#avr Logs
Dec 26 2017
#avr Calendar
01:25 AM nuxil: mornings!
01:28 AM nuxil: whats been going on here ?
02:17 AM nuxil: so when i do a compile. it complains about the asm stuff. same as yesterday. https://pastebin.com/euAL3Uy6 but it gives a error like:
02:17 AM nuxil: /tmp/ccCDmZHO.s:317: Error: number must be positive and less than 32
02:18 AM nuxil: wtf does that mean
02:18 AM nuxil: it related to the asm code by the looks of it since it says. Assembler messages:
02:29 AM nuxil: who got an idea?
02:36 AM Thrashbarg: nuxil: try running GCC with -S to stop the compile process before it's assembled, then look at line 317
02:36 AM Thrashbarg: there should be a .S file there
02:37 AM nuxil: ok tanks. will try
02:56 AM antto: Emil i know it's a bot
02:56 AM antto: i can talk to a wall if i want
04:25 AM nuxil: damn sbis
04:27 AM nuxil: old atmega8 code used this in the asm.
04:27 AM nuxil: :"r" (ad0),"r" (ad1),"r" (ad2),"e" (signal),"I" (_SFR_IO_ADDR(PORTA)), "I" (_SFR_IO_ADDR(SPCR))
04:27 AM nuxil: but that gave me error.
04:28 AM nuxil: main.c:98:92: error: 'SPCR' undeclared (first use in this function)
04:28 AM nuxil: so i changed it to.
04:28 AM nuxil: SPDR0
04:28 AM nuxil: i mean SPCR0
04:29 AM nuxil: but that gives me error Error: number must be positive and less than 32
04:29 AM nuxil: .
04:30 AM nuxil: and its related to sbis. cos sbis will only work on lower registers.
04:30 AM nuxil: and it seems like spcr0 is a higher one.
04:31 AM * nuxil goes to look for a new register.
04:39 AM nuxil: bad. i'll just use portd register :p
04:39 AM nuxil: *bah
04:40 AM nuxil: yay. and it compiles
04:42 AM antto: it compiles! SHIP IT!
04:43 AM theBear: with that attitude yer'll get a job programming fer ms in no time, could be main team or fixes+updates dept, they both right into that rule
04:45 AM antto: i learnt this from #c++ ;P~
04:47 AM nuxil: see.. one more reason no to C++ :p
04:47 AM nuxil: jk
04:48 AM antto: >:(
04:55 AM Jartza: is there actually reason for inline asm?
04:55 AM nuxil: yes
04:55 AM Jartza: hard to tell without seeing more of the code
04:57 AM nuxil: scroll up
04:57 AM nuxil: and you'll see the paste
04:58 AM nuxil: and to understand more about dds. read this :p file:///F:/Coding/C/C-AVR/Atmega324/Siggen/Direct%20Digital%20Synthesis%20(DDS)%20technology.html
04:58 AM nuxil: ups
04:59 AM nuxil: http://www.hit.bme.hu/%7Epapay/sci/DDS/start.htm
05:00 AM nuxil: portA is outputed to a R2R network
05:02 AM nuxil: so as you see. the resolution is dependand on how many cycles you do.
05:03 AM nuxil: in this case its a 24bit phase accumulator
05:04 AM antto: hm.. i'm doing a 32bit digital oscillator with some funky 64bit maths on the atmega2561
05:04 AM antto: in C++
05:05 AM antto: but it's used for a synchronization clock, no waveforms/LUTs/DACs..
05:06 AM antto: nuxil isn't reading from PROGMEM a bit slower?
05:08 AM nuxil: <nuxil> So i have a question. the signal tables are defined with sections. like
05:08 AM nuxil: <nuxil> const uint8_t sinewave[] __attribute__ ((section (".MySection1"))) = {...}
05:08 AM nuxil: <nuxil> and in my makefile i need to put LDFLAGS += -Wl,-section-start=.MySection1=FooBar
05:08 AM nuxil: <nuxil> but cant PROGMEM be used instead ? like
05:08 AM nuxil: <nuxil> static const uint8_t sinewave[] PROGMEM {...}
05:08 AM nuxil: got no real good aswer on that. :p
05:09 AM nuxil: other than that __attribute__ ((section (".MySection1"))) might have been some old stuff and PROGMEM has replaced its method
05:09 AM antto: not sure about sections, but progmem is a way to tell the compiler to keep that data as part of the flash program and read it directly from there
05:09 AM antto: this could save you RAM
05:12 AM nuxil: lpm. takes 3 cycles anyway.
05:13 AM antto: not sure, maybe sections were used due to something related to the inline asm
05:14 AM nuxil: as suggested. the code is old. and PROGMEM might have not been in the avr lib yet. idk.
05:14 AM antto: poshiburu
06:19 AM Jartza: .MySection1 is probably defined in linker script and as arguments for linker to place data into specific address
06:26 AM Jartza: nuxil: what are you generating?
06:39 AM nuxil: Jartza, yea. that what it does. i checked the map file with progmem and with section. main difference is where data started. with section i used 3a00 and the data started there. with progmem it started at 3a30
06:43 AM nuxil: Jartza, will be generating lots of frequencys :p
07:04 AM polprog: i remember i was checking out menus of some function gen in some workshop once. one of the presets was "Quake". on the scope ot looked like music but im not sure
07:04 AM polprog: some funny easter egg would be nice :P
07:10 AM nuxil: quake on oscillioscope has been done :p
07:10 AM nuxil: https://www.youtube.com/watch?v=gky7GqblIdU&ytbChannel=null
07:11 AM nuxil: dont know if i could handle playing like that .p
08:34 AM polprog: nuxil: have you considered copying the wave table entry to ram at the beginning? that could save you 1 cycle since lpm is 3 cycles and ld is 2
08:39 AM nuxil: hmm. the array has a size of 256 element. and each element is a byte.
08:47 AM nuxil: not sure how to do it with ld. feel to optimize the code :p
08:51 AM polprog: you can even make the array copying part in C... anyway. LD is used similarly like lpm iirc
08:51 AM polprog: just a suggestion if you need even faster code
08:51 AM polprog: ;)
08:56 AM nuxil: anyway. the table should fit in the ram. since it has 2k sdram.
08:57 AM nuxil: but i'll look into that later. atm. working on the isr routine for my rotaty encoders to vary the frequence.
08:58 AM nuxil: luckely i can reuse lot of the code from my atiny85 :)
10:04 AM polprog: ncie
10:04 AM polprog: nice*
10:04 AM polprog: i need a function generator as well
10:11 AM nuxil: dont you have a atmega16 ?
10:12 AM nuxil: DDS V2 uses a atmega16. but same asm code.
10:14 AM nuxil: plan is to put it in a box like this https://www.elfadistrelec.no/no/skrivebordsboks-svart-217-138-82-mm-abs-rnd-components-rnd-455-00105/p/30064326#
10:14 AM nuxil: with bnc connectors
10:15 AM nuxil: ofc i must drill hole for lcd and stuff :p
10:16 AM nuxil: so it might look like a swiss chees after im done with it
12:09 PM rue_bed: sympatix, hey, where you at, I lost ya
12:20 PM polprog: nuxil: i have a bunch of atmegas. Time and will on the other hand ;)
12:20 PM polprog: that's a nice box
12:27 PM rue_bed: theBear, isn't audio rack different somehow than computer rack?
12:27 PM rue_bed: there must be some kinda hacked narrow standard for a rack thats about half a computer rack width
01:10 PM polprog: audio racks are a bit different but ive seen brackets that you can screw on to adapt one gear to another. i dont remember which rack was wider
01:22 PM rue_mohr: if you can adapt the audio racks, then the computer ones are wider
01:22 PM rue_mohr: </logic>
01:22 PM rue_mohr: wait
01:22 PM rue_mohr: ilogic
01:23 PM rue_mohr: no I read stuff intot hat, blarg
01:23 PM rue_mohr: NOT NOW i'M COING!
01:23 PM rue_mohr: CODING
01:23 PM rue_mohr: having a problem, looks like I might have a typo...
01:24 PM polprog: ive seen brackes that can adapt one type to another, not sure which types exactly
01:25 PM rue_mohr: oh I see, there are two of us coding at once, and each is using a different method to solve the same thing...
01:27 PM polprog: hmm there's a 19" rack which is, well, 19" wide and then there's "etsi" rack which is 21" wide
01:29 PM rue_mohr: this is sweet, did you know the area of a polygon will be positive or negitive depending on if its defined clockwise or counterclockwise?
01:33 PM nuxil: https://gyazo.com/7da12982c784670fd59bf256ef63d06b
01:34 PM nuxil: i wounder if i have enuf resistors for my r2r network :p
01:35 PM polprog: holy shit, all tht. i feel sorry for you
01:36 PM polprog: also, nice fluke
01:36 PM nuxil: its ok
01:36 PM polprog: ive seen some cool paralell DACs, why are you using an r2r
01:38 PM nuxil: better to read this . https://www.tek.com/blog/tutorial-digital-analog-conversion-%E2%80%93-r-2r-dac
01:40 PM nuxil: i am using a r2r network sine portA is my output. all pins are signal
01:42 PM nuxil: so in sort. if you got a paralell digtal signal you want to convert to a analoge signal. use a r2r :)
01:48 PM antto: if you want 16bit, use two SIPOs ;P~
01:48 PM antto: and MOAR resisterz
01:50 PM nuxil: hehe
01:50 PM polprog: :P
01:51 PM antto: abuse the SPI for it
01:57 PM rue_mohr: your playing with dac too eh?
01:57 PM rue_mohr: I
01:58 PM rue_mohr: got some 0603 and 0805 resistors for mine, they are about the size of a dip16
01:58 PM rue_mohr: did one as a sip
01:58 PM rue_mohr: nuxil, I was playing with filtered pwm
01:59 PM nuxil: :)
01:59 PM rue_mohr: using the adc for feedback and using apid loop
02:00 PM rue_mohr: but, I suppose with the pwm output also driving low, the responce is so fast, that the closed loop cant add much
02:00 PM rue_mohr: I also have some code that internally models an RC filter, for a 1 bit digital output
02:01 PM rue_mohr: not tried it yet, but after playing with the filtered pwm...
02:01 PM rue_mohr: atleast with a feedback loop you can get 10 bits resolution fro a 8 bit pwm timer
02:02 PM rue_mohr: http://ruemohr.org/~ircjunk/images/p1090546.jpg
02:02 PM rue_mohr: http://ruemohr.org/~ircjunk/images/p1090545.jpg
02:02 PM rue_mohr: http://ruemohr.org/~ircjunk/images/p1090544.jpg
02:02 PM rue_mohr: http://ruemohr.org/~ircjunk/images/p1090543.jpg
02:03 PM nuxil: nice.
02:03 PM nuxil: what kind of scope is that :p
02:03 PM rue_mohr: pid isn't needed, as it has no inductance, but I was playing
02:03 PM rue_mohr: thats a cheapo chineese DSO138
02:04 PM polprog: i need to play with feedback loops as well
02:04 PM rue_mohr: for now, its my 'digital scope'
02:04 PM rue_mohr: polprog, electronic or mechanical systems?
02:04 PM polprog: electronics
02:04 PM polprog: i dont have enough bench space for mechanical, nor the tools
02:04 PM nuxil: if youre want dds on 1pin. take a look at the tinytune lib :)
02:05 PM rue_mohr: you know baout the really cheap 2 channel 16 bit adcs from china right?
02:06 PM polprog: no?
02:06 PM nuxil: why buy one when you can build one :p
02:08 PM polprog: hmm the tinytune lib makes really adorable sounds
02:09 PM polprog: still a suqre wave but very pleasant as opposed to raw suqare
02:09 PM rue_mohr: ds1287?
02:09 PM rue_mohr: nope
02:10 PM Tom_L: rue_mohr, did you get my note on the ballscrew?
02:11 PM rue_mohr: I cant find the part number
02:11 PM rue_mohr: haha
02:11 PM rue_mohr: Tom_L, no
02:11 PM Tom_L: read the robotics log
02:11 PM rue_mohr: Tom_L, are you gonna stand guard for the next spamming?
02:12 PM rue_mohr: cant just be me
02:12 PM Tom_L: no, i'm gonna go out and get my parts that are ready
02:12 PM Tom_L: he just sent me a txt
02:12 PM rue_mohr: polprog, http://ruemohr.org/~ircjunk/projects/newpower/p1070573.jpg
02:12 PM rue_mohr: top left chip
02:12 PM rue_mohr: :)
02:12 PM rue_mohr: all you have to do is read the number
02:13 PM rue_mohr: oh PT8211?
02:13 PM polprog: rue_mohr: off a 640x480 image? might be difficult ;)
02:14 PM polprog: what am i looking at? some psu?
02:14 PM rue_mohr: yes, thats it pt8211, 100pcs for $10
02:14 PM polprog: wow
02:14 PM polprog: that's cheap
02:14 PM rue_mohr: https://www.aliexpress.com/item/50PCS-PT8211-S-PT8211-SOP8/2024710373.html
02:14 PM rue_mohr: the snag is
02:15 PM rue_mohr: they output +-2.5V centred at 2.5V
02:15 PM polprog: i think i got a reel of 300 74hct573 latches
02:15 PM rue_mohr: nic
02:15 PM rue_mohr: e
02:15 PM rue_mohr: how come you went for thruhole?
02:15 PM polprog: output? so is it an adc or dac?
02:15 PM rue_mohr: you can solder a 0805 or an 0603 across 0.1" pads
02:16 PM rue_mohr: its dac
02:16 PM rue_mohr: 16 bits
02:16 PM rue_mohr: serial
02:16 PM polprog: <@rue_mohr> how come you went for thruhole?
02:16 PM polprog: me?
02:16 PM rue_mohr: ya
02:16 PM rue_mohr: (the resistors)
02:16 PM polprog: it's nuxils resistors
02:17 PM rue_mohr: p1090521.jpg
02:17 PM nuxil: hey. i got like 2k of variouse thruhole resistors. got to use them for something.
02:17 PM rue_mohr: just a sec
02:18 PM rue_mohr: http://ruemohr.org/~ircjunk/images/p1090521.jpg
02:18 PM rue_mohr: 8 bit dac
02:18 PM rue_mohr: SIP10
02:19 PM rue_mohr: I have another one on a thruplated board, that I put the other resistors on the backside
02:19 PM polprog: oh nice. i have a resistor ladder i used to use with a led bar for debugging. saved a lot of wiring on the breadboard
02:20 PM polprog: back when i was debugging by outputting an 8 bit int on a port and watching blinkenlights. i still do sometimes
02:20 PM rue_mohr: ok, your all distracting me from taking over the world
02:21 PM rue_mohr: ok, I have successfully determined if a polygon is cw or ccw
02:21 PM Jartza: nuxil: generating lots of frequencies, but what kind of signal?
02:22 PM nuxil: rue_mohr, https://www.youtube.com/watch?v=AWzhJoPY_7Q
02:22 PM nuxil: Jartza, sinus, tri, rev tri, squar.
02:22 PM nuxil: i mean saw
02:23 PM nuxil: and rew saw
02:23 PM rue_mohr: pff
02:23 PM Jartza: just wondering if some dds chip would be easier :)
02:23 PM rue_mohr: I'm not that kinda person, I just plan to outlive everyone
02:24 PM polprog: nuxil: will you also generate a cosinus wave?
02:24 PM polprog: ;)
02:24 PM nuxil: lolz
02:24 PM rue_mohr: I got a good one
02:24 PM rue_mohr: for generating 60Hz sinewave for AC
02:24 PM rue_mohr: its a rom, the bits are pwm
02:25 PM nuxil: Jartza, yea i guess a dedicated chip would make life alot easyer. but whats fun with that :p
02:25 PM rue_mohr: as you walk thru the addresses, it generates 8 bit pwm sine wave
02:25 PM polprog: a dedicated atmega for dds and then another one for frontend?
02:25 PM Jartza: nuxil: just a thought, if that's for something "real" :)
02:25 PM rue_mohr: I can put 0 degrees, 90 degrees, 120 and 240 degrees all on the same chip
02:25 PM polprog: you wouldnt want your button press to interrupt the generation
02:25 PM Jartza: but if it's just for playing and fiddling, then sure
02:26 PM rue_mohr: so I can generate single, two (90 degrees for running capcitor run motors), or three phase things
02:26 PM rue_mohr: and in pwm, so I can use the pwm to operate the driver directly
02:26 PM rue_mohr: then filter it
02:26 PM rue_mohr: (57Khz)
02:31 PM Jartza: http://www.analog.com/en/products/rf-microwave/direct-digital-synthesis/ad5932.html
02:31 PM Jartza: that's pretty nice chip :)
02:32 PM rue_mohr: https://www.aliexpress.com/item/AVR-DDS-Function-DDS-Signal-Generator-Module-Kits-Sine-Triangle-Square-Wave/32789267445.html
02:33 PM rue_mohr: they still want a fair bit of money for the kits
02:33 PM rue_mohr: I wonder how the dma+spi of an stm32 could perform
02:34 PM polprog: dma would work wonders
02:34 PM rue_mohr: https://www.aliexpress.com/item/Programmable-Microprocessors-AD9833-Sine-Square-Wave-DDS-Signal-Generator-Module-Drop-shipping/32687211805.html
02:34 PM polprog: you just populate the ram area, and let it run.
02:34 PM rue_mohr: from a sine wave, you can generate everything
02:34 PM polprog: how?
02:34 PM Jartza: although, I have no need for signal generator anymore, I have Tsunami
02:35 PM polprog: sorry for stupid questions, my knowledge on analogue stuff is pretty shallow
02:35 PM rue_mohr: well, first you make it square wave, then you can use an integrator to make it triangle
02:35 PM rue_mohr: or use edge detectors to make sawtooth
02:35 PM polprog: hmm
02:35 PM polprog: makes sense
02:35 PM polprog: make it a quare wave with a comparator for example?
02:35 PM rue_mohr: 0 MHz to 12.5 MHz output frequency range
02:35 PM rue_mohr: 28-bit resolution: 0.1 Hz at 25 MHz reference clock
02:35 PM rue_mohr: hmm
02:36 PM rue_mohr: yes
02:36 PM [1]MrMobius is now known as MrMobius
02:36 PM rue_mohr: VOUT Maximum: 0.65V
02:36 PM rue_mohr: VOUT Minimum: 38mV
02:36 PM rue_mohr: ^^ that bit sucks
02:37 PM rue_mohr: but at 25Mhz, you dont have much time to change the voltage anyhow
02:37 PM rue_mohr: prolly fixable with a video amp
02:37 PM polprog: this is still partly voodoo for me
02:38 PM polprog: i have to do a lot of reading :P
02:38 PM rue_mohr: I'v wondered about a pll on a sinewave osc, and using the microcontroller to set the pll ref and rate
02:39 PM polprog: i was thinking about analog generators but with digital pots so a micro can tune them
02:39 PM polprog: never messed with analog pots though
02:40 PM Jartza: polprog: many dds generators even use built-in comparator to generate square wave out of the sine they generate :)
02:40 PM polprog: well generating a square wave out of a wavetable would be pretty pointless
02:40 PM polprog: :D
02:40 PM rue_mohr: I have an HP one, they generate a square wave and then push it thru shapers :(
02:40 PM polprog: sharpers?
02:41 PM rue_mohr: yea, they mull it into a triangle wave and then mull that into a sine wave
02:41 PM rue_mohr: sdrawkcab ssa
02:42 PM polprog: oh. interesting
02:43 PM rue_mohr: ok I need to work on breaking an ear off a polygon
02:43 PM rue_mohr: [ monty python, parrot skit ]
02:44 PM rue_mohr: waiting on 77 deliveries from china
03:40 PM polprog: goodnight
04:21 PM Ameisen: Why is the atmega1284 faster than the 2560, but also has less memory?
04:21 PM Ameisen: There's no 'obviously better' chip than the 2560
04:25 PM cehteh: faster as in what?
04:26 PM cehteh: better speed grades?
04:37 PM rue_mohr: uh, all the chips are made in different speed grades, usually topping out at 20Mhz
04:37 PM rue_mohr: the ones that cant sustain 20Mhz are just rated lower
05:03 PM Ameisen: cehteh - I mean, 'faster' isn't particularly relative
05:03 PM Ameisen: the 1284 is objectively faster. It also less SRAM and flash, though.
05:03 PM Ameisen: 20MHz vs 16MHz
05:05 PM nuxil: my atmega324p supports 20mhz
05:06 PM nuxil: full name is atmega324p-20pu i think.
05:07 PM nuxil: 0 - 20MHz @ 4.5 - 5.5V :)
05:09 PM cehteh: yes, the full name matters, there are different variants
05:11 PM nuxil: yes
05:12 PM nuxil: ATmega324PV is 1/2 as slow as just ATmega324P, PV can only do 10Mhz :p
05:50 PM rue_shop3: actaully, you can prolly overclock it to 20 just fine
05:50 PM rue_shop3: same die
05:50 PM rue_shop3: they just decided that it should only run at 10
05:50 PM rue_shop3: I'd like to see how that grading test works
05:51 PM rue_shop3: maybe there is test firmware
05:51 PM rue_shop3: I wonder if they overclock it and scale it back a percentage
05:52 PM Tom_L: the center of the disk probably produces different results than the edges
05:53 PM rue_shop3: floppy disk?
05:53 PM Tom_L: silicon disk
05:53 PM Tom_L: wafer
05:53 PM rue_shop3: lens flare? :)
05:53 PM Tom_L: different cooling rates, who knows
05:54 PM Tom_L: i got about a foot of slide leftover that i cut off
05:55 PM rue_shop3: oo
05:55 PM Tom_L: probably about the same on the screw
05:56 PM rue_shop3: thats a lot of rail
05:56 PM rue_shop3: more than enough for a 3d printer
05:56 PM Tom_L: i cut the 1000mm set in half
05:56 PM Tom_L: that will become Y
05:56 PM rue_shop3: hey, but I'm interested if there is any detectable lash in the balldrive when you have the chance
05:57 PM rue_shop3: there is some indicators that they are selftightening, and some that they aren't
05:57 PM Tom_L: it will be a while before i test that
05:57 PM rue_shop3: sure
05:57 PM Tom_L: these have backlash nuts
05:57 PM Tom_L: so it better be close to zero
05:57 PM rue_shop3: so their the double ball nut ones?
05:58 PM Tom_L: i guess, they had 4 access holes in em
05:58 PM rue_shop3: hmm
05:58 PM rue_shop3: for the most part, they are way over my budget
05:58 PM Tom_L: i'm not gonna take it off and look
05:58 PM rue_shop3: no
05:58 PM Tom_L: your budget has exceptions for tooling... i know it does
05:59 PM rue_shop3: hmm, your not wrong
05:59 PM Tom_L: this was one of those times for me
05:59 PM rue_shop3: well thats a nice machine
05:59 PM Tom_L: notice most of it is made from scrap though
05:59 PM rue_shop3: my goal is just kinda *another* machine
06:00 PM Tom_L: but why just make *another* machine?
06:00 PM rue_shop3: yes, but you could just as easy go buy that same material new
06:00 PM Tom_L: just to take up more space
06:00 PM rue_shop3: cause the more I do the more I learn
06:00 PM rue_shop3: which means I can do more, and learn more
06:00 PM Tom_L: i observe and learn
06:00 PM rue_shop3: I do to, I'm watching you carefully
06:01 PM rue_shop3: iirc your about $300 in?
06:01 PM Tom_L: and for the most part, this is how the big boys build em
06:01 PM rue_shop3: yea, thats not me tho
06:01 PM Tom_L: the rails were more
06:01 PM Tom_L: maybe 5
06:01 PM Tom_L: total
06:01 PM rue_shop3: I was surprised that they haas machines use that type of slider
06:01 PM Tom_L: alot of newer ones do
06:01 PM rue_shop3: I was really expecting dovetai
06:01 PM rue_shop3: l
06:01 PM Tom_L: but they don't get them from china
06:02 PM rue_shop3: :)
06:02 PM rue_shop3: cant be many other places they are made nowadays
06:02 PM Tom_L: sure there are
06:02 PM Tom_L: just cost alot more
06:02 PM rue_shop3: germany
06:02 PM Tom_L: we probably make em here i'd bet
06:02 PM rue_shop3: the usa?
06:02 PM Tom_L: yeah
06:03 PM rue_shop3: mmm, dunno about that, everyone farms everything out
06:03 PM Tom_L: or ground to a better presision with higher quality metal at least
06:03 PM Tom_L: my bud said the screw was pretty hard
06:04 PM Tom_L: once you get thru the surface hardening it was easier
06:04 PM rue_shop3: ooo good thing I did a loose fit on this printed part, its pretty tight
06:04 PM Tom_L: it was pretty thick. i can see it
06:05 PM rue_shop3: the hardening?
06:05 PM Tom_L: yes
06:05 PM rue_shop3: it looks like china has two price grades, I'd be interested to know what the difference is
06:06 PM Tom_L: if you ask on linuxcnc they can tell you the different grades they come in
06:06 PM Tom_L: i didn't ask, i knew what i was getting
06:06 PM Tom_L: high precision are ground screws, these are rolled
06:07 PM Tom_L: but they're still pretty good
06:07 PM rue_shop3: well, I think you got them cheaper than the chineese ones too
06:07 PM Tom_L: remember the goal?
06:07 PM Tom_L: better than sherline?
06:07 PM rue_shop3: but I cant, I"m in canada and shipping is always more than product
06:07 PM rue_shop3: yea
06:07 PM rue_shop3: if I use a belt drive, I have to be carefull of the lash in the gearbox
06:10 PM Tom_L: don't use that little stuff they use on repraps
06:10 PM Tom_L: it won't last
06:36 PM rue_shop3: "nor will standard threaded rod as a drive"
06:36 PM rue_shop3: but I got one machine thats been running since 1998 and one thats got, prolly over 600 hrs now, and ... they are still going
06:53 PM [1]MrMobius is now known as MrMobius
08:31 PM rue_shop3: http://ruemohr.org/~ircjunk/art/A4988-5.png
08:32 PM rue_shop3: ^^^ copy that image everywhere, the government dosn't want you to have it!
08:32 PM rue_shop3: :)
09:01 PM [1]MrMobius is now known as MrMobius
09:50 PM _ami_: rue_shop3, whats that?>
09:52 PM rue_mohr: a nice clean line diagram of a A4988 stepper driver module
09:53 PM rue_mohr: the ONLY clean line digram pinout of it ON THE WHOLE FREAKING DAMN INTERNET
09:53 PM rue_mohr: and I just drew it today
09:53 PM rue_mohr: there is one thats close in the google image results, but its about 100x320 pixels, and its a 404
09:54 PM rue_mohr: it only exists in the unobtainium of the google cache
09:54 PM rue_mohr: if google is going to escort you to the site, wTF wont they check if it was a 404 and purge it from the database?
10:06 PM Ameisen: Trying to calculate the lowest latency possible to read from an SD card that is connected via an SPI sd card reader
10:06 PM Ameisen: presuming raw, binary data
10:07 PM Ameisen: I want to use it as an extra data storage medium for processed data
10:07 PM Ameisen: but I need to be able to stream from it fast enough to use the data realtime
10:07 PM tpw_rules: that depends on a random number generator
10:07 PM tpw_rules: but i mean you can stream raw audio
10:08 PM tpw_rules: also i don't think you mean latency, you mean badwidth
10:08 PM tpw_rules: if you have the ram, a double buffer of 512 bytes should cover it
10:24 PM rue_mohr: the problem that I ran into was the trasnfer rate of the cards at really tiny read sizes
10:24 PM rue_mohr: it was slower than the 22khz wave I was trying to play
11:08 PM [1]MrMobius is now known as MrMobius
11:55 PM day__ is now known as day