|
Bloody posted:define embedded systems and meaningful information. 100ms is anywhere between a lifetime or an instant information is generated by a scanning lidar system, i have to pick information out of the return pulses and use that to calculate information about the sensor's location and motion and cross-reference that with information from other sources embedded systems meaning anything down to and including single-threaded microcontrollers
|
#
?
Jun 2, 2016 20:15
|
|
|
|
| # ? May 23, 2024 12:57 |
|
|
Star War Sex Parrot posted:why hasn't someone designed a strict subset of C or new language for embedded applications that takes inspiration from stuff like JPL's coding standards and MISRA-C ada?
|
|
#
?
Jun 2, 2016 20:15
|
|
|
Bloody posted:ada?
|
|
#
?
Jun 2, 2016 20:16
|
|
|
Star War Sex Parrot posted:why hasn't someone designed a strict subset of C or new language for embedded applications that takes inspiration from stuff like JPL's coding standards and MISRA-C i mean part of the problem is that people generally want more abstraction and features, not less. why would i want a language that enforces, say quote:Rule 16 (use of assertions) ...when there are often perfectly legitimate reasons not to?
|
|
#
?
Jun 2, 2016 20:17
|
|
|
code:
|
|
#
?
Jun 2, 2016 20:17
|
|
|
quiggy posted:i mean part of the problem is that people generally want more abstraction and features, not less. why would i want a language that enforces, say cuz sometimes you want to be really sure you really have zero bugs
|
|
#
?
Jun 2, 2016 20:18
|
|
|
i like ada
|
|
#
?
Jun 2, 2016 20:20
|
|
|
quiggy posted:i mean part of the problem is that people generally want more abstraction and features, not less. why would i want a language that enforces, say 
|
|
#
?
Jun 2, 2016 20:21
|
|
|
Star War Sex Parrot posted:because we're flying space probes 3 billion miles from Earth i mean i get why those standards exist, and i picked that one for a reason. sanity checks are a good thing, and certainly when you're building a spacecraft to operate 3 billion miles from earth you want them everywhere. but imagine trying to write in a language that actually enforced those checks. you'd feel like you're writing INTERCAL where you have to pepper PLEASE statements every so often or else the compiler will yell at you for not obeying its rules, even though sometimes you might have a legitimately good reason to not have an assertion in an 11-line function
|
|
#
?
Jun 2, 2016 20:22
|
|
|
Star War Sex Parrot posted:
|
|
#
?
Jun 2, 2016 20:23
|
|
|
quiggy posted:loaded up a rust tutorial because bossman is out today and im bored because their types can't really be expressed properly, and so that the compiler can check the format string matches the given args
|
|
#
?
Jun 2, 2016 20:33
|
|
|
quiggy posted:any language where you have to explicitly declare self/this or w/e is loving awful, I'm a huge prick This seems like a very strange reason to decide that a language is "loving awful". In any case, from what I've heard Rust is not quite there yet for embedded stuff, so you are probably gonna be using C (or maybe ada or something idk). Which is fine, that's the sort of environment where C is kind of nice.
|
|
#
?
Jun 2, 2016 20:40
|
|
|
assembly is rad
|
|
#
?
Jun 2, 2016 20:54
|
|
|
HoboMan posted:assembly is rad literally lmao if you don't code everything in raw machine language
|
|
#
?
Jun 2, 2016 20:56
|
|
|
i use verilog so i can create my own machine language
|
|
#
?
Jun 2, 2016 20:57
|
|
|
MALE SHOEGAZE posted:most plangers don't really use runtime dynamism, they use dynamism to make writing code easier. you can use Marcos to allow dynamic language features without being dynamic at runtime Does p-lang mean python/php/perl/poop? Or does it mean procedural language?
|
|
#
?
Jun 2, 2016 21:53
|
|
|
HoboMan posted:assembly is rad I learned myself a little bit of assmebly this past week messing around with that MSP430 CTF thingy. I have to agree. It is pretty rad and debugging was rad and following the registers and whatnot was totally tubular.
|
|
#
?
Jun 2, 2016 21:55
|
|
|
gonadic io posted:because their types can't really be expressed properly, and so that the compiler can check the format string matches the given args when are they going to make rust dependently typed
|
|
#
?
Jun 2, 2016 21:56
|
|
|
Finster Dexter posted:Does p-lang mean python/php/perl/poop? If poop==JavaScript, then it's this one.
|
|
#
?
Jun 2, 2016 22:04
|
|
|
p-langs are all the bad langs that start with p: python, php, perl, pruby, pjavascript
|
|
#
?
Jun 2, 2016 22:07
|
|
|
gonna start calling it p-yava-script thanks
|
|
#
?
Jun 2, 2016 22:22
|
|
|
Share Bear posted:gonna start calling it p-yava-script thanks 
|
|
#
?
Jun 2, 2016 22:42
|
|
|
pajamascript
|
|
#
?
Jun 2, 2016 22:47
|
|
|
: kramers into thread shouting erratically : apparently some goons were talking about swift error handling in here, if anyone actually gives a gently caress they can ask about it, gently caress reading previous pages tho
|
|
#
?
Jun 2, 2016 22:56
|
|
|
rjmccall posted:: kramers into thread shouting erratically : there was some poo poo about exceptions and eithers both existing
|
|
#
?
Jun 2, 2016 23:05
|
|
|
ctps: i need a new thing to do in haskell or something this parser project is degenerating into solving the problem vs learning the lang
|
|
#
?
Jun 2, 2016 23:06
|
|
|
HoboMan posted:assembly is rad pyf asm I like arm, flags only being set when you ask for it and branchless conditional execution are both neat. also even gas uses the standard arm syntax instead of dumb at&t bs.
|
|
#
?
Jun 2, 2016 23:07
|
|
|
best asm is tis-100 actual answer mips i guess? its the one ive used the most
|
|
#
?
Jun 2, 2016 23:08
|
|
|
quiggy posted:best asm is tis-100 lol sorry for your loss
|
|
#
?
Jun 2, 2016 23:12
|
|
|
Bloody posted:lol sorry for your loss ive never done much asm, the mips is only from a class i took on it in undergrad so w/e
|
|
#
?
Jun 2, 2016 23:13
|
|
|
msp430 asm is trash atmel asm is slightly less trash arm asm is pdeece idk anything about any cisc asm
|
|
#
?
Jun 2, 2016 23:13
|
|
|
quiggy posted:best asm is tis-100 the only mips I've ever used had hilarious design flaws like explicit delay slots. why yes I'd love to write code where the instruction after every branch op is executed even if the branch was taken, that is certainly a thing programmers should have to deal with
|
|
#
?
Jun 2, 2016 23:25
|
|
|
nobody actually writes mips in the real world. its a prank academia plays on undergrads
|
|
#
?
Jun 2, 2016 23:30
|
|
|
GameCube posted:nobody actually writes mips in the real world. its a prank academia plays on undergrads 
|
|
#
?
Jun 2, 2016 23:46
|
|
|
overheard this the other day:quote:i have had a screen session open for a year and im hosed without it so moments later the server in question was hard-powered down. hes hosed
|
|
#
?
Jun 2, 2016 23:55
|
|
|
hackbunny posted:no, I get it, every line of code in C++ is, basically, a monad. it takes all variables, functions, etc. visible in the scope as inputs, and copies them to outputs, possibly modifying them, sometimes introducing new variables, sometimes erasing them if it's the end of a scope. I remember this from my computer science fundamentals course, every line of code being a function over all the variables that returns all the variables, with modifications, was one of the things I remember the best, but could never find a practical use for where would a terrible programmer go to learn more things like this that isn't a university? which 25 year old text book do i read?
|
|
#
?
Jun 3, 2016 00:07
|
|
|
Soricidus posted:f# without the manual declaration/file order micromanagement bullshit would basically be the perfect pl at first I thought this was a silly restriction but then I realized it never bothered me in Lisp
|
|
#
?
Jun 3, 2016 01:54
|
|
|
Soricidus posted:pyf asm best asm is Motorola 680x0, by far it's super easy to read and write and modify and all wonderfully orthogonal and also very complete in terms of both functionality and addressing modes extra bonus: install a debugger on an old Mac running the original OS (pre System 7) and you can see the source code, modulo symbolic constants, since virtually the entire OS was actually written in asm—and later versions of MacsBug included tables of constants they applied to listings too lots of classic Mac developers eschewed source level debugging prior to PowerPC because they could just as easily debug the generated code using MacsBug or TMON
|
|
#
?
Jun 3, 2016 02:05
|
|
|
ASM to mouth would be a good twit name or namechange
|
|
#
?
Jun 3, 2016 02:33
|
|
|
|
| # ? May 23, 2024 12:57 |
|
|
quiggy posted:any language where you have to explicitly declare self/this or w/e is loving awful, I'm a huge prick Being explicit about scoping prevents a lot of dumb errors, though.
|
|
#
?
Jun 3, 2016 02:43
|
|
