#avr Logs
Nov 17 2018
#avr Calendar
12:10 AM rue_mohr: jesseg, I need to unify the name format of my gui callback functions, ideas?
12:30 AM rue_mohr: YES!!!! I HAVE ACHIVED CAM!!!!!
12:40 AM jesseg: rue_mohr, sorry I'm afraid I'm fresh out of ideas. I'm not even sure I understand the question!
12:40 AM jesseg: Congratulations on your achievement though!
12:41 AM rue_mohr: jesseg, the callbacks from the gui
12:41 AM rue_mohr: all the functions need to be renamed
12:42 AM rue_mohr: something that signifies them as callbacks
12:43 AM jesseg: oh, I didn't realize you were using callbacks
12:43 AM jesseg: what events trigger the callbacks?
12:44 AM rue_mohr: motif
12:44 AM rue_mohr: / xt
12:44 AM rue_mohr: all the clicks, menus,
12:44 AM rue_mohr: the program is basically two parts
12:44 AM jesseg: oh... right.... the ui... doh...
12:45 AM rue_mohr: one that sets up an event interface, and one that is all the event handlers
12:45 AM jesseg: I dunno, I never was good with names. What about something droll like appending _callback to the ends of th enames? :P
12:45 AM rue_mohr: http://www.cambam.info/screenshots/img/timing-pulley.png
12:46 AM rue_mohr: ho hum, I'll get closer, I will
12:46 AM rue_mohr: I have no idea how to process holding tabs
12:47 AM jesseg: hey do you think a gear generation feature would be something cool to add to my open source cad program?
12:48 AM jesseg: I plan to put object array generation features for things like circular hole patterns and such
12:49 AM rue_mohr: na, too much work for the payoff
12:51 AM jesseg: unless one day I want to cut a wooden gear
12:51 AM rue_mohr: woodgears does a good one
12:51 AM rue_mohr: I was going to write one, its evil
12:52 AM rue_mohr: I made the 2d library for it
12:52 AM jesseg: I think I'd definitely try to write my own before paying for such a simple thing
12:52 AM jesseg: especially since I already have the user interface
12:55 AM rue_mohr: one of the next things I need to do is to design a really complex gui pane for setting all the properties of the toolpath operation
12:55 AM rue_mohr: lucky enough, I'm only starting with drilling holes
01:02 AM jesseg: Good ngiht
01:02 AM jesseg: night :P
01:54 AM polprog: rue_mohr: no, i dont have a CNC
01:54 AM polprog: and damn i have so much shit to do here
01:54 AM polprog: disk cloning failed yesterday, bloody ntfs
01:55 AM polprog: i need to check what settings did linux use when i was formatting my external
01:55 AM polprog: and see if not disabling the image slicing helps
01:56 AM polprog: because it ran for 2h, i stayed up late to let it finish
01:56 AM polprog: then it finished and tried to write a 100G file onto ntfs
01:57 AM polprog: and it shat the bed saying no space left kn device. but there is 700G free there
01:57 AM polprog: so i suppose that it has to do with cluster size in ntfs.
01:58 AM polprog: i hope enabling slicing will work
03:42 AM vmt: journalling filesystems are so messy
03:43 AM vmt: fatxx is the best
03:43 AM polprog: fat12
03:43 AM polprog: ftw
03:43 AM vmt: well no, not 12
03:44 AM vmt: it's super annoying to cross byte boundaries
03:44 AM polprog: 2/10 not fucked up enough
03:44 AM polprog: how about NTFS but on Sun partitioned disk?
03:45 AM polprog: Sun partition table is just pure instanity
03:45 AM vmt: no ntfs anywhere
03:45 AM vmt: i hope reiser gets assfucked in prison
03:46 AM vmt: for having conceived such a monstrosity clusterfuck filesystem
03:46 AM polprog: but how will i complete my bill gates shrine if i dont use ntfs
03:46 AM polprog: lol reiser
03:46 AM polprog: he killed his wife if i remember correctly?
03:46 AM vmt: did he ever go to prison? I guess he did
03:46 AM vmt: well, he was charged with it anyway
03:47 AM vmt: recite the winapi every night and salvation will be coming your way
03:47 AM vmt: (jokes aside, writing stuff for winapi beats writing stuff for posix 11/10 times, for documentation)
03:47 AM vmt: well, because of the docs*
03:49 AM polprog: the posix core is pretty good
03:49 AM polprog: as in docs
03:49 AM polprog: the rest is average, depending on the topic
03:49 AM vmt: average being an overstatement
03:50 AM polprog: definitely im not gonna be scrolling man pages, i look up docs online
03:50 AM vmt: well, usually you won't find better docs than the man pages
03:50 AM polprog: have to disagree
03:50 AM vmt: in which case it doesn't exactly matter where you look them up
03:50 AM vmt: okay, give me an example
03:50 AM polprog: manpages online is more of a formatting issue
03:51 AM polprog: example
03:51 AM polprog: using termios for posix/linux serial stuff
03:51 AM vmt: what are you saying
03:52 AM vmt: unless we're talking about having outdated man pages in your system
03:52 AM polprog: i dont thing reading man pages would give me any idea how to properly write a serial interfacing program
03:52 AM polprog: they are good for looking up stuff that you know how works, but need to remind yourself the syntax
03:52 AM vmt: but that's outside of POSIX-domain really, in a lot of senses
03:52 AM polprog: its a reference, not a tutorial
03:53 AM vmt: exactly
03:53 AM polprog: and more times i need a tutorial instead of a reference
03:53 AM vmt: well, the tutorials are *NOT* documentation
03:53 AM polprog: yeah
03:53 AM vmt: basically my original argument was MSDN vs man pages (wrt. posix)
03:53 AM polprog: yeah
03:53 AM vmt: and winapi is one of the best documented systems i've ever come across
03:54 AM polprog: its probably the most used API in the world
03:54 AM vmt: which?
03:54 AM polprog: Winapi
03:56 AM vmt: maybe
03:56 AM polprog: think about it
03:57 AM polprog: what language do you program windows in and what language linux in
03:57 AM vmt: well, it depends. windows has a strong push for using C#
03:57 AM polprog: exactly
03:58 AM polprog: and linux has tons of different languages in use
03:58 AM vmt: whichever happens to be the most trendy
03:58 AM polprog: most of them use their own apis an libraries, and the posix api is hidden deep down below them
03:59 AM polprog: how many people who program on linux know what ioctl does
03:59 AM polprog: i guess not much because they write python or whatever
03:59 AM vmt: well i guess you're right. but when i speak of winapi, i meant the c-interface, strictly
03:59 AM vmt: mean*, even
04:00 AM polprog: okay
04:00 AM vmt: technically i guess it has a different name. win32 i guess. should have been more clear
04:00 AM polprog: still i guess C# api is similar to it
04:00 AM polprog: yeah the win 32 subsystem
04:01 AM vmt: i guess winapi is an umbrella term covering all of their apis
04:01 AM vmt: ...for all languages
06:37 AM kakimir: Severity Code Description Project File Line
06:37 AM kakimir: Error unknown type name 'size_t' SmartCharger C:\Users\Development\Documents\Atmel Studio\7.0\SmartCharger\SmartCharger\main.c 110
06:38 AM kakimir: nice
06:38 AM kakimir: there should be clipboard monitor in windows
06:38 AM kakimir: but hey is it true there is no size_t?
06:41 AM kakimir: forgot it was in stddef.h :|
07:25 AM cehteh: i guess you are the only one using atmel studio :D
11:32 AM rue_mohr: its a good thing you dont need it
12:00 PM nohit: not everyone is a programming wizard like you rue
12:03 PM rue_mohr: while in the shower it occured to me that debuggers are a means to reverse engineer software
12:03 PM rue_mohr: meaning that using a debugger means your reverse engineering the software you wrote to try to figure out how it works
12:04 PM rue_mohr: it makes a lot more sense if your debugging someone elses code, kinda
12:04 PM rue_mohr: like a closed source library
12:06 PM nohit: :D
12:34 PM rue_mohr: awe, there is no entry widget thats set up to only take numebrs
12:34 PM rue_mohr: jesseg, you have a entry field filter in your code dont cha?
01:03 PM rue_shop3: jesseg, can you give me the link to you cam program gain
01:04 PM rue_shop3: jesseg, can you give me the link to you cam program again
01:14 PM nohit: do you have something in development rue ?
01:18 PM nohit: nice. maybe i underestimated you
04:34 PM Rickta59: i'm trying unsuccessfully to find an older version of the dragon_fw.zip on the atmel/microchip site .. anyone know by change where i might find one?
04:35 PM Rickta59: i loaded atmel studio 7.0 .. it wanted to upgrade my firmware .. then it failed
04:35 PM Rickta59: so i tried loading 6.2 and doing a manual update from there
04:35 PM Rickta59: still doesn't recognize the dragon
04:36 PM nohit: polprog do you have old dragon firmware ?
04:41 PM Rickta59: it seems to be stuck in upgrade mode
04:41 PM Rickta59: it shows up as AVRBLDR in lsusb
04:42 PM Rickta59: i do not have the old firmware
04:42 PM Rickta59: as i was mostly using it on linux
04:42 PM Rickta59: and then i tried atmel studio 7 for the first time in a vm
04:43 PM Rickta59: at one point ( 5 years ago) I had the old avr studio stuff
04:43 PM polprog: nohit: i think mine has 6.2 but i have no idea how to rip it
04:44 PM Rickta59: i found it in there .. but atfw.exe -t avrdragon -a dragon_fw.zip didn't seem to do anything
04:44 PM Rickta59: c:\program ... \avrdragon\dragon_fw.zip
04:59 PM [1]MrMobius is now known as MrMobius
06:16 PM Evidlo: I'm thinking about using an attiny85 to sample an X and Y voltage at 25Hz and plotting a line on a SPI LCD using the USI. Should be able to do that, right?
06:19 PM LeoNerd: Almost certainly. If clocked at 8MHz internal oscillator, that still leaves you 320,000 instructions per sample
07:07 PM rue_mohr: unles you use c++ or java, yea, you should be fine
08:28 PM Rickta59: arg .. that was painful ..
08:28 PM Rickta59: i ended removing all the atmel studio stuff
08:28 PM Rickta59: and then all the drivers
08:29 PM Rickta59: then installed the Atmel USB Drivers install 7.0.888
08:29 PM Rickta59: https://www.microchip.com/mplab/avr-support/avr-and-sam-downloads-archive
08:29 PM Evidlo: what makes atmel studio better than a text editor and Makefile?
08:29 PM Rickta59: i was using linux ..
08:29 PM Rickta59: i made the mistake of saying 'yes' update my driver when i went to try out atmel studio 7
08:30 PM Rickta59: then i used the atfw.exe -t avrdragon -a dragon_fw.zip from 6.2
08:30 PM Rickta59: * then uninstalled atmel studio .. and drivers shutdown windows vm and went back to linux
08:30 PM Rickta59: i really wanted to check out the debugger
08:31 PM Rickta59: but i can do that with avarice and avr-gdb
08:31 PM Rickta59: * looks at clock .. hmm .. this started a couple of hours ago
08:56 PM rue_mohr: why
08:56 PM rue_mohr: any code editor and avrdude
08:56 PM rue_mohr: and gcc-avr
08:57 PM rue_mohr: atmel studio is just pain, nomatter how or what your trying to use it on
08:57 PM rue_mohr: nobody ever comes here and says "WOW, THAT ATMEL STUDIO IS JUST AWESOME!!!"
08:58 PM rue_mohr: NOBODY
10:37 PM Rickta59: why? because I wanted to see what the latest atmel studio 7 offered, I want to use those new attiny416 chips and its updi (sp?) debugger, and I don't think that is going to work on linux. I haven't bought the attiny dev board yet, but I did have an avr-dragon and an atmega644p lying around, so I thought I'd see what the debugger offeres
10:39 PM Rickta59: and it basically chocked, but that could have been because it was running in a vm. When it was trying to update the firmware, i'm sure it was monkeying around with usb getting it to reenumerate and it timed out and failed
10:39 PM Rickta59: basically choked
10:39 PM Rickta59: probably would have been fine if i was running a native windows machine
10:41 PM Rickta59: my desire would be to just use the attiny dev board on linux. i'm pretty sure avarice hasn't been updated in years so i really wouldn't have a clue on how to make the latest debugger thing work
10:45 PM rue_mohr: so question
10:45 PM rue_mohr: why do you need to reverse engineer code you wrote?
10:46 PM Rickta59: i'm lazy
10:46 PM rue_mohr: huh
10:46 PM Rickta59: i like to have a debugger
10:46 PM Rickta59: maybe i'll never need it, but if i'm going to use a chip, i want to know it is there
10:47 PM rue_mohr: http://paste.debian.net/1052169/
10:47 PM rue_mohr: I didn't have to single step or breakpoint that code once...
10:47 PM rue_mohr: those for loops have almost __max_evil__
10:48 PM Rickta59: ok .. what does x code have to do with embedded chips?
10:48 PM Rickta59: sometimes it is just easier to go into gdb and type:
10:48 PM rue_mohr: I dont ahve to single step or breakpoint the code I write for chips either
10:48 PM Rickta59: x/16x 0
10:49 PM Rickta59: x/h *0x66
10:49 PM Rickta59: you are amazing
10:49 PM Rickta59: i'm an idiot
10:49 PM rue_mohr: aparently the bar isn't really that high
10:49 PM Rickta59: i like a debugger, you don't ...
10:49 PM rue_mohr: I dont need one...
10:50 PM rue_mohr: I tend to do a few tests when I'm checking things
10:50 PM rue_mohr: dont need a debugger for it
10:50 PM Rickta59: i like watchpoints on my stack
10:50 PM Rickta59: i like not having to have a serial port running to get info
10:51 PM Rickta59: i like to examine shit in memory without writing code to do it
10:51 PM rue_mohr: I'd like to take on a small project of someones, to see what people are up to and how I'd attack it vs them
10:52 PM rue_mohr: small, cause, time.
10:53 PM rue_mohr: in the meantime, I seem to have created the most advanced number entry field that motif has ever seen...
10:54 PM rue_mohr: ***HAHAHA!!! I HAVE CONQUERED THE STONE AGE!!! ***
10:54 PM rue_mohr: :)
10:55 PM Rickta59: why would anyone be writing native X in the age of webbrowsers
10:57 PM rue_mohr: becasue I dont like apps that take 3 seconds to respond to a drag event
10:58 PM rue_mohr: actauly, that number entry field works better than anything in python, or any app
10:58 PM rue_mohr: lots of people try to validate the entry after you hit enter or soetmhing
10:59 PM rue_mohr: one of the program I tried actaully replaces the field with "Error" if its not valid
10:59 PM rue_mohr: pronterface will just revert the value to whatever was valid
10:59 PM rue_mohr: librecad will just silent abort whatever was going on
11:00 PM rue_mohr: I wonder what calc.exe does if you enter an invalid string for a number
11:01 PM rue_mohr: if it lets you
11:06 PM rue_mohr: I need th speed
11:06 PM rue_mohr: I am abnormally fast on a computer
11:06 PM rue_mohr: I can click a button so fast, that the graphic of the button changes, but the thing its supposed to do doesn't happen
11:06 PM rue_mohr: it happens a lot
11:07 PM rue_mohr: windows and linux
11:07 PM rue_mohr: I'm guessing that if the mouse button isn't still down after the button image is changed, the handler is never called
11:07 PM rue_mohr: its quite frustrating
11:07 PM rue_mohr: nobody I'v ever heard of has ever had it happen to them
11:09 PM rue_mohr: my patience with a computer lasts about 470ms, after that I'm waiting and borred
11:38 PM day__ is now known as day