#linuxcnc | Logs for 2015-03-20

Back
[00:01:14] <furrywolf> woot!
[00:01:17] <furrywolf> zeeshan: still around?
[00:02:35] * furrywolf pokes zeeshan hard
[00:06:11] <furrywolf> zeeshan: http://imgur.com/P0qKL0h
[00:07:04] <furrywolf> bah, see if I help you again.
[00:11:17] <zeeshan> sorry
[00:11:24] <zeeshan> was putting one of the pieces of my bridge together :-)
[00:11:34] <zeeshan> lol what is this
[00:12:06] <furrywolf> something that looks like it could be depth information...
[00:12:20] <zeeshan> howd you get it
[00:13:03] <furrywolf> since the file was a megabyte too large, I tried searching the end of it to see if there might be anything identifiable in the extra data... just a quick loop that chopped off the end of the file and ran "file" on it.
[00:13:24] <furrywolf> most times file found random garbage, like 5884886: foo.foo: Atari 68xxx CPX file (version 0058)
[00:13:43] <furrywolf> but, one of them looked very promising... 5885255: foo.foo: PC bitmap, Windows 3.x format, 400 x 300 x 32
[00:14:03] <furrywolf> cut out that part of the file, converted to png.
[00:15:36] <furrywolf> it looks like their files are a bunch of other things concatenated together...
[00:16:09] <zeeshan> hm
[00:16:14] <zeeshan> towards the end of the file?
[00:16:27] <furrywolf> yes, after the end of the tiff data
[00:16:39] <zeeshan> i feel like they have a matrix there
[00:16:49] <zeeshan> that defines a color map.
[00:16:51] <zeeshan> for the overlay
[00:17:09] <furrywolf> the image I pasted looks like they're using RGBA to store information that looks much like depth information
[00:17:26] <zeeshan> hm.
[00:18:08] <furrywolf> found more
[00:18:12] <zeeshan> i can't visually correlate that image
[00:18:15] <zeeshan> w/ that color map
[00:19:42] <furrywolf> that's because the color data isn't being used as colors, and 1/4 of the data got tossed in the bitmap-png conversion. it's not actually a bitmap, it's a bitmap being used to store longs, I think.
[00:20:05] <furrywolf> http://imgur.com/DYWfPSk next data block after the bitmap
[00:20:20] <zeeshan> HEY
[00:20:22] <furrywolf> 6365335: foo.foo: JPEG image data, JFIF standard 1.01
[00:20:23] <zeeshan> thats my original image!
[00:20:24] <zeeshan> :P
[00:20:34] <zeeshan> that i put the color map on.
[00:20:45] <zeeshan> and its the high rez version too
[00:20:47] <zeeshan> interesting.
[00:20:53] <furrywolf> well, it's saved in the file too. :P
[00:20:57] <zeeshan> hahaha
[00:21:00] <zeeshan> dude youre the best
[00:21:07] <zeeshan> you need to tell me wtf youre doing!
[00:21:25] <furrywolf> I'm doing brute-force searching for stuff in files using standard unix commands. :P
[00:21:53] <zeeshan> haha
[00:22:06] <furrywolf> for i in `seq 6365255 6366255` ; do tail -c +$i 001a.tif >foo.foo ; echo -n "$i: " ; file foo.foo ; done | grep -v "foo.foo: data"
[00:22:40] <zeeshan> i should hook up my raspberry pi 2
[00:22:43] <zeeshan> so i can run that :p
[00:22:44] <furrywolf> the first number is where I guesstimated the end of the bitmap data was, then for that byte offset, and a thousand ones after it, it chops the file there, then sees if "file" thinks it's something other than random data.
[00:23:33] <furrywolf> file returns lots of random things since it's being fed essentially random garbage, so you have to sift through them to see what looks useful
[00:23:50] <furrywolf> 6365333: foo.foo: DBase 3 index file
[00:23:51] <furrywolf> 6365335: foo.foo: JPEG image data, JFIF standard 1.01
[00:23:51] <furrywolf> 6365345: foo.foo: raw G3 data, byte-padded
[00:23:51] <furrywolf> 6365346: foo.foo: Targa image data - Map 17694 x 27000
[00:23:51] <furrywolf> 6365347: foo.foo: PDP-11 UNIX/RT ldp
[00:23:51] <furrywolf> 6365365: foo.foo: TIFF image data, big-endian
[00:23:51] <furrywolf> 6365371: foo.foo: Applesoft BASIC program data
[00:23:51] <furrywolf> 6365379: foo.foo: TTComp archive data
[00:23:52] <furrywolf> 6365380: foo.foo: VAX-order2 68k Blit mpx/mux executable
[00:23:52] <furrywolf> etc :P
[00:24:10] <zeeshan> yes but when you find that
[00:24:21] <zeeshan> do you just extract that portion of the image?
[00:24:33] <furrywolf> now, one of those looked promising... 6365335: foo.foo: JPEG image data, JFIF standard 1.01. that looks like something it might actually be, while a vax executable seems unlikely.
[00:24:33] <furrywolf> yep
[00:25:00] <furrywolf> and what do you know, it's your image. :)
[00:25:35] <zeeshan> there must be one image
[00:25:38] <zeeshan> whre its just a color map
[00:25:42] <zeeshan> none of the original image in there
[00:25:59] <furrywolf> I think the first one I pasted is the data you're looking for.
[00:26:23] <zeeshan> the zebra lines dont match with the color map though
[00:26:53] <furrywolf> it's not a color map. it's depth information being stored as a bitmap in a non-image fashion
[00:27:17] <zeeshan> depth as in?
[00:27:27] <zeeshan> depth of the image?
[00:27:30] <zeeshan> 8 bit etc?
[00:29:05] <furrywolf> dunno
[00:29:30] <furrywolf> might even be floats...
[00:29:32] <zeeshan> i know it has a specific image with a color map
[00:29:38] <zeeshan> cause ive seen it in the softwarte
[00:29:41] <zeeshan> i just cant save it!
[00:29:56] <zeeshan> it literally is a color map with none of the image shown.
[00:30:06] <furrywolf> I'm poking at the hex now trying to figure out what it means
[00:30:08] <zeeshan> color contour plot
[00:31:07] <furrywolf> 4-byte floats with a 2-byte offset from the start of the file gives something that looks like numbers
[00:31:43] <furrywolf> 000036 0.931202 0.93291354 0.93428344 0.9361171
[00:31:44] <furrywolf> 000046 0.9376292 0.93777454 0.93797505 0.93813
[00:31:44] <furrywolf> etc
[00:31:49] <zeeshan> what address is that
[00:31:56] <zeeshan> 36 is the address?
[00:32:17] <furrywolf> 0x36 from 5885255 in the tiff file
[00:32:44] <zeeshan> 59CD47 address ok
[00:33:03] <furrywolf> the big number is decimal, the small one hex. I'm mixing programs here. :P
[00:34:11] <furrywolf> you'll want to display it as shorts
[00:34:21] <furrywolf> grr, floats
[00:34:22] * furrywolf is tired
[00:34:32] <zeeshan> my hex editor
[00:34:34] <zeeshan> only shows hex! :P
[00:40:33] * furrywolf tries to figure out imagemagick
[00:40:49] <zeeshan> ??!
[00:41:07] <furrywolf> trying to make it do something complicated, and not succeeding
[00:41:58] <zeeshan> fyi
[00:42:18] <zeeshan> im lost :)
[00:43:31] <Jymmm> furrywolf: like?
[00:43:53] <furrywolf> Jymmm: like onvert -size 400x300 -depth 32 -define quantum:format=floating-point gray:foo4.foo foo4.png
[00:44:12] <Jymmm> you mean convert
[00:44:15] <zeeshan> thats that little image
[00:44:20] <zeeshan> with the zebra pattern?
[00:44:28] <furrywolf> that's a copy+paste missed character
[00:44:32] <Jymmm> k
[00:44:51] <Jymmm> so resize and greyscale it
[00:45:00] <Jymmm> ?
[00:45:06] <furrywolf> ok, so you know less than I do.
[00:46:02] <Jymmm> furrywolf: But at least I know where to get the solutions from. You REALLY need to learn to be less condensending.
[00:48:33] <furrywolf> I get no-such-file-or-directory errors, _after_ processing, even with stdin.
[00:48:46] <furrywolf> I'm thinking I have a imagemagick bug.
[00:48:57] <zeeshan> what exactly is
[00:49:01] <zeeshan> "strip" in a tiff?
[00:49:05] <zeeshan> matlab saysL:
[00:49:10] <zeeshan> Number Of Strips: 600
[00:49:44] <zeeshan> http://pastebin.com/RUHy1Vd6
[00:51:59] <furrywolf> seriously. it works with display, but not with convert. wtf?
[00:52:56] <furrywolf> I have a command that works with display. I change "display" to "convert", add an output file to the end, and... it doesn't work.
[00:54:41] <furrywolf> argh... ok, I think I got it. it's reading an infinite number of input files, then complaining it can't find more.
[00:54:43] <zeeshan> ive never used imagemagick before
[00:56:07] <furrywolf> hrmm, got something, but it's not right yet.
[00:56:21] <furrywolf> http://imgur.com/veBDMZN
[00:56:30] <zeeshan> hey!
[00:56:41] <zeeshan> :-)
[00:56:55] <furrywolf> or maybe that is right
[00:57:13] <zeeshan> is that the zebra image
[00:57:14] <zeeshan> converted?
[00:57:22] <furrywolf> looking at your original tiff with colors again, I thought it was smooth, but it actually does have sharp transisitions between levels
[00:57:23] <furrywolf> yes
[00:58:17] <furrywolf> does that look at all like your correct depth map?
[00:58:34] <zeeshan> im trying to compare
[00:58:43] <zeeshan> i think it is
[00:59:03] <zeeshan> yea im pretty sure it is
[00:59:06] <zeeshan> its just missing the color info! :P
[00:59:10] <furrywolf> it could be doing some filtering, and this data is more-raw than what it used for the color map
[00:59:30] <zeeshan> what i dont understand is
[00:59:34] <zeeshan> why is this image so much smaller?
[00:59:38] <furrywolf> ok, so despite the microscope owner's claim, I can now put your depth info in an excel file if you want. :P
[00:59:48] <Cromaglious_> I don't wanna walk across the living room
[00:59:48] <furrywolf> because the microscope's 3d resolution is less than its image resolution
[01:00:00] <zeeshan> so how is it overlaying?
[01:00:04] <zeeshan> it must be interpolating
[01:00:12] <furrywolf> yes, and probably some smoothing too
[01:00:21] <zeeshan> the question is
[01:00:32] <zeeshan> what do those pixels rpresent size wise
[01:00:47] <zeeshan> i can make the assumption that white = 0.00
[01:00:53] <furrywolf> that's probably in some of the other data that's essentially random.
[01:00:58] <zeeshan> and red = black
[01:01:00] <zeeshan> = max height.
[01:01:55] <zeeshan> you know the color map i saw?
[01:01:58] <zeeshan> i wish i had a pic of it.
[01:02:00] <zeeshan> it was huge
[01:06:49] <ssi> https://pbs.twimg.com/media/CAhKh3mUwAAedO_.jpg:large
[01:06:51] <ssi> cabane struts in
[01:06:55] <ssi> and that's all I can possibly do tonight :(
[01:09:53] <furrywolf> bbl, it's waaay past my bedtime, and now I have to be up for work in less than a night's sleep.
[01:10:59] <furrywolf> now you know how your microscope puts its files together. :)
[01:11:05] <zeeshan> hehe
[01:11:07] <zeeshan> thanks man
[01:11:48] <furrywolf> http://filebin.net/1owlk62ew6/foo5.csv
[01:18:42] <Cromaglious_> ssi nice, i see highwing...
[01:37:47] <zeeshan> ssi i cant wait to see this thing built :p
[01:37:57] <zeeshan> http://i.imgur.com/ps9DodR.jpg
[01:38:02] <zeeshan> main section of bridge is done! :P
[01:38:16] <zeeshan> still letting the glue dry
[01:39:01] <Jymmm> zeeshan: bought a kit?
[01:39:09] <zeeshan> no
[01:39:28] <Jymmm> stole a kit?
[01:39:35] <zeeshan> designed and laser cut my own
[01:39:41] <zeeshan> for fun
[01:39:52] <Jymmm> ah. didn't realize you had a laser.
[01:39:57] <zeeshan> i dont
[01:39:59] <zeeshan> school does :)
[01:40:03] <Jymmm> gotcha
[01:40:13] <zeeshan> ive been cutting jobs for students, so i cut my own stuff when im free
[01:40:37] <Jymmm> cool
[01:40:48] <Jymmm> you should have made a useless box =)
[01:40:49] <zeeshan> i want a laser cutter at home!
[01:40:58] <zeeshan> pandora's box
[01:41:16] <Jymmm> https://www.youtube.com/watch?v=aqAUmgE3WyM
[01:42:18] <zeeshan> haha
[01:43:00] <Jymmm> HAHAHAHAHAHAHA https://www.youtube.com/watch?v=UkgoSOSGrx4
[01:43:56] <zeeshan> rofl!
[01:44:22] <Jymmm> Now that's talent!!!
[01:50:44] <Jymmm> zeeshan: I hope you have a hamster or pet rat, then it could be Attack of the giant rat on the brooklyn bridge!
[01:50:57] <zeeshan> my fatty cat
[01:51:00] <zeeshan> was trying to climb it
[01:51:41] <Jymmm> zeeshan: If you want to get rid of some of the black cut marks, use orange waterless hand cleaner
[01:51:55] <zeeshan> Jymmm: for some reason i really like the dark look!
[01:52:01] <zeeshan> it contrasts well
[01:52:20] <zeeshan> but thanks for the tip, ill try it out, cause i know some students want it gone
[01:52:32] <zeeshan> ive been teling them to sand it off with 200 grit
[01:52:34] <Jymmm> If wont eliminate them, just make then dark brown instead of black
[01:52:38] <zeeshan> ah
[01:53:12] <Jymmm> black and beige look like shit. Dk brown and beige are a nice complimentary contrast.
[01:53:45] <zeeshan> looks are always very opinionated :)
[01:53:51] <Jymmm> sanding , yeah, no.
[01:54:05] <zeeshan> sanding restores the color
[01:54:06] <Jymmm> no burnt black looks like shit,
[01:54:18] <zeeshan> i dont think that :)
[01:54:31] <Jymmm> nobody asked you to think
[01:54:38] <Jymmm> ;)
[01:54:40] <zeeshan> =D
[01:55:05] <Jymmm> Now, find me a 1 watt candela bulb
[01:55:15] <zeeshan> why so little
[01:55:25] <Jymmm> I want a glow, not light
[01:55:34] <zeeshan> afraid of the dark?
[01:55:35] <zeeshan> :-)
[01:55:38] <Jymmm> and very low to no heat
[01:56:00] <Jymmm> 7w gets too hot
[01:56:29] <Jymmm> afraid of the dark? heh hardly.
[02:08:09] <Jymmm> LEave it to HP to make their walwarts negative tip just to be different.
[02:26:04] <Cromaglious> they're not the only ones
[02:35:38] <Jymmm> No, but they're what I have a surplus of that makes them useless to me atm.
[03:13:27] <Deejay> moin
[03:45:58] <zeeshan> R: 252, G: 30, B:30 ; R:0 G:41 B:228
[03:46:01] <zeeshan> er
[03:47:46] <archivist> purple
[04:48:42] <XXCoder1> royal purple
[09:41:17] <MacGalempsy> mornin
[09:46:09] <MacGalempsy> after a week of painting, its time to go home and work on the cnc some
[10:35:04] <ssi> MORN
[10:35:23] <archivist> moan
[10:37:19] <pcw_home> mourn
[10:48:19] <Tecan> http://i.imgur.com/FUaJwRn.jpg http://i.imgur.com/rQqDqEM.jpg
[10:59:23] <malcom2073> water wheel?
[11:00:27] <Tecan> pelton wheel and now im printing http://www.thingiverse.com/thing:595380
[11:00:56] <Tecan> not even metal bearings, just add cd's
[11:16:32] <archivist> good luck with those double helix bearings
[11:18:43] <malcom2073> Interesting
[11:25:23] <malcom2073> Tecan: gonna run yours at 80krpm? :P
[11:25:26] <malcom2073> That sounds pretty cool though
[11:25:45] <archivist> 80
[11:26:06] <archivist> I think that K is over optimistic :)
[11:26:13] <malcom2073> heh
[11:26:23] <malcom2073> There's a project for my new lathe
[11:34:21] <Jymmm> pencil sharpener?
[11:35:43] <malcom2073> I should mount a sharpener in the chuck, and make a video of it, just to piss people off
[12:16:59] <monkeyisl> what's mixed decay mode?
[12:17:23] <cradek> in what?
[12:17:38] <monkeyisl> drv8825 stepper driver
[12:18:10] <cradek> the first hit for "drv8825 mixed decay mode" explains it
[12:18:12] <monkeyisl> it has mixed mode.. and slow fast.. but i'm working on small size 3d printer.. .. i don't know what it really means.
[12:18:23] <monkeyisl> got it
[12:19:57] <cradek> well the doc explains what they are, but I don't know how you know which to choose
[12:20:44] <archivist> there is another doc with trade offs
[12:21:18] <archivist> hate google search links TI_35753A67E389B351B9654868_.pdf
[12:22:16] <monkeyisl> tahnksss
[13:37:17] <_methods> https://www.youtube.com/watch?v=jyTA33HQZLA
[13:55:19] <CaptHindsight> _methods: LOL exactly
[13:57:20] <CaptHindsight> anyone work with investment casting materials? Anyone know if they are no longer water soluble after the first baking at ~300F
[13:59:58] <andypugh> I don’t know, but I would guess that they would be technically still soluble until vitrified.
[14:02:54] <zeeshan> http://i.imgur.com/N2MkR7j.jpg
[14:03:02] <zeeshan> cat caught attacking the bridge
[14:03:13] <CaptHindsight> I've been looking over all this lost wax casting using SLA vs wax and the fastest way is just printing the investment casting without wax or polymers
[14:03:38] <andypugh> Not a lot different from Prometal RCT
[14:03:56] <andypugh> (which prints sand for conventional casting)
[14:04:07] <andypugh> How do you bind the investment?
[14:04:16] <CaptHindsight> I have a water soluble polymer for SLA but why bother
[14:04:19] <furrywolf> zeeshan: did you send my csv file to the person who said it couldn't be done? :)
[14:04:27] <zeeshan> furrywolf: no
[14:04:31] <zeeshan> im still trying to go through the image myself
[14:04:41] <zeeshan> pretty sure that was the thumbnail version of the real image
[14:05:03] <zeeshan> i searched for jpg tags
[14:05:07] <zeeshan> and i got the original image
[14:05:14] <CaptHindsight> andypugh: was mainly looking at small detailed parts like jewelry
[14:05:24] <furrywolf> I've accounted for most of the bytes in the file now. I don't think there's any large things like images left...
[14:05:31] <andypugh> So, use sand then a shrinking ray :-)
[14:05:51] <CaptHindsight> I haven't seen a powder printer for casting
[14:06:07] <andypugh> I have been wondering whether simple water and plaster of paris would work
[14:06:22] <zeeshan> so you found out that the main image is 5,760,000 bytes?
[14:06:28] <zeeshan> and jpg looks like its in compressed format
[14:06:29] <archivist> I did some lead castings in plaster
[14:06:30] <zeeshan> so i cant figure out the bytes.
[14:06:35] <furrywolf> you have your colored image, the height data, and the original image... there's probably also metadata for things like units.
[14:06:46] <CaptHindsight> yes, but with some tweaks to powder and binder it could be much higher res
[14:06:51] <zeeshan> furrywolf: that scale bar is also stored in the image file somewhere.
[14:07:01] <zeeshan> i have two versions of the image in tif format
[14:07:13] <zeeshan> color and non color.
[14:07:20] <CaptHindsight> plain Di Water will seep into plaster of paris
[14:07:36] <furrywolf> the main image is something like that size, then you have the 400x300x4+header height data, then you have the original jpeg.
[14:07:40] <zeeshan> and that jpg is stored in the exact same location
[14:07:59] <zeeshan> what type of file was the header+height data?
[14:08:36] <furrywolf> a non-standard windows bitmap, with floating point greyscale data.
[14:08:38] <CaptHindsight> but if you control the surface tension of the liquid binder and the porosity of the plaster then the resolution can be tightly controlled
[14:08:55] <zeeshan> starting with 46 4F 52 4D ?
[14:09:05] <furrywolf> didn't look. lol
[14:09:18] <furrywolf> the floats are in the range of 0 to 1
[14:10:14] <zeeshan> man how do you view floats!!! :P
[14:10:15] <furrywolf> imagemagick couldn't figure out that it was greyscale floats instead of rgba, hence the original zebra-stripe image as you called it. had to strip the header and tell imagemagick to read raw greyscale data.
[14:10:23] <tjtr33> CaptHindsight, long ago i used plaster of paris, poured around wax models. put into pottery kiln to remove wax, poured silver in using centrifugal pour, dropped whole plaster chink into bucket of water, and reahed in and pulled uot the sprued jewelry
[14:10:23] <furrywolf> od
[14:10:39] <furrywolf> od -t f4
[14:10:49] <tjtr33> CaptHindsight, the detail was incredible, i hand finger prints on first try
[14:11:09] <CaptHindsight> tjtr33: yes, i want to skip the wax mold step
[14:11:21] <CaptHindsight> no wax
[14:11:27] <tjtr33> print plaster?
[14:11:29] <CaptHindsight> just print the plaster molds
[14:11:35] <archivist> I had writing on my lead items from the mold, detail was good
[14:11:38] <tjtr33> ah, cleever
[14:11:55] <CaptHindsight> https://www.youtube.com/watch?v=ixuO-4IBRHg better version of this
[14:13:10] <FinboySlick> Hehe, the image processing of the camera is making me feel drunk.
[14:13:15] <CaptHindsight> more like this only with plaster https://youtu.be/LFGnjkoeDng?t=2m36s
[14:13:49] <tjtr33> CaptHindsight, stillwatching 1st, print head spritzes water?
[14:14:28] <CaptHindsight> tjtr33: yes, they used an old thermal inkjet
[14:14:58] <CaptHindsight> I can print several liters per hour at 1200 dpi
[14:15:07] <tjtr33> thats really interesting, JT should looksee, he's into casting now
[14:15:32] <CaptHindsight> yes, made me think it all over again just recently
[14:16:08] <archivist> just rebooting firefox, is that the starch printer
[14:16:39] <CaptHindsight> archivist: plaster
[14:16:49] <CaptHindsight> but any power is possible
[14:16:55] <CaptHindsight> power/powder
[14:17:38] <archivist> I have one of the early 3d ball bearing samples
[14:17:40] <furrywolf> lol. the ebay seller who sent me the 50W light that draws 20W, and didn't package it at all so it arrived smashed... offered me a $3 refund.
[14:17:45] <furrywolf> time to get someone's ebay account dinged.
[14:17:48] <CaptHindsight> plus all you really need to do is print the interface or inner surfaces of the mold where the metal will come into contact
[14:18:16] <CaptHindsight> furrywolf: release the hounds!
[14:20:05] <tjtr33> hmmm graphite? would it hold up during sink EDM? well rez might be poor ( we grind them now and thats pretty smooth & precise )
[14:20:42] <furrywolf> isn't graphite what's usually used for sink edm?
[14:21:00] <CaptHindsight> tjtr33: make micro graphite tips?
[14:21:03] <tjtr33> yah, thus the thought
[14:21:23] <CaptHindsight> we talked about that last year
[14:21:28] <CaptHindsight> it's on my list
[14:21:42] <CaptHindsight> we can make micron sized graphite tips
[14:21:51] <tjtr33> i have a pile ( no list )
[14:22:30] <furrywolf> do you have scrap graphite? I'm looking for some decent-sized sheets or rods to try as electrolysis electrodes.
[14:22:34] <CaptHindsight> I started forgetting so I had to start writing them down
[14:23:25] <Jymmm> dollar store pencils?
[14:23:41] <furrywolf> "decent-sized".
[14:23:48] <Jymmm> subjective
[14:24:01] <furrywolf> ok... 3" by 18" by 1/2". how's that?
[14:24:06] <CaptHindsight> the biggest pain might be removing the plaster from very tight areas
[14:24:21] <Jymmm> better
[14:24:34] <Jymmm> golf club shafts?
[14:24:55] <furrywolf> aren't they like 1% graphite and 99% plastic and marketing?
[14:25:04] <Jymmm> nfc
[14:25:31] <CaptHindsight> http://www.amazon.com/Fine-Extruded-Graphite-Rod-Pack/dp/B000RC0SZ8
[14:25:45] <Jymmm> just a suggestion where you might find some locally (even broken for free) locally
[14:26:07] <furrywolf> I don't think graphite composites are conductive or suitable for electrodes.
[14:26:29] <CaptHindsight> yes they can be
[14:26:42] <CaptHindsight> depends on loading and particle shape
[14:26:47] <furrywolf> I don't think the ones in consumer products are. :)
[14:27:00] <CaptHindsight> see above :)
[14:27:00] <Jymmm> http://www.beckergraphite.com/stock.html
[14:27:05] <Jymmm> rod and sheet
[14:27:46] <CaptHindsight> home depot polyester resin and 50% carbon black
[14:27:47] <tjtr33> furrywolf, figure 40-50cents/cubic inch for low quality graphite. but dealers wont talk to you for low quantity ( set up is the expense ) as a dealer for cut offs, or a mold shop for used scrap
[14:28:01] <Jymmm> No Minimum Orders
[14:28:22] <furrywolf> hrmm, $20 for 12x12x1/2... that's not entirely unreasonable.
[14:28:52] <Jymmm> http://www.beckergraphite.com/overstock1.html#keenarcelectrodes
[14:29:05] <tjtr33> best look for used electrodes, if they work, then consider buying new ;)
[14:29:30] <furrywolf> I won't order from them though, as I have a strict policy of not dealing with spammers.
[14:29:33] <Jymmm> tjtr33: did you see the link I posted with NO MINIMUM ORDER?
[14:29:48] <tjtr33> yes, it aint edm graphite fer sure
[14:29:55] <Jymmm> ah
[14:30:01] <tjtr33> so i am not familiar with it at all
[14:30:46] <furrywolf> I was reading that if you use graphite for electrolysis rust removal anodes, you end up with about the tenth the sludge, and you don't need to replace them constantly...
[14:30:49] <tjtr33> toyo tanso, le carbinee lorraine, UCAR , POCO those i know
[14:31:07] <CaptHindsight> http://www.thecarystore.com/containers-products/raw-materials-and-specialty-chemicals-pigments-carbon-black-furnace-and-channel-blacks just mix with resin
[14:33:00] <tjtr33> boy thatd be a helluva place to work, you cant contain that dust
[14:33:09] <CaptHindsight> heh
[14:34:14] <andypugh> My dad has worked at a carbon black factory a couple of times. Apparently the trick is to wash it off with olive oil.
[14:35:11] <_methods> off your lungs lol
[14:35:21] * furrywolf would think the trick was to read the help wanted ads
[14:35:33] <FinboySlick> furrywolf: Well, since surface area matters for electrolysis, you pencil leads might be an idea assuming you arrange them well.
[14:36:30] <furrywolf> required effort is a factor too.
[14:36:32] <CaptHindsight> carbon black + boating polyester resin in a heatsink mold
[14:37:11] <CaptHindsight> http://img.diytrade.com/cdimg/1013190/10596799/5/1253856143/aluminium_heatsink.jpg as a form
[14:37:13] <tjtr33> good tip andypugh it also can get hot chili oils off your fingers
[14:37:42] <CaptHindsight> thats why I wear nitrile gloves
[14:37:56] <furrywolf> I've never had problems with hot peppers...
[14:40:18] <FinboySlick> Had to prepare a bunch of habanero once. First time I experienced heat from a pepper through the skin. My hands felt hot for a good 3 days.
[14:40:41] <_methods> yeah be careful when you go to the bathroom
[14:40:47] <furrywolf> habaneros aren't even hot.
[14:40:58] <DaViruz> and when rubbing your eyes.
[14:41:02] <furrywolf> I should have a nice harvest of trinidad moruga scorpions this year. :)
[14:41:07] <CaptHindsight> I hate it when you wash your hands 4 times and then 2 hours later rub your eye and it still burns
[14:41:16] <_methods> yeah
[14:41:34] <XXCoder1> good morning
[14:41:39] <FinboySlick> furrywolf: Oh I have fresh trinidad butch t here. I know hot. But the habanero I had to handle in quantity.
[14:41:49] <CaptHindsight> XXCoder1: late night?
[14:41:55] <furrywolf> I ordered some Reaper starts for this year... will see how they turn out.
[14:42:04] <XXCoder1> just slept in lol
[14:42:15] <XXCoder1> normally wake at 10:30
[14:42:35] <CaptHindsight> i was up before the sun today
[14:42:38] <DaViruz> just spray pepper spray in the mouth
[14:42:39] <furrywolf> I never got any butch ts... the plants always died before fruiting. the morugas are doing much better.
[14:43:04] <CaptHindsight> DaViruz: lol, might as well right
[14:43:14] <FinboySlick> Besides, heat isn't everything. Chili has to taste good too. Habanero and red savina taste good.
[14:43:33] <furrywolf> I don't like the habanero taste. I don't grow them for a reason.
[14:43:37] <DaViruz> CaptHindsight: if you're hell bent on having the hottest pepper in your mouth you might as well :)
[14:43:49] <furrywolf> for flavor, serrano, rooster spur, and fatali are high on my list.
[14:43:58] <furrywolf> aji crystals are good too
[14:44:07] <XXCoder1> sraracha all way for me lol
[14:44:29] <furrywolf> sriarcha SUCKS. it's like fucking sugar in a bottle.
[14:44:32] <XXCoder1> except in soylent. sraracha magic dont work there
[14:44:49] <FinboySlick> sraracha is fermented, no?
[14:45:13] <furrywolf> sriracha is too sweet to eat.
[14:45:21] <XXCoder1> not for me
[14:45:43] <furrywolf> sugar's like the third ingredient...
[14:45:59] <XXCoder1> you sure it isn't first? lol
[14:46:00] <FinboySlick> Well, peppermaster makes a delicious sweet goatpepper chili sauce. There's some pretty amazing sweet heat stuff.
[14:46:42] <furrywolf> I grew goat's weed twice, and neither time did I like the peppers.
[14:46:54] <CaptHindsight> tjtr33: maybe print this http://www.reynoldsam.com/product/ultracal-30/
[14:47:09] <furrywolf> for a sweet sauce, I'd probably start with rocottos... but I don't like sweet sauces.
[14:47:55] <tjtr33> CaptHindsight, just finished the 2nd video, good info
[14:48:25] <CaptHindsight> how fine do they make sands for molds?
[14:48:30] <furrywolf> absolutely no reason to ruin something by putting sugar in it.
[14:48:49] <tjtr33> hmm print historic plaster renovations.
[14:49:36] <CaptHindsight> tjtr33: yeah large flatbed printers are easier to build than routers since there's no cutting
[14:49:43] <furrywolf> I'm not going to have a lot of varieties this year... going to be too busy for much gardening.
[14:49:49] <CaptHindsight> no chatter no vibration
[14:50:13] <furrywolf> no cutting force
[14:53:33] <zeeshan> Dimensional measurements can be made on the microscope by just clicking the area to be measured with the mouse. Measurement data is stored with the image file for easy information sharing, and results can even be exported as a CSV file.
[14:53:35] <zeeshan> FU microscope!
[14:53:41] <zeeshan> i know its stored in the image
[14:53:44] <zeeshan> i need the dATA!!
[14:53:52] <zeeshan> software doesnt let me export 3d data ;[
[14:54:12] <CaptHindsight> yeah, why i build my own
[14:54:14] <XXCoder1> is it jpg?
[14:54:22] <zeeshan> no. tif
[14:54:32] <furrywolf> why does it say "can be exported as a csv file" if it can't?
[14:54:32] <XXCoder1> tif lemme look if tif supports comments
[14:54:38] <CaptHindsight> can you get the raw tiffs?
[14:54:44] <furrywolf> XXCoder1: it's not a tiff file. it just pretends to be.
[14:54:44] <zeeshan> it is the raw tiff i have
[14:54:51] <XXCoder1> oh
[14:54:56] <zeeshan> its got a bunch of shit inside the tiff file
[14:55:19] <XXCoder1> furrywolf: well some image files do support comments so it might be legal still
[14:55:21] <furrywolf> your depth info is 400x300 single-precision floats after the tiff data
[14:55:33] <furrywolf> XXCoder1: no, it's just a proprietary format.
[14:55:38] <XXCoder1> ok
[14:55:55] <zeeshan> furrywolf: i don't see how it correlates to the original image
[14:55:58] <furrywolf> from what I can tell, they simply concatenated a bunch of different file types... despite the .tiff extension, it has a tiff, a windows bitmap, and a jpeg in it. :)
[14:56:09] <furrywolf> ?
[14:56:14] <zeeshan> 1600 x 1200 original image
[14:56:19] <CaptHindsight> http://www.axner.com/raw-materials.aspx casting supplies
[14:56:20] <XXCoder1> bad programming
[14:56:25] <zeeshan> 400x300 depth info image, scale factor of 4.
[14:56:33] <XXCoder1> to me it repersents middle finger to users
[14:56:38] <furrywolf> right, and?
[14:57:33] <zeeshan> pixel 1 in the 400x300 = what pixel in the 1600x1200 ?
[14:59:07] <XXCoder1> * all locs by 4?
[15:00:36] <zeeshan> furrywolf: which tiff data?
[15:00:40] <zeeshan> theres two tiff images in there
[15:00:41] <zeeshan> and a jpg
[15:00:48] <zeeshan> one starts with II* header, the other MM*
[15:03:26] <jdh> tiff is more of a wrapper. you can use jpeg inside
[15:04:21] <furrywolf> the depth data corresponds perfectly
[15:04:31] <zeeshan> how
[15:04:32] <furrywolf> I just made my own colorized version
[15:05:03] <furrywolf> loading the png I made, using gimp's gradient filter, overlaying it with the original jpeg at 50%ish transparency
[15:05:20] <furrywolf> http://imgur.com/M6pmnK5
[15:05:55] <zeeshan> how does that work
[15:06:02] <zeeshan> you have a 400x300 image for depth
[15:06:04] <furrywolf> obviously I used a different color setup, but it's the same thing
[15:06:05] <zeeshan> thats what i dont get.
[15:06:22] <furrywolf> ... "scale factor of 4"
[15:07:03] <zeeshan> you just resized the image by 4x
[15:07:07] <zeeshan> and overlayed it
[15:07:14] <furrywolf> yes
[15:07:27] <zeeshan> i have a shit load of these images to go throughs
[15:07:30] <tjtr33> CaptHindsight, https://www.youtube.com/watch?v=6HnPjuokcVo
[15:07:31] <zeeshan> so i need to know the steps
[15:07:34] <zeeshan> so i can replicate it
[15:07:34] <tjtr33> i remember finding this place back in the 90's. i could _print_ these friezes :) they used latex molds from real temples. they let them weather in a damn garden to make them look old
[15:07:44] <furrywolf> and mapped it to a gradient instead of greyscale
[15:07:52] <tjtr33> damp
[15:08:21] <zeeshan> that zebra image
[15:08:22] <furrywolf> if you have a shitload of them, first step is to automate the file-splitting process
[15:08:31] <zeeshan> you converted it to png i noticed
[15:08:40] <furrywolf> the zebra image is useless
[15:08:42] <zeeshan> oh
[15:08:51] <furrywolf> http://imgur.com/veBDMZN is it converted properly
[15:09:14] <zeeshan> okay lemme try playing with that first in gimp to see the process
[15:09:19] <XXCoder1> zee is it some project or pay job?
[15:09:20] <furrywolf> the zebra image is what happens when you read greyscale floats as RGBA. :)
[15:09:31] <zeeshan> XXCoder1: im trying to analyze some data for masters
[15:09:36] <zeeshan> the scope is being a lamer
[15:09:49] <zeeshan> and the people who bought it are being useless
[15:09:49] <XXCoder1> cool
[15:10:05] <XXCoder1> maybe thats your REAL test lol
[15:10:10] <XXCoder1> adopt to crappy software
[15:10:14] <zeeshan> did you uselinear interpolation
[15:10:17] <zeeshan> when scaling
[15:10:17] <furrywolf> in gimp, open the greyscale depth image, resize to 1600x1200, change mode to rgb, select a gradient you like, gradient map, copy, paste onto original jpeg image, new layer, set transparency where you want.
[15:10:28] <furrywolf> I think cubic is the default
[15:18:24] <zeeshan> how do you apply a damn gradient
[15:18:26] <zeeshan> in gimp
[15:19:30] <furrywolf> after changing the image mode to rgb from grayscale, and selecting the gradient you want on the gradients tab, colors -> map -> gradient map
[15:19:53] <zeeshan> pretty!
[15:20:01] <zeeshan> okay now i see the loss of info
[15:20:05] <zeeshan> a bit better. its not too bad
[15:20:26] <zeeshan> okay now the question is
[15:20:32] <zeeshan> you said it was a floating point data
[15:20:37] <zeeshan> greyscale image.
[15:20:38] <renesis> jymmm: omg hwy 130 was amazing, easily one of my fav drives
[15:20:58] <zeeshan> so im assuming BLACK = minimum depth
[15:21:01] <zeeshan> and white = maximum depth
[15:21:06] <zeeshan> of that makes any sense.
[15:21:24] <renesis> made it to the observatory too late, chatted with a groundskeeper there
[15:21:32] <furrywolf> black and white are arbritary. it's stored as values from 0 to 1 it looks like.
[15:21:53] <zeeshan> okay so 0 would be min value of scale bar
[15:21:57] <zeeshan> and 1 would be max value of it
[15:22:09] <zeeshan> i can make this assumption, analyze a cross section of it in the software and validate
[15:22:14] <renesis> but upside to that was dropping into san jose during sunset
[15:22:46] <zeeshan> furrywolf: i appreciate your help, youre the best :-)
[15:24:03] <renesis> but yeah all the youtube vids missing the hill climb stuff after the del purto canyon to 130 fork
[15:24:35] <renesis> best part, worth the extra hour for sure
[15:27:40] <furrywolf> hrmm, there's no compression... does the windows bitmap data always start at the same offset in every file? is that why they put the huge black border, to make it the same size, so they can get the data from a fixed offset?
[15:30:22] <furrywolf> I've been trying to find an easy way to find what the last byte of a tiff image is, but if it's always in the same spot, that's much easier.
[15:30:38] <furrywolf> but, for now, bbl. unless you want to start paying a consulting fee, I need to get done useful projects. :)
[15:34:02] <furrywolf> once you have the bitmap, this might come in handy (the line I use): convert -depth 32 -size 400x300+54 -define quantum:format=floating-point gray:foo7.foo[0] png:foo7.png
[15:34:54] <XXCoder1> woot my van's flasher module finally arrived. its specifically supports leds
[15:34:59] <XXCoder1> so it flashs properly now
[15:35:06] <XXCoder1> damn long wait
[15:35:53] <renesis> the leds are higher current, blinks too fast?
[15:35:53] <zeeshan> the jpg looks like its in the same location
[15:35:58] <zeeshan> i can search for the file signature for jpg
[15:36:00] <zeeshan> extract jpg
[15:36:10] <XXCoder1> renesis: it dont use enough current so triggers dont work right
[15:36:12] <furrywolf> that tells imagemagick that you have raw greyscale data, in 32-bit floating point, of 400x300 pixels, starting 54 bytes from the start of the file (skipping bitmap header), that you only want the first image (so it doesn't turn the garbage after it into a second image), and save it as a png.
[15:36:13] <zeeshan> then search for file signature of your specially found depth thing
[15:36:23] <XXCoder1> so it blinks fast
[15:36:28] <renesis> oh, oposite
[15:36:58] <XXCoder1> yeah. orginial was half broken so it randomly flashes fast or just on
[15:37:02] <furrywolf> on the image you sent me, the bitmap is at 5885255 or 5885254 depending on how you count. (tail counts differently, go figure)
[15:37:06] <XXCoder1> bumping module changes stuff lol
[15:37:20] <zeeshan> furrywolf: there is no unique signature?
[15:37:23] <malcom2073> I cut my first thing ever on a lathe! woohoo!
[15:37:32] <renesis> neat
[15:37:42] <XXCoder1> malcom2073: way to go. was it a good dilio? ;)
[15:37:55] <malcom2073> XXCoder1: Haha, no, this is a lathe, not a 3d printer :P
[15:37:58] <furrywolf> try convert -depth 32 -size 400x300+5885308 -define quantum:format=floating-point gray:001a.tif[0] png:foo8.png
[15:38:07] <XXCoder1> lathes makes great round things ;)
[15:38:11] <XXCoder1> rod even
[15:38:21] <malcom2073> I took an aluminum rod, and make it slightly rounder!
[15:38:22] <zeeshan> 308?
[15:38:24] <furrywolf> for the file you sent me, that grabs the depth info in a single command. if it's at the same place in other files, it will work on other files too...
[15:38:26] <zeeshan> 5885308?
[15:38:38] <XXCoder1> malcom2073: nice
[15:38:44] <XXCoder1> guess you got plans for it
[15:38:47] <zeeshan> id like to do this all in matlab
[15:38:51] <furrywolf> bitmap starts at 5885254, plus 54 bytes to skip the bitmap header and extract the raw data
[15:38:51] <malcom2073> Not uh.... really.
[15:38:54] <zeeshan> cause im gonna make it spit out the data
[15:38:57] <furrywolf> I know zero about matlab.
[15:38:59] <malcom2073> It was a really good deal, and I figure I'll eventually need one since I have a mill now
[15:39:03] <zeeshan> i know
[15:39:06] <zeeshan> thats why im asking the steps :)
[15:39:18] <XXCoder1> malcom2073: plan to automate it with cnc stuff?
[15:39:31] <malcom2073> XXCoder1: Nah not this one, I do want to stick a DRO on it though
[15:39:32] <furrywolf> do you have another tiff handy?
[15:39:51] <Cromaglious> I really need to cnc my lathe, though I need to get a qctp for it and make a milling attachment or get one.
[15:40:29] <furrywolf> zeeshan: so we still haven't negotiated my consulting fee... :P
[15:40:35] <malcom2073> I'm planning on a 4th axis for my mill, so that'l take care of most cncing of roundish stuff
[15:41:19] * furrywolf will take payment in cash, working deepcycle batteries, good indian food, or sex toys.
[15:41:25] <zeeshan> ROFL
[15:41:45] <zeeshan> furrywolf: guess what
[15:42:01] <zeeshan> i do!
[15:42:02] <zeeshan> ;p
[15:42:08] <furrywolf> oh, and tooling! especially tooling. or mesa boards.
[15:42:12] <zeeshan> i'll send you some toolz
[15:42:25] * furrywolf needs a 7i76e
[15:42:27] <zeeshan> you have a lathe yea?
[15:42:54] <furrywolf> I have a shoptask lathe/mill combo machine, plus a cnc sherline mill, and non-cnc sherline lathe.
[15:43:07] <furrywolf> do you have another tiff uploaded somewhere?
[15:43:07] <zeeshan> can it take .625"x.625" tools?
[15:43:11] <zeeshan> yea second.
[15:43:22] <furrywolf> hrmm. good question. I don't remember if it's 5/8 or 1/2 that it likes.
[15:43:26] <furrywolf> the sherline takes 1/4. :)
[15:43:36] <zeeshan> cute
[15:44:08] <furrywolf> I'll check next time I'm at the storage unit where it is
[15:44:16] <zeeshan> https://www.dropbox.com/s/3qafe4d83iud302/016a.tif?dl=0
[15:44:21] <zeeshan> i have a few tools i dont need
[15:44:29] <zeeshan> will go through em and see if you can use em
[15:44:39] <furrywolf> anything that takes indexable carbide would be nice
[15:44:57] <furrywolf> ... is that the same image? because it sure looks similar.
[15:45:04] <zeeshan> nope
[15:45:06] <zeeshan> its diff :)
[15:45:15] <zeeshan> look @ the scale bar
[15:45:17] <furrywolf> how about one that actually looks different? :P
[15:45:28] <zeeshan> ok sec
[15:45:38] <furrywolf> you know, so I can tell them apart when looking at the files. :P
[15:45:49] <XXCoder1> lol
[15:45:59] <furrywolf> what are these pointy bumps anyway?
[15:46:03] <zeeshan> https://www.dropbox.com/s/94s6po23a5nfzky/025a.tif?dl=0
[15:46:10] <XXCoder1> mace points heh
[15:46:14] <zeeshan> i indented my shape memory polymer
[15:46:18] <zeeshan> this is the evolution of its recovery
[15:46:28] <furrywolf> lol
[15:47:22] <XXCoder1> http://hackaday.com/2015/03/20/exposing-private-facebook-photos-with-a-malicious-app/
[15:47:40] <XXCoder1> this is why I never upload pictures I dont wanna show to public
[15:47:41] <furrywolf> wow, malware can do bad things? I had no idea!
[15:48:12] <zeeshan> furrywolf: can you tell me the signature of the bit map
[15:48:17] <XXCoder1> uh Im not saying it does not do bad shit
[15:48:25] <XXCoder1> im saying do not upload private shit
[15:48:35] <XXCoder1> because nothing is truly secure
[15:48:45] <zeeshan> first two bytes
[15:48:50] <zeeshan> or a couple of bytes
[15:49:01] <furrywolf> http://imgur.com/TQYYj3w
[15:49:08] <zeeshan> nice!
[15:49:27] <furrywolf> so, yeah, same offset.
[15:49:47] <zeeshan> it must have a signature!
[15:49:50] <furrywolf> all I did was change the filename to get that. :)
[15:49:51] <XXCoder1> man wish Im skilled on hacking. guy got another 10k payout
[15:50:03] <XXCoder1> for total of 22k. not bad for couple months work
[15:50:25] <furrywolf> 000000 42 4d 36 53 07 00 00 00 00 00 36 00 00 00 28 00
[15:50:31] <furrywolf> first bytes of the bitmap file
[15:50:38] <furrywolf> (000000 being offset, not bytes)
[15:50:43] <zeeshan> oh you're right
[15:50:45] <zeeshan> its at 059cd40
[15:51:00] <zeeshan> its at 059cd46
[15:51:01] <zeeshan> i mean
[15:51:28] <furrywolf> convert -depth 32 -size 400x300+5885308 -define quantum:format=floating-point gray:025a.tif[0] png:foo9.png gave that last image
[15:51:56] <zeeshan> hm , maybe i can use imagmagick to convert them
[15:52:00] <zeeshan> and then just load em up in matlab.
[15:52:03] <renesis> xxcoder1: they change TOS too much, they were always just a data farm
[15:52:08] <zeeshan> but i'dreally like to figure it out
[15:52:11] <zeeshan> why are you using 32 bits?
[15:52:14] <zeeshan> for the depth
[15:52:18] <XXCoder1> renesis: yeah
[15:52:22] <zeeshan> the main tiff is stored as 24 bits
[15:52:58] <XXCoder1> renesis: would be bad if anyone ever figured how to access anyone messages data
[15:53:07] <furrywolf> because it's a bitmap of single-precision floats, which are 32 bites
[15:53:08] <furrywolf> bits
[15:53:22] <XXCoder1> because everyone on fb said something embarring in least once for sure.
[15:53:23] <renesis> i just use it to follow event promoters because they stopped updating caldar sites
[15:53:42] <XXCoder1> https://gordsgarage.wordpress.com/2015/03/18/vacillating-vertical-pendulum/ nice
[15:53:45] <renesis> not me, i dont socialize
[15:53:45] <zeeshan> gotcha
[15:53:58] <furrywolf> you might not be able to read it with anything that tries to read it as a windows bitmap... I'm not sure windows bitmaps ever supported floats...
[15:54:04] <XXCoder1> malcom2073: your next project? heh
[15:54:07] <furrywolf> I suspect it's a non-standard extension
[15:54:21] <zeeshan> 42 4d 36 53 07 20 20 20 20 20 36 20 20 20 28 20 20 20 90 01 20 20 2c 01 20 20 01 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20
[15:54:22] <renesis> if you dont tell it your high school it cant figure out who you are
[15:54:24] <XXCoder1> yeah bitmap isnt very complex
[15:54:32] <zeeshan> that didnt paste right.
[15:54:33] <zeeshan> wtf
[15:54:36] <XXCoder1> renesis: not me. my name is completely unique
[15:54:39] <renesis> its pretty dumb
[15:54:48] <renesis> oh
[15:55:10] <renesis> i guess thats diff
[15:55:16] <XXCoder1> my friend name is also unique. because his dad used name normally used as last name as first name
[15:55:22] <zeeshan> http://i.imgur.com/OisqXXz.png
[15:55:38] <zeeshan> the highlighted part looks like its a bunch of crap before the actual data?
[15:56:37] <zeeshan> yea that doesnt seem to change between the files
[15:56:55] <furrywolf> that crap is either the tiff image directory for the thumbnail, the exif data, or metadata.
[15:57:24] <Cromaglious> yea! 60 degree engraving bits arrived
[15:57:27] <furrywolf> somewhere your units are stored
[15:57:51] <XXCoder1> wonder if it is stucture data
[15:58:11] <XXCoder1> maybe its saying next is bitmap then blah blah size so on
[15:58:40] <zeeshan> it must
[15:58:46] <andypugh> All TIFF files start with “42”
[15:58:48] <zeeshan> the image has the scale stored in it
[15:58:49] <furrywolf> it could well be. it could also be part of the tiff file.
[15:59:05] <malcom2073> XXCoder1: wtf is that?
[15:59:12] <XXCoder1> yo yos
[15:59:26] <andypugh> Then the next digits tell you what tags are there and where to find them. It’s pretty clever as a file format.
[15:59:30] <zeeshan> furrywolf: when i delete everything before the 42
[15:59:32] <malcom2073> lol
[15:59:35] <zeeshan> i get that zebra like image again lol
[16:00:23] <furrywolf> as I said, it's not a standard windows bitmap. it's an extension to the format.
[16:00:34] <zeeshan> yes you said its a floating point
[16:00:51] <XXCoder1> I hate it when programmers hack up format then use same extension
[16:00:56] <XXCoder1> why not make up somehting like
[16:00:59] <zeeshan> so if i extract the next 400x300 bits from it
[16:01:02] <XXCoder1> like foo.fuk
[16:01:04] <zeeshan> that is technically my image?
[16:01:17] <furrywolf> you want the next 400x300 floats, which are 4 bytes each.
[16:01:18] <zeeshan> starting at 5885308
[16:01:26] <furrywolf> yes
[16:01:44] <XXCoder1> malcom2073: wow it looks like commerical product at end
[16:01:45] <zeeshan> and you distinguish the columns from rows
[16:01:46] <XXCoder1> very good
[16:01:54] <malcom2073> Lol indeed
[16:01:54] <zeeshan> by using the fact that 400x4 bytes per ROW
[16:02:24] <furrywolf> yep
[16:03:44] <zeeshan> 61207C looks like is the last location
[16:06:28] <furrywolf> is there a better way than tail -c to get only the last part of a file? dd bs=1 is ungodly slow.
[16:06:34] <LeelooMinai> Hmm, any tips on testing modbus in debian? It looks like libmodbus is installed, now what? Are there some apps that can do something useful over it? Like reading/setting some registers?
[16:06:36] <jdh> tiff data is normally stored in strips. libtiff is pretty friendly
[16:07:12] <furrywolf> jdh: this has nothing to do with tiffs.
[16:08:05] <furrywolf> jdh: to recap, he has a proprietary file output by some software. the file consists of, at least the parts I've identified, a tiff image, a windows bitmap, and a jpeg image, concatenated together, as well as some as-yet-unidentified data.
[16:08:56] <jdh> sounds like a tiff
[16:09:27] <furrywolf> no, they're separate files concatenated. the extra stuff is not contained in any tiff directories.
[16:09:37] <andypugh> Yes, tiff is a package for all sorts of possible data
[16:09:57] <zeeshan> furrywolf: it worked!
[16:10:03] <andypugh> furrywolf: You have walked the tiff tag tree?
[16:10:13] <furrywolf> yes, I know what tiff is. Again, the other files are not contained within the tiff structure.
[16:10:14] <furrywolf> andypugh: yes
[16:10:21] <furrywolf> with about four different programs, just to be sure.
[16:10:25] <furrywolf> including tiffdump
[16:10:31] <jdh> and didnt hit a nonstandard tag?
[16:10:38] <XXCoder1> yeah I found that strange that programmer didnt use tiff features
[16:10:41] <zeeshan> andypugh: theres propietary file structure
[16:10:43] <zeeshan> within the tiff
[16:10:48] <andypugh> Sorry. Would you like instructions on how to suck eggs next ? :-)
[16:10:51] <furrywolf> the only nonstandard tag is some exif data, at least that I could find.
[16:10:57] <zeeshan> furrywolf: !!!
[16:11:09] <zeeshan> so basically i use matlab to extract data from those 2 addresses.
[16:11:14] <zeeshan> or to be safe, i search for that header,
[16:11:22] <zeeshan> and offset from there to extract the data.
[16:11:39] <XXCoder1> andypugh: ironic thing is that most old people would need instructions as we dont suck eggs anymore :P
[16:11:54] <XXCoder1> that saying is very old. probably older than most civilizions
[16:12:28] <zeeshan> furrywolf: i have a q.
[16:12:37] <furrywolf> they seem to follow fixed byte positions... this works to extract them: if=025a.tif ; convert -depth 32 -size 400x300+5885308 -define quantum:format=floating-point gray:$if[0] $if.png ; tail -c +6365335 $if >$if.jpg
[16:12:47] <zeeshan> im a bit confused about what you were talking about when you were mentioning that it's not a standard bitmap
[16:12:52] <zeeshan> thats why its showing the wrong pattern
[16:13:20] <zeeshan> you're saying that each 4 bytes of the data represents a floating point number
[16:13:30] <zeeshan> which basically tells you the grayscale value.
[16:13:54] <furrywolf> windows bitmaps specify RGBA data, standard 0-255 values. Your bitmap specifies single-precision floating points. it's cutting the 4 bytes of the float up and treating them like color channels, giving interesting but useless results.
[16:14:08] <zeeshan> ah
[16:14:26] <XXCoder1> I guess it looks weird when opened in paint or something
[16:14:50] <furrywolf> I'm not an expert on windows bitmaps, but googling couldn't find any mention of floating-point ones, and imagemagick doesn't complain about anything odd with the header, just reads it anyway...
[16:15:21] <XXCoder1> bitmap is literally just bit map. header than color pixels
[16:15:36] <zeeshan> i think its like RGB RGB RGB RGB
[16:15:40] <XXCoder1> it is
[16:15:43] <zeeshan> in order for each pixel
[16:15:56] <furrywolf> the oneliner above will split your images, assuming they all have the same fixed offsets the files so far have had, and you have imagemagick and tail...
[16:16:03] <furrywolf> but you still need to find where your units are.
[16:16:14] <zeeshan> units i can grab manually from the scale for now
[16:16:16] <SpeedEvil> Pirate bitmaps go ARGB
[16:16:24] <zeeshan> whats 'a?
[16:16:31] <XXCoder1> alpha
[16:16:36] <zeeshan> ??
[16:16:42] <LeelooMinai> transparency
[16:16:43] <XXCoder1> transparenacy level lol
[16:17:13] <furrywolf> I'm thinking the depth information is normalized, so if your images have different scales, the data still shows 0-1 anyway... you need to find where in the file your scale is stored.
[16:17:38] <furrywolf> the data without the scale isn't useful to compare between images
[16:17:39] <zeeshan> yea it is definitely normalized
[16:17:53] <zeeshan> well if you look at the scale bar
[16:17:57] <zeeshan> the max value is the only one that changes.
[16:18:01] <zeeshan> the min value remains at 0
[16:18:15] <zeeshan> youre like 10 steps ahead of me
[16:18:28] <furrywolf> it's unlikely the scale bar is stored in the file as a scale bar... more likely there's a single field somewhere in one of the blocks of unidentified data with the scale factor.
[16:19:31] <furrywolf> I may know more about files, but I'm sure you know more about shape memory polymers... closest thing I have to one is my tempurpedic mattress. :P
[16:19:32] <zeeshan> it has to be
[16:19:37] <furrywolf> bbl, need to get some stuff done before the rain hits
[16:19:37] <zeeshan> cause i have another image of the EXACT same size
[16:19:43] <zeeshan> that has no scale bar :
[16:19:54] <jdh> I store calibration values in a struct in a tag for scalung
[16:20:35] <furrywolf> I'll poke at it more tonight, but I can't waste non-raining daylight. bbl.
[16:21:57] <zeeshan> thanks!
[16:25:02] <furrywolf> if you can manually change the scale without changing anything else, save the image, change the scale, save it again, then compare for differences between the two.
[17:01:23] <XXCoder1> angry birds the movie jeez
[17:01:29] <XXCoder1> 3 years ago it'd have made sense
[17:06:02] <CaptHindsight> animated or adult?
[17:06:49] <XXCoder1> 3d
[17:07:26] <XXCoder1> http://thenextweb.com/media/2015/03/19/rovios-3d-angry-birds-movie-will-have-cost-more-than-160m-by-the-time-it-lands-next-year/
[17:07:38] <XXCoder1> 160 mil. doubt they will clear that
[17:08:15] <malcom2073> I'd pirate it
[17:17:18] <CaptHindsight> refractory silicas are available is avg diameters of 7-12 nanometer, that should be good enough resolution :)
[17:21:17] <XXCoder1> lol
[17:21:48] <XXCoder1> I can't wait to get next paycheck. I will buy cnc router frame with it
[17:21:52] <CaptHindsight> if you really need batter than that well tough
[17:23:12] <XXCoder1> wtf https://www.lollipuff.com/blog/329/diy-how-to-reuse-reset-illumask-light-therapy-mask
[17:24:10] <malcom2073> Lol, light therapy
[17:24:13] <XXCoder1> more than one wtf really.
[17:25:14] <malcom2073> Many wtfs
[17:26:06] <XXCoder1> same guy who hacked it also said it dont work
[17:26:13] <XXCoder1> made skin extra dry
[17:26:24] <XXCoder1> https://www.lollipuff.com/blog/327/nuface-and-illumask-review-do-these-beauty-gadgets-work
[17:26:32] <malcom2073> Shock, a bright hot light makes your skin dry?
[17:26:36] <malcom2073> I'm amazed.
[17:26:40] <malcom2073> And annoyed that I didn't think of it first
[17:26:50] <XXCoder1> easy
[17:26:57] <XXCoder1> same design, use elsewhere
[17:27:06] <malcom2073> Oh! Sunglasses with lights in them
[17:27:08] <malcom2073> decrease uh....
[17:27:08] <XXCoder1> HMMM light therpy for balls for better sperm
[17:27:12] <malcom2073> bags under the eyes
[17:27:13] <XXCoder1> manhood sells.
[17:27:14] <malcom2073> yeah
[17:27:26] <malcom2073> Lol
[17:27:32] <XXCoder1> if you make millions, get me a decent CNC mill
[17:27:37] <XXCoder1> vm10 is fine
[17:27:41] <XXCoder1> *hurco
[17:27:47] <malcom2073> Sure
[17:27:54] <XXCoder1> and stuff to run it of course lol
[17:28:04] <malcom2073> So what should I make? The yoyo seems a bit complicated for afirst project :P
[17:28:11] <malcom2073> Oh yeah, software'll have to be included :)
[17:28:21] <XXCoder1> dilios is easiest project I can think lol
[17:28:52] <malcom2073> As in David Wenham?
[17:29:07] <malcom2073> :P
[17:29:20] <XXCoder1> dunno on that reference
[17:29:27] <malcom2073> google dilios
[17:29:36] <malcom2073> First 10 results.
[17:29:36] <malcom2073> :P
[17:29:50] <XXCoder1> fine lol a sec
[17:30:50] <XXCoder1> If I recall though, if you want more testronine without pills or anything, you need to let sun shine on dick and balls
[17:31:10] <XXCoder1> but if you design uv ball light for sessions it would do same I guess?
[17:34:06] <XXCoder1> https://hackadaycom.files.wordpress.com/2015/03/terminal.jpg?w=800&h=800 man
[17:34:14] <XXCoder1> old
[17:40:24] <XXCoder1> lol charactor named dilios
[17:49:59] <Nick001-shop> <PCW> Around?
[17:57:06] <Nick001-shop> questions about 7i76 card
[17:58:16] <micges> Nick001-shop: ask, maybe I can help
[17:59:11] <Nick001-shop> with input pins like for limit switches - is there voltage involved?
[18:00:46] <micges> what do you want to do?
[18:02:24] <Nick001-shop> imit swith on machine
[18:02:33] <Nick001-shop> limit
[18:03:14] <Nick001-shop> Also, what is field power – is that power driving an output signal to say a solenoid or relay with a 24 VDC coil?
[18:03:40] <micges> ah
[18:04:27] <micges> field power is supply used to drive outputs and sense inputs
[18:04:31] <zeeshan> Nick001-shop: yes, and also powers the chips for inputs
[18:04:33] <zeeshan> :P
[18:05:27] <micges> if you use field 24V supply you need to put > 16V on inputs to get true and all outputs will be 24V
[18:05:46] <PCW> Yes for 7I76 filed I/O inputs its typically +24V --> switch --> input
[18:06:22] <Nick001-shop> Then I hook one side of the limit switch to say output 017 and the other side to ground and the computer should see it
[18:07:41] <zeeshan> no
[18:07:47] <zeeshan> +24vdc to limit switch
[18:07:57] <zeeshan> and the other side to input of mesa
[18:10:28] <Nick001-shop> Looks like I should have a terminal strip with the 24 VDC as I have five limit type switches to deal with
[18:11:22] <Nick001-shop> Does the 7i76 switch that voltage to TTL to protect the computer
[18:13:08] <Deejay> gn8
[18:13:15] <PCW> the 24V I/O is completely isolated from the computer
[18:13:39] <PCW> no common connections
[18:15:19] <Nick001-shop> Does the 24 VDC have to go through TB1 field power connector or can I use a 24 VDC power supply separately
[18:23:52] <PCW> the 24VDC for the switches must have a common ground with the field power
[18:24:13] <PCW> normally just one power supply would be used
[18:28:01] <Nick001-shop> Right,but does it have to be connected up to the field power plug or can I use a separate terminal strip to drive all the stuff I need to drive. Like five limit switches and five or six solenoid's controlling air devices
[18:37:35] <PCW> You can connect however you like as long as the field power common is the same as the power supply common
[18:42:51] <Nick001-shop> Ok - I know they're dumb quetions but I'm more mechanical and still trying to catch up with electronics. I wired my whole shop up but I'm way behind on these itty bitty electronics.
[18:44:43] <Nick001-shop> If I drive an output to operate a solenoid coil at 24 VDC I guess I should have a connection to the field power connector
[18:44:45] <LeelooMinai> I am connecting GS2/spindle to linuxcnc: http://i.imgur.com/5WV8Q1m.png
[18:44:48] <PCW> Not dumb, better to ask than blindly connect things that either dont work and waste time or worse ,damage something
[18:45:37] <LeelooMinai> But the GS2 is "deceived" and uses, say, 2400 for 24000 rpm , that is divided by 10. Can I include that factor in the pin assignment?
[18:45:58] <LeelooMinai> That is, can I do: net gs2-RPM spindle-vfd.speed-command <= motion.spindle-speed-out / 10.0 ?
[18:46:24] <PCW> with solenoids you have to make sure the current is within the 7I76s ratings and you need a flyback diode across the solenoid coil
[18:46:46] <Nick001-shop> I have generated smoke a few times by just going ahead and connecting it up
[18:47:15] <LeelooMinai> And I guess together with: net gs2-at-speed motion.spindle-at-speed <= spindle-vfd.at-speed * 10 Are those valid?
[18:47:40] <Nick001-shop> What is the current rating on the 7i76 and what is a flyback diode
[18:48:13] <PCW> +24V --> 7I76 field power
[18:48:15] <PCW> 7I76 output --> solenoid coil+
[18:48:16] <PCW> 24V common --> solenoid coil -
[18:48:29] <Tom_itx> flyback diode prevents the coil from sending voltage back to the control
[18:48:34] <LeelooMinai> Nick001-shop: Flyback diode is to protect cuircuits from inductive "kickback" - that is very high voltages that coils can generate
[18:48:35] <Tom_itx> current*
[18:49:24] <Tom_itx> when you disengage the coil there is a current spike from the coil
[18:49:33] <Tom_itx> the diode supresses it
[18:49:38] <LeelooMinai> i.e. Never cut coils that are in operation with scissors:)
[18:49:55] <Nick001-shop> Will a standard common diode do the job. I know the voltages 24 VDC but what current will 7176 handle
[18:50:09] <PCW> The 7I76 outputs are OK up to about 60 mA without a flyback
[18:50:10] <PCW> beyond that, a flyback diode is required
[18:50:24] <Tom_itx> preferrably a fast diode that will handle 2x the current
[18:51:03] <PCW> so 24V relays are normally OK without a flyback but solenoids are usually more than 60 mA
[18:51:23] <Tom_itx> it's good practice
[18:52:21] * LeelooMinai sighs and loads HAL manual
[18:52:31] <PCW> the 7I76 outputs are protected but the inductive energy is dissipated in the output drivers if no flyback diode is used
[18:52:39] <XXCoder1> wow http://www.theguardian.com/world/2015/mar/20/france-decrees-new-rooftops-must-be-covered-in-plants-or-solar-panels
[18:52:57] <PCW> (rather than the relay/solenoid coil)
[18:53:04] <LeelooMinai> Is HAL linuxcnc specific? I read there is also linux HAL, that was replaced with udev... are those separate concepts?
[18:53:40] <PCW> different
[18:53:45] <LeelooMinai> XXCoder1: lol...
[18:53:57] <XXCoder1> whats funny about it
[18:54:11] <PCW> well same concept but unrelated uses
[18:54:17] <LeelooMinai> What's funny about unable to have normal roof... I don't know...
[18:54:29] <LeelooMinai> Seems a bit excessive? :)
[18:54:38] <XXCoder1> only for businesses
[18:54:47] <XXCoder1> but its better for envorment reallt.
[18:54:52] <XXCoder1> well gonna go laters
[18:55:03] <LeelooMinai> PCW: I see
[18:56:24] <Nick001-shop> Let me rephrase the question - hooking up a 24 VDC power supply to the card, what should be the maximum amperage output of that power supply.
[18:56:55] <LeelooMinai> "You can have your bussiness here, but your building must look like a big mushroom, have solar roof and elf garden replica in the front"
[18:57:03] <Tom_itx> depends what you are powering with it
[18:57:24] <Tom_itx> calculate the current draw from all your 24v devices and add a safety margin to it
[18:59:05] <Nick001-shop> But how much amperage will the 7i76 card handle
[18:59:47] <Tom_itx> that should be in the pdf
[19:01:32] <PCW> max output driver total is 5.6A but its very unlikely you have 5A of control power
[19:02:04] <PCW> sorry 2.8A not 5.6
[19:02:28] <Tom_itx> most coils are in the mA range
[19:02:34] <Tom_itx> low*
[19:02:41] <LeelooMinai> Hmm, cannot find how to scale pin values inside hal
[19:03:15] <LeelooMinai> I think I need a "component" even for simple expressions
[19:03:18] <PCW> scaling is often done by the I/O component
[19:03:39] <LeelooMinai> Right, so I have to route the pin through one of them
[19:04:03] <PCW> what are you scaling?
[19:04:34] <LeelooMinai> GS2 VFD RMP - I have it setup scaled by 10, because they have 9999 limit and my spindle is 24000
[19:04:40] <LeelooMinai> RPM*
[19:04:55] <Nick001-shop> The Bridgeport boss control has a 1.4 amp 24 VDC power supply to drive the solenoid's and relays originally. I guess the card can handle everything I have to hook up to it. Thank you for the time and information.
[19:05:06] <LeelooMinai> So I need to multiply it by 10 for HAL and back by 10 the other way
[19:05:28] <PCW> how are you communicating ti the VFD?
[19:05:39] <LeelooMinai> Through USB to rs485 converter
[19:05:53] <LeelooMinai> I used modbuspoll command line tool - seems to work fine
[19:06:02] <LeelooMinai> modpoll* to test it that is
[19:06:23] <PCW> Ahh not sure if the modbus component has any scaling
[19:06:54] <LeelooMinai> So basically there's a port created by FTDI RS232 that is on that converter ($5:) called /tty/USB0 and I guess it will work fine with that HAL componenet for GS2 they have there
[19:06:57] <Tom_itx> what about 'scale' or 'gain'??
[19:07:18] <LeelooMinai> I see componenets like mult2 - I think it should do, no?
[19:07:31] <LeelooMinai> O, raight, there's scale too
[19:08:13] <LeelooMinai> /dev/ttyUSB0 *
[19:08:49] <LeelooMinai> I used that command line tool to switch to 38400 instead of 9600 - seems to work fine
[19:09:57] <LeelooMinai> btw, if anyone needs super-cheap way to talk to modbus rs435 using usb, I have this: http://www.ebay.ca/itm/USB-to-RS485-TTL-Serial-Converter-Adapter-FTDI-interface-FT232RL-75176-Module-Ne-/161264238508?pt=LH_DefaultDomain_0&hash=item258c18ffac
[19:10:14] <LeelooMinai> 485*
[19:11:22] <LeelooMinai> + I bought cheap 4 wire rj11 cable, cut it in half, and connected to the GS2
[19:11:44] <LeelooMinai> Total cost $6:)
[19:13:59] <zeeshan> LeelooMinai:
[19:14:07] <zeeshan> http://wiki.linuxcnc.org/cgi-bin/wiki.pl?VFD_Modbus
[19:14:18] <zeeshan> see example under sumitomo
[19:14:36] <LeelooMinai> Right, scale, ok
[19:27:04] <furrywolf> green roofs are a good idea, until you actually have one.
[19:27:56] <LeelooMinai> Or you are a Hobbit:)
[19:29:06] <furrywolf> you have to design for a stupidly high loading, they leak and are a nightmare to fix, the soil can erode, poor selection of plants results in dieouts or needing to fertilize, mow, or otherwise constantly maintain, the whole "absorbs rainwater" thing is bunk - during a heavy rain most of it just runs off, while other times you have to water, creating its own environmental issues, etc, etc.
[19:29:46] <LeelooMinai> I had a feeling it's like that
[19:30:41] <LeelooMinai> At least solar panels do not require a lot of maintenance
[19:30:52] <furrywolf> the plants that don't require maintenance and watering are the same ones all your neighbors consider weeds
[19:31:23] <furrywolf> I'm waiting for a HOA to start sending people around with a ruler to measure how long you've let your roof grow, and fine you. :)
[19:32:40] <LeelooMinai> Or people jailed for using fake grass on roofs
[19:33:45] <furrywolf> they attract animals which most people don't want (birds are great, until they cover your car and everything else in shit, racoons, etc), many plants worsen allergies, fertilizer runoff creates pollution issues in stormwater, they cost many times more to build (stronger framing, waterproofing, etc), they attract ants and termites,...
[19:34:10] <LeelooMinai> Hmm, I wonder there are docs for all those built-in components
[19:34:40] <LeelooMinai> They just list them and what...
[19:34:56] <Tom_itx> there are descriptions
[19:35:10] <LeelooMinai> For few only as I see
[19:39:36] <LeelooMinai> In the manual I see something, say, like this: "maj3 Compute the majority of 3 inputs." ... so name and descrption, but no pin names or examples
[19:41:05] <LeelooMinai> Found something by man maj3 - weird
[19:57:59] <LeelooMinai> I see they have some PyVCP (whatever that is) example there with some speedometer for a spindle. Is that normal way of doing this kind of display?
[19:59:41] <andypugh> LeelooMinai: Every HAL component has a man page
[20:00:33] <LeelooMinai> Right, found it. Kind of strange since those componenets are specific for linuxcnc, but man scale worked
[20:00:35] <andypugh> And all those pages appear as links towards the bottom of; http://www.linuxcnc.org/docs/html/]
[20:01:37] <LeelooMinai> Hmm, did not see that. First link I got for "lunuxcnc components" was this: http://linuxcnc.org/docs/html/hal/components.html
[20:03:05] <andypugh> That page you linked needs to either auto-generate or disappear, it gives an innaccurate description of what there is.
[20:03:29] <LeelooMinai> Right, it confused me a bit - if I saw that one you linked it would be much better
[20:08:49] <Cromaglious__> now to figure out how to make a slightly regualated power supply out of a yoocnc spindle side of the power supply
[20:09:42] <Cromaglious__> byeya PCW
[20:15:18] <tjtr33> LeelooMinai, rely on the man pages, they are from the code for the system you have.
[20:15:19] <tjtr33> else remember about old linuxcnc docs, they are like the HitchHikers guide "though it contains much that is apocryphal, or at least wildly inaccurate"
[20:16:45] <LeelooMinai> Hmm, I knew it won't just work... http://i.imgur.com/uZIwP5X.png
[20:17:14] <LeelooMinai> I used this: http://linuxcnc.org/docs/html/hal/pyvcp_examples.html#gs2-rpm-meter
[20:17:33] <LeelooMinai> To hopefully display that spindle GUI
[20:20:25] <LeelooMinai> So... where is the schema file for that pyvcp xml format?
[20:20:50] <LeelooMinai> Cannot find it somehow.
[20:21:57] <LeelooMinai> Since the error message does not tell me the line in the xml.
[20:22:40] <toastydeath> if anyone's been through engineering or physics, I have a basic fluids question that I don't think I'm setting an integral up right on
[20:22:44] <toastydeath> full workup: http://imgur.com/a/RRyRJ
[20:30:30] <malcom2073> Man, it facinates me that there is math that will determine that, but I'll be damned if I'll ever understand it
[20:32:13] <zeeshan> toastydeath:
[20:32:33] <zeeshan> are you treating v and h as constants?
[20:33:14] <toastydeath> h is a constant, v is a function of of the position
[20:33:49] <toastydeath> i deal with V in the middle of page 2
[20:34:56] <zeeshan> i did this so long ago :/
[20:35:00] <zeeshan> forgot it
[20:35:04] <zeeshan> i just know its couette flow
[20:35:17] <toastydeath> like, i'm 99% sure that this is not me cocking up my physics/fluids
[20:35:18] <zeeshan> and the velocity profile is linear :P
[20:35:21] <toastydeath> but rather me not setting up the integral
[20:35:38] <toastydeath> I'm almost _positive_ that M = F * d is the incorrect way to start this
[20:36:15] <toastydeath> i.e. that it should be int[ F * d dA ]
[20:36:32] <toastydeath> and then something ELSE has to happen with that "A" in the force equation
[20:37:16] <andypugh> it’s been an awfuly long time, but I would be considering rings of width dr
[20:37:43] <tjtr33> LeelooMinai, the vbox has no 'width', look at pyvcp_widgets.py in class pyvcp_vbox(Frame) it does inherit Frame but unless a feature is specific in the new class, it often 'just dont work' (TM)
[20:38:17] <LeelooMinai> Wonderful... I am just removing those widths to see what happens
[20:38:24] <andypugh> dM = Frdf where df = mu(r.dr)v(r)
[20:38:28] <tjtr33> thats one of the reasons for the new renderes like GL and Qt for linuxcnc panels
[20:38:57] <toastydeath> andypugh, unfortunately that doesn't solve my problem, but I think it indicates that my assumption - that I need to have a dA term
[20:38:58] <toastydeath> is correct
[20:39:00] <LeelooMinai> New like planned development, or there are new ones?
[20:39:01] <tjtr33> the underlying 'schema' is really TclTk ;)
[20:39:10] <toastydeath> but I have no goddamn idea what to do with the area term in the force equation
[20:39:46] <andypugh> set it up in terms of dr. Then A = pi.r.dr
[20:40:14] <toastydeath> let me reframe
[20:40:32] <toastydeath> I have a function dependent on area, integrating BY area
[20:40:39] <tjtr33> LeelooMinai, already existing toolkits like JT's http://gnipsel.com/linuxcnc/gladevcp/index.html
[20:40:47] <toastydeath> whether I do A = pi r dr or double integration is equivalent
[20:40:54] <toastydeath> because the fundamental problem is still there
[20:41:05] <toastydeath> wtf do I do with an area term when the area is infinitesmal
[20:41:10] <andypugh> It’s almost identical to the problem of calculating the moment of inertia (or, for that matter, the area of a dis)
[20:41:47] <andypugh> I am almost certain that it is a single integral in r from 0 to R
[20:42:23] <toastydeath> andypugh, it doesn't matter if it's a single or double integral (because in this case they're equivalent
[20:42:25] <andypugh> Consider a stack of rotating rings
[20:42:25] <zeeshan> if you treat it like a ring
[20:42:27] <zeeshan> its a single integral
[20:42:29] <toastydeath> guys
[20:42:32] <zeeshan> if you treat it like an element like hes doing it
[20:42:37] <zeeshan> its a double integral
[20:42:51] <zeeshan> cause its a function r, theta
[20:42:59] <toastydeath> either way I do it, i either have a ring of infinitesmal volume, or an element of infintesmal volume
[20:43:10] <zeeshan> but with ring its axisymmetric so its just a function of r
[20:43:12] <toastydeath> and I am still stuck with a function dependent on area
[20:43:17] <zeeshan> (which is much eaasier to intrrepret!)
[20:43:29] <toastydeath> maybe I'm overthinking it, and the area is just plain macroscopic area
[20:43:38] <toastydeath> i.e. pi r^2
[20:43:55] <andypugh> And why is that a problem? The whole point of calculus is to love your infinitessimals
[20:44:03] <zeeshan> lol
[20:45:11] <toastydeath> because "dr dr dtheta" with a double integral is problematic
[20:45:12] <andypugh> Formulate the equation in torque for a tiny thin ring of radius r and width dr. Then integrate from 0 to R
[20:45:20] <toastydeath> sigh
[20:45:28] <toastydeath> what i am doing is _exactly_ equivalent to that
[20:45:56] <toastydeath> because a double integral when you proceed around theta first, collapses to the single integral you describe
[20:45:59] <andypugh> OK, so then you need to be much better at maths than me, and solve the equation :-)
[20:46:01] <toastydeath> (which is what i did)
[20:46:26] <toastydeath> ty for looking at it, though - I've been banging my head for three days so all eyes are greatly appriciated
[20:46:45] <toastydeath> i am going to head to the bar with attractive friends and drown my math sorrows, bbl and ty all
[20:46:50] <tjtr33> LeelooMinai, sorry the Qt stuff is machinekit http://doc.qt.io/qt-5/gettingstartedqml.html
[20:47:35] <LeelooMinai> Ok, I think I will not go there - just make this work.
[20:48:49] <LeelooMinai> I wonder how machinists deal with all this. I am a programmer and eventually will succeed, but... :)
[20:50:36] <andypugh> I am neither a machinist nor a programmer. I am not even an engineer. What am I doing here?
[20:51:05] <LeelooMinai> No idea:p
[20:51:16] <Tom_itx> you play with metal
[20:51:54] <andypugh> Ah, yes, I almost have a qualification in metallurgy
[20:52:00] <andypugh> Good point
[20:52:50] <LeelooMinai> I a, just trying to imagine some old-school machinist saying "Hey ho, I will now load linuxcnc and control this bugger" :)
[20:54:06] <andypugh> I think you under-estimate them.
[20:55:11] <LeelooMinai> For their sake, I hope so:)
[20:58:34] <andypugh> If you read the old model-engineering magazines the people are just the same, only the tools have changed
[21:01:43] <Bushman> ave
[21:01:45] <tjtr33> model engineering magazine since 1898 ( slightly older than Dr. Who :)
[21:03:23] <tjtr33> used to travel 4 hrs to goto Evanston IL to get issues in 60's & 70's, then got subscription ( and a job to pay for it )
[21:08:06] <andypugh> toastydeath: I get this: https://picasaweb.google.com/lh/photo/TlS3KSrOuMUxXDixD3JYZyRcCUDYouMZNo_-aopP-bY?feat=directlink (rigor = low)
[21:09:52] <andypugh> Which seems suspiciously too much simpler than yours.
[21:10:33] <zeeshan> where is mr ohm coming from
[21:10:44] <andypugh> angular velocity
[21:10:45] <zeeshan> we usually use small omega
[21:10:53] <zeeshan> to capital :-)
[21:11:11] <andypugh> I couldn’t remember little omega. It’s been too long
[21:11:27] <zeeshan> and btw, envelopes make great paper for math work! :P
[21:13:18] <zeeshan> andypugh: symmetry makes great simplification
[21:13:21] <andypugh> It feels right, linear in separation and speed, but cubic in radius
[21:13:49] <zeeshan> im pretty sure your solution is correct
[21:14:00] <LeelooMinai> Ok, I have lame spindle panel visible... now what? Do I control spindle using G-codes?
[21:14:09] <zeeshan> you're using a line element
[21:14:21] <zeeshan> and evaculating it over the entire radius
[21:14:54] <zeeshan> LeelooMinai: you dont need the panel at all
[21:14:57] <zeeshan> to run your spindle
[21:15:09] <andypugh> LeelooMinai: What do you _want_ your spindle panel to do?
[21:15:16] <zeeshan> when you associate the right hal pins, the +/- in axis makes your spindle work
[21:15:20] <zeeshan> aka m03 etc
[21:15:59] <LeelooMinai> Yes, well, I wired the panel - it's supposed to dpisplay RPM, I know. I just want to control the spindle nwo
[21:16:15] <andypugh> zeeshan: I don’t think I am even doing that. I am simply treating it as a 1D problem and ignoring the geometry of my individual “rings”
[21:16:18] <LeelooMinai> O, +/- I see
[21:16:39] <LeelooMinai> Wait, those are zomm buttons:)
[21:16:40] <andypugh> LeelooMinai: MDI M3 S1000000000
[21:16:59] <zeeshan> you can see the rpm in hal
[21:17:02] <zeeshan> without the panel :p
[21:17:23] <LeelooMinai> andypugh: Is that spindle on command?
[21:18:25] <LeelooMinai> Whhat is that weird S10000000 ?
[21:19:12] <jdh> he has a really fast spindle?
[21:19:34] <andypugh> M3 = spindle on, S1000000000 = 1 meeelion RPM
[21:19:41] <LeelooMinai> Right...
[21:19:58] <LeelooMinai> Can I tell linuxcnc that everything is homed without actually homing?
[21:20:10] <andypugh> No, and yes
[21:20:25] <jdh> you can tell it you don't have to home
[21:20:31] <andypugh> You can’t do it now. But you can configure shuch that you could have done :-)
[21:20:56] <tjtr33> and you dont have to home to use the spindle ( none of the above )
[21:21:16] <LeelooMinai> Well, that M03 command ended with complaint that I have to home first...
[21:21:46] <tjtr33> ? wow didnt know that was wired , good to know
[21:21:49] <LeelooMinai> "can't issue MDI commands when not homed"
[21:21:55] <andypugh> http://www.linuxcnc.org/docs/html/config/ini_config.html#sub:TRAJ-section
[21:22:07] <tjtr33> oh right, but hal can be used instead of mdi
[21:22:23] <tjtr33> you're just testing right?
[21:22:25] <andypugh> OK, you can’t MDI then. But you can click the spindle buttons in Axis (if you are using Axis)
[21:23:06] <andypugh> LeelooMinai: That link_ NO_FORCE_HOMING
[21:23:35] <andypugh> But homing is better, much better. You should at least have an ambition to have home switches.
[21:24:24] <LeelooMinai> I have all that, just want to test spindle only
[21:25:01] <LeelooMinai> O, it works
[21:25:06] <LeelooMinai> Magic
[21:25:07] <tjtr33> weeeeeee
[21:25:27] <LeelooMinai> Well, the lame panel doesn't but the command seems to
[21:25:31] <andypugh> Does it run the other way too?
[21:25:42] <jdh> spindle via modbus only or digital IO?
[21:25:48] <andypugh> Lame panels are lame and un-necessary.
[21:25:50] <tjtr33> you can even run the pyvcp panel without axis, without linuxcnc
[21:26:12] <LeelooMinai> I did m03 s1000 and then s8000 and it changed speed
[21:26:26] <LeelooMinai> So I guess now I will try reverse direction
[21:26:39] <andypugh> M4 S1000 should run the other way, but might not do
[21:26:42] <LeelooMinai> Do I have to issure stop first?
[21:26:49] <andypugh> No, no need.
[21:27:08] <andypugh> But m5 to stop if you want to
[21:27:28] <LeelooMinai> Seems ok - slowed down then changed dir
[21:27:43] <LeelooMinai> Now, the real test:)
[21:27:46] <LeelooMinai> s24000
[21:27:51] <andypugh> That’s good. Quite often reversing takes work.
[21:28:21] <LeelooMinai> Hmm, I wonder if it's supposed to sound like that
[21:28:46] <tjtr33> m5 stop dubious sound. then got a tacho? or display on vfd?
[21:29:29] <LeelooMinai> Yes, vfd has display
[21:31:04] <LeelooMinai> Do those spindles sound like coffee grinders? :)
[21:31:32] <tjtr33> empty or crunching beans?
[21:32:02] <LeelooMinai> Well, it's not a brutal sound - but it's not completely smiith
[21:32:04] <LeelooMinai> smooth
[21:32:24] <tjtr33> where's pastesound.prg ?
[21:32:39] <LeelooMinai> At 24000 it sounds like a sand was inside or something like that:)
[21:33:08] <LeelooMinai> I guess those are bearings, but not sure what is normal or not
[21:33:11] <andypugh> Under no load that _might_ be normal
[21:33:46] <LeelooMinai> Should I try to open it and check if the bearings are greased?
[21:35:17] <andypugh> No, you probably shouldn’t. Unless you know the bearing preload settings, how much grease there should be, and what grade of grease.
[21:35:32] <jdh> does your display show rpm? Mine doesn't seem to
[21:35:53] <LeelooMinai> No, but I did not dobule check everything yet
[21:35:53] <zeeshan> got an encoder?
[21:35:55] <tjtr33> maybe run with load a bit before disassembly (friction) just long enuf to get the sound or not get the sound
[21:37:00] <LeelooMinai> Maybe it needs to run for a bit like engines
[21:38:27] <LeelooMinai> Is it easy to attach encoder to a spindle like that? I don't see where it could go as on the back there's only a cooling fan
[21:39:29] <andypugh> The fan has blades. What are they made of? They sound like an encoder to me.
[21:39:53] <LeelooMinai> What? :p
[21:40:07] <Tom_itx> inductive pickup if they're ferrous
[21:40:09] <tjtr33> its easy to mount encoder like a tool ( and its just for testing/calibration )
[21:40:25] <LeelooMinai> You think the fan rotates at the spindle speed?
[21:40:41] <Tom_itx> it rotates at the motor speed i'd bet
[21:40:47] <tjtr33> mount encoder shaft in the collet/chuck
[21:41:11] <LeelooMinai> I meant for permanent setup
[21:41:24] <tjtr33> nm
[21:41:58] <tjtr33> well once you know the calibration is right, then its a waste of an encoder
[21:42:29] <LeelooMinai> I don't think the spindle is following the frequency/current/etc. exactly
[21:43:04] <tjtr33> andy wrote special comps the linearize the speed given enough samples (iirc)
[21:43:47] <LeelooMinai> I just like the direct approach of an encoder I guess - not much wrong can go there
[21:43:55] <andypugh> Yes, “lincurve” can compensate that, or closed-loop PID control.
[21:44:46] <tjtr33> maybe use a tachometer rpm = volts , cheaper than encoder, more classic too
[21:45:10] <tjtr33> typical like 7V/krpm
[21:45:34] <LeelooMinai> Right, I will look into all options later.
[21:45:49] <LeelooMinai> For today I will only try this needle display to work
[21:46:32] <Jymmm> You can now buy 3D printer from home depot... http://www.homedepot.com/p/Dremel-Idea-Builder-3D-Printer-3D20-01/205448581
[21:47:10] <LeelooMinai> I am kind of tired of all those 3d printers tbh:)
[21:48:08] <Jymmm> We all are
[21:48:55] <Jymmm> They even had samples there. Like I've always said... glorified glue gun.
[21:49:48] <LeelooMinai> People will buy them, then be trolled on materials, like the printer manufactures do with cartridges, and they will gather dust:)
[21:50:27] <Jymmm> NAh, ppl will buy from overseas, it's nothign special, just plastic rod
[21:50:41] <malcom2073> I dunno, I find mine extremely useful
[21:50:43] <CaptHindsight> what about emergencies?
[21:50:51] <LeelooMinai> Well, the manufacturers can make them special:)
[21:51:08] <LeelooMinai> Soecial cartridges etc. - for the consumer market
[21:51:26] <Jymmm> CaptHindsight: ones that can wait 6-12 hours?
[21:51:52] <CaptHindsight> I was planning on going out tonight but I forgot to grow out a mustache the past two weeks. No problem http://www.thingiverse.com/thing:253394 and 3 hours later I'm ready to hit the town
[21:52:07] <Jymmm> LeelooMinai: Nah, look below it, just spools of plastic
[21:52:15] <zeeshan> lol CaptHindsight hahahaha
[21:52:21] <LeelooMinai> Right, 3d-print a carriege for someone who just broke the neck, so he/she can be transported to a hospital:)
[21:52:39] <tjtr33> haha and i was listening to Poirot
[21:53:09] <Jymmm> CaptHindsight: crazy glue and a pair of cisors, done in 60 seconds.
[21:53:10] <malcom2073> Calling people who use 3d printers that, is the same as me calling people who use cnc machines gun nuts. Just because you can use a device to make something, doesn't mean everyone will.
[21:54:23] <malcom2073> Seems that gross misunderstanding of a technology isn't just for the masses :P
[21:54:38] <CaptHindsight> http://www.thingiverse.com/thing:203101 you can't buy these
[21:54:51] <Jymmm> malcom2073: Of ot
[21:55:00] <LeelooMinai> Or you can print a Yoda head - like 10k other people:)
[21:56:18] <Jymmm> s a 480K one, then no problem, it's no longer a glue gun. I'm not looking at the "technology" I'm looking at the final result. The sample I have, which I would suspect would be "the best" aint so pretty inless you are a kid and making a goft for mothers day or something along those lines.
[21:56:27] <Jymmm> $80K
[21:56:37] <malcom2073> Jymmm: No, you're looking at the media
[21:56:58] <LeelooMinai> I call them 3D Blobbers in my mind for some reason - no idea why:)
[21:57:03] <Jymmm> malcom2073: I dont give a poo. it's crap, period.
[21:57:56] <Jymmm> I'm barly flexing it and it's delaminating.
[21:58:02] <malcom2073> Jymmm: Then you're doing it wrong.
[21:58:15] <LeelooMinai> flex gentler
[21:58:35] <Jymmm> malcom2073: Its a glue gun, if you dont like that name, oh well.
[21:59:06] <Jymmm> LeelooMinai: I am, that's the problem.
[21:59:09] <CaptHindsight> https://youtu.be/YN8NZJYboHg?t=1m54s a useful 3d printer
[21:59:59] <malcom2073> I dont' care for name calling, but if that's your thing so be it
[22:00:42] <malcom2073> It just amuses me the amount and source of negativity
[22:01:17] <LeelooMinai> It's to counterbalance amount of naivness? :)
[22:01:38] <malcom2073> Anything that hits mainstream media has a decent amount of both carried along with it heh
[22:02:32] <CaptHindsight> malcom2073: it's just that >90% of the parts printed by FDM are just useless
[22:03:24] <LeelooMinai> Plastic jewelery for small girls works though:)
[22:03:37] <malcom2073> CaptHindsight: It's unfortunate that has been your experience
[22:03:42] <malcom2073> I try for a significantly smaller failure rate :)
[22:03:44] <CaptHindsight> toys stores should have them
[22:04:14] <LeelooMinai> Barbie 3D Printer Deluxe
[22:04:16] <CaptHindsight> malcom2073: I develop 3d printers for a living, just not FDM
[22:04:26] <malcom2073> Why does it anger you guys that people are printing trinkets in their living rooms?
[22:05:11] <LeelooMinai> $1k to print trinkets - I don't know:)
[22:05:37] <malcom2073> LeelooMinai: People spend significantly more than that, to do significantly less :P
[22:05:49] <CaptHindsight> it's also the waste of time and resources on hype
[22:05:56] <Jymmm> malcom2073: That's NOT helping your glue gun case btw.
[22:06:02] <LeelooMinai> Yes, well, doesn't mean I should admire it:)
[22:06:14] <malcom2073> Jymmm: I don't have a case
[22:06:40] <Jymmm> malcom2073: You seem pretty admant on defending the anti-gluegun campaign
[22:06:49] <malcom2073> Jymmm: I have nothing to prove. My 3d printers save me an unbelievable amount of time and money, that's all the justification I need.
[22:06:53] <malcom2073> I'm trying to understand all the anger
[22:07:03] <CaptHindsight> I don't think it angers anyone. It's just an easy target for amusement
[22:07:12] <Jymmm> There is no anger, just no love either.
[22:07:25] <malcom2073> So it's trolling?
[22:07:38] <LeelooMinai> In my case it's not anger at all, ore like "meh, this again":)
[22:07:40] <Jymmm> It's calling what it is.
[22:08:06] <malcom2073> Jymmm: Calling it a glue gun with the intention of it being an insult isn't just calling it what it is, don't treat me that stupid heh
[22:08:09] <CaptHindsight> same for people that go to Katy Perry concerts
[22:08:27] <LeelooMinai> Well, it's pretty close to a glue gun...
[22:08:58] <CaptHindsight> filament fetishists
[22:09:13] <Jymmm> Glue gun = thermal polymer dispenser. $1K 3d printer, glorified thermal polymer dispenser.
[22:09:21] <malcom2073> This has devolved into exactly where I expected it to, so nevermind heh
[22:10:09] <CaptHindsight> when you hear and read the hype about fdm printers
[22:10:27] <Jymmm> Glue gun = manual thermal polymer dispenser. $1K 3d printer = automated thermal polymer dispenser.
[22:10:33] * LeelooMinai imagines malcom2073 petting his 3d printer and saying "Don't worry, you are not really a glue gun"
[22:10:38] <CaptHindsight> 3d printing actually has lots of great applications, just not fdm
[22:11:52] <tjtr33> malcom2073, you're absolutely right about this channel being very neagtive. not often helpful or praising. and its true that toolmakers wont look kindly at one of those extruders. both are true.
[22:12:20] <LeelooMinai> One should make one of those Hitler-meme videos about getting a 3D printer imho
[22:13:17] <malcom2073> tjtr33: It's not even the not being helpful, it's the attitude of distain for anyone who does 3d printing that shocks me. I guess I shouldn't be THAT shocked, this is the internet heh
[22:13:23] <malcom2073> Shouldn't expect miracles of deceny
[22:13:26] <malcom2073> decency*
[22:14:11] <LeelooMinai> I don't see how not mocking questionable use of technology is decent:)
[22:14:39] <Jymmm> malcom2073: Quit bitchin because some think the results of a $1k FDM is shit, else prove me wrong is something you've created that is exceptional in quality
[22:14:46] <CaptHindsight> spend a day in #reprap and then come back to this discussion, I dare you
[22:14:55] <malcom2073> CaptHindsight: I spent quite a few days in there.
[22:15:02] <malcom2073> I agree, they're not helping the case
[22:15:33] <CaptHindsight> it's anti-science, facts, intellectual etc
[22:16:22] <CaptHindsight> it's more poking fun at the absurdity of FDM changing the world
[22:16:28] <Jymmm> malcom2073: Seriously, show pics of things you've made than have came out exceptional in quality that you could sell to retail (as example)
[22:16:41] <LeelooMinai> I never was there - is that channel like @arduino?
[22:16:44] <malcom2073> Jymmm: Why would I do that? That's not what I use it for
[22:16:47] <CaptHindsight> they hijacked the term 3D printing when FDM is just a small subset
[22:16:51] <zeeshan> its a precise glue gun!
[22:17:15] <zeeshan> its like calling LeelooMinai's machine a dremel
[22:17:20] <zeeshan> :p
[22:17:23] * zeeshan hides
[22:17:56] <LeelooMinai> Drenel dremels
[22:17:59] <Jymmm> malcom2073: Your in a channel with ppl that deal in thousands of an inch tolerance, and tryng to finger out why we call them glues gluns? Thats like someone with an airsoft going into S&W "why do you call it a toy?"
[22:18:01] <tjtr33> the low level desire to build is good. i admire that part. but the items produced by reprap et al are not pleasing to machinists. and machinists are not always politically correct
[22:18:26] <zeeshan> i like the parts reprap makes
[22:18:30] <CaptHindsight> #reprap is more a den of snake oil, #arduino is more hobby electronics
[22:18:33] <zeeshan> you can make useful covers out of em!
[22:19:19] <Jymmm> malcom2073: And note I say $1K 3d printers, not 3d printing itself.
[22:19:30] <LeelooMinai> Too bad most arduino people do not know much about eletronics at all:)
[22:19:33] <malcom2073> Jymmm: you're assuming the owner of the airsoft pistol is angry that it's unable to shoot and kill people :P
[22:19:44] <malcom2073> My printer was under $1k
[22:20:16] <Jymmm> malcom2073: I never said an airsoft couldn't be lethal.
[22:20:16] <malcom2073> But if your point is that they can't make injection moulded quality parts, I agree.
[22:20:29] <malcom2073> However, that's not what you said.
[22:20:49] <Jymmm> malcom2073: it's an autmated glue gun,
[22:21:10] <CaptHindsight> http://www.3ders.org/articles/20150309-boxzy-an-all-in-one-3d-printer-laser-etcher-cnc-machine-launches-on-kickstarter.html is another example
[22:21:24] <malcom2073> Jymmm: There's no reason to continue talking about it.
[22:21:25] <CaptHindsight> $2500 for this
[22:21:35] <malcom2073> So I'll drop it.
[22:21:54] <LeelooMinai> CGGC - Computer Glue Gun Control - Machining:)
[22:22:15] <malcom2073> Amusingly enough, reprap is what got me into CNC in the first place :)
[22:22:21] <jdh> we have a laser sinter one at work. you have to send the result out to be ground if you want anything with a good finish
[22:22:26] <zeeshan> ROFL LeelooMinai
[22:26:43] <CaptHindsight> the majority of FDM suppliers are snake oil, why do they deserve any respect?
[22:27:14] <CaptHindsight> that's one of the problems, they should be called out for their scams
[22:27:57] <Jymmm> CaptHindsight: I like the blonde in the link you gave
[22:28:05] <CaptHindsight> but I guess that's the Amerikan way now, if it makes a profit then it's ok
[22:28:52] <Jymmm> CaptHindsight: Don't you know, we are a disposable society and we buy disposable stuff, like cars and such
[22:31:36] <furrywolf> I do think the usefulness of glue-gun 3d printers is rather overstated... but I also think they're very useful in getting people interested in manufacturing, design, etc.
[22:40:51] <zeeshan> hi furrywolf
[22:41:02] <zeeshan> http://i.imgur.com/QSAN0SX.png
[22:41:04] <zeeshan> tada
[22:41:18] <CaptHindsight> is working?
[22:41:29] <zeeshan> if you multiply by the max scale value by that bitmap matrix
[22:41:32] <zeeshan> it seems to give good values
[22:41:32] <zeeshan> yes
[22:41:35] <furrywolf> did you extract scale data?
[22:41:43] <zeeshan> no
[22:41:48] <zeeshan> ill do it on monday or tomorrow
[22:41:57] <zeeshan> ill generate an image w/ same conditions, except scale
[22:43:05] <furrywolf> try to make them as identical as possible
[22:43:06] <zeeshan> the bad news is
[22:43:10] <zeeshan> i can't extract strain from them
[22:43:18] <furrywolf> ?
[22:43:21] <zeeshan> cause i noticed, the stuff moves in the xy plane too. not just z.
[22:43:29] <zeeshan> and i have no dots or anything to track.
[22:44:04] <zeeshan> the lines look like they stay fairly stationary and only move in Z.
[22:44:17] <zeeshan> maybe i can use them to get an "overall" strain
[22:44:20] <CaptHindsight> zeeshan: how did you end up making the patterns?
[22:44:26] <CaptHindsight> spray?
[22:44:34] <zeeshan> CaptHindsight: this is w/ optical microscope
[22:44:45] <zeeshan> no pattern yet
[22:44:49] <furrywolf> laser etch a grid pattern into your material, see how it distorts when pressed?
[22:44:59] <zeeshan> furrywolf: thats the big challenge on this matterial
[22:45:02] <zeeshan> you can't do that
[22:45:11] <zeeshan> you severely change the film properties
[22:45:12] <furrywolf> why not? :P
[22:45:17] <CaptHindsight> zeeshan: post pics when you do if possible
[22:45:22] <furrywolf> film? you have a weird material.
[22:45:30] <zeeshan> its a sandwich of materials
[22:45:33] <zeeshan> film, adhesive, metal
[22:45:44] <zeeshan> CaptHindsight: will do :)
[22:45:54] <zeeshan> CaptHindsight: might not be until summer till i get back to that work again
[22:46:20] <CaptHindsight> zeeshan: can I visit this summer?
[22:46:31] <zeeshan> ofcourse!
[22:46:32] <furrywolf> LOL! the ebay seller that offered me a $3 refund for my broken light, after I opened a case on ebay, is now offering me $8.
[22:46:42] <furrywolf> maybe he thinks if he pisses me off more, I'll be nicer?
[22:46:52] <zeeshan> how much was the light
[22:47:08] <furrywolf> $27
[22:47:17] <zeeshan> furrywolf: you get full credit for discovering the depth info in the file
[22:47:23] <zeeshan> i really wanan rub it in that guys face on monday
[22:47:36] <furrywolf> heh
[22:49:30] <furrywolf> 59cd10 1792.6854 1344.514 35.643032 0
[22:49:33] <furrywolf> 66d5c0 -2.912565e-25 -23183.521 35.64874 9148902
[22:49:39] <furrywolf> grr, wrong paste
[22:49:45] <furrywolf> 59cd10 1792.6854 1344.514 35.643032 0
[22:49:49] <furrywolf> 59cd10 1792.6854 1344.514 152.03207 0
[22:50:11] <furrywolf> recognize a number there?
[22:50:39] <zeeshan> hjahaha
[22:50:41] <zeeshan> HAHJAHAH
[22:50:51] <CaptHindsight> https://www.kickstarter.com/projects/1853707494/pancakebot-the-worlds-first-pancake-printer
[22:50:55] <zeeshan> wait, which file is this
[22:51:02] <zeeshan> 001a.tif?
[22:51:10] <furrywolf> one is 001a, one is 025a
[22:51:44] <zeeshan> for 25a the 35.64 is the max scale value.
[22:51:50] <furrywolf> yep
[22:51:56] <furrywolf> and?
[22:52:24] <zeeshan> i dont see a 152 something in there
[22:52:29] <zeeshan> nm i do.
[22:52:30] <zeeshan> :)
[22:52:42] <zeeshan> that must b e the scale!
[22:52:48] <zeeshan> for x y and z.
[22:52:49] <furrywolf> same offset in both files, even.
[22:52:49] <furrywolf> yep
[22:53:01] <furrywolf> I don't know about x and y... those numbers could be anything. but it's possible.
[22:53:12] <zeeshan> whatd you do? use od to convert to float and search for the number?
[22:53:17] <furrywolf> yep
[22:53:34] <furrywolf> od -Ax -t f4 001a.tif | grep "152\.03"
[22:54:07] <zeeshan> nice :)
[22:54:24] <zeeshan> so the starting address is
[22:54:34] <zeeshan> erm lemme search it
[22:54:49] <furrywolf> the starting address is the first thing on the line.
[22:54:51] <furrywolf> in hex.
[22:56:17] <furrywolf> echo "Z scale:"$(od -An -t f4 -j 0x59cd18 -N 4 001a.tif)"um"
[22:58:02] <zeeshan> im pretty sure the 0 refers to min values.
[22:58:05] <zeeshan> *value
[22:58:30] <furrywolf> identifying a non-changing "0" in an unknown file is pretty impossible.
[23:01:22] <furrywolf> if=025a.tif ; convert -depth 32 -size 400x300+5885308 -define quantum:format=floating-point gray:$if[0] $if.png ; tail -c +6365335 $if >$if.jpg ; echo "Z scale:"$(od -An -t f4 -j 0x59cd18 -N 4 $if)"um" added that to my one-line image splitter
[23:02:06] <furrywolf> that address, by the way, is in the space between the end of the tiff file and the start of the bitmap
[23:05:40] <ssi> eesh
[23:05:56] <zeeshan> they must have a crap load of info there
[23:06:15] <zeeshan> its only a couple bytes before the start
[23:09:38] <jdh> that is one spiffy convert line
[23:12:05] <zeeshan> hes advanced linux user! :p
[23:12:21] <jdh> indeed
[23:12:25] <zeeshan> its crazy how strong those commands are
[23:17:26] <tiwake> I looked around and couldent find an open source washing machine controller
[23:20:10] <furrywolf> ... open source washing machine controller?
[23:20:21] * furrywolf hands tiwake a timer with a clear plastic window stuck on the back
[23:23:43] <furrywolf> http://fw.bushytails.net/broken%20screw%20head%2001.jpg first time I've had a screw break in the middle of the head, instead of the narrow shaft
[23:24:00] <furrywolf> (warning: macro mode abuse)
[23:37:20] <tjtr33> tiwake, include a notifier ( email or SMS ). then move onto other machines that silently wait for you. Hal could do it!
[23:48:07] <Jymmm> furrywolf: Your pentac camera takes some really great shots
[23:50:23] <furrywolf> I've had a few, and have been quite happy with them. a couple of them fit in an altoids tin. :)
[23:52:12] <Jymmm> I'm wanting another DSLR (last one got stolen), but not sure I want to spend that much this time.
[23:52:45] <furrywolf> I can't afford a dslr.
[23:52:53] <furrywolf> the pentax is a mini point-and-shoot.
[23:53:09] <Jymmm> Yeah, I looked it up after you lightbulb photo
[23:53:44] <Jymmm> Pentax Optio S10
[23:54:19] <Jymmm> http://www.amazon.com/Pentax-Optio-Digital-Camera-Optical/dp/B0013V5F14