sounds broken
Posted by renesis at 09:41 | permalink | 0 comments
just about freezing for a couple months
doesnt snow a lot
k
Posted by renesis at 09:28 | permalink | 0 comments
yeah fuck college
professor posts bunch of lectures, and i feel guilty devoting time to watching them because not rest and not worth points
been hot here
people bitching a lot, im like guys it was 110F like a month ago, chill
inland northern california
along the 5/99, AKA that cali asshole corridor
s/that/the
kent: its mostly farmland
oh
it gets cold here
Posted by renesis at 09:23 | permalink | 0 comments
also in keil you can right click and see referenced constants and functions
wich helps a ton in figuring out how the diff source files are linked
thats how i figured out a lot of it is air, no code
like in reality, usually its as simple as you want it to be
damn
so its awesome when my boxers have become so worn that the fabric is like, gauze, very light open mesh
that period has been exceed, now they are just boxers with a hole
hey dont smack about my light airy plaid boxers
Posted by renesis at 09:18 | permalink | 0 comments
so like, you can totally redesign your hardware while running pretty much same code
or change stuff like uart settings or initial pwm states
its some lazy as fuck shit
i feel guilty using it
i dunno
and my experience is with keil
and i am careful about the options after regen
if you regen and open project, and keil is open, shit can get weird
like, what clobbers what, because keil will see the file changes and prompt
and then the other keil will do the same if you dont close it
if im being paranoid, i will close keil before doing anything, but you dont need to
like, its the kind of thing that happens when you are working fast, which is kind of exactly when you dont want shit to fuckup
so that and the constants thing, everything else is like, CUBEMX OMGFUCKYA!
shit just works
Posted by renesis at 09:12 | permalink | 0 comments
also i dont think the constants section in the generated code is for your constants
its for like, constants you define inside cubemx
so if you dont backport those into cubemx and regenerate code, itll eat your constants
thats the only thing about it thats pissed me off
yes! thats why it has the sections marked off for your code
Posted by renesis at 09:07 | permalink | 0 comments
right there is a laziness element
even having done this shit manually in c and asm, i just pop out cube code, check the code it generates, shrug and move on
if i was doing this 60 hours a week and getting paid, id prob want things a bit more concise
also im am trying harder to just go with the abstraction
also the 1ms timebase interrupt is in there somewhere
so you can throw code in that where it says to and have easy realtime control
that plus a few counters or state machines and you have a timed system running
lazy and effective
Posted by renesis at 09:02 | permalink | 0 comments
theres the full datasheets (whatch out for those short data breif ones), theres HAL programming manual, also there is like asm and CMSIS manuals
its mostly just function specifications
so its a very good reference, but like the example code it wont fully explain stuff
the datasheets have all the peripheral and architecture information
and st has a lot of app notes with clear explanations of peripherals and how to code for them
its a lot of stuff
but right, the HAL manual and the datasheet to look up stuff you dont get in the function/constant descriptions is good start
and ya cubemx generated c code is messy as fuck
like, layered and very spread out
well itll compile fine
Posted by renesis at 08:57 | permalink | 0 comments
well, you can look in the library to see what it does, it might just be hitting a bit that says its handling the interrupt
the HAL shit is like, procedural made to be future proof
like, we might want to put code there in 10 years so just make everything the same
like, some of the stuff, you will dig two or three functions deep and its literally nothing
this is why i usually have the manual open
i try and figure stuff out from that more than example code
that has full explanations, example code almost never does
jezus fuck, theres like 4
Posted by renesis at 08:52 | permalink | 0 comments
gpf: i think the issue is its a diff peripheral
like, there is ext interrupts on gpio, but i think you can also set them up nmi, i dunno where the reg is tho
EXTernal Interrupt 2
you can have lots
jero32: sure itll just popout an ISR for the pin
interrupt service routine
yeah
right theres hall stuff right below the cubemx stuff
use a pastabin
but right, cubemx is basically a built template generator for stm32 HAL libs
*build
Posted by renesis at 08:47 | permalink | 0 comments
jero32: there is a source file that equates the vector addresses to function names
its basically like bunch of jmp function() or similar
so you just have to know the function name and put your code in it
jero32: dunno for sure without the compiler open
and its maybe not the same in keil and gcc
it might be stuff in the linker script
yeah
cubemx generates them all
and then tells you where to put code
right so look for the ISR functions
and i think there is a reg for that?
Posted by renesis at 08:41 | permalink | 0 comments
pretty sure
yes on both, and pretty sure gpio can all interrupt but i dunno, maybe exceptions, generally they do
i wish i could wipe that tiva stuff from my brain
Posted by renesis at 08:16 | permalink | 0 comments
but there arent many occations where you cant look shit up if you arent sure
also, the term here is 'freshmans dream' but there are a lot of concept that can fuck you if you use them in the wrong place
like, things you can do with multipliers you cant do with exponents
that takes practice, because thats not something you fuckup because you can't check, yu fuck that up because you dont think you need to
Posted by renesis at 06:25 | permalink | 0 comments
yeah its weird im prob better at math now than 95% of the population, still feel like im not nearly as good as i should be
mostly with me, its a time issue
ya pretty much
but hand calculation, i know i make enough mistakes at little thing that i tend to go slow, check along the way
like, irl you are probably going to be using tools, like if im doing math, im doing it once in a spreadsheet or program
like, i will be testing on some diff equation stuff, but then fuckup on some exponent thing even tho i get the diff equation concept
well you need the math to verify shit
Posted by renesis at 06:20 | permalink | 0 comments
sounds right
later on there are situations where simplifying will fuck you
because obvious calcellation wont be obvious later
in terms of a math class about factoring, less terms usually
im not very good with remembering definitions and rules
Posted by renesis at 06:15 | permalink | 0 comments
you can take one x out of everything
i wouldnt go further than (x^2 + x)/(x^3 + 1)
er
(x^2 + x)/(x^3 - 1)
sure and mine doesnt work for 1 or -1
shrug
trying to figure out what simply means can be difficult if you dont know what the goal is
*simplify means
Posted by renesis at 06:10 | permalink | 0 comments
yeah
so here the progression is what youre doing (precalc), differential calc, integral calc, vector/3d calc (i didnt have to take this), diff equations, sometimes linear algebra
no it stays the same in math coarses
thats basically what math courses are, hey we figured this shit out, now here are a ton of odd corner cases to confuse you
in engineering, you apply math in a practical way
and a lot of the corner cases never come up
but sometimes, an entire concept will be built around some corner case
right
Posted by renesis at 06:05 | permalink | 0 comments
so basically all of engineering is built around it
right
youll do it enough eventually you will
yes
yeah theyve forgotten what its like to not understand
yup
https://www.youtube.com/user/patrickJMT
so like, lots of diff people end up going back to this guy
multiple schools, lots of classes, at some point some students like I FOUND THIS GUY HE CAN EXPLAIN THE SHIT OUT OF EVERYTHING IN THIS CLASS
and its always patrickjmt vids
how many more math classes do you have to take after?
Posted by renesis at 06:00 | permalink | 0 comments
which math?
calc1 and calc2 werent very hard i just didnt like all the homework
diff eq was fucked
?
what are you studying?
review/
?
factoring?
sounds like precalc, its basically reviewing all the shit from before thats going to come up again
so factoring, exponents and trig
right
like, were good at polynomial math
Posted by renesis at 05:55 | permalink | 0 comments
http://www.reuters.com/article/us-usa-trump-russia/u-s-belatedly-begins-to-comply-with-russia-sanctions-law-idUSKBN1CV3LF
heh @ US having problems getting the US to comply with Russian sanctions
Posted by renesis at 04:22 | permalink | 0 comments
every fuckin election, this bullshit, plus gerrymandering, plus supression
everyone knows, too, the thing with rove looking like a fool on tv because he didnt know the fix was cancelled is known by everyone in politics
Posted by renesis at 03:19 | permalink | 0 comments
https://www.theregister.co.uk/2017/10/26/voting_server_georgia_wiped/
"You could just go to the root of where they were hosting all the files and just download everything without logging in," he said. He also noted the files had been indexed by Google, making them readily available to anyone looking in the right place.
That same email also notes that backups of the server data were also deleted more than a month after the initial wipe - just as the lawsuit moved to a federal court.
im sorry republicans can wave american flags around all they want, but theyre like the perfect definition of shit americans
Posted by renesis at 03:14 | permalink | 0 comments
Top | Add to Technorati Favorites