Back
[01:08:59] <rue_house> most of my thoughts revolve on wishing I had time to sit down and think about it
[01:09:29] <rue_house> most of my thoughts revolve on wishing I had time to sit down and work on it, just like the petrified pommigranite on my kitchen counter
[01:14:28] <Jak_o_Shadows> haha.
[01:18:31] <rue_house> well, gee, with the canadian dollar becomming worthless, I'm not inclined to buy ANYTHING
[01:18:59] <rue_house> if its gonna keep dropping I should get a bitcoin wallet made up tho
[01:20:38] <Jak_o_Shadows> Yeah, the AUD isn't very good
[01:21:04] <rue_house> you have bitcoin?
[01:23:08] <Jak_o_Shadows> Nah
[03:52:18] <hypodyne> howdy
[09:29:44] <rue_house> morning
[09:30:54] <jhylands> indeed it is
[17:52:09] <robotustra> https://scontent-yyz1-1.xx.fbcdn.net/hphotos-xlf1/v/t1.0-9/1495535_10151950926227621_95143154_n.jpg?oh=b78916e4ade36ffcc599b8588c15c8a9&oe=57100E3A
[18:32:27] <Snert_> Snowhenge
[18:35:56] <veverak> qnice
[18:36:17] <veverak> ROS handling raspi cam got much smaller delay than rpi and mjpg
[18:36:20] <veverak> just what I expected :)
[18:41:24] <jhylands> how big of a delay do you get with rpi and mjpg?
[18:41:43] <veverak> so small you can't tell how big it is
[18:42:08] <jhylands> yeah, that was my experience too
[18:42:18] <jhylands> so how did ROS get much smaller than that?
[18:43:05] * veverak usually got ~2s with mjpeg
[18:43:09] <veverak> or really notcieable
[18:43:36] <veverak> (setup is that it streams from rpi over wifi to my laptop)
[18:59:42] <jhylands> yeah, I use mjpg-streamer to do that with a couple 720p webcams
[18:59:57] <jhylands> with no noticable lag
[21:12:53] <wildmage> hey
[21:34:00] <rue_shop3> jhylands, 3 seconds?
[21:34:07] <rue_shop3> if you want no lag you have to use analog
[21:34:13] <jhylands> no you don't
[21:34:23] <jhylands> if you want no lag don't use motion
[21:34:26] <rue_shop3> cause apparenlty EVERYONE who writes video code buffers atleast one frame
[21:34:39] <jhylands> well, no noticeable lag
[21:35:14] <rue_shop3> I could not get anything over ethernet to less than 3 seocnds
[21:35:26] <jhylands> you didn't try hard enough :-)
[21:36:51] <jhylands> I've got it set up to stream 2 usb webcams over wifi, 720p, with lag measured in low 100-ms
[21:37:08] <rue_shop3> well, all I have to do is assemble the output filter of my power converter and I can test it
[21:37:19] <rue_shop3> jhylands, wow
[21:37:30] <jhylands> this is the key:
[21:37:31] <jhylands> https://github.com/jacksonliam/mjpg-streamer/archive/master.zip
[21:37:38] <rue_shop3> hmm
[21:37:44] <rue_shop3> I"ll play later
[21:38:03] <jhylands> here's my build instructions:
https://pastebin.mozilla.org/8856615
[21:38:35] <jhylands> and here's the real magic:
[21:39:10] <jhylands> https://pastebin.mozilla.org/8856616
[21:39:19] <jhylands> (the last line)
[21:52:53] <rue_shop3> I need to get on writing software for my power supply
[21:53:11] <rue_shop3> hmm, need brain food
[22:33:18] <wildmage> anything cool?
[22:34:49] <rue_shop3> hahaha
[22:35:06] <rue_shop3> the stage 1 flter of my power supply has a resonance of 222Khz
[22:35:16] <rue_shop3> the smps runs at 277.5Khz
[22:35:22] <rue_shop3> er, 227.5
[22:35:24] <rue_shop3> anyhow
[22:35:30] <rue_shop3> er, 227.5khz
[22:35:35] <rue_shop3> *smirk*
[22:35:44] <rue_shop3> wildmage, what you been up to?
[22:46:56] <wildmage> finished school
[22:47:02] <wildmage> now trying to find my role in the work force
[22:47:12] <wildmage> rue_shop3, you have to call me Dr. Wildmage now
[22:47:14] <rue_house> heh
[22:47:24] <rue_house> thought you finished a long time ago
[22:47:29] <wildmage> yeah, i should have
[22:47:36] <robotustra> wildmage: you HAVE TO MAKE MONEY
[22:47:38] <wildmage> i had a hard time completing my results
[22:47:52] <wildmage> robotustra, or I can go back to school! i'm really good at it
[22:47:58] <wildmage> :P
[22:48:16] <rue_house> did you steal tech from a university you can make a spinoff company with?
[22:48:26] <wildmage> sure, there are lots of things I can do
[22:48:32] <wildmage> i don't trust my work ethic yet to pull it off
[22:48:36] <wildmage> so I'm working on that first
[22:48:56] <wildmage> hence why i'm on IRC :)
[22:49:03] <robotustra> wildmage: show me your paper
[22:49:45] <wildmage> it's kind of bad
[22:49:50] <wildmage> but it is 200 pages
[22:49:55] <robotustra> and what?
[22:50:06] <wildmage> i submitted a paper to a conference and i'll know if it gets accepted this friday
[22:50:24] <robotustra> doctor doctor gimme a... paper
[22:50:37] <wildmage> http://digitallibrary.usc.edu/cdm/ref/collection/p15799coll3/id/566721
[22:51:10] <wildmage> developing a touch based mapping approach from the ground up was no picnic
[22:51:11] <z64555> boop
[22:51:42] <z64555> was it a fancy dinner?
[22:51:52] <robotustra> I mean where is your PhD thesises?
[22:52:02] <wildmage> z64555, it was more like a liquid diet, giving you the shits every night
[22:52:13] <wildmage> robotustra, i just gave you the link
[22:52:15] <z64555> hmmm, yes.
[22:52:24] <robotustra> ah ok
[22:53:24] <robotustra> wildmage: you know, if you where in USSR you would cut a half of your pages
[22:53:42] <z64555> oh neat, a snake bot
[22:53:43] <wildmage> what does that mean?
[22:54:08] <wildmage> z64555, yes, but it's not a snake robot dissertation. the snake robot scientists always criticize my papers because of that
[22:54:20] <robotustra> it means that we don't put aknowlegement to the PhD papers
[22:54:29] <wildmage> the snake robot is kind of impractical, but the general approach is sound
[22:55:08] <z64555> gotcha
[22:55:10] <wildmage> robotustra, ahh, yes, that's kind of a tradition in every western disseration
[22:55:18] <wildmage> well, US dissertationos
[22:55:29] <rue_house> wildmage, "titanoboa"
[22:55:36] <rue_house> seen it?
[22:56:04] <wildmage> rue_house, i believe they're extinct
[22:56:33] <wildmage> robotustra, did you study in USSR?
[22:56:44] <rue_house> heh, the robot
[22:56:51] <rue_house> I'v seen it, its getting big
[22:57:26] <rue_house> https://www.youtube.com/watch?v=7xUFHzE7EdU
[22:57:46] <robotustra> wildmage: yeah, I did my PhD in Belarus
[22:57:52] <robotustra> former USSR
[22:58:10] <robotustra> it was computational physics
[22:58:35] <wildmage> ahhh, I've worked with belarussians before
[22:58:42] <wildmage> very competent
[22:58:58] <robotustra> "Now you are a doctor, not that doctor who can help people" (C)
[22:59:09] <wildmage> It's okay everyone, I'm a doctor
[23:00:12] <rue_house> every one gonna get scared off when you pick up a drill tho
[23:00:52] <robotustra> too many wavy lines in the paper, assert
[23:01:10] <rue_house> wildmage, the switching converter runs at 227.5Khz,from a 34V source, the target max output ripple is 5mv, design an output LC fitler
[23:01:36] <z64555> not even an R?
[23:01:53] <rue_house> R is from 0 to infinity
[23:02:23] <wildmage> i haven't done circuits since 1999
[23:02:24] <rue_house> or, if you like 0.25 ohms to 500kohms
[23:02:34] <rue_house> good, it'll be a good refresher
[23:02:40] <robotustra> wildmage: is it theoretical paper only?
[23:03:16] <wildmage> robotustra, yes, all algorithms. we did a lot of experiments in simulation. Current work now is to make it more solid, give it a broader treatment so we can turn it into a journal article
[23:03:42] <robotustra> but not into real robot?
[23:03:55] <wildmage> well, I think that is very doable
[23:04:01] <wildmage> i just need a very simple robot
[23:04:10] <wildmage> with a couple arms on a wheeled base
[23:04:18] <robotustra> on the base of arduino of cause
[23:04:25] <wildmage> of course
[23:04:44] <wildmage> although there is a lot of computation so it should be either cloud-based or running a powerful onboard embedded processor
[23:05:02] <robotustra> like this?
https://www.youtube.com/watch?v=6LS5nyMGA6U
[23:05:05] <wildmage> running the computation remotely would probalby be the easiest
[23:05:46] <wildmage> robotustra, although I wouldn't use a snake. That adds too many complications, and like I said, the robot snake scientists will hate it
[23:06:05] <wildmage> i want to avoid robot snakes if at all possible since that's not my main contribution
[23:06:33] <robotustra> https://www.youtube.com/watch?v=YKlCzcH8oWk
[23:06:36] <wildmage> robotustra, i don't think that robot gives feedback. I will need to provide the sensor feedback back to the base station
[23:06:42] <robotustra> so you are into robot worms?
[23:07:02] <wildmage> actually, any articulated robot will do for my approach
[23:07:16] <wildmage> if it has joints with joint sensors, my algorithms will work
[23:07:29] <robotustra> ok
[23:07:38] <wildmage> just a stick on a motor will be good enough
[23:07:45] <wildmage> although it will not sense very much
[23:07:48] <wildmage> the more joints the better
[23:07:50] <robotustra> I'm trying to avoid sensors in joints
[23:08:10] <wildmage> robotustra, how do you control them then?
[23:08:28] <wildmage> servos? they have a potentiometer built-in
[23:08:32] <z64555> yup
[23:08:36] <wildmage> you can add wires to get the feedback
[23:08:42] <robotustra> I'm planning to have current feedback from any motor
[23:08:58] <robotustra> I'm not using servos they are too inefficient
[23:09:04] <wildmage> that's a good approach... although it limits the time of locomotions you can do
[23:09:05] <robotustra> just current sense
[23:09:11] <wildmage> s/time/type/
[23:09:25] <robotustra> yeah, wheels
[23:09:39] <wildmage> current sensing can also be used for detecting collisions
[23:09:51] <robotustra> the robot will be
http://robotustra.ca/my-robot
[23:10:16] <robotustra> I'll use it to prevent burning of circuits
[23:10:36] <robotustra> if one of motors will go crazy
[23:10:51] <robotustra> I will just shut down it :)
[23:11:59] <robotustra> here what's done
http://robotustra/my-robot-done
[23:12:52] <robotustra> arms will have a lot of joints, but the toplogy is a bit different from linear
[23:13:07] <wildmage> robotustra, very nice
[23:13:55] <robotustra> we can consider humaoid like arm as a snake :)
[23:14:02] <wildmage> yes
[23:14:12] <wildmage> there are also now soft robots
[23:14:23] <wildmage> squishy worms and tentacles
[23:14:30] <wildmage> that are in dire need of a sensing approach
[23:14:36] <wildmage> which I intend to deliver
[23:14:45] <robotustra> ok,
[23:14:46] <wildmage> robotustra, that's a lot of work you've done
[23:15:17] <robotustra> and it's just about 15-20% only
[23:15:33] <wildmage> i like the chameleon eyes
[23:15:47] <robotustra> I mean, it's about 50% of hardware is done
[23:16:08] <robotustra> after I'll finish hardware I think I'll programm it 1-2 years :)
[23:17:03] <robotustra> need to complete everything in between base and torso
[23:17:08] <robotustra> and arms
[23:17:09] <wildmage> https://www.youtube.com/watch?v=Pwymtmj3ouU
[23:17:09] <wildmage> https://www.youtube.com/watch?v=_HABjn7qLSY
[23:17:28] <wildmage> okay, I have to put my kids to bed, be back
[23:17:43] <robotustra> I have to put myself into bed too
[23:17:46] <robotustra> see you
[23:18:35] <wildmage> доброй ночи
[23:18:50] <robotustra> спасибо
[23:18:59] <robotustra> пока
[23:19:36] <wildmage> i never learned cyrrilic typing :P
[23:20:10] <z64555> well I definitely have a bs degree. Lol
[23:21:31] <z64555> sheesh. I need to go over passive circuits again, and take better notes!
[23:23:49] <z64555> oh well, at least I know how to take notes now
[23:24:20] <wildmage> z64555, what did you get your BS in?
[23:27:23] <z64555> take a guess :)
[23:28:48] <z64555> EEEN. I've taken the FE and passed, but it doesn't help that I haven't done much of anything with the knowledge for 4 years or so
[23:29:39] <z64555> GPA wasn't too hot, and the job market was looking for folks to replace their retiring seniors. No place for rookies such as myself.
[23:36:31] <z64555> not my biggest worry, thankfully. I've got other things going on. :)
[23:57:34] <wildmage> z64555, what is an EEEN?
[23:58:15] <wildmage> electrical engineering? Those guys can work literally anywhere
[23:59:56] <z64555> electrical and electronic engineering, they grouped them together at some point
