#robotics | Logs for 2016-02-24

Back
[00:04:33] <MarkX> G is for G Code?
[00:04:34] <MarkX> :P
[00:06:33] <rue_shop3> heh
[00:08:35] <MarkX> H is for hysteresis
[00:21:01] <z64555> T is for Tin-whiskers
[01:06:55] <flyback> rue_shop3, about that btw
[01:07:02] <flyback> newer hd's have pussy magnets
[01:07:05] <flyback> thin and weak
[01:07:17] <flyback> so older hd's you decide to scrap have the most PAINFUL magnets
[01:07:30] <flyback> I have quarter size and thick ones cause instant blood blisters from old scsi hd's
[01:07:42] <rue_house> instant you say
[01:08:03] * rue_house throws flyback two magnets in succession
[01:08:07] * anonnumberanon churtles
[01:08:26] * rue_house throws a 3rd magnet at flyback
[01:09:18] <flyback> BMCC
[01:09:24] <z64555> blood blisters? You sure you're not a little heavy on the hemaglobin?
[01:09:33] * flyback DUKES OF HAZARD JUMPS OVER rue_house
[01:09:49] <flyback> z64555, SOME OF THOSE scsi ones have like 65lbs pull
[01:09:56] <flyback> yuou try to pry one out and they snap back
[01:10:11] <z64555> Nice.
[01:10:46] <flyback> rue_bed, if you like entertaining kids with magnets
[01:10:49] <z64555> so apply an electromagnet to it to reduce its field a bit
[01:10:50] <anonnumberanon> What do I do to use quadcopter motors in water?
[01:10:53] <flyback> nothing beats the copper pipe and a nib
[01:10:58] <flyback> every time I show a kid they go nuts
[01:11:10] <z64555> anonnumberanon: get a submarine
[01:11:34] <anonnumberanon> z64555, re-read my question carefully.
[01:11:48] <z64555> :P I know, I'm just being cheeky
[01:11:55] <z64555> real answer in a bit
[01:12:36] <z64555> Most motors are not sumbersable, so one of the things to do besides water-profing your electronics is to make a housing for the motor that's water tight as well
[01:12:38] <anonnumberanon> I wasn't asking what I can build with it, but obviously how it wouldn't blow up, break, hurt me, hurt people, discolor the water in which it is dipped, cause other damage, make me lose money, etc etc...
[01:13:20] <flyback> why quads in water
[01:13:24] <flyback> quads are a cunt
[01:13:30] <flyback> if you lose a blade it's a brick
[01:13:33] <z64555> motors
[01:13:41] <flyback> use skies if you want to land a quad in water
[01:14:07] <z64555> The housing of the motor must have some heat sink piping, though
[01:14:11] <anonnumberanon> my cousin wants me to build his quad submersible and I've seen it done on Youtube somewhere so I said I'll do it.
[01:14:22] <flyback> ah
[01:14:34] <flyback> i'd go bldc
[01:14:45] <flyback> advantage is you can get a motor that doesn't need seals
[01:15:00] <flyback> the coil is sealed and the magnet and shaft are what's in the water
[01:15:09] <flyback> like better aquarium power head pumps
[01:15:12] <z64555> Yeah... but I have high doubts of the durability of the coils
[01:15:15] <Triffid_Hunter_> anonnumberanon: use computer fans for water, even if you soak the stator in epoxy, the aluminium bodies of normal motors won't last long in water
[01:15:19] <flyback> why
[01:15:25] <flyback> the coil is perm potted and sealed
[01:15:34] <anonnumberanon> https://www.youtube.com/watch?v=FC9EJhs0pc0
[01:16:01] <z64555> sealed with what?
[01:16:01] <flyback> the field goes thru the casing
[01:16:09] <flyback> instead of a stupid seal that wears out
[01:16:35] <Triffid_Hunter_> z64555: epoxy
[01:16:37] <flyback> glue, epoxy
[01:16:40] <flyback> the coil is fixed
[01:17:04] <anonnumberanon> this video shows motors that are probably the same than standard bldc, it's just the electronics that are all amphibianized and stuff
[01:17:09] <flyback> go bldc or switched reluctance
[01:17:11] <z64555> hm, not on the one's I've seen
[01:17:31] <flyback> lookup aquarium powerhead pump
[01:17:33] <z64555> the only thing protecting the coils are their insulation
[01:17:36] <flyback> look at the impeller
[01:17:43] <flyback> I can't open browser doing disk main
[01:19:04] <anonnumberanon> it looks like a regular quad that just had all its electronics encased. It actas as a submarine and can take off from the surface.
[01:19:31] <flyback> z64555, https://www.google.nl/search?q=aquarium+powerhead+impeller&client=browser-ubuntu&hs=bDq&channel=fe&hl=en&source=lnms&tbm=isch&sa=X&ved=0ahUKEwjcg8bH6I_LAhWHlx4KHfs3AL0Q_AUIBygB&biw=960&bih=651
[01:19:35] <z64555> pretty much
[01:20:09] <flyback> http://www.petsolutions.com/images/products/15516530c.jpg
[01:20:18] <flyback> the electromagnet is perm sealed
[01:20:24] <flyback> no rotating seals to breech etc
[01:21:34] <flyback> I bet someone has done them
[01:25:22] <z64555> so you're looking for an in-runner BLDC, preferably rated for water
[01:28:58] <z64555> and you might need a gearbox
[01:29:31] <anonnumberanon> It should be just splash-proof.
[01:29:40] <anonnumberanon> z64555, did your quad fly?
[01:29:48] <anonnumberanon> or just drawing phase for now?
[01:30:16] <z64555> Oh it flew
[01:30:22] <z64555> straight into the wall
[01:31:27] <z64555> The mechanical aspects were pretty good, but we had a trouble matching the motor power with battery weights (we were paying out of pocket and were broke college kids)
[01:31:45] <z64555> and some damn fool insisted on assembler until the last day
[01:35:39] <anonnumberanon> what a faggot
[01:36:06] <anonnumberanon> man i had to lead a team of 6 and 4 were chinese and never did anything to make the project go
[01:36:10] <z64555> bleh. At least I got to keep the thing
[01:36:40] <anonnumberanon> only one dude was really good with computers and learning so i knew C and taught him C but he learned C++ really and did some pretty cool things for the project
[01:36:55] <z64555> cheers
[01:37:01] <anonnumberanon> we made it fly barely (with PID stabilization) at like the day it was due
[01:37:43] <z64555> lol.
[01:38:11] <z64555> I need to pick up a helicopter transmitter and a reciever for it
[01:38:17] <anonnumberanon> that's what got me into embedded
[01:38:25] <z64555> so at the very least I can fly it manually
[01:38:41] <anonnumberanon> does it stabilize on a stand? do that first
[01:39:30] <z64555> Yeah, it tried to. But the stand I had for it put the pivot on the bottom
[01:39:39] <z64555> I need to set up a fork stand
[01:39:39] <anonnumberanon> lol
[01:40:02] <z64555> and use its arms as a pivot
[01:40:38] <anonnumberanon> watch my videos for some ideas, about $40 of wood from Home Depot to build it, and rods at the end of each arm, under the motors, that fit into the stand
[01:40:54] <anonnumberanon> https://www.youtube.com/watch?v=01HExZuUz9s
[01:41:43] <z64555> heh, I got a video of it in action
[01:41:57] <z64555> Yeah, mine's unique
[01:41:58] <anonnumberanon> show!
[01:42:09] <z64555> it doesn't use square tubing for its arms, but round bar stock
[01:42:34] <anonnumberanon> what did you use for brain?
[01:42:54] <z64555> https://www.facebook.com/allen.babb.1/videos/200727363326050/?l=4186910995976629021
[01:42:59] <z64555> MiniDragon
[01:43:13] <z64555> Motorola HC9S12
[01:43:27] <anonnumberanon> It's considered "successful" due to the fact that the craft is not trying to act like a seesaw of death...
[01:43:29] <anonnumberanon> lol
[01:43:54] <anonnumberanon> ill add you okay?
[01:44:02] <z64555> sure
[01:44:29] <z64555> Valentin?
[01:44:34] <anonnumberanon> yeah
[01:44:44] <z64555> k. added
[01:45:22] <z64555> I don't have many fun things on my facebook page yet. That'll change soon, hopefully. :)
[01:46:01] <anonnumberanon> ah, A&M
[01:49:17] <z64555> Kingsville is a good campus. My only wish while I was there was that I was more serious about studying
[01:49:59] <anonnumberanon> You still got time, if only while waiting until you get hired somewhere.
[01:52:57] <anonnumberanon> Bachelor is like a survey degree. Teaches you to work hard for grades though and that's a good start, then you need to become specialized in something you like so that you can tell people you're good at "X" and be able to back it up.
[01:53:08] <z64555> :)
[01:57:08] <anonnumberanon> why did you use Motorola HC9S12 ?
[01:57:56] <z64555> It was what we had on-hand, and we had experiance working with it
[01:58:26] <z64555> We also didn't know what the uC market looked like, much less how to show for stuff
[01:58:43] <anonnumberanon> ah did you guys learn assembly with that platform?
[01:59:06] <z64555> ArduCopter was *just* starting out at the time, and its price range was outside what we could afford, IIRC
[01:59:14] <z64555> and yes, that's what we learned assembler on
[01:59:33] <anonnumberanon> yeah or "ardupilot"
[01:59:46] <anonnumberanon> when did you graduate?
[02:00:25] <z64555> 2011? 2012? in was in the winter
[02:00:53] <anonnumberanon> ah it's been a while, did you get a job or still looking?
[02:01:07] <anonnumberanon> I've been looking since May.
[02:01:27] <z64555> not in my field, I'm holding off on that until I can get something other than FE as a qualification
[02:01:46] <z64555> Most of the opening's I've seen for Junior and Senior level engineers.
[02:02:25] <z64555> homefront issues haven't been helpful. :/
[02:03:01] <anonnumberanon> thing i wish they would have taught us in engineering is how to flip burgers
[02:04:10] <z64555> heh.
[02:04:46] <z64555> burgers are easy to cook, they just take a lot of time
[02:05:08] <z64555> and *everybody* has their own opinion on how they're supposed to be
[02:05:35] <anonnumberanon> hahaha
[02:06:09] <anonnumberanon> i bet you can optimize the "how they're supposed to be" so that customers don't complain much
[02:07:26] <z64555> Yeah. give them more booze. :P
[02:09:41] <anonnumberanon> give them religion, i hear it's the opium of the people, ain't opium better than burgers?
[02:10:20] <z64555> i dunno
[02:12:01] <z64555> Well, like I said before, I scored the quad as a... condolence prize, I guess
[02:12:55] <z64555> So once I finally decide to get back to working on it, I'll be using it as a job qualifier
[02:13:34] <anonnumberanon> You can't work on things right now?
[02:13:43] <anonnumberanon> It's great for forgetting about this shitty world
[02:14:34] <z64555> eh, I don't have lab space for it. As you might've seen some of my posts, I'm in the process of de-junking my smallish room
[02:15:20] <z64555> I do have a spacious front yard full of brush, however :)
[02:15:46] <anonnumberanon> Meh, that's good space to build a workshop !
[02:19:12] <z64555> lol
[02:38:14] <z64555> dangit, I lost my focus on my programming project
[02:38:18] <z64555> lol.
[02:39:35] <anonnumberanon> what was it about?
[02:42:25] <z64555> Some initial stages of a controls input overhaul, I was organizing and doccumenting stuff
[02:42:35] <z64555> This is for FreeSpace2 Open
[02:43:47] <deshipu> https://www.youtube.com/watch?v=rVlhMGQgDkY
[02:43:52] <deshipu> Boston Dynamics
[02:44:05] <z64555> yep
[02:45:09] <z64555> anonnumberanon here beat ya to it, deshipu
[02:45:09] <deshipu> I can't believe they are still at it
[02:45:12] <anonnumberanon> this thing has incredibly torquey hip motors
[02:46:13] <z64555> a lot of folk are interested in bipedal motion, not to mention machine learning
[09:25:24] <rue_house> I hate the way boston dynamics walkers cant mix in static balance, they have to dance
[10:25:47] <deshipu> actually spot can walk statically stable gaits
[10:26:26] <deshipu> and it seems to me that in that latest video Atlas doesn't dance
[10:27:38] <deshipu> while handling those boxes
[10:54:28] <SpeedEvil> http://syruptrap.ca/2016/02/haggard-canadarm-1-working-assembly-line-job-at-ford-factory/
[10:55:24] <deshipu> poor little fella
[10:56:01] <deshipu> the photo seems somewhat... um... electronically enchanced
[10:58:26] <SpeedEvil> well - yes - canadarm can't work in 1g
[10:58:43] <myself> but... that's the body seam line, turn signals aren't installed there.
[10:59:10] <myself> I don't think that's oakville anyway, ceiling is too low.
[11:00:46] <SorcererX> http://imgur.com/GtCB9hK
[11:19:36] <myself> That's shocking, they hand-wired that instead of turning the stick around and having a little board for the connectors?
[11:22:19] <SorcererX> if labour is cheap enough, I guess that is what happens
[11:27:47] <myself> or the scam is impromptu enough
[11:28:56] <Snert_> so the bolts are just to add weight?
[11:29:15] <SorcererX> looks like it, yeah
[11:29:22] <doomlord> lol
[11:29:26] <Snert_> that's damned ballsy
[11:29:33] <Snert_> I bet it's a hoax
[11:30:00] <SorcererX> nah, I've had friends that've gotten similar stuff
[11:30:27] <SorcererX> I've seen PSUs where they've added scrap metal inside to add weight instead of putting in real components
[11:30:45] <doomlord> those new boston dynamics videos are rather impressive
[11:30:48] <myself> yeah, I've personally gotten some of those
[11:34:09] <Snert_> luckily I've never gotten any bogus goods out of china yet
[11:34:25] <Snert_> well, not anymore than the usual bogus.
[11:35:21] <myself> my new abandon thread gif: https://www.youtube.com/watch?v=rVlhMGQgDkY#t=2m27s
[11:44:14] <Snert_> http://bostondynamics.com/img/SandFlea%20Datasheet%20v1_0.pdf
[11:51:57] <MarkX> afternoon
[11:52:42] <doomlord> is motor/artificial muscle tech improving?
[11:53:39] <Snert_> if ali would deliver my miniature gear motors with encoders I'd tell you :)
[15:10:15] <MarkX> so going back to this code i found >> https://github.com/WRidder/MotionProfileGenerator/blob/master/cpp/MotionProfile.cpp << taking a look at the method on line 118, we start to do the actual math to see if we need to accelerate, de-accelerate or maintain our max velocity
[15:12:23] <MarkX> at line 146, 151 and 154 we calculate our current position based on what we've done with the velocity over time.
[15:13:03] <MarkX> since i have an actual position being calculated using the potentiometer and the math provided by anniepoo_ , there is no need for current position calculations provided right?
[15:15:14] * z64555 tries to make a mental simulation
[15:16:35] <anniepoo_> howdy
[15:17:14] <MarkX> hey anniepoo_
[15:17:43] <z64555> hm, I need more coffee
[15:18:38] <anniepoo_> 8cD
[15:18:47] <anniepoo_> 1 min, checking on laser
[15:22:20] <SpeedEvil> LASER
[15:22:29] <MarkX> dang being called out. take care everyone
[15:23:33] <vicarion> conversations on here can be a bit sporadic. are there other engineering/tech related channels you guys talk on when you're not talking here?
[15:29:12] <z64555> MarkX: when you get back, it doesn't appear you have any sort of output for that class.
[15:30:00] <z64555> well, there's ::update, but I think at the moment the class is more of a simulation than a controller
[15:34:38] <anniepoo_> yup LASER
[15:35:18] <anniepoo_> RECI 6090 90watt CO2 laser
[15:35:20] <anniepoo_> 8cD
[15:35:58] <anniepoo_> it'll go through 1/2" of plywood if you push it
[15:36:00] <anniepoo_> 8cD
[15:43:01] <SpeedEvil> All LASERs will go through 1/2" plywood
[15:43:06] <SpeedEvil> (at ~100m/s)
[16:18:30] <deshipu> more on Atlas: http://spectrum.ieee.org/automaton/robotics/humanoids/boston-dynamics-marc-raibert-on-nextgen-atlas
[16:18:47] <deshipu> vicarion: sure, all of them
[16:19:24] <deshipu> vicarion: #arduino, ##hackaday, #esp8266, #raspberrypi, etc.
[16:20:34] <SpeedEvil> ##electronics ##mechanics ##engineering #linuxcnc #mutirotor #multiwii #highaltitude
[16:25:38] <deshipu> mutilator ;)
[16:38:21] <akem> they should sell intel inside to the japanese
[16:40:42] <akem> what my friend was telling me and i agree, anyway...
[16:41:11] <vicarion> deshipu: thanks man. can you expand on #etc?
[16:45:31] <z64555> lol, not #etc
[16:45:43] <z64555> etc. as in "many others that I don't remember at the moment"
[16:45:49] <vicarion> i meant it #metaphorically
[16:46:00] <z64555> #orly
[16:47:17] <vicarion> but thanks for suggestions - i need to learn a lot more about mechanics and engineering
[16:48:01] <vicarion> like, why are the hip joints on that boston dynamics biped set so far back?
[16:48:55] <z64555> might have something to do with the center of mass of the torso
[16:50:26] <vicarion> i'm pretty sure in humans the neck, hips, knees and ankles all line up in a neutral standing stance. i don't know how it is for other bipeds
[16:55:51] <z64555> try weaing a backpack full of physics books
[16:56:07] <z64555> or maybe my calculus book, that was a nice thick one
[16:57:12] <vicarion> you'd lean forwards, right? but would you pivot forwards at the hips, or would the forwards leaning start lower down?
[16:58:42] <vicarion> i notice seriously fat people tend to hunch forwards - i presume that's partly to counter the extra weight in their abdomen
[17:00:57] <z64555> no, that's just bad posture
[17:01:30] <z64555> at least I think it is
[17:01:42] <z64555> But going back to the backpack, you'd lean forwards at the hip
[17:02:16] <z64555> This is to distribute the center of mass of your torso over the pivot point of the hips
[17:02:36] <z64555> Alternatively
[17:03:15] <z64555> You could lean a little bit more forwards, and have the CoM over the center of your feet
[17:03:19] <z64555> bending the legs
[17:05:09] <z64555> Moving the hip rearwards has the same effect as us leaning forwards
[17:05:31] * z64555 's back aches at the thought
[17:08:40] <vicarion> except it's permanent, rather than to just compensate for the robot carrying something
[17:11:04] <z64555> ok you lost me
[17:16:14] <vicarion> i mean, why not design the robot so its torso mass is symmetrical, and put the hip in the middle?
[17:17:55] <z64555> I dunno then
[17:29:29] * z64555 is unhelpful
[17:53:24] <flyback> http://www.networkworld.com/article/3037088/researchers-make-low-power-wi-fi-breakthrough.html
[17:57:49] <z64555> so they centralized the signal generator into a separate device
[17:58:54] <z64555> hm, although. magazine articles have gotten things wrong or otherwise misworded
[17:58:59] * z64555 looks for sources
[18:02:17] <Snert_> yea, it wasn't too clear.
[18:02:45] <z64555> http://passivewifi.cs.washington.edu/#abstract
[18:02:51] <z64555> Video there makes it crystal clear
[18:02:52] <Snert_> xmttr power is xmttr power, seperating the analog from the digital ?
[18:03:03] <z64555> It's a parastic device
[18:03:26] <z64555> You have one wifi device that generates a carrier, and the parasitic devices modulate the carrier
[18:04:09] <z64555> You know how a stop light works?
[18:04:29] <z64555> or a traffic signal. You have a light bulb per color
[18:04:37] <z64555> That's like the wifi devices now
[18:04:55] <Snert_> ok. same amount of power would be used no matter which light is lit
[18:04:58] <z64555> With the parasitic wifi, You have like a shutter
[18:05:06] <z64555> with the lightbulb always on
[18:06:01] <z64555> So, you'd reduce possibly 2 or 3 lightbulbs to just one, but still have 3 signals
[18:06:20] <z64555> not a perfect analogy, sorry
[18:12:33] <Snert_> hmmm.
[18:12:44] <Snert_> less gadgets everywhere.
[18:13:09] <Snert_> they'd have to commercialize somehow.
[18:13:45] <z64555> Nah, it's the old Smart House they keep trying to cash in on
[18:15:34] <Snert_> wallwarts, dongles, doHickeys ....all that.
[18:15:52] <Snert_> I'd have to not see it or know it was there.
[18:16:55] <Snert_> and if it doesn't have a potential value to the next home buyer I may sell my house to then it's less of what I'd want
[18:18:25] <Snert_> be that as it may.... backscatter communication is not unheard of. They just applied it to 802.11 I suppose.
[18:18:39] <z64555> And made a chip
[18:19:00] <z64555> that actually works. Lol
[18:19:21] <SpeedEvil> The beacon is 160kbits outwards, and the nodes can 'send' normal packets.
[18:19:56] <SpeedEvil> The beacon sender recieves packets 'for' the nodes, and transcodes to this 160kbits/s
[18:20:03] <SpeedEvil> for when they need inbound
[18:21:41] <Snert_> I think that's all worth it and all, but I have no problem with battery life on my phone and plugging it in daily to charge.
[18:22:49] <z64555> Nah, this is targeted for IoT
[18:23:13] <SpeedEvil> yeah -it's irrelevant to phones
[18:23:15] <z64555> Might see it in some cheaper phones
[18:23:23] <SpeedEvil> other than maybe body-networking
[18:23:50] <SpeedEvil> RFID, for example.
[18:24:32] <SpeedEvil> Or stuff like battery operated weather sensors
[18:24:32] <Snert_> and not all new and wonderful gadgets make it into the marketplace
[18:24:38] <z64555> RFID, yes, body-networking bluetooth would be better
[18:24:49] <SpeedEvil> It may be cheaper than bluetooth
[18:25:29] <SpeedEvil> Losing most of the RF stages helps
[18:25:36] <z64555> Great thing about bluetooth is that it's more difficult to snoop on the transmissions
[18:25:52] <SpeedEvil> To a degree, yes
[18:26:03] <z64555> Also, if you're in a noisy environment, the bluetooth can select a less noisey band
[18:26:18] <SpeedEvil> the beacon would pick the band
[18:27:28] <z64555> hm.
[18:27:56] <z64555> Something I just thought of, what's to prevent the wifi devices to go into backscatter mode?
[18:28:04] <SpeedEvil> lack of beacon
[18:28:11] <SpeedEvil> if you mean conventional wifi
[18:29:09] <z64555> so have one act as the beacon when not transmitting
[18:29:20] <SpeedEvil> Implementing it in - say - ESP8266 might be nice.
[18:29:36] <SpeedEvil> It could signal in a low power way over the 160khz channel wakeups
[18:30:04] <SpeedEvil> And if the signal is adequate transmit passively
[19:29:24] <z64555> alright. I'm doing it
[19:29:52] <z64555> I wonder if I have any of my source code for the quadrotor around
[19:33:24] <z64555> Assuming I could *find* it
[19:36:19] <z64555> blah
[19:36:27] * z64555 writes down everythign from memory
[20:06:32] <z64555> Anybody have a recommendation for a fixed-point maths library?
[20:15:30] <ace4016> hrm, it's not terrible hard to make yourself, though i guess if someone else made one, it's better not to reuse...
[20:19:41] <z64555> true, but I'd like to have fewer things I have to worry about going horribly wrong. :)
[20:20:33] <z64555> I've already done a few fixed point implementations myself
[20:23:05] <z64555> its, uh... basically just int16_t[2] for a fixed32_t
[20:24:10] <z64555> not too terrible to do in C__
[20:24:13] <z64555> *C++
[21:38:33] <MarkX> hey z64555 sorry for the long delay
[21:38:47] <z64555> no worries
[21:39:12] <MarkX> ya it looks like that code is not so good
[21:39:19] <MarkX> i think i will make my own from scratch
[22:05:01] <z64555> hope I don't get too distracted by implementing the fixed point API...
[22:05:27] <MarkX> just stay off reddit :P
[22:16:31] <z64555> well, aside from the multiply and divide functions, I want a few functions to convert to like float32, etc.
[22:17:04] <z64555> now... having a seperate FPU is possible, so I could have true floating point support
[22:17:44] <z64555> heh. or I could just use the uC that has an FPU to begin with
[22:17:59] <z64555> that'd probably be easier for me for now
[22:18:03] <z64555> so Imma do that
[22:33:18] <z64555> but first, my spanish lessions for tonight
[22:43:19] <z64555> done
[22:47:23] <z64555> ok, got my PID module coded. decided to stick with floats for the time being
[22:47:59] <z64555> That means I can't use the 9S12, and have to learn how to program the LM4F120
[22:49:52] <z64555> hm.
[22:50:04] <z64555> next up would be to code the AHRS
[22:52:14] <z64555> or maybe on the engine controllers...
[23:49:13] <anonnumberanon> z64555, why use such exotic microcontrollers, what challenge are you trying to pass that would require better than a cheap 8 bit avr?
[23:49:52] <z64555> uh, quadrotor?
[23:50:10] <z64555> With some waypoint guidance
[23:51:48] <z64555> If it was just manually controled, I'd stick with a simpler uC
[23:51:55] <anonnumberanon> ah okay
[23:52:12] <anonnumberanon> Didn't know you were adding waypoints.
[23:53:01] <anonnumberanon> Just making some simple graphics 3D here with SDL. Trying to make my array of pixels for 3D objects. I can do one in a format but then to display it I have to convert that format maybe.
[23:53:37] <z64555> array of pixels? or voxels?
[23:54:17] <anonnumberanon> I don't know what voxels are.
[23:54:49] <z64555> voxels are a graphical element with at least position and color
[23:55:14] <z64555> pixels are what's on your computer screen
[23:55:34] <z64555> and then there's texels, which are color
[23:55:39] <anonnumberanon> Well if you have an array you have positions.
[23:55:42] <z64555> A texture is made up of texels
[23:56:04] <z64555> hm or rather, a map is made up of texels
[23:56:18] <z64555> nah, like X,Y,Z coordinates
[23:56:31] <anonnumberanon> ah yeah then that's different, more data
[23:56:52] <anonnumberanon> I think i have texels then.
[23:58:07] <z64555> SDL 2.0 uses OpenGL, so, I'd look into their data types
[23:58:46] <z64555> https://wiki.libsdl.org/CategoryPixels
[23:59:08] <z64555> although, it seems they have their own API for some simple graphics
[23:59:13] <anonnumberanon> I need to understand my ray tracer's format first.
[23:59:22] <anonnumberanon> To see what it corresponds to in SDL.