#avr Logs

Oct 15 2019

01:10 AM rue_bed: --
01:51 AM rue_bed: hmm, I need to capture two edges relative to each other
01:52 AM rue_bed: but in hardware at 16Mhz resolution
01:52 AM rue_bed: faster than an isr can do
01:56 AM rue_bed: hmm, on mega88
01:57 AM rue_bed: maybe I need a different avr
01:57 AM rue_bed: what has two input compares?
01:57 AM rue_bed: I wonder where that feature chart went
01:59 AM rue_bed: hmm, 328 also only has 1
02:00 AM rue_bed: not mega32
02:00 AM rue_bed: hmm, I might need another idea
02:02 AM rue_bed: at 16Mhz, dividing by 256...
02:03 AM rue_bed: 62.5kHz rollover
02:03 AM rue_bed: hypothetically I can count rollovers going that fast
02:04 AM rue_bed: hmm, and I only need them from 1 counter
02:05 AM rue_bed: dont ya just hate it when your in bed and you need a peice of paper to think an idea thru?
02:05 AM rue_bed: ok, so I have an 8 bit counter clocked from a 16Mhz source
02:05 AM rue_bed: with two latches off it
02:05 AM rue_bed: edge triggered
02:06 AM rue_bed: ok
02:06 AM rue_bed: I need one latch to trigger on rising edges
02:06 AM rue_bed: that one is period?
02:06 AM rue_bed: yes
02:06 AM rue_bed: you need a flag too
02:07 AM rue_bed: I can use the interrupt pin
02:07 AM rue_bed: ok
02:07 AM rue_bed: so phase
02:07 AM rue_bed: yea, thats... also rising edge?
02:07 AM rue_bed: yes, but you need to deal with reset
02:07 AM rue_bed: software?
02:08 AM rue_bed: I suppose if you clear the rollover counter and snapshot the counter, that works
02:08 AM rue_bed: the counters are snapshot for you
02:08 AM rue_bed: well, this seems quite plausable
02:08 AM Miyu is now known as hackkitten
02:09 AM rue_bed: I think a good question, is if a ripple counter is ok
02:09 AM rue_bed: yea, if so, '393
02:09 AM rue_bed: whats it good to/;
02:09 AM rue_bed: ?
02:09 AM rue_bed: 50Mhz
02:09 AM rue_bed: k, so 16Mhz is nothing
02:10 AM rue_bed: maybe we could do 16 bits?
02:10 AM rue_bed: there is nothing synchronous?
02:11 AM rue_bed: there might be, but I dont know it
02:11 AM rue_bed: question?
02:11 AM rue_bed: yes?
02:11 AM rue_bed: stm32 have two input compare channels?
02:11 AM rue_bed: hah you dog
02:11 AM rue_bed: well, its worth asking
02:14 AM rue_bed: hey it says both timers have 4 capture channels
02:14 AM rue_bed: well damnit
02:16 AM rue_bed: er, yea dude, yea
02:16 AM rue_bed: yea, 4 input compare channels
02:16 AM rue_bed: we need to switch micros
02:16 AM rue_bed: well damnit
02:16 AM rue_bed: yea, sorry
02:17 AM rue_bed: well, good news tho
02:17 AM rue_bed: yea?
02:17 AM rue_bed: we wrote it all in portable C :)
02:17 AM rue_bed: yea, this change is gonna hurt tho
02:17 AM rue_bed: yea, but its a good hurt, right?
02:17 AM rue_bed: your just evil
02:17 AM rue_bed: sorry
02:22 AM * Thrashbarg checks his ignore list
02:23 AM Thrashbarg: either I had someone ignored for some reason or rue_bed has completely lost it :P
02:26 AM rue_bed: its all good, were just working out the problems with my filter profiler
02:27 AM rue_bed: I'm using an isr to capture one of the timer values, and its causing phase measurement to drift/jump by multiples of 2 degrees
02:27 AM rue_bed: but none of the avr I have, have got more than one input capture on a timer
02:28 AM rue_bed: it was pointed out that the stm32 might have better hardware, and it does, which means porting my project, which sucks
02:28 AM rue_bed: but will fix the issue
02:28 AM rue_bed: accept the 3.3V thing
02:28 AM rue_bed: whaaaa!?
02:29 AM rue_bed: no worries, their 5V tollerant.....
02:29 AM rue_bed: }:/ hmm
02:29 AM rue_bed: will the outputs be ok tho
02:29 AM rue_bed: did you ever do serial with your stm32s?
02:29 AM rue_bed: I think so...
02:30 AM rue_bed: well, hmm
02:30 AM rue_bed: your not sleeping
02:30 AM rue_bed: yea, you are keeping me up
02:30 AM rue_bed: its your fault for still needing that job
02:31 AM rue_bed: you could help me with that ya know
02:31 AM rue_bed: stm32
02:31 AM rue_bed: yea, port project, ugh
02:31 AM rue_bed: what did we use stm32 on so far
02:31 AM rue_bed: I dunno, where do all my avrs go?
02:31 AM rue_bed: heh, yea, I'm not sure
02:32 AM rue_bed: we must have used an stm32 on something
02:32 AM rue_bed: I reaaaaaly dont know
02:32 AM rue_bed: oh well there was the hexapod
02:32 AM rue_bed: yea, ok
02:37 AM rue_bed: hmm most of my stm32 stuff is just tests
02:37 AM rue_bed: k
05:42 AM Ashleeee is now known as Ashleee
10:52 AM nohit: --
10:57 AM Fuchikoma: ++
07:43 PM aborazmeh_ is now known as aborazmeh
11:55 PM day__ is now known as day