true, but it's something a lot of modern MCU's lack (multiply), and theyre still known as 'computers'
23:16:31
##forth
<xentrac>
Yes, analog circuits are still preferred to digital computers for things where digital computers (or, as we call them today, "computers") are too slow for. Submillimeter RF demodulation, for example.
23:16:46
##forth
<lispmacs[work]>
around the 1970s, hybrid computers become popular, where a digital computer controlled an anlog computer. This allowed the digital computer to handle some mundane tasks rapidly like setting up initial conditions
23:16:58
##forth
<xentrac>
tpbsd: but any modern MCU can multiply by running a multiplication subroutine, and therein lies the enormous difference between analog calculating devices and digital computers.
23:18:05
##forth
<lispmacs[work]>
I do not believe that a computer has to be "general purpose" in the sense of being able to perform any program possible, in order to be useful. For several decades, electronic analog computer were used by power companies, aircraft companies, and such like, for simulation, to design things like control grids and wing shapes
23:18:58
##forth
<tpbsd>
xentrac, that kind of clain will send most programmers into fits because if a mcu doesnt have a hardware multiplier, the operation usually has a high clock cycle requirement abd is very wasteful
23:19:23
##forth
<xentrac>
tpbsd: there's an enormous difference between "wasteful" and "impossible"
23:19:49
##forth
<tpbsd>
lispmacs[work], agreed, not to mention Soviet hand cranked mechanical ballistic 'computers'
23:20:17
##forth
<lispmacs[work]>
the Norden Bomb sight is a very good example of an analog computer doing critical computations in real time
23:21:13
##forth
<tpbsd>
xentrac, that depends solely on the time available for the computing task, in some cases "wasteful" and "impossible have the same meaning
23:21:18
##forth
<lispmacs[work]>
for those not in a big hurry, I would recommend this stimulating old training video:
23:21:27
##forth
<xentrac>
lispmacs[work]: I agree that analog calculating devices are useful. Electronic analog circuits are used in every switching power supply, on almost every motor, in every speaker driver, in every 2.4GHz radio, etc., and are very useful. They just aren't called "analog computers" any more, even though they're more complex versions of the same circuit designs people called "analog computers" in the 01950s
23:21:33
##forth
<xentrac>
and 01960s
23:21:50
##forth
<lispmacs[work]>
from EIA, a designer of hybrid computers
23:22:33
##forth
<lispmacs[work]>
well, many (most?) people today think that a smartphone is not computer, but analog computers are computers
23:22:43
##forth
<xentrac>
Well, and especially 01940s, which was really the heyday of analog computers.
23:22:47
##forth
<tpbsd>
I have never heard a switch mode PSU called "analog computers", I call bullshit on that claim
23:22:47
##forth
<lispmacs[work]>
regardless of popular ignorance
23:23:28
##forth
<tpbsd>
and I specialized in PSU design being a real electronics technician all my life
23:23:29
##forth
<xentrac>
tpbsd: You seem to have misunderstood the claim: I claimed that those are *not* called "analog computers" any more. Is that the claim you're "calling bullshit on"?
23:24:35
##forth
<lispmacs[work]>
I don't know of anyone advocating for going back to the analog computing paradigm of the 1970s. But it is an interesting and enlightening subject to study. There is some modern research as well into the idea of electronic analog accelerator cards and FPAAs, since electronic analog computer is much more electrically efficient that digital computing, where it can be used
23:24:42
##forth
<tpbsd>
xentrac, they never were called that name. Switchmode was only possible with the invention of the transistor, so the technology is not that old
23:25:23
##forth
<lispmacs[work]>
I can reference also several modern books (post 2000) discussing analog computers, and some modern products called analog computers
23:25:26
##forth
<xentrac>
tpbsd: actually relay-based switchmode power supplies go back to Tesla
23:25:58
##forth
<xentrac>
lispmacs[work]: People who think that a smartphone is not a computer just don't know what a computer is. I'm falling back on ontological coherency again here: if a Burroughs mainframe is a "computer", and a 40-year-old car's ECU is a "computer", and an IMSAI is a "computer", and my laptop is a "computer", on what basis can you exclude my smartphone?
23:26:20
##forth
<lispmacs[work]>
a smartphone is a computer
23:26:27
##forth
<xentrac>
the FPAA is a brand name for a product that failed 20 years ago
23:27:00
##forth
<tpbsd>
lispmacs[work], analog computers are a beautiful thing in my opinion. In this day and age of 'retro' where lots of people build 6502 and other old 8 bit 'computers', few build 'analog computers' which is a shame imho
23:27:32
##forth
<xentrac>
tpbsd: the circuits that are used in modern SMPSs for feedback control would easily be recognized as "analog computers" by the folks that made the EIA video lispmacs[work] linked above
23:28:49
##forth
<tpbsd>
xentrac, so what, the fact remains that a switchmode psu is *not* any kind of computer, it's a psu
23:29:03
##forth
<lispmacs[work]>
I'm not saying you have to believe in the future of analog computers or invest in product stock, just saying that an analog computer is very much a real thing, and there are many examples of them in fairly recent history. Everything from a small, very specialized module like the Norden Bombsite, to analog computers of the 1970s that could simulate many aspects of aircraft physics
23:29:10
##forth
<xentrac>
tpbsd: I agree, and I would broaden that claim to "analog computers" in general
23:29:51
##forth
<lispmacs[work]>
one could, of course, make a distinction between reprogrammable and non-reprogrammable analog computers
23:30:07
##forth
<xentrac>
the Norden Bombsight, EIA's "hybrid computers", etc., are not "computers" in the way that a cellphone, a 4004, or a mainframe is a computer
23:30:20
##forth
<lispmacs[work]>
the 1970s analog computers were definitely reprogrammable, though, as I said, not able to run every program imaginable
23:30:28
##forth
<xentrac>
you can add some CD4051s to an analog "computer" to make it "reprogrammable"
23:30:36
##forth
<tpbsd>
the Norden bombsight was definitely a computer
23:30:49
##forth
<xentrac>
they weren't able to run any programs; they were only able to do analog calculations
23:31:08
##forth
<lispmacs[work]>
The 1970s analog computers were reprogrammable by having a patch patch where you could change the interconnections of the modules to change the computation
23:31:18
##forth
<lispmacs[work]>
I meant to type "patch panel"
23:31:24
##forth
<xentrac>
right, a patch panel
23:31:33
##forth
<xentrac>
CD4051s are probably a better option
23:31:46
##forth
<xentrac>
gate current injection isn't a concern in this context
23:32:47
##forth
<tpbsd>
anyway, I have better things to do than waste time arguing the meaning of computer with xentrac, so see you guys later! My plang2 neovim plugin is calling me.
23:32:55
##forth
<xentrac>
have fun!
23:33:11
##forth
<xentrac>
the US navy also called their analog mechanical gear-based naval gun aiming systems "fire control computers" until at least the 01970s
23:33:50
##forth
<lispmacs[work]>
yes, those were definitely computers, though I think not reprogrammable
23:34:11
##forth
<xentrac>
that's an extremely eargrayish definition of "computer"
23:34:20
##forth
<xentrac>
do you know about eargrayishness?
23:34:50
##forth
<cleobuline>
on dit ordinateur chez nous
23:35:15
##forth
<lispmacs[work]>
when I say computer, I mean a machine used for performing computations
23:35:32
##forth
<xentrac>
that's true. an analog "computer" is not an ordinateur
23:35:43
##forth
<xentrac>
by "computations" do you mean numerical calculations?
23:36:34
##forth
<lispmacs[work]>
xentrac: yes, though not saying that the computation has to be discrete, nor to have exact accuracy
23:37:22
##forth
<xentrac>
then your definition excludes the laptop I am typing this on, because it is not used for performing numerical calculations
23:37:23
##forth
<lispmacs[work]>
an electronic analog computer, e.g., might have only an accuracy of 1% due to component tolerances
23:37:39
##forth
<xentrac>
1% was typical for mechanical differential analyzers, yes
23:38:00
##forth
<xentrac>
electronic analog calculating machines based on op-amps can often reach 0.1%
23:38:07
##forth
<lispmacs[work]>
xentrac: you might not be using it for mathematical operations, but it is constantly performing numerical calculations
23:38:14
##forth
<xentrac>
§2.3.1
23:38:55
##forth
<lispmacs[work]>
e.g., calculating memory jumps, or where to display things on the screen
23:38:59
##forth
<skvery>
PPM multiplication of analog signals can be 10 ppm.
23:39:18
##forth
<xentrac>
skvery: what's PPM multiplication?
23:39:20
##forth
<skvery>
*PWM
23:39:23
##forth
<xentrac>
oh, PWM
23:39:34
##forth
<xentrac>
right, because it's time-domain rather than voltage-domain
23:40:47
##forth
<lispmacs[work]>
digital analog computers (DDAs) do away with the problem of inaccurancy from component tolerances, but they have to deal with the issue of accumulator register accurance (i.e., limited bits). This was more of a concern back in the 1970s
23:40:54
##forth
<xentrac>
lispmacs[work]: that seems like a question about how it's implemented, not what it's used for. is a car "used for performing numerical calculations" because the gears in its manual transmission spin in integer ratios to one another?
23:41:24
##forth
<xentrac>
DDAs had already mostly fallen out of use by the 01970s; my grandfather worked on them in the 01950s
23:42:02
##forth
<lispmacs[work]>
xentrac: I'm not highly knowledgeable on transmissions, but it sounds like an example of a non-reprogrammable analog computer
23:42:33
##forth
<lispmacs[work]>
maybe similar, say, to analog computers that used to navigate missle flight
23:43:01
##forth
<xentrac>
that extends your definition of "computer" to not just Model T Fords but also grandfather clocks
23:43:27
##forth
<xentrac>
yeah, he was working on the missiles
23:43:38
##forth
<xentrac>
(and spy planes)
23:44:35
##forth
<xentrac>
you really only need about 12 bits to match the voltage precision of most analog circuits, though, as skvery points out, their time precision can be very much finer
23:45:26
##forth
<xentrac>
it seems like your definition of "computer" might also include things like a steelyard or bismar
23:45:50
##forth
<lispmacs[work]>
a grandfather clock is, I'd say, a very limited example of a digital computer, with simulation time equal to real time. You could, for example, speed up the grandfather clock, to calculate the number of hours given a certain number of seconds
23:46:18
##forth
<lispmacs[work]>
by replacing the regulator
23:47:08
##forth
<xentrac>
The reason I think it's worth spending some attention on getting the terminology right is this:
23:47:11
##forth
<xentrac>
> Suppose I wanted to argue that mice were larger than grizzly bears. I note that both mice and elephants are “eargreyish”, meaning grey animals with large ears. We note that eargreyish animals such as elephants are known to be extremely large. Therefore, eargreyish animals are larger than noneargreyish animals and mice are larger than grizzly bears.
23:47:18
##forth
<xentrac>
> As long as we can group two unlike things together using a made-up word that traps non-essential characteristics of each, we can prove any old thing.
23:47:42
##forth
<xentrac>
(quoted from above link)
23:48:55
##forth
<lispmacs[work]>
well, labels can be fuzzy and ambiguous, for sure, and you will get yourself quickly into difficults trying to come up with a perfectly precise system of categories that has absolutely not contradictions or grey areas. But for sure, analog computers are justly called computers, both for practical and historical reasons
23:49:12
##forth
<xentrac>
It seems like, in order to come up with a definition of "computer" that encompasses both Vannevar Bush's mechanical differential analyzers and 8051 microcontrollers, you either have to broaden it out until it also encompasses Model T Fords and grandfather clocks, or you have to make it a very eargreyish definition
23:49:27
##forth
<lispmacs[work]>
I don't mind broadening it out
23:50:01
##forth
<lispmacs[work]>
we can of course further categorize from there, e.g., reprogrammable vs. non-reprogrammable computers.
23:50:22
##forth
<xentrac>
The problem with such a broad definition is that there isn't very much you can say truthfully about such a broad category
23:50:41
##forth
<lispmacs[work]>
simulation modeling computers vs. algorithmic computers
23:51:27
##forth
<lispmacs[work]>
well, somehow, people talked about "analog computers" for about 40 years, and they managed to get by writing a lot of stuff that blows the brains of most people
23:52:06
##forth
<xentrac>
it blows their brains in many cases because they mistakenly think that what those people are talking about is an analog version of what we currently call "computers"
23:52:21
##forth
<xentrac>
but what we use those for isn't numerical calculation, mostly
23:52:25
##forth
<lispmacs[work]>
well, people need to expand their horizons of knowledge
23:52:48
##forth
<xentrac>
I'd put the "analog computers" era as 01890 to 01960 more or less, though there were a few stragglers
23:52:59
##forth
<cleobuline>
hello world in brainfuck : ++++++++++[>+++++++>++++++++++>+++>+<<<<-]>++.>+.+++++++..+++.>++.<<+++++++++++++++.>.+++.——.——–.>+.>.
23:53:20
##forth
<xentrac>
for the first 50 of those years people didn't use the term "analog computers" because there weren't digital computers to contrast them with
23:53:30
##forth
<xentrac>
like "landline phone"
23:54:16
##forth
<xentrac>
Tatjana van Vark has constructed a lovely harmonic analyzer you'd probably enjoy looking at
23:59:48
##forth
<xentrac>
she explains how the harmonic analyzer works