#avr Logs

Apr 18 2018

#avr Calendar

12:09 AM _ami_: polprog, good afternoon
01:04 AM polprog: if someone opped me via chanserv OP command but im not in the list in ACCESS, if i deop myself i wont be able to op myself back?
01:12 AM Casper: exactly
01:12 AM Casper: same as if they would have done a /mode #chan +o nick
01:12 AM Casper: except that need the person to be actually opped
01:21 AM polprog: ok
03:00 AM Haohmaru: does the avrisp2 clone have some problem with xmega with 128K flash size?
03:01 AM Haohmaru: i'm trying to put a bootloader, and it's either broken (bad code, bad compiler/linker arguments), or it doesn't get flashed correctly
03:02 AM Haohmaru: "avrdude: verification error, first mismatch at byte 0x20000 0x0d != 0xff"
03:07 AM Haohmaru: datasheet says the boot section starts at 10000, i tried that (it flashes and verifies) but doesn't work, i then tried 20000 because that should be the offset in bytes (instead of words) and i can't even verify it
03:07 AM Haohmaru: :~(
03:12 AM Haohmaru: hm, i should write the bootloader (despite the verification error) and then try to extract all of the flash to see what's been written where
03:15 AM * Haohmaru blames Emil
03:35 AM Haohmaru: i don't even get any data when i extract the flash x_x
03:35 AM Haohmaru: i guess it's all FF
03:36 AM Haohmaru: (the app section at least)
03:38 AM Haohmaru: oh wait, since the boot section is dedicated (unlike atmegas) avrdude defines that as "boot", not "flash" .. so maybe.. maybe..
03:48 AM Haohmaru: nope
03:48 AM Haohmaru: and even tho i've set the BOOTRST fuse, if i flash my normal firmware - it runs as usual
03:53 AM Haohmaru: hm, i can flash the bootloader with -U boot:w:... if i've compiled the bootloader with -Wl,--section-start=.text=0x00000, then it verifies
03:53 AM Haohmaru: but this seems a bit wrong
04:32 AM Haohmaru: i tried converting the elf to bin, i get an 8MB binary x_x
04:32 AM Haohmaru: :~(
04:36 AM AndrevS: Are there any fuses encoded in the elf file? And do they map to a certain memory address that would sit at 8 MB? Then I suppose, this would result in the firmware, a lot of zeroes, and then the fuses.
04:37 AM Haohmaru: yes, there are fuses defined in the actual code, so they surely go into the elf file
04:37 AM Haohmaru: the binary begins with stuff that looks like a bootloader tho
04:37 AM Haohmaru: i'll look into that
04:41 AM Thrashbarg: off topic, is there a Z80 emulator core written in C which does cycle by cycle emulation?
05:18 AM nuxil: Thrashbarg, ? http://www.oshonsoft.com/z80.html, and set a breakpoint on each instruction :p
05:20 AM Thrashbarg: I want to simulate the video section of a small computer I want to make and need a cycle accurate core to run it with :P
05:20 AM Thrashbarg: *clock cycle
05:22 AM * nuxil slaps Thrashbarg. its not 1976 anymore, get a stm32 or something
05:23 AM Thrashbarg: everyone's done STM32
05:23 AM nuxil: except me :p
05:23 AM Haohmaru: and me
05:23 AM Thrashbarg: mmgmm
05:23 AM Thrashbarg: mhmm
05:23 AM Thrashbarg: *
05:23 AM nuxil: i still got this nucleo in its orginale package :p
05:26 AM AndrevS: ST makes fine microcontrollers.
05:28 AM AndrevS: I've been using STM32F103 and L151. The Noric nRF52 are also worth looking at, even if you're not using the radio. The great thing about the nRF series is all digital peripherals can be mapped to any gpio pin.
05:29 AM LeoNerd: Mmmmm crossbar
05:29 AM LeoNerd: The PIC24s have that. I'd reeeeally like one in an ATmega chpi
05:29 AM LeoNerd: *chip
05:30 AM LeoNerd: Some of the ATtiny chips are starting to go that way; e.g. most of the newer tinies have a REMAP register that lets you move some pins around, or the flexible timer output pins. It's a good start, but I feel more could be done
05:31 AM AndrevS: I haven't looked at PICs yet. Even though I have some laying around waiting.
05:48 AM Haohmaru: added -R .eeprom -R .fuse -R .lock -R .signature, now the binary is the same size as the .data section
05:51 AM polprog: we should develop an AVR simulator on STM32
06:05 AM Haohmaru: when the simulated program crashes, the simulator should print: "Error, ur program crashz0riz3d, blame Emil!"
06:09 AM polprog: rofl
06:36 AM polprog: if (errno == ERTFM || errno == EMIL)
10:18 AM nuxil: ERTFM is redundant
10:18 AM nuxil: :p
10:19 AM polprog: https://xkcd.com/293/
01:44 PM polprog: https://m.imgur.com/gallery/0ozsD1V
01:44 PM polprog: <3
01:45 PM polprog: nuxil, Jartza ^^
04:01 PM dStruct: polprog: lol at the meme from like 5 hours ago :/
04:02 PM polprog: so its "lol" and ":/" at once
04:02 PM polprog: agreed. heh
04:05 PM dStruct: I think that pretty much sums it up hehe
04:06 PM dStruct: so I'm trying to step up my 5v DO to 12v for my clock signal out, however I'm having some very weird results
04:10 PM dStruct: I'm guessing 300ohm is too much limit and I'm not getting enough current maybe, I have a 0-5v square wave (50% duty, 2.5v avg), going into a 2222 base, 12v on collector, emitter to probe and 1k to GND, max I can get is 2-2.5v
04:27 PM Casper: What a close fit! https://i.imgur.com/cXXZR88.jpg https://i.imgur.com/oKyAtbZ.jpg https://i.imgur.com/nB57oE6.jpg <=== I don't know how come it actually does close
04:32 PM dStruct: Casper: nice, I like Asus products
04:33 PM dStruct: fwiw, http://www.schematics.com/project/5v-to-12v-signal-booster-99504/
04:38 PM polprog: 2222 is a BJT
04:39 PM polprog: i dont know its typical beta but i would use a mosfet
04:39 PM polprog: 2n7000 for example
04:39 PM polprog: with mosfets at your speeds you can forget about base currents etc.
04:40 PM polprog: and since your system is most likely cmos then its a bit simpler to use mosfets as well
04:43 PM Casper: dStruct: saw the clearance between the sink and the video card?
04:46 PM polprog: holy crap casper
04:46 PM Casper: polprog: saw the side of the case?
04:47 PM polprog: saw all 3
04:47 PM Casper: it DOES close!
04:49 PM polprog: heres mine lol
04:50 PM polprog: https://puu.sh/yF7U2/b8adc9398a.jpg
04:50 PM Casper: hehe
04:50 PM polprog: 4 core 4ghz in a 20 year old athlon case
04:50 PM polprog: rip airflow
04:50 PM Casper: :)
04:50 PM Casper: 14 cores
04:51 PM polprog: and the psu fan shat itself and makes noise all the time. gotta replace it soon
04:51 PM polprog: wow 14
04:51 PM polprog: thats a weird number for cores
04:51 PM polprog: why not 16
04:51 PM cehteh: https://pbs.twimg.com/media/DJO66zUWAAIHzov.jpg :D
04:52 PM dStruct: Casper: yeah that's nuts
04:53 PM dStruct: polprog: the 2222 is just what I had a couple of laying around, I should really be using a 3904
04:53 PM Casper: polprog: maybe a 16 cores with 2 cores disabled, so if any 2 of the 16 are faulty they can still sell it
04:53 PM polprog: maybe
04:53 PM Casper: like the ps3 is a 8 cores with 1 disabled, and advertise it as a 7<
04:54 PM polprog: dStruct: oh ok. my next thought was the buffer chip
04:54 PM polprog: cehteh: thats not bad at all mate
04:54 PM Casper: https://i.imgur.com/zZddSjH.jpg <=== missing parts, like ram
04:54 PM cehteh: with mine the hdds where not cabled when i took that pic
04:55 PM cehteh: but that was only for transfering some data, later i removed some hdd's
04:55 PM cehteh: only 6 drives now
04:56 PM cehteh: and you dont see the 64GB ECC RAM :D
04:59 PM Casper: https://i.imgur.com/was9jNs.jpg <=== my nas, with 10hds
05:00 PM Casper: time to close shop
05:00 PM Casper: and attempt to kill my car engine
05:00 PM dStruct: polprog: ahh I figured it out, I had the damn resistor on the wrong side of the circuit, I had to move it from the GND side of emitter, to the collector side where the 12v comes in
05:02 PM polprog: oh. yeah that makes sense
05:02 PM polprog: so what, using two transistors to make a 5-12 v level shifter?
05:02 PM polprog: it should work.
05:03 PM dStruct: it's probably going to end up being two, I have 1 in there now and get 6v ish
05:03 PM dStruct: max23x series chips would have probably worked too if I had them laying around
05:04 PM polprog: surely they would
05:05 PM polprog: i suggest you take a look at the whole maxim converter range, but apart from max232 which is the most popilar, there are also quad input ones for example
05:07 PM dStruct: I have, that's why I mentioned the 23x as in they have a few 232-like chips, very handy
05:09 PM polprog: indeed
05:09 PM polprog: oh i didnt notice the x lol
05:09 PM polprog: well, time to sleep. haha
05:12 PM polprog: okay ive see it all now
05:12 PM polprog: https://stackoverflow.com/questions/592322/php-expects-t-paamayim-nekudotayim
05:12 PM polprog: php has an error code in HEBREW
05:21 PM dStruct: polprog: have a good one :)
09:40 PM rue_shop3: polprog, whats wrong with an old case?
11:02 PM polprog: rue_shop3: nothing in general, thi one may have a bit of space probles but it works well anyway
11:27 PM day__ is now known as day
11:47 PM rue_mohr: ah
11:49 PM polprog: rue_mohr: have you seen the photo?
11:51 PM rue_mohr: jigawatt heatsinks dont fit in old cases
11:51 PM rue_mohr: unles you bought a proper full size atx case back in the day
11:59 PM polprog: not sure if its a full size atx case.. 17"x16"