Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
jackdh4x0r
Jun 8, 2005

I've got an eighteen minute window.

Sweevo posted:

i dumped the ROM from my cars ECU, disassembled the code, and have been slowly working through trying to figure out what it does

what's weird is that even though it's pure 8052 assembly you can see where different people worked on it because the code is so different

Dude, I am/was working with some guys on the BOSCH LH-jetronic stuff (pre-motronic), it's 8052 as well.

Another guy did the hard work of the disassembly and stuff, once I had it I made a bunch of tunerpro defs and actually wrote a rudimentary datalogger that would poo poo out a signal that could be read with an OBD-II K-Line reader

https://www.youtube.com/watch?v=VvwqEg-GXLs

It never worked well, I eventually went Megasquirt after blowing a few headgaskets doing this on poorly hacked factory management

https://www.youtube.com/watch?v=TlhIkL4xOR4

Adbot
ADBOT LOVES YOU

Sweevo
Nov 8, 2007

i sometimes throw cables away

i mean straight into the bin without spending 10+ years in the box of might-come-in-handy-someday first

im a fucking monster

got any links?

trying to follow what the hell is going on with the rpm calculations is driving me crazy.

Stay Safe
Sep 1, 2008

by FactsAreUseless
I dance (usually locking) and post on comedy forums. I also spend the majority of my time learning norwegian, swedish, korean and other languages

i spend at least an hour or two working out and reading (not at the same time) and eventually want to take singing or violin lessons, and maybe learn how to do yoga properly

if I don't have time for dinner or breakfast I just eat raw fruits and veggies

i dont really have any "projects" except for academic projects but i do enjoy my hobbies

Stay Safe fucked around with this message at 18:07 on Jul 23, 2013

jackdh4x0r
Jun 8, 2005

I've got an eighteen minute window.

Sweevo posted:

got any links?

trying to follow what the hell is going on with the rpm calculations is driving me crazy.

here is the wiki site we setup. I had to cancel my vhost so it is hosted by another guy
now, and somehow some of the downloads got corrupt. Unless they have fixed them by now.

PM me if you want something specific from there; i have most of it in my dropbox somewhere

Also, are you talking about RPM calculations on the data tables? On LH2.4 at least, Bosch did it a really funny way, probably to save bits or something. Can't remember exactly how it works but feel free to hit me up.

SRQ
Nov 9, 2009

cheese-cube posted:

post some stories im bored

I'll post the next one I write (I end up losing most of them) if you promise to understand early on that they're all mostly terrible first draft works and not to make fun of me :ohdear:.

Cold on a Cob
Feb 6, 2006

i've seen so much, i'm going blind
and i'm brain dead virtually

College Slice
i have absolutely nothing to contribute to this thread but i'm glad it was posted and maybe it'll pressure me to get off my duff and have something to contribute later :unsmith:

CamH
Apr 11, 2008

jackdh4x0r posted:

Dude, I am/was working with some guys on the BOSCH LH-jetronic stuff (pre-motronic), it's 8052 as well.

Another guy did the hard work of the disassembly and stuff, once I had it I made a bunch of tunerpro defs and actually wrote a rudimentary datalogger that would poo poo out a signal that could be read with an OBD-II K-Line reader

https://www.youtube.com/watch?v=VvwqEg-GXLs

It never worked well, I eventually went Megasquirt after blowing a few headgaskets doing this on poorly hacked factory management

https://www.youtube.com/watch?v=TlhIkL4xOR4

sup dude? I don't have anything useful to contribute but I run megasquirt 2 extra on a turbo 240Z so high five to a fellow megasquirter or w/e

Sweevo
Nov 8, 2007

i sometimes throw cables away

i mean straight into the bin without spending 10+ years in the box of might-come-in-handy-someday first

im a fucking monster

jackdh4x0r posted:

here is the wiki site we setup. I had to cancel my vhost so it is hosted by another guy
now, and somehow some of the downloads got corrupt. Unless they have fixed them by now.

PM me if you want something specific from there; i have most of it in my dropbox somewhere

Also, are you talking about RPM calculations on the data tables? On LH2.4 at least, Bosch did it a really funny way, probably to save bits or something. Can't remember exactly how it works but feel free to hit me up.

thanks, that gives me some reading material for this evening.

i meant how the code actually works out RPM from the flywheel pickup. timing is accurate to 0.75 degrees, but i can't work out how as the timing teeth are 6 degrees apart. i guess it's different on the jetronic because it only does fuel, whereas the motronic does ingition too. the timing code is all interrupt-driven and has multiple tooth counters going on, and can compensate for mis-counted timing teeth. its really hard to follow

i figured out the table-reading code a while back. it's not how I would have written it, but it makes sense once you get your head round it. it reads like they were trying to save ROM space. the early motronics only had 8K so that makes sense, but mine has 32K and is 40% empty so they could have done it much more simply. i guess they just stuck with code they knew worked?

hubris.height
Jan 6, 2005

Pork Pro
these are all really cool projects and just kind of make me wish for a little more technical know-how.

for example my most advanced creation so far is a command prompt based RSS-reader for me to use at work so the wall of text makes it look like i'm doing something productive. it was fun and cool but it could have been so much better executed and i feel overwhelmed by the idea of putting more time into it.

MORE CURLY FRIES
Apr 8, 2004

hubris.height posted:

these are all really cool projects and just kind of make me wish for a little more technical know-how.

u have access to the internet so just decide a small thing you wanna learn and find some things that are easy to do while learning it and do them

Moo Cowabunga
Jun 15, 2009

[Office Worker.




youtube can be good for this

echinopsis
Apr 13, 2004

by Fluffdaddy
see if you can learn to ejaculate without touch your penis (either directly or indirectly)


im sure the internet can teach

tef
May 30, 2004

-> some l-system crap ->

hubris.height posted:

these are all really cool projects and just kind of make me wish for a little more technical know-how.

:siren: HEY GUESS WHAT, MOST PEOPLE THINK THIS :siren:

when you learn more you can see further and realise how alone you are

quote:

it was fun and cool but it could have been so much better executed

YOU ARE IN A WORLD OF POSSIBILITIES, BUT YOU ARE FEELING SADFACE FOR
SOME UNSUBSTANTIATED EXISTENTIALISM AND A CRUSHING AMOUNT OF SELF WORTH

> check inventory

YOU HAVE A COLLECTION OF FUTILE APHORISMS

- PERFECT IS THE ENEMY OF THE GOOD
- HINDSIGHT IS 20/20
- SOMETIMES THE FIRST STEP TO BEING GOOD AT SOMETHING IS BEING LESS BAD

> feel ok

YOU STOP MAKING A SAD FACE AND IMAGINE YOU ARE PETTING A CATTE

quote:

and i feel overwhelmed by the idea of putting more time into it.

it's ok, just abandon it and do something else. if you work on one thing for all of your life, you may end up like ted nelson.

GameCube
Nov 21, 2006

is that losethos's real name

e: oh wow i looked that guy up and i'm pretty sure i have more respect for losethos

GameCube fucked around with this message at 21:26 on Jul 23, 2013

MORE CURLY FRIES
Apr 8, 2004

tef posted:

YOU ARE IN A WORLD OF POSSIBILITIES, BUT YOU ARE FEELING SADFACE FOR
SOME UNSUBSTANTIATED EXISTENTIALISM AND A CRUSHING AMOUNT OF SELF WORTH

> check inventory

YOU HAVE A COLLECTION OF FUTILE APHORISMS

- PERFECT IS THE ENEMY OF THE GOOD
- HINDSIGHT IS 20/20
- SOMETIMES THE FIRST STEP TO BEING GOOD AT SOMETHING IS BEING LESS BAD

> feel ok

YOU STOP MAKING A SAD FACE AND IMAGINE YOU ARE PETTING A CATTE

MORE CURLY FRIES
Apr 8, 2004

can that be the yosbbs welcome message

SRQ
Nov 9, 2009

I just got a neat idea I'll pen as a short story sometime this weak, look forward to SRQ Presents: Amateur Science Fiction 2000.
Also I'm gonna collect all the world building notes I made when I was bored one day so you can laff or whatever.

tef
May 30, 2004

-> some l-system crap ->
HELLO I HAVE SPENT MORE THAN A DECADE PROGRAMMING AND I STILL FELL DEAD INSIDE, BUT THAT'S OK

i often hear people saying the following things, and they're sort of anti-learning thoughts

- i want to know the right way to do it, i don't want to waste my time
- i often abandon my projects after playing with them and they stop being fun
- if only i could do cool stuff like other people

the reality is you learn how to program by making mistakes and then learning from them. so you need to go and make some goddam mistakes and terrible code if you wanna be a better programmer. don't worry about how many mistakes you make, just try and avoid making them over and over. it can take a while to learn from the experience. along the way you will have to learn new libraries, new platforms and languages, but that's ok. you'll be poo poo at them too for a while.

:siren: BITE SIZE MORAL INCOMING :siren: >> if you didn't make any mistakes, you probably haven't learned anything from the experience. <<

we have yet to find a shortcut to being better at something without just goddam practicing it all the time. so try these three weird old tippes.

- write more code
- read more code
- edit more code

the best way to do this is as a side effect, rather than a goal. you will end up doing these anyway. so if you are bored of your recent toy, make a new one. if you don't know what you want to code, pick one at random and spend an hour trying things. look at something else someone else has done and make a simpler, cut down version that's possible to implement in an afternoon.

the easiest way to start doing this is aim low, and then aim lower still.

i wrote a game once. my original objective was "draw a loving box on the screen", then i figured i'd change the code to handle keypresses, changing colours, moving, animating, and finished the game by taking a bunch of small steps first. if you get something working as quickly as possible, then you'll find yourself just sinking time into the project.

OH poo poo HERE COMES ANOTHER MORAL you learn by copying others, and you'll learn quicker if you learn from other peoples mistakes.

write some crappy project in an hour, maybe play on it for an afternoon, perhaps even the weekend, upload it to github, and post it here and ask peopel to tell you some unhelpful bullshit about how they'd write it. sometimes it can be useful. go and look at someone's github and poke around the code and ask yourself "OH JESUS CHRIST WHAT IS THIS I DON'T EVEN"

check it out anyway and see if you can change something.



tl;dr you won't make anything cool whining about how you're not doing it. you won't make anything cool until you do a bunch of uncool stuff first. go out and write terrible code, it's ok.

Bloody
Mar 3, 2013

existentialism has been tearing me apart for months now send help

GameCube
Nov 21, 2006

never heard of this ted nelson chap

quote:

Nelson founded Project Xanadu in 1960 with the goal of creating a computer network with a simple user interface. The effort is documented in his 1974 book Computer Lib / Dream Machines and the 1981 Literary Machines. Much of his adult life has been devoted to working on Xanadu and advocating it.

The Xanadu project itself failed to flourish, for a variety of reasons which are disputed. Journalist Gary Wolf published an unflattering history on Nelson and his project in the June 1995 issue of Wired calling it "the longest-running vaporware project in the history of computing".[1] Nelson expressed his disgust on his website,[2] referring to Wolf as "Gory Jackal", and threatened to sue him. He also outlined his objections in a letter to Wired,[3] and released a detailed rebuttal of the article.[4]

it's been almost two more decades, and this is xanadu as of last week

https://www.youtube.com/watch?v=1yLNGUeHapA&t=196s

hubris.height
Jan 6, 2005

Pork Pro

tef posted:


it's ok, just abandon it and do something else. if you work on one thing for all of your life, you may end up like ted nelson.

this is legit the best advice i've gotten. i haven't stopped programming or let my frustration interrupt with trying to make new little programs, but i've been trying to learn things like list and matrix comprehensions and poo poo from that dumb coursera site, and it makes my brain feel like its going to explode. just pain in the rear end to follow through technical jargon i dunno i just felt like complaining because i have almost nothing in life to complain about


i will continue to work at it because of your curt but accurate words tef thanks

Bloody posted:

existentialism has been tearing me apart for months now send help

this is a good thing

SRQ
Nov 9, 2009

Bloody posted:

existentialism has been tearing me apart for months now send help

Nothing you ever do will matter, furthermore your happiness/sadness is of no relevance to the universe.
So be happy and enjoy your 80 odd years of life however you want to, be happy you get to have a meaningless 80 years instead of nothing.

SRQ
Nov 9, 2009

why can't i access the shares on the win2k server i thought this would be easy with the total lack of security
ugh wtf

hubris.height
Jan 6, 2005

Pork Pro

Werthog 95 posted:

never heard of this ted nelson chap


it's been almost two more decades, and this is xanadu as of last week

https://www.youtube.com/watch?v=1yLNGUeHapA&t=196s

sworph lol

GameCube
Nov 21, 2006

tef get back here. explain this poo poo

why did the french knight this guy

Deacon of Delicious
Aug 20, 2007

I bet the twist ending is Dracula's dick-babies

tef posted:

if you don't know what you want to code, pick one at random and spend an hour trying things

this is the part i can never figure out, the blank screen is my biggest enemy

i am already good at the making mistakes part, getting in over my head part, and making mistakes part

hubris.height
Jan 6, 2005

Pork Pro

Werthog 95 posted:

tef get back here. explain this poo poo

why did the french knight this guy

he sworphed in and explained that they were trying to juxtapose convention into the new world and instead they should xanadize their interfaces and no one laughed

tef
May 30, 2004

-> some l-system crap ->
here is the thing i forgot to mention, i find it harder and harder to hack on side projects, all of my ideas are insane and too grand in scale to even begin with, and often boil down to "SCREW THIS LEGACY BULLSHIT I'M STARTING FROM SCRATCH". anyway despite this i have managed to have some side projects along the way but most of them i haven't touched in months.

- buttbot

so when #cobol first started everyone was writing bots so i dragged out some perl code I wrote when i was 16 and took flowenol's suggestion "write a bot that says butt", and then played around with it for a week before pushing it to google code. a bunch of people hacked on it, comics got drawn, basically it's the most significant piece of code i've written based on how many people it made giggle.

- a million toy language projects

i've written lots of parsers, some interpreters, some vms, and lots of bits and pieces to understand how platforms fit together.

- hyperglyph/rpc poo poo

i went mad at work and decided to explore representational state transfer as an ipc mechanism and it kinda worked, making p-lang rpc. i don't think it's incompatible with more contract based ipc though so i might revisit that at some point.

- logo interpreter

i wrote this during my day job as an excuse to learn javascript and also because i resented the day job. it was fun but logo is a terrible language underneath, and has that same sort of xfig like arcanery about it. it has a lot of weird idioms that seem a bit odd to modern eyes.



right now really i want to write my own ipc system with it's own data description language, transport and application protocol, and then of course a distributed file system, and then my own browser for it, with my own little declarative interface language. because i've gone mad.

jackdh4x0r
Jun 8, 2005

I've got an eighteen minute window.

Sweevo posted:


i meant how the code actually works out RPM from the flywheel pickup. timing is accurate to 0.75 degrees, but i can't work out how as the timing teeth are 6 degrees apart. i guess it's different on the jetronic because it only does fuel, whereas the motronic does ingition too. the timing code is all interrupt-driven and has multiple tooth counters going on, and can compensate for mis-counted timing teeth. its really hard to follow


With LH Jetroinc, the ignition is an entirely separate computer... and the code is pretty much exactly what you are describing. scary, and interrupt driven. Lots of NOP slides, not sure if they are left over from debugging but they looked to be timing specific.

Microcontroller programming for real time systems owns bones, respect++ for anyone who does it. When I was writing the ASM to output the variables for the datalogger above, I had to be very careful to make sure that poo poo didn't take too long... but at the same time I had to wait for the UART to be ready before sending another byte so I didn't clobber the previous one. And if poo poo took too long the car would run funny and sometimes the watchdog timer would reset the ECU. I never got it perfect because I blew the motor up.... three times. And Volvo redblocks are pretty stout motors; I definitely did some silly poo poo.

It was fun, I wish I had the motivation to dick with that stuff like I used to. Car is sitting in the grass with some kind of fuel pump problem that I haven't troubleshot yet. :effort:


CamH posted:

sup dude? I don't have anything useful to contribute but I run megasquirt 2 extra on a turbo 240Z so high five to a fellow megasquirter or w/e

:hfive: ms2/extra 4 lyfe, i have a big 'ol 88 town car that I am gonna megasquirt one day just cause.

tef
May 30, 2004

-> some l-system crap ->

Deacon of Delicious posted:

this is the part i can never figure out, the blank screen is my biggest enemy

the best way to design something is to add constraints until the solution is obvious. sometimes the real problem is actually asking the right question, rather than finding the right answer.

so let's start: i wanna make something

- that can be working in an hour
- in a language i know
- that does something on screen
- that maybe reacts to keyboard input

so you write a boring program maybe in javascript that draws a butt to the canvas, and when you push a button poop comes out.

now you have some code to edit, you can start to shape it.

- maybe instead of a butt, it should be a catte, and when you press a key it purrs.
- maybe i should be able to have multiple cats
- maybe i should be able to get them to chase my mouse pointer

then maybe you start something else.

tef
May 30, 2004

-> some l-system crap ->

Werthog 95 posted:

tef get back here. explain this poo poo

why did the french knight this guy

basically he was a pioneer and inspired and influenced a lot of people who tried to get a little bit towards what he wanted, like mr tim berners lee, and others.

tef
May 30, 2004

-> some l-system crap ->

hubris.height posted:

this is legit the best advice i've gotten. i haven't stopped programming or let my frustration interrupt with trying to make new little programs, but i've been trying to learn things like list and matrix comprehensions and poo poo from that dumb coursera site, and it makes my brain feel like its going to explode. just pain in the rear end to follow through technical jargon i dunno i just felt like complaining because i have almost nothing in life to complain about

maybe instead of learning language tricks and linear algebra you could pick up a library like pygame, or http://kivy.org/ or open frameworks, or supercollider, or erlang.

it can be useful to learn about nicer ways to write code later on, but most of them are just shorter ways of writing the same dumb code you would write anyway, so just write the dumb code and worry later.

jackdh4x0r
Jun 8, 2005

I've got an eighteen minute window.

tef posted:

right now really i want to write my own ipc system with it's own data description language, transport and application protocol, and then of course a distributed file system, and then my own browser for it, with my own little declarative interface language. because i've gone mad.

shine on you crazy diamond!

also your AV is great, reminds me of being a bored highscooler and using qbasic to draw odd fractal-ish things in my "keyboarding" class.

Deacon of Delicious
Aug 20, 2007

I bet the twist ending is Dracula's dick-babies

tef posted:

so you write a boring program maybe in javascript that draws a butt to the canvas, and when you push a button poop comes out.

i bet butt-based programming would get a lot more students interested in programming

i want to make a computer go "boop" and i should stop worrying that i don't know how to do that. use something like supercollider to make a boop and go from there

GameCube
Nov 21, 2006

tef posted:

basically he was a pioneer and inspired and influenced a lot of people who tried to get a little bit towards what he wanted, like mr tim berners lee, and others.

and then he pissed and moaned that they didn't do it RIGHT apparently

SRQ
Nov 9, 2009

tef posted:

here is the thing i forgot to mention, i find it harder and harder to hack on side projects, all of my ideas are insane and too grand in scale to even begin with, and often boil down to "SCREW THIS LEGACY BULLSHIT I'M STARTING FROM SCRATCH". anyway despite this i have managed to have some side projects along the way but most of them i haven't touched in months.

- buttbot

so when #cobol first started everyone was writing bots so i dragged out some perl code I wrote when i was 16 and took flowenol's suggestion "write a bot that says butt", and then played around with it for a week before pushing it to google code. a bunch of people hacked on it, comics got drawn, basically it's the most significant piece of code i've written based on how many people it made giggle.


holy poo poo you made buttbot??
you're a fuckin' hero

tef
May 30, 2004

-> some l-system crap ->

Deacon of Delicious posted:

i want to make a computer go "boop" and i should stop worrying that i don't know how to do that. use something like supercollider to make a boop and go from there

super collider, pure data, a whole bunch of weird rear end music programming languages. supercollider is a weird rear end smalltalk with a reactive programming system thrown atop. pure data is a graphical language to wire up beeps and boops. supercollider is a bit low level but you could take someone's code and play around with it. probably easiest to try making a simple drum machine.

tef
May 30, 2004

-> some l-system crap ->

jackdh4x0r posted:

shine on you crazy diamond!

also your AV is great, reminds me of being a bored highscooler and using qbasic to draw odd fractal-ish things in my "keyboarding" class.

right now i'm wondering about how to make a gigantic shared sandbox for code, basically multi user turtle graphics i guess.

i.e you can sit in your browser and write some logo like things, and then you can send the url to a friend and they can edit and play with it. then you can connect together over webrtc or indirectly via the server, and draw cocks on each other's screens. maybe let the turtles talk to each other and communicate i guess.

i guess then it would be neat to let them run the code on a physical turtle, drawing cocks on the floor, or maybe on a quadcopter running around. i guess the earlier step is really building a sandbox for the real world. then you could get them to build actual robots, rather than using prebuilt ones, learning about controls and electronics.

turtles all the way up

poemdexter
Feb 18, 2005

Hooray Indie Games!

College Slice
tef, your advice is gold and i completely agree 100%. i pretty much do all those things in game dev projects. i'm not a designer so much of my projects are:

1. play a new indie game with an interesting mechanic.
2. try to write a clone of that game.
3. get enough done to feel like i've implemented it correctly and is playable and then shelf it.

i'm working on sharpening my design chops since i feel i can code up almost anything given enough time, i just don't know how to make a game fun. when i was brand new to game dev with XNA, i pretty much did your steps.

1. get a thing on screen.
2. move a thing with keypresses.
3. make it collide with other things via checking intersecting boxes.
4. add <interesting thing i've not tried to implement before>

my buddy gets stuck in a rut because he never tries to code, he just wants to read about how to be a good coder. the absolute best way to learn is to code code code and then yell at yourself a month later when you realize it's all poo poo and then do it better.

game development is pretty much the only thing keeping me in software development and sane because enterprise java web development is nothing but solved problems and spring/hibernate stacks and boring BI poo poo.

Adbot
ADBOT LOVES YOU

SRQ
Nov 9, 2009

I can never focus much on software projects, but I love building dumb computers for dumb purposes (then tearing them down for dumber purposes)
My job in the Canadian Reserves will literally be that so hey cool.

"Radio tower is broken, we've got wire and wooden poles go make a new one"

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply