#emc | Logs for 2006-06-15

[00:14:32] <CIA-8> 03flo-h 07HEAD * 10emc2/src/po/de.msg: updated
[00:27:08] <CIA-8> 03jepler 07HEAD * 10emc2/src/hal/utils/ (scope_disp.c scope_usr.h):
[00:27:08] <CIA-8> fix bugs in trace-selecting code.
[00:27:08] <CIA-8> record whether the trace or its baseline is selected.
[00:27:08] <CIA-8> make dragging the baseline move it up and down, snapping to the "divs".
[00:27:08] <CIA-8> mekd dragging the trace change the scale.
[00:29:46] <jepler> "mekd"? should learn how to type.
[00:30:03] <cradek> yeah I wondered about mekd.
[00:39:22] <Rugludallur> Does anyone know of anything that would prohibit you from connecting one pid output and one axis motor pos to the stepgen-position-cmd at the same time ?
[00:46:07] <dmessier> '
[00:46:23] <jepler> Rugludallur: yes, an input pin can only have one signal connected. Depending on your goal, you probably want to use a "mux" or a "sum2" block to combine the two inputs.
[00:46:43] <Rugludallur> ok thx
[00:47:38] <Rugludallur> I am linking in a Torch Height Control for a Z axis and I keep getting ferror
[00:48:58] <jepler> that's because the feedback will differ from the commanded Z position
[00:49:10] <fenn_shop> you could: 1) increase the ferror or 2) (hang on)
[00:49:58] <cradek> if you add something to emc's position output, subtract that same thing back out of the feedback
[00:50:02] <jepler> hmph, I broke something in halscope before I made that checkin
[00:50:19] <Rugludallur> yup, thats what I figured, I can hook up both components to the same fb which I have done, but I was not sure if signals commanded via one signal would be propegated to the other signal via the pin
[00:50:56] <fenn> arg i hate not being able to start keystick through ssh
[00:52:54] <CIA-8> 03jepler 07HEAD * 10emc2/src/hal/utils/scope_disp.c:
[00:52:54] <CIA-8> re-fix selection
[00:52:54] <CIA-8> improve trace scaling by dragging on the display area (drag away from the
[00:52:54] <CIA-8> baseline to scale one way, drag towards to scale the other)
[00:57:45] <jepler> whee
[00:57:47] <jepler> now I think it's right
[00:58:05] <cradek> yay
[01:01:54] <fenn> Rugludallur: what do you have hooked up to axis.2.f-error?
[01:02:42] <fenn> oh damn we really need a hal gui
[01:02:47] <Rugludallur> I have stepgen.2.position-fb
[01:03:09] <Rugludallur> and pid.0.feedback
[01:05:38] <Rugludallur> sorry, I misread your q
[01:06:31] <Rugludallur> nothing hooked up to the axis.2.f-error
[01:08:01] <fenn> ok i couldnt remember if it was just there for use with halscope
[01:08:47] <Rugludallur> does it give the diff of cmd/fb ?
[01:10:53] <fenn> ah it's a parameter
[01:12:01] <fenn> that doesn't make any sense
[01:16:36] <fenn> Rugludallur: could you put your hal file on pastebin.com?
[01:17:57] <Rugludallur> will do
[01:17:58] <Rugludallur> just a sec
[01:18:04] <Rugludallur> Im cleaning it up
[01:22:19] <Rugludallur> http://pastebin.com/709840
[01:23:37] <Rugludallur> There are some errors, which are not related to the problem but you should be able to see what I was trying to do
[01:27:54] <fenn> hang on i'm still parsing the .hal file :)
[01:28:35] <Rugludallur> :D
[01:31:02] <fenn> this constant stuff won't be in the final implementation will it?
[01:32:49] <CIA-8> 03jepler 07HEAD * 10emc2/src/hal/utils/scope_disp.c: left shift-drag on trace move baseline. it's hard to grab the baseline you want when a bunch of them are on the same div
[01:38:13] <fenn> Rugludallur: i'm not sure what you're trying to do with the destZ up/down stuff because you can't know the position you want to be at ahead of time
[01:38:55] <fenn> if you knew the position you wouldnt need the THC right?
[01:45:11] <fenn> Rugludallur: Loads the dallur-advanced.hal file first to set up motion and stepgen - this is probably telling Z to stay put, but your constant.0 is telling Z to move somewhere
[01:45:38] <fenn> is Rugludallur still there?
[01:45:41] <Xaver> yup
[01:45:46] <Xaver> got disconnected but im back
[01:46:28] <fenn> ok - dallur-advanced.hal has axis.0 hooked up to stepgen right?
[01:46:54] <fenn> and motion is expecting axis.0 to stay put, so it throws an ferror when you make Z move with constant.0
[01:47:15] <Xaver> in this case axis 2 is hooked up to stepgen and
[01:47:16] <Xaver> pid
[01:47:22] <fenn> sorry, axis.2 i mean
[01:47:43] <Xaver> so when I use constant to make pid move axis2 stepgen is probably give ferror
[01:48:12] <Xaver> I was not sure if one cmd signal would update the other via the pin or not
[01:48:48] <fenn> no a read pin will only read
[01:49:01] <fenn> a pin is just a pointer to a register (the signal)
[01:49:51] <fenn> you have to switch between the two sources of control, or add them together
[01:50:00] <fenn> otherwise they will fight and get confused
[01:50:49] <fenn> that is what jepler was talking about using mux or sum2 blocks
[01:51:08] <Xaver> yp
[01:51:10] <Xaver> yup
[01:51:22] <Xaver> I'm working on MUX,
[01:52:02] <fenn> ok, so.. what passes control of the Z axis over to the THC?
[01:52:03] <Xaver> I can use the bit signal from the thc to select the active one and since
[01:52:24] <Xaver> z is never moved from gcode there should be no conflict
[01:52:37] <Xaver> Xaver is now known as Rugludallur
[01:53:34] <fenn> if Z is always slaved to the THC then it shouldnt be hooked up to axis.2
[01:53:57] <Rugludallur> well I still want to be able to jog
[01:54:17] <fenn> do "halcmd show func"
[01:54:32] <fenn> funct
[01:55:36] <fenn> i mean, show thread
[01:56:06] <fenn> the order of functions in the thread determines who has control of the axis
[01:56:12] <Rugludallur> ok
[01:56:38] <fenn> if motion-controller is first, the gcode output will win and Z will stay at 0
[01:57:37] <fenn> if pid (?) is first it will win and move the axis, but you will get an ferror because motion didn't expect the axis to move
[01:57:57] <fenn> er, the other way around
[01:58:01] <fenn> whoever is last wins
[01:58:42] <Rugludallur> hold on, I'm commenting out all the mux :D
[01:59:06] <fenn> no you need the mux to switch between "THC" mode and "jog" mode
[01:59:18] <fenn> you dont want either one to win, you want them to cooperate :)
[01:59:29] <fenn> mux lets you control which one has control of the axis
[01:59:45] <Rugludallur> yes, but im commenting out to show threads
[01:59:50] <Rugludallur> cause there is an error :P
[02:02:54] <Rugludallur> stepgen.update-freq is first so pid should not move the axis however if I jog the axis pid will give error
[02:03:25] <Rugludallur> makes sense, since both fb are hooked up and only one cmd can connect to the pin, .....
[02:05:01] <fenn> i think turning up ferror would be good just so you can see what's going on
[02:05:21] <fenn> you can monitor ferror with halscope and it will still work
[02:05:35] <fenn> or halmeter even
[02:05:39] <Rugludallur> yup
[02:05:57] <Rugludallur> I turned ferror up to 1500 earlier , still got ferror as soon as I jogged for some reason
[02:06:13] <fenn> turn up min-ferror in the ini file
[02:06:21] <Rugludallur> yup
[02:06:38] <Rugludallur> hmm min-ferror
[02:06:45] <Rugludallur> I turned up ferror , sorry
[02:06:51] <fenn> FERROR only applies at the maximum velocity
[02:07:01] <Rugludallur> ahh
[02:07:05] <fenn> the total error allowed is MIN_FERROR + velocity*FERROR
[02:07:14] <fenn> or something like that
[02:07:17] <Rugludallur> :D thanks
[02:07:25] <Rugludallur> for the expl.
[02:07:49] <cradek> max(min_ferror, (velocity/maxvel)*ferror)
[02:07:49] <fenn> MIN_FERROR + velocity*FERROR/MAX_VELOCITY
[02:08:12] <Rugludallur> One though, if I mux the cmd for the movement, would I not also have to mux the fb and switch it with the cmd to avoid ferrors ?
[02:09:56] <fenn> this is a stepper motor right? so why do you even have feedback? ( cradek is that the normal way of doing it? )
[02:10:10] <fenn> * fenn digs through sample configs
[02:10:38] <Rugludallur> I think there is always an "emulated" feedback
[02:11:51] <fenn> right
[02:11:54] <cradek> stepgen provides feedback
[02:12:27] <fenn> what is pid feedback supposed to do?
[02:15:06] <fenn> Rugludallur: why are you using pid again?
[02:15:32] <Rugludallur> well it's the only way I know of for me to
[02:15:42] <fenn> cant you just hook a constant to stepgen?
[02:16:13] <Rugludallur> yes
[02:16:21] <Rugludallur> now that I am starting to get to know blocks better
[02:16:43] <Rugludallur> create mux with 0 and 10 and control with the bit signal
[02:16:54] <Rugludallur> hook it up to stepgen
[02:17:18] <Rugludallur> So there is a better way to do this :D
[02:17:59] <fenn> no i might just be confused
[02:18:10] <Rugludallur> my inital problem was that pid was the only way i could find to take the bit signal from the parallel port and use it create velocity
[02:18:52] <fenn> what happens if you hook that bit signal directly to stepgen.2.position-cmd?
[02:19:22] <fenn> 1 = up 0 = down?
[02:19:28] <fenn> or is it more complex than that?
[02:19:33] <Rugludallur> I have 2 bit signals
[02:19:50] <Rugludallur> 2 pins
[02:19:55] <Rugludallur> each can be low or high
[02:20:41] <Rugludallur> and I think position-cmd only takes actual positions in mm/" so I would need to convert one bit to 300 and the other to 0
[02:20:42] <Rugludallur> or so
[02:21:00] <Rugludallur> and the input into position-cmd has to be float if I am not mistaken
[02:21:46] <fenn> i see
[02:23:16] <CIA-8> 03jepler 07HEAD * 10emc2/src/hal/utils/ (scope_disp.c scope_trig.c): display and allow manipulation of trigger level and direction in the display
[02:33:36] <CIA-8> 03compile-farm 07BDI-TNG (2.4.18-rtai) * 10emc2head/: build FAILED ; see http://linuxcnc.org/compile_farm/emc2head_slot3_log.txt
[02:33:55] <CIA-8> 03compile-farm 07BDI-4.20 (2.6.10-adeos) * 10emc2head/: build FAILED ; see http://linuxcnc.org/compile_farm/emc2head_slot5_log.txt
[02:35:07] <jepler> uh oh
[02:36:21] <jepler> oh, it's a "duh" error
[02:36:24] <jepler> I'll have it fixed shortly
[02:38:15] <CIA-8> 03jepler 07HEAD * 10emc2/src/hal/utils/scope_disp.c:
[02:38:15] <CIA-8> snap to half-divs in addition to divs
[02:38:15] <CIA-8> fix for pre-c99 compilers
[02:38:15] <CIA-8> tweak of arrow size for small windows
[02:39:46] <Rugludallur> fenn: if I am understanding this correctly than axis generates velocity, directs it into stepgen which makes dir/steps which I send out of the parallel port
[02:41:11] <jepler> the input of stepgen is position. if you want to input velocity, then you want freqgen
[02:41:58] <Rugludallur> yes sorry, my mistake
[02:42:23] <Rugludallur> thank you
[02:46:15] <jepler> so you may want to use freqgen for the "Z" axis .. get a velocity of -1, 0 or 1 from the 2 bits and that's the input of freqgen
[02:46:24] <CIA-8> 03compile-farm 07BDI-TNG (2.4.18-rtai) * 10emc2head/: build FAILED ; see http://linuxcnc.org/compile_farm/emc2head_slot3_log.txt
[02:46:56] <jepler> argh
[02:47:22] <fenn> aaarrrggghhh! :)
[02:48:09] <fenn> http://www.cs.unm.edu/~dlchao/flake/argh/index.html
[02:48:38] <CIA-8> 03jepler 07HEAD * 10emc2/src/hal/utils/scope_disp.c: remove unused variables
[02:48:54] <jepler> interesting
[02:48:58] <CIA-8> 03compile-farm 07BDI-4.20 (2.6.10-adeos) * 10emc2head/: build FAILED ; see http://linuxcnc.org/compile_farm/emc2head_slot5_log.txt
[02:51:58] <skunkworks> bed time - night'
[02:52:03] <jepler> see you skunkworks
[02:59:04] <CIA-8> 03compile-farm 07BDI-TNG (2.4.18-rtai) * 10emc2head/: build FAILED ; see http://linuxcnc.org/compile_farm/emc2head_slot3_log.txt
[02:59:07] <CIA-8> 03compile-farm 07BDI-4.20 (2.6.10-adeos) * 10emc2head/: build FAILED ; see http://linuxcnc.org/compile_farm/emc2head_slot5_log.txt
[02:59:51] <CIA-8> 03jepler 07HEAD * 10emc2/src/hal/utils/scope_disp.c: remove extra semicolon
[02:59:52] <jepler> aha, now I spotted it!
[03:05:38] <jepler> I'm off to bed too .. if the farm is still partially red in the morning, somebody bug me about it.
[03:05:59] <fenn> nite nite
[03:19:13] <CIA-8> 03compile-farm 07BDI-TNG (2.4.18-rtai) * 10emc2head/: build PASSED
[03:19:35] <Rugludallur> I have to get some sleep, thanks for all the help fenn, jepler and cradek
[03:26:03] <CIA-8> 03compile-farm 07BDI-4.20 (2.6.10-adeos) * 10emc2head/: build PASSED
[03:27:17] <fenn> whee!
[04:44:32] <Jymmm> whee?
[05:39:42] <A-L-P-H-A> who's where? why?
[05:39:53] <A-L-P-H-A> what? when? how?
[06:29:41] <Jymmm> here and now
[06:49:18] <A-L-P-H-A> playing that silly game... scorched 3d against hte computer
[06:51:55] <Jymmm> http://www.netaffilia.com/ad/electronics/frys/i/2006/06/14/17419.html
[06:57:38] <A-L-P-H-A> HEY. that's a nice deal.
[06:57:55] <Jymmm> I read one sites reviews, not bad - no paraport on it
[06:58:11] <A-L-P-H-A> PCI parallel port then... $20CDN
[06:58:32] <Jymmm> Yeah. Do you know what SLI is?
[06:58:42] <A-L-P-H-A> when you connect multiple cards together.
[06:58:43] <Jymmm> something about video I think
[06:58:46] <A-L-P-H-A> forget what it stands for.
[06:58:52] <Jymmm> ???????
[06:58:53] <A-L-P-H-A> works for video cards, and something else too...
[06:59:04] <Jymmm> explain
[06:59:12] <A-L-P-H-A> SLI when you connected two PCIe (or PCI) cards together with a ribbon or cable.
[06:59:28] <A-L-P-H-A> Scalable Link Interface
[06:59:32] <A-L-P-H-A> http://en.wikipedia.org/wiki/Scalable_Link_Interface
[06:59:40] <Jymmm> looking...
[06:59:44] <A-L-P-H-A> 3dFX used to do it...
[07:00:03] <Jymmm> link two crts to "split" the video in half ?
[07:00:06] <Jymmm> like
[07:00:33] <A-L-P-H-A> nono.
[07:00:59] <A-L-P-H-A> it's like one card, does one frame, the other card does the next... OR... one card does one half of the frame, and the other card does the other half.
[07:01:52] <Jymmm> Oh... like dual cpu's, but for video
[07:02:14] <A-L-P-H-A> yeah
[07:02:21] <Jymmm> got it.
[07:03:00] <Jymmm> Not a gamer, so I could care less =)
[07:03:19] <Jymmm> I've used my 16MB ATI AIW for 6+ years thus far =)
[07:04:40] <A-L-P-H-A> heh... you couldn't even play scorched 3d earth I think.
[07:04:41] <Jymmm> it has RAID too, but now I here some mobo's are using BIOS RAID - bastards
[07:04:55] <A-L-P-H-A> they ALL are.
[07:04:59] <A-L-P-H-A> it's a cheap solution...
[07:05:04] <A-L-P-H-A> it's software raid...
[07:05:04] <Jymmm> Oh I could... I can run SolidWorks
[07:05:09] <A-L-P-H-A> nvidia, intel, and via.
[07:05:25] <A-L-P-H-A> yeah, with text mapping, 3d water?
[07:05:28] <Jymmm> This mobo is HW raid (in my current system), but was one of the firsts too
[07:07:12] <Jymmm> I'm just wondering if the onboard video is shared memory and if emc would choke on it.
[07:07:39] <Jymmm> reaidng manual now...
[11:48:25] <jepler> oh good, the farm is green
[11:50:36] <alex_joni> yup
[11:51:11] <alex_joni> jepler: short question
[12:21:42] <alex_joni> jepler: any experience with minicom?
[12:29:32] <jepler> alex_joni: nope
[12:30:44] <alex_joni> ok.. thanks ;)
[12:30:57] <alex_joni> or maybe you know another program to save a log from a /dev/ttySx
[12:35:13] <jepler> beyond doing slip and ppp back in the day I know next to nothing about serial on unix
[12:35:26] <alex_joni> heh.. ok ;)
[12:35:30] <jepler> I'd say something that probably doesn't work like "stty 19200 < /dev/ttyS0; cat /dev/ttyS0 > log"
[12:35:31] <alex_joni> not much different here
[12:36:23] <jepler> * jepler looks
[12:36:27] <jepler> there's not even a serial port on this laptop
[12:37:02] <alex_joni> heh
[12:37:12] <skunkworks> telnet://towel.blinkenlights.nl/
[12:37:18] <alex_joni> skunkworks: yeah
[12:37:23] <alex_joni> not exactly what I'm after
[12:37:36] <skunkworks> :)
[12:37:53] <skunkworks> got thats been up forever.
[12:38:00] <skunkworks> god
[12:42:17] <alex_joni> right, but there's an ipv6 one now, in colour
[12:45:14] <alex_joni> jepler: that seems to work :D
[12:45:20] <alex_joni> any way to append to the file?
[12:45:28] <alex_joni> instead of rewrite?
[12:48:47] <skunkworks> jepler: what are your thoughts on matts tp?
[12:49:36] <skunkworks> is that part of bdi or what? (I don't know who matt is)
[12:49:48] <alex_joni> skunkworks: matt helped with the emc on the BDI
[12:50:03] <alex_joni> he sent some patches (including a new driver called quickstep)
[12:50:59] <skunkworks> cool
[12:53:09] <skunkworks> what kind of insentive would it take for him to help on emc2 - some sort of signing bonus?
[12:53:34] <alex_joni> he already looked at emc2 and decided to work on emc1-bdi
[12:53:46] <skunkworks> oh - one of those :)
[12:54:16] <skunkworks> Atleast he seems to be helpful
[12:54:29] <skunkworks> throwing out ideas
[12:59:30] <jepler> skunkworks: yeah, and if you install a bdi system or go to the secret URL, you can get a .tar.gz of the source
[12:59:41] <jepler> skunkworks: I like Matt Timmermans, he seems like a good guy
[13:00:04] <jepler> it looks like it would be at most a weekend of work to hook up his trajectory planner to emc2
[13:06:53] <skunkworks> Does the tp look promising? what does it do differnt?
[13:07:19] <alex_joni> skunkworks: lookahead it seems
[13:07:55] <skunkworks> isn't that sort of what jeplers addition does?
[13:08:10] <skunkworks> as far as blending multible segments.
[13:08:11] <alex_joni> jepler might know better ;)
[13:08:15] <skunkworks> yah
[13:08:17] <skunkworks> :)
[13:08:48] <skunkworks> alex - fenn is also running your kernel
[13:09:22] <Lerneaen_Hydra> 'afternoon
[13:09:25] <alex_joni> yah, I read them logs
[13:10:58] <Lerneaen_Hydra> alex_joni: there was something I was going to fix with the wiki, do you remember what it was?
[13:11:17] <Lerneaen_Hydra> IIRC it was the edge & back angle, and what they are based on
[13:21:28] <alex_joni> Lerneaen_Hydra: think you spoke with cradek
[13:21:53] <Lerneaen_Hydra> alex_joni: yes, but I just wanted to make sure I remembered correctly ;)
[13:22:45] <alex_joni> Lerneaen_Hydra: unforch I don't really know
[13:22:52] <alex_joni> maybe you can see it in the logs
[13:22:55] <alex_joni> logger_aj: bookmark
[13:22:55] <alex_joni> See
[13:22:58] <Lerneaen_Hydra> yeah, I'll check
[13:23:11] <Lerneaen_Hydra> that's the wrong day
[13:23:18] <Lerneaen_Hydra> is the previous day saved too?
[13:24:07] <alex_joni> Lerneaen_Hydra: sure
[13:24:12] <alex_joni>
[13:24:29] <Lerneaen_Hydra> yep, it was the angle
[13:24:39] <Lerneaen_Hydra> where it's measured from
[13:25:03] <Lerneaen_Hydra> oh, I read in the log that the chips.ngc file didn't compensate for gouging
[13:25:38] <Lerneaen_Hydra> if there is a wireframe or solid model of it I can generate gcode that compensates for the tool radius and doesn't gouge
[13:31:57] <Lerneaen_Hydra> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Lathe_Advanced_Features
[13:32:08] <Lerneaen_Hydra> if you look at the image in gouging protection
[13:32:13] <Lerneaen_Hydra> is that understandable?
[13:32:58] <alex_joni> Lerneaen_Hydra: probably not for me.. but others might understand
[13:33:09] <SWPadnos> I don't think anyone has a model for chips
[13:33:11] <Lerneaen_Hydra> ok
[13:33:13] <alex_joni> kidding.. looks great
[13:33:32] <SWPadnos> it was probably made from a bitmap, using height mapping (then manually fiddled with)
[13:33:37] <Lerneaen_Hydra> SWPadnos: hmm, thats to bad. how was chips.ngc generated?
[13:33:41] <Lerneaen_Hydra> SWPadnos: oh...
[13:33:45] <Lerneaen_Hydra> alex_joni: ;)
[13:41:32] <jepler> there's a 3d "tux penguin" model here: http://tuxaqfh.cvs.sourceforge.net/tuxaqfh/tuxaqfh/models/tuxedo.ac?revision=1.1&view=markup
[13:42:01] <Lerneaen_Hydra> what is the file-format?
[13:42:07] <alex_joni> .ac
[13:42:09] <Lerneaen_Hydra> .ac? I don't recognise that
[13:42:49] <cradek> wow, what a terrible file format
[13:42:56] <alex_joni> http://www.ac3d.org/
[13:43:04] <SWPadnos> looks like it's missing some textures though
[13:43:14] <Lerneaen_Hydra> not that it matters in this case
[13:43:20] <SWPadnos> actually, for a 3D model, that's not a bad format
[13:43:28] <cradek> you specify the texture file for every triangle!
[13:43:34] <SWPadnos> consider that diff and cvs can actually be used
[13:43:51] <SWPadnos> well, some grouping would be nice ;)
[13:43:57] <cradek> if you rename the texture file you'll get a million line diff
[13:44:03] <alex_joni> lol
[13:44:10] <SWPadnos> heh - there is that
[13:44:22] <cradek> but it looks really easy to read naively
[13:44:45] <SWPadnos> unfortunately, cvs and diff can't tell when there's a change that doesn't change anything, such as reordering two independent items
[13:45:48] <skunkworks> nice
[13:45:57] <cradek> http://timeguy.com/cradek-files/stl-to-dxf/stl-to-dxf.py
[13:46:05] <cradek> a very quick hack to this will convert it to dxf 3dfaces
[13:46:14] <cradek> then toolpath could load it
[13:46:50] <Lerneaen_Hydra> ooh, sounds nice
[13:46:52] <jepler> cradek: get cracking!
[13:47:02] <skunkworks> toolpath?
[13:47:03] <Lerneaen_Hydra> but I can't run .py on my windows boxxen
[13:47:12] <SWPadnos> sure you can ;)
[13:47:21] <cradek> skunkworks: toolpath is the program I wanted to show everyone at workshop but never did
[13:47:26] <Lerneaen_Hydra> not right now, without adding stuff ;)
[13:47:35] <SWPadnos> http://www.python.org/download/windows/
[13:47:49] <skunkworks> cool - I remember you mentioning that
[13:47:56] <SWPadnos> I saw it, it's quite impressive
[13:49:08] <Lerneaen_Hydra> so.. who wants to make a dxf for lazy little me? ;)
[13:49:16] <Lerneaen_Hydra> oh, btw, cradek, check the wiki
[13:49:17] <Lerneaen_Hydra> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Lathe_Advanced_Features
[13:49:22] <cradek> ok I will soon
[13:50:13] <jepler> http://blog.modernmechanix.com/2006/06/14/worlds-first-calculator-watch-the-hp-01/
[13:51:42] <cradek> because of the original price those are quite uncommon today
[13:52:00] <cradek> I could have sworn they made a solid gold one too, but this article doesn't say that
[13:53:15] <cradek> http://www.vintagecalculators.com/html/pulsar_calculator_watch1.html
[13:53:29] <cradek> maybe I'm thinking of this one ($3950 for gold)
[13:55:08] <alex_joni> http://www.stack.nl/~crypton/pguin/Lpenguin.zip
[13:55:21] <alex_joni> that's a 3D tux in 3ds
[13:55:50] <Lerneaen_Hydra> ooh, thanks
[13:56:32] <alex_joni> wasn't exactly easy to find, they changed the URL
[13:57:18] <Lerneaen_Hydra> eugh
[13:58:29] <alex_joni> http://www.acmesystems.it/?id=21
[13:58:31] <Lerneaen_Hydra> alex_joni: he's one'eyed :/
[13:58:45] <alex_joni> really?
[13:58:50] <alex_joni> I didn't open it..
[13:58:50] <SWPadnos> pirate tux arrrrrrrr
[13:58:58] <alex_joni> harr harr
[13:59:50] <Lerneaen_Hydra> hmm, could you test opening it to see if my program is b0rked?
[13:59:50] <alex_joni> I like this one: http://www.stack.nl/~crypton/pguin/TuXperience~1.jpg
[13:59:57] <alex_joni> Lerneaen_Hydra: nothing here to open it
[14:00:15] <SWPadnos> heh - nice image
[14:00:36] <Lerneaen_Hydra> sitting was not b0rked, but it's harder to machine
[14:00:52] <Lerneaen_Hydra> unless you've got a 5 axis machine with a small small bit
[14:01:03] <skunkworks> isn't that saying linux sucks but windows sucks more?
[14:01:14] <alex_joni> lol
[14:01:35] <Lerneaen_Hydra> I'll test sitting for now
[14:05:33] <alex_joni> Lerneaen_Hydra: what SW are you using?
[14:05:45] <Lerneaen_Hydra> hmm, I think I'm going to try cradek's script and the .ac
[14:05:53] <cradek> http://timeguy.com/cradek-files/emc/tuxedo.png
[14:05:57] <Lerneaen_Hydra> I'm having trouble getting the file as a .dxf
[14:06:08] <cradek> here's what he looks like in toolpath (without the textures)
[14:06:16] <Lerneaen_Hydra> alex_joni: edgecam for the cam bit
[14:06:17] <alex_joni> cradek: pretty close
[14:06:18] <alex_joni> :D
[14:06:25] <alex_joni> lol
[14:06:31] <cradek> maybe not real useful?
[14:06:39] <alex_joni> cradek: is that the .ac?
[14:06:43] <alex_joni> or the .3ds ?
[14:06:49] <cradek> the ac
[14:06:52] <Lerneaen_Hydra> the .3ds is much higher res
[14:06:54] <cradek> I guessed how to convert it to dxf
[14:06:55] <Lerneaen_Hydra> hmm
[14:07:00] <alex_joni> ok.. not really usefull
[14:07:07] <Lerneaen_Hydra> that was far too low-re
[14:07:08] <cradek> it could be a bowling pin just as easily as tux
[14:07:13] <Lerneaen_Hydra> indeed
[14:07:51] <cradek> I thought the file seemed awfully small.
[14:09:38] <cradek> however we could cut this "tux" on a lathe...
[14:09:39] <cradek> haha
[14:12:04] <alex_joni> lol
[14:12:10] <alex_joni> even thread it to screw a cap on
[14:12:26] <cradek> then we could draw eyes/beak/feet on him with a sharpie
[14:13:07] <alex_joni> I'm getting a converter from 3ds to dxf now
[14:16:03] <alex_joni> wohoo.. it's working, but quite big
[14:16:09] <alex_joni> 11 MB dxf
[14:16:51] <Lerneaen_Hydra> alex_joni: you converted the 3ds to dxf?
[14:16:54] <cradek> Lerneaen_Hydra: I understand the picture now, but I'm not sure which values would go in the tool table
[14:17:11] <cradek> is there a standard or does it not matter because there's normally a gui?
[14:17:27] <Lerneaen_Hydra> usually there is a gui with boxes for the different values
[14:17:43] <Lerneaen_Hydra> a csv list would be doable (IMO)
[14:17:48] <alex_joni> cradek: http://dsplabs.cs.upt.ro/~juve/emc/sitting.dxf
[14:18:32] <Lerneaen_Hydra> what a strange format
[14:19:46] <cradek> alex_joni: can you scale him down? he looks like he's 24" tall or so
[14:20:11] <cradek> but the good news is toolpath can read it
[14:20:12] <alex_joni> cradek: not really..
[14:20:21] <cradek> ok I can try to do it in autocad
[14:20:24] <alex_joni> the converter I used is pretty crippled :D
[14:20:57] <alex_joni> cradek: http://dsplabs.cs.upt.ro/~juve/emc/standing.dxf
[14:21:19] <cradek> throughput is very slow today, can you gzip?
[14:21:34] <alex_joni> Originally written by Keith Rule, Crossroads 3D is no longer developed (since 1998) ...
[14:21:44] <alex_joni> hang on
[14:21:59] <cradek> still don't have the entire first fil
[14:22:01] <cradek> e
[14:22:18] <alex_joni> I'm gzipping
[14:22:25] <alex_joni> tux.tar.gz
[14:22:27] <alex_joni> has both
[14:22:30] <alex_joni> 2MB something
[14:22:51] <cradek> thanks
[14:22:58] <alex_joni> I'm getting about 2-3MB/sec ;)
[14:23:03] <alex_joni> feels fast
[14:23:05] <cradek> 30K/s here
[14:23:09] <alex_joni> :/
[14:23:25] <Lerneaen_Hydra> where's the gziped file?
[14:23:34] <cradek> http://dsplabs.cs.upt.ro/~juve/emc/tux.tar.gz
[14:23:34] <alex_joni> http://dsplabs.cs.upt.ro/~juve/emc/tux.tar.gz
[14:23:38] <Lerneaen_Hydra> haha
[14:23:52] <alex_joni> wonder who was faster
[14:23:53] <alex_joni> :D
[14:24:03] <SWPadnos> apparently it is Adsl ;)
[14:25:04] <alex_joni> no, I meant with that URL paste
[14:25:08] <alex_joni> logger_aj: bookmark
[14:25:08] <alex_joni> See
[14:25:23] <cradek> something's not right about the file
[14:25:31] <alex_joni> which one?
[14:25:36] <alex_joni> the tar?
[14:25:48] <alex_joni> tar -cvzf tux.tar.gz sitting.dxf standing.dxf
[14:26:34] <Lerneaen_Hydra> I could open it
[14:26:36] <alex_joni> md5sum tux.tar.gz
[14:26:36] <alex_joni> 54e7725ed205d74c3ecd5e8cdf46ac6c tux.tar.gz
[14:27:13] <cradek> http://timeguy.com/cradek-files/emc/tux.png
[14:27:19] <Lerneaen_Hydra> alex_joni: any chance of a more low-res version, that one is really making my cam app stutter
[14:27:21] <cradek> sorry I mean the dxf file itself
[14:27:28] <cradek> he has one eye
[14:27:37] <cradek> and parts of the eye are missing
[14:27:43] <alex_joni> cradek: I think Lerneaen_Hydra mentioned that too
[14:27:48] <alex_joni> It must be the 3ds
[14:27:55] <Lerneaen_Hydra> yep
[14:28:04] <cradek> darn
[14:28:11] <cradek> he looks good other than that
[14:28:11] <alex_joni> but the sitting one should be good
[14:28:45] <Lerneaen_Hydra> I definetly need something more low-res...
[14:28:49] <cradek> Lerneaen_Hydra: does your cam program do A axis?
[14:29:00] <Lerneaen_Hydra> it can do 6 axes
[14:29:02] <Lerneaen_Hydra> or wait
[14:29:03] <Lerneaen_Hydra> 5
[14:29:21] <Lerneaen_Hydra> so it can do XYZAB
[14:30:00] <cradek> in sitting, his eyes are still different but at least he has two
[14:30:15] <Lerneaen_Hydra> haha
[14:30:59] <alex_joni> lol
[14:31:02] <alex_joni> cradek: can you put a snapshot?
[14:31:07] <cradek> http://timeguy.com/cradek-files/emc/tux2.png
[14:31:24] <alex_joni> oops :D
[14:32:10] <alex_joni> ok.. I'm heading home
[14:32:13] <alex_joni> later everyone
[14:32:30] <Lerneaen_Hydra> see you
[14:34:23] <cradek> ok I see
[14:34:29] <cradek> jepler pointed out that some of him is inside-out
[14:34:39] <cradek> alex's translator must get that wrong sometimes
[14:35:05] <jepler> the original could be wrong too
[14:35:13] <cradek> I bet he will look right if we turn off culling
[14:35:20] <jepler> glDisable(GL_CULL_FACE)
[14:35:27] <cradek> of course toolpath could be wrong too
[14:35:38] <jepler> yeah that's certainly possible
[14:35:51] <jepler> it would have been an obvious optimization to make, back when it was python code that did the scan-conversion and zbuffering
[14:36:13] <jepler> but that should also be easy to "fix"
[14:36:28] <jepler> cradek: had a chance to play with halscope after my last changes?
[14:36:34] <cradek> not yet
[14:36:43] <cradek> but if they do what the log messages say, they'll be great
[14:36:57] <jepler> you should try it; I need your praise
[14:37:06] <cradek> ok
[14:38:38] <skunkworks> being reconised by your peers is great :)
[14:39:42] <jepler> in fact, all of you should praise me
[14:39:48] <jepler> or else I'll have some sort of fit and quit contributing
[14:40:01] <Lerneaen_Hydra> * Lerneaen_Hydra praises jepler
[14:40:30] <SWPadnos> all hail jepler
[14:40:37] <jepler> whee
[14:40:45] <SWPadnos> can we go home now?
[14:40:56] <cradek> haha
[14:41:07] <cradek> now he's all there, but some of him is dark and some light
[14:41:11] <cradek> he looks stupid
[14:41:16] <SWPadnos> jepler?
[14:41:23] <jepler> SWPadnos: yeah, you can take the rest of the week off
[14:41:27] <jepler> but be back in the mines bright and early on monday
[14:41:29] <SWPadnos> thanks
[14:41:37] <SWPadnos> bright mines, got it!
[14:42:14] <cradek> http://timeguy.com/cradek-files/emc/tux3.png
[14:42:45] <cradek> some of him is definitely inside-out
[14:43:03] <cradek> even though it's lit wrong, I think toolpath will generate correct paths over this
[14:43:14] <cradek> maybe I'll try it later
[14:43:43] <jepler> glLightModeli(GL_LIGHT_MODEL_TWO_SIDE, 1)
[14:43:52] <jepler> or something like that
[14:43:53] <cradek> I'll have to figure out how to scale him down first, he's 420" tall
[14:45:04] <cradek> although the sitting model isn't going to be any good from the front - his feet cover up a lot of stuff
[14:51:24] <Lerneaen_Hydra> wheh
[14:51:29] <Lerneaen_Hydra> that was a long compil
[14:51:46] <Lerneaen_Hydra> 10 minutes for solidworks to convert from dxf to a parasolid
[14:52:15] <SWPadnos> it was probably trying to recognize features, so you can modify them parametrically
[14:52:23] <Lerneaen_Hydra> yep
[14:52:34] <Lerneaen_Hydra> it made all features loft's, extrudes, and so on
[14:52:47] <SWPadnos> much easier on non-organic models ;)
[14:53:21] <Lerneaen_Hydra> indeed
[14:55:31] <Lerneaen_Hydra> brb
[14:56:43] <Lerneaen_Hydra> that was probably the worst ram-bloat firefox has ever given me
[14:57:10] <Lerneaen_Hydra> it ate 3 gb of memory (1.5 gb physical, 1.5gb virtual)
[15:00:47] <Lerneaen_Hydra> I don't think that CAD/CAM like organic models too much... :(
[15:00:59] <Lerneaen_Hydra> I'm getting extremely severe performance hits
[15:01:18] <Lerneaen_Hydra> even though I've got good hardware
[15:01:42] <Lerneaen_Hydra> (2x3 ghz AMD64, 2gb ram, fast motherboard etc etc)
[15:04:11] <SWPadnos> aws it solidworks that was the hog or firefox?
[15:04:15] <SWPadnos> was
[15:10:11] <Lerneaen_Hydra> I'mnot sure actually, I had practically 0 free memory after I closed solidworks, and I freed lots after closing firefox, but I got the feeling that solidworks was eating the memory
[15:12:24] <SWPadnos> that seems more likely than FF
[15:12:43] <SWPadnos> if only SolidWorks were truly multithreaded
[15:12:44] <Lerneaen_Hydra> FF tends to only eat memory in the 50-150mb range
[15:32:26] <jepler> eek, for the much simpler "tuxedo.ac" model toolpath generates an 90line g-code file
[15:32:39] <jepler> * jepler twiddles his thumbs while waiting for axis to load it
[15:33:00] <Lerneaen_Hydra> jepler: haha, the code I would generate with edgecam would probably be in the 20k-40k region
[15:35:22] <jepler> but with g64 p.01 (inches) it's run at a pretty good speed by emc2
[15:35:26] <Lerneaen_Hydra> * Lerneaen_Hydra hopes that axis has loaded the file now
[15:35:41] <Lerneaen_Hydra> g64? scaling factor?
[15:35:49] <jepler> no, path tolerance
[15:36:13] <Lerneaen_Hydra> the 90 line program, was that the one that could have been a bowling pin?
[15:36:22] <Lerneaen_Hydra> or the standard chips.ngc
[15:36:32] <jepler> oops
[15:36:34] <jepler> I meant 90kline
[15:36:46] <jepler> for the one that could be a bowling pin
[15:36:54] <Lerneaen_Hydra> cradek: are you there? if so, is there anything else you were wondering about lathes?
[15:37:07] <Lerneaen_Hydra> jepler: eugh, that sound ineffective
[15:37:12] <cradek> no, not right now
[15:37:32] <jepler> http://emergent.unpy.net/files/sandbox/bowlingpin.png
[15:38:09] <Lerneaen_Hydra> jepler: why is it only doing one pass so close to the surface?
[15:38:44] <skunkworks> its probably tuning multible level cuts
[15:39:08] <Lerneaen_Hydra> tuning?
[15:39:58] <skunkworks> sorry
[15:40:38] <skunkworks> if your making it out of a solid block - it is making 2 passes to get rid of the material
[15:41:06] <Lerneaen_Hydra> oh
[15:41:20] <Lerneaen_Hydra> it's just that the first pass seems too close to the top
[15:41:48] <Lerneaen_Hydra> assuming the block isn't much thicker than the finished part
[15:42:39] <skunkworks> hard to tell from the perspective - but it almost seems 50 - 50
[15:43:28] <bill203> this dental mill closed for cheap: http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=7626230885
[15:43:37] <bill203> I figured it would go much higher.
[15:44:18] <alex_joni> all hail jepler
[15:46:07] <Lerneaen_Hydra> * Lerneaen_Hydra all hail jepler...
[15:50:51] <alex_joni> at times like these I realize how addictive we are of electricity
[15:52:09] <alex_joni> power went out, luckily I have juice in the laptop, and a backup internet connection (runs on batteries too)
[15:56:29] <jepler> cradek: oh, I had a completely wrong scale on Yacc in that halscope
[15:56:55] <alex_joni> * alex_joni retracts the hails
[15:57:09] <alex_joni> jepler: kidding, I'm sure it's great
[15:57:29] <skunkworks> I hail you all
[15:57:29] <alex_joni> I just tried halscope a few days ago, after a long time.. and went like .. whoa
[15:58:15] <skunkworks> and on the 7th day - jepler said - "let there be blending"
[15:58:41] <alex_joni> let there be colours
[16:04:01] <SWPadnos> let there be a 64-bit, SMP realtime kernel that works with the NVidia drivers ;)
[16:06:11] <jepler> make it open source while you're at it
[16:07:22] <alex_joni> nice
[16:07:30] <alex_joni> this is also very interesting: http://www.asterisk.org/about
[16:07:33] <cradek> I never did check the video in my laptop
[16:07:49] <cradek> it's old ATI and works with acceleration, it would be nice to know if we can get those old cards for desktops too
[16:07:53] <alex_joni> SWPadnos: how's the latency test?
[16:07:56] <SWPadnos> oh, wait. do I have to *write* it?
[16:08:11] <bill203> our office phones are on Asterisk
[16:08:11] <SWPadnos> about an hour right now
[16:08:18] <SWPadnos> (upgrading the big box to dapper :) )
[16:08:59] <alex_joni> bill203: really? looks like a great thing
[16:09:06] <alex_joni> using digium hardware?
[16:09:28] <bill203> I'm not involved in managing it, so I dont know about the hardware.
[16:09:48] <alex_joni> bill203: ok, no sweat
[16:09:51] <bill203> just that if I dial a number with extra digits, it wont go thru. :-|
[16:11:46] <alex_joni> heh
[16:11:51] <alex_joni> misconfigured probably
[16:12:39] <bill203> wouldn't suprise me.
[16:15:01] <Rugludallur> I have asterisk to, had it for years and there are no problems with extra numbers, it's just misconfig
[16:15:21] <bill203> ahh.
[16:15:42] <Rugludallur> and we also decided agains having extension for an outside line
[16:15:47] <Rugludallur> you just dial the number directly
[16:46:52] <Lerneaen_Hydra> so, what goes on in the world of an EMC developer?
[17:43:44] <alex_joni> this is soo cool: http://www.deviantart.com/view/34244097/
[17:47:50] <Lerneaen_Hydra> alex_joni: that was really cool
[17:55:30] <skunkworks> Petr sounds like he has a rt latency issue
[17:58:04] <skunkworks> cradek: is toolpath something your thinking of selling? do you have any specs on it?
[17:58:23] <cradek> no and no
[17:58:35] <cradek> it's gpl
[17:59:04] <cradek> the cvs is already maybe anon/public
[17:59:09] <cradek> please play with it if you want
[17:59:23] <cradek> I haven't worked on it for several years
[17:59:29] <skunkworks> wow
[17:59:51] <skunkworks> is this another 80% example :)
[18:00:03] <cradek> definitely
[18:01:56] <skunkworks> how would I find it?
[18:03:02] <cradek> umm just a minute
[18:03:09] <skunkworks> or I could just do a cvs toolpath or whatever
[18:03:13] <skunkworks> ?
[18:04:22] <cradek> http://unpy.net/cgi-bin/viewcvs.cgi/toolpath/
[18:04:37] <cradek> you can use "download this directory in tarball" at the bottom
[18:06:00] <skunkworks> thanks chris
[18:06:19] <cradek> last I checked it builds very easily on breezy
[18:08:00] <cradek> if you're already running AXIS you probably have everything you need to use toolpath
[18:08:47] <skunkworks> I just lost my test computer here at work - going to have to find another. :(
[18:09:16] <cradek> look under that pile of papers on the far end of your desk - that's where I always lose things
[18:09:27] <CIA-8> 03awallin 07HEAD * 10documents/lyx/emc2/Master_HAL.lyx: now includes hal_halui.lyx
[18:10:38] <cradek> yay we have a documentation guy now
[18:10:43] <cradek> that's terrific
[18:11:09] <skunkworks> http://www.electronicsam.com/images/desk.jpg
[18:11:17] <skunkworks> where exactly?
[18:11:43] <cradek> maybe try under the phone?
[18:11:53] <skunkworks> :)
[18:11:57] <cradek> it's hard to tell from here
[18:18:42] <CIA-8> 03jepler 07HEAD * 10axis/tcl/axis.tcl: remove debugging statement
[18:25:23] <cradek> skunkworks: 'sudo python setup.py install' should do it
[18:26:35] <SWPadnos> cool. the dapper upgrade failed on my dual opteron
[18:27:12] <cradek> cool?
[18:27:20] <jepler> one thing to watch out for is that both the CVS version of toolpath and whatever version of axis you use will both try to install 'minigl.so', and maybe other files .. so installing toolpath will overwrite files from emc2-axis.deb
[18:27:21] <skunkworks> cradek: thanks again
[18:27:24] <SWPadnos> well - interesting, at least
[18:27:53] <cradek> jepler: urg.
[18:28:23] <skunkworks> ok.
[18:28:34] <alex_joni> SWPadnos: what kind of upgrade?
[18:28:46] <SWPadnos> breezy -> dapper
[18:28:57] <skunkworks> it isn't like I am running anything mission critical. If something breeks - something breaks
[18:29:12] <skunkworks> thats part of the fun :)
[18:29:13] <alex_joni> did you run dist-upgrade?
[18:29:25] <SWPadnos> nope - "upgrade to dapper" in the package updater
[18:29:34] <alex_joni> ok, should be the same thing
[18:29:37] <alex_joni> what happens?
[18:29:37] <SWPadnos> yep
[18:29:45] <SWPadnos> well, a couple of interesting things.
[18:30:01] <SWPadnos> first, there was a problem with the package (or file) "hevea"
[18:30:04] <SWPadnos> whatever that is
[18:30:46] <SWPadnos> hmmm - actually, that was the second thing. the first dialog that popped up was for something else. I can probably find it in some log
[18:31:05] <alex_joni> hevea - translates from LaTeX to HTML, info, or text
[18:31:19] <SWPadnos> the hevea dialog said "the upgrade will abort now", so I didn't click the ok button, since there were still things going on in the terminal window of the updater
[18:31:40] <alex_joni> * alex_joni agrees
[18:31:49] <alex_joni> aborting an upgrade sounds like a bad idea
[18:31:58] <SWPadnos> oh right - the first error was postgres related, so I let it go on
[18:32:11] <SWPadnos> since I don't care about postgres clustering crap
[18:33:28] <alex_joni> heh
[18:33:34] <Jymmm> SWPadnos :but but but that'll slow down the cataloging of your p0rn collection w/o a pg cluster"
[18:33:40] <SWPadnos> anyway, I waited until there was a question in the terminal - which was related to gdm
[18:35:46] <SWPadnos> phone
[18:38:22] <CIA-8> 03awallin 07HEAD * 10documents/lyx/emc2/hal_drivers.lyx: m5i20 notes: driver details, fpga bug.
[18:39:04] <SWPadnos> so, I told it to go ahead and replace my gdm conf file, since I know what the changes were
[18:39:37] <SWPadnos> well, I couldn't do that until I dismissed the hevea error dialog (clicked OK)
[18:40:31] <SWPadnos> at that point, the package names under the progress bar went lightning fast, until it got to GDM, where it popped up a dialog asking if It should replace the gdm.conf file ;)
[18:41:13] <SWPadnos> I said OK again, and it continued on, until there was some X related error, at which point it actually decided to exit, and the update process went away.
[18:42:51] <alex_joni> argh
[18:42:59] <alex_joni> that's why I like apt-get dist-upgrade better
[18:46:38] <skunkworks> we have a test system up running dapper and mysql 5 - so far so good.
[18:51:10] <alex_joni> AHAHAHA http://www.youtube.com/watch?v=fIEVMyW6fLI
[18:53:53] <alex_joni> cradek, jepler: did you move AXIS cvs?
[18:56:42] <alex_joni> or rather change the script? I got an diff for the last AXIS commit.. not that I'm complaining
[19:14:11] <cradek> it sends to the list now I guess
[19:14:28] <jepler> I changed it to send to the list
[19:14:31] <jepler> the CVS is still in the same place
[19:15:59] <alex_joni> jepler: ok, no worries
[19:16:30] <cradek> alex_joni: what's the significance of the song that's in all this stuff you find?
[19:17:06] <alex_joni> that's some finish song
[19:17:17] <alex_joni> and this part doesn't have much of a meaning
[19:17:44] <cradek> is it common for people to make fun of the particular song for some reason?
[19:17:46] <alex_joni> http://www.noside.com/nsd6010note.html
[19:17:51] <alex_joni> guess it sounds nice..
[19:18:11] <alex_joni> "On its release in Finland, this song entered the radio charts - a rare occurrence for traditional music. The humorous story is told in the Savo dialect of Eastern Finland."
[19:20:03] <jepler> alex_joni: people who make fun of my president should be shot like the terrists they are
[19:20:18] <Jymmm> bush sucks
[19:21:50] <alex_joni> terrists? :D
[19:22:38] <SWPadnos> arbo-tree-ists
[19:22:50] <SWPadnos> according to our prezzident
[19:23:05] <skunkworks> nuclure
[19:23:14] <SWPadnos> nukular?
[19:24:20] <skunkworks> trying to spell it the way he pronounces it
[19:24:27] <SWPadnos> me, too
[19:24:57] <skunkworks> :)
[19:25:24] <cradek> noo-kya-lerr
[19:25:52] <SWPadnos> dum-ass-ular ;)
[19:26:18] <alex_joni> you'll upsed dmess
[19:26:22] <alex_joni> upset even
[19:26:38] <SWPadnos> well, he's a lawless canuck, so that doesn't matter to us USians
[19:26:45] <skunkworks> Maybe he has changed - For a while there was a syllable missing
[19:27:02] <skunkworks> at least how I heard it
[19:27:04] <SWPadnos> I always rememer it with an extra syllable
[19:27:09] <SWPadnos> remember, too
[19:27:34] <SWPadnos> and now lots of people pronounce it that way, because "he's the prezzident, he must be right"
[19:28:28] <jepler> I experience a terrible confusion whenever I pronounce the word anymore
[19:28:50] <skunkworks> I wonder how he pronounces Molecular
[19:29:02] <SWPadnos> he doesn't
[19:29:03] <Jymmm> molecure
[19:29:11] <Jymmm> mole-cure
[19:29:14] <SWPadnos> "chemical"
[19:29:16] <skunkworks> :)
[19:30:01] <alex_joni> did you see the one where he's talking about installing new fences on the border to mexico?
[19:30:18] <SWPadnos> should start soon, I believe the budget for that passed
[19:30:59] <alex_joni> but did you see the movie? that was soo hilarious
[19:31:24] <SWPadnos> nope. but they're all funny (if you can suspend your disbelief long enough)
[19:31:27] <alex_joni> http://youtube.com/watch?v=y4VQnBQxCtM&search=bush%20fence
[19:31:32] <alex_joni> WATCH IT
[19:31:50] <cradek> until you realize again that he's the president
[19:31:59] <alex_joni> watch the back of that shot
[19:32:03] <alex_joni> ROFLMAO
[19:32:17] <SWPadnos> heh - that is funny
[19:32:22] <giacus> hello
[19:32:34] <SWPadnos> cradek, right - that's the disbelief you have to suspend
[19:32:50] <SWPadnos> once it hits you that what he says matters, it gets very depressing very fast
[19:32:58] <cradek> yeah.
[19:33:27] <jepler> I still think that folks like you, who are against protecting america's families, should be allowed to die without medical attention when the bird flus come
[19:33:31] <giacus> http://www.giacus.org/files/0001.jpg
[19:33:34] <giacus> haha
[19:33:49] <SWPadnos> the vird flus are coming, the bird flus are coming!
[19:33:57] <giacus> please notice the laptop running Axis :D
[19:34:09] <SWPadnos> I guess that would be the german version - ze vird fluz!
[19:35:30] <jepler> yay axis
[19:36:05] <alex_joni> we are sinking that ze vird fluz is koming
[19:36:24] <giacus> hi alex_joni
[19:36:35] <giacus> what virus ?
[19:36:35] <alex_joni> giacus: got a 3D model of the penguin?
[19:36:58] <giacus> I received a mail marked as **spam** form ML
[19:37:06] <giacus> alex_joni: yeah :)
[19:37:13] <giacus> going to draw a lathe ..
[19:37:17] <alex_joni> giacus: can you make a g-code out of it?
[19:37:19] <cradek> haha that's very nice giacus
[19:37:24] <giacus> :D
[19:37:26] <Jymmm> giacus: Is that Mr Potato Head ?
[19:37:27] <alex_joni> we want a new 3D-Chips
[19:37:49] <giacus> alex_joni: not sure I can do that :(
[19:38:00] <giacus> just playng around
[19:38:08] <Jymmm> heh... looks like an operating room
[19:39:14] <skunkworks> their making babies :)_
[19:39:31] <alex_joni> giacus: what format do you have it in?
[19:39:37] <alex_joni> can you export to dxf?
[19:39:55] <alex_joni> 3ds?
[19:40:14] <ValarQ> crap?
[19:40:16] <ValarQ> :o)
[19:40:17] <skunkworks> sorry - They are making babies
[19:40:17] <giacus> Blender 2.4, let me check ..
[19:42:19] <giacus> alex_joni: yeah, DXF and amny others
[19:42:31] <Lerneaen_Hydra> if anyone can get a solidmodel (.x_t, .sldprt etc etc) then I can make a nice new version of chips (nose cutter compensation, roughing and then finishing cut etc. etc)
[19:42:33] <alex_joni> can you put an dxf online?
[19:42:41] <giacus> of that scene ?
[19:42:53] <alex_joni> no, only the tux
[19:43:06] <alex_joni> Lerneaen_Hydra: try to work a format with giacus
[19:43:43] <Lerneaen_Hydra> giacus: could you export to a parasolid model?
[19:43:46] <ValarQ> what cam system do you use to convert dxf files to emc2 gcode?
[19:44:19] <Lerneaen_Hydra> I use edgecam
[19:44:22] <giacus> alex_joni: I can, just give me a bit of time
[19:44:26] <alex_joni> bbl
[19:45:31] <giacus> Lerneaen_Hydra: just 3 days playng with blender, tryng to animate some scene
[19:45:36] <Lerneaen_Hydra> giacus: prefferably a file that's not too big (as long as the parasolid is under 5-8mb I think I should be able to handle it without too much trouble)
[19:45:39] <ValarQ> Lerneaen_Hydra: does it run on UNIX?
[19:45:46] <Lerneaen_Hydra> unfortunately not :(
[19:45:49] <ValarQ> :(
[19:45:56] <giacus> Lerneaen_Hydra: ok
[19:46:00] <Lerneaen_Hydra> I don't know if you can run it in wine/cedega
[19:46:05] <Lerneaen_Hydra> possibly
[19:46:20] <ValarQ> i can always go my own way and make a CrapCAM i guess...
[19:46:56] <ValarQ> to bad the dxf format is so weird
[19:49:06] <Lerneaen_Hydra> I suppose I could do an approximation of tux with some simple commands in solidworks (loft, extrude, revolve) but I'm not sure the result would be any good
[19:51:36] <giacus> Lerneaen_Hydra: I'll try to get a good model first and test toolpath, I'll tell you later
[19:51:44] <Lerneaen_Hydra> ok
[19:52:04] <Lerneaen_Hydra> That will have to be sometime tomorrow though, I have to get some sleep now
[19:52:26] <giacus> was impressed by this movie yestarday http://orange.blender.org/
[19:52:36] <giacus> an open source movie
[19:52:40] <giacus> very good
[19:53:01] <Lerneaen_Hydra> g'night all
[19:53:55] <giacus> Lerneaen_Hydra: G'Night
[20:39:29] <kerry_> kerry_ is now known as rayh
[21:01:22] <K4ts> hello
[21:05:44] <giacus> hey K4ts
[21:10:43] <K4ts> smack giacus
[21:10:49] <K4ts> :-)
[21:12:44] <giacus> ;*
[21:15:24] <giacus> doh: http://www.repubblica.it/2006/06/sezioni/scienza_e_tecnologia/gates-lascia/gates-lascia/gates-lascia.html
[21:15:40] <giacus> should gates leave the M$ in 2008 ??
[21:15:52] <K4ts> ahah
[21:15:57] <giacus> leggi ..
[21:16:02] <K4ts> is the gag
[21:16:05] <K4ts> ahah
[21:16:53] <giacus> Ray Ozzie will be the new saint
[21:53:27] <alex_joni> nice picture.. thought I'd share (warning big) http://tmp.4chan.org/hr/src/1150342413086.jpg
[22:35:31] <K4ts> night
[23:06:33] <robin_sz> meep?
[23:27:18] <guess> logger_aj: bookmark
[23:27:18] <guess> See
[23:27:33] <minimess> bonjour tous
[23:31:30] <SWPadnos_> SWPadnos_ is now known as SWPadnos
[23:31:40] <guess> guess is now known as skunkworksB
[23:33:06] <minimess> Who was involved in the puppy distro?? alex??
[23:33:45] <SWPadnos> chinamill, I think
[23:34:45] <skunkworksB> roltek: ?
[23:36:02] <SWPadnos> the coolcnc iso is from chinamill, I'm pretty sure
[23:36:14] <SWPadnos> there was another puppy iso that alex may have done
[23:36:47] <skunkworksB> SWPadnos: Figure out the upgrade?
[23:36:53] <minimess> will it only multi session to a dvd ..... or possibly a cdrw as well.. it coughed on a cd-r earlier
[23:37:19] <SWPadnos> well, it works, but there are still issues with several tetex related packages
[23:37:38] <SWPadnos> it should be useful only as a single-session disc, I think
[23:37:47] <minimess> this one is neat enuf... but issues trying to sell a machine to a 14 yr old that cant get msm
[23:38:28] <skunkworksB> didn't alex_joni make a live ubuntu cd with emc2?
[23:38:31] <minimess> gaim want ssl stuff
[23:38:45] <SWPadnos> yes, there's an ubuntu liveCD with emc2
[23:38:51] <minimess> really...
[23:39:01] <SWPadnos> http://www.cncgear.com/EMC/
[23:39:09] <SWPadnos> several ISOs there
[23:40:23] <minimess> right on
[23:40:29] <minimess> thx
[23:40:34] <SWPadnos> sure
[23:41:01] <minimess> this ssl stuff i think is NOt doable...
[23:41:13] <SWPadnos> ?
[23:42:30] <SWPadnos> hi Les
[23:42:33] <minimess> gaim ssl.. msn... nightmare
[23:42:47] <les_w> hi swp
[23:42:57] <SWPadnos> $anything ... (ms)$anything ... nightmare ;)
[23:43:08] <les_w> haha
[23:43:21] <les_w> I didn't get much done today
[23:43:33] <minimess> i agree... but the kids arent all convinced yet
[23:43:34] <les_w> just trying to match impedances better
[23:43:45] <SWPadnos> strange. it's usually the other way around
[23:43:53] <SWPadnos> wrt kids. not impedance atching
[23:43:59] <les_w> haha
[23:45:43] <minimess> no doubt... the youger one just likes Chips... the other 2 are a harder sell
[23:45:43] <les_w> I drew up some 4th order equivalent circuits and need to do some SPICE runs
[23:47:40] <SWPadnos> are you doing monte carlo (or similar) statistical evaluation, or just simple frequency graphs?
[23:48:19] <skunkworksB> * skunkworksB thinks swpadnos just did a quick web search ;)
[23:48:31] <SWPadnos> nope
[23:48:43] <les_w> well I am synthesizing electrical reactance with daubs of epoxy and tiny pieces of foil on piezos
[23:48:46] <SWPadnos> I is a engineer, remmemmer?
[23:48:57] <skunkworksB> just ribbing you. I think I know a lot but you guys blow me away.
[23:49:14] <SWPadnos> ;)
[23:49:17] <minimess> go for broke les... ; )
[23:49:18] <les_w> easier to just do spice runs with parameters the amnalyzer spits out
[23:49:43] <SWPadnos> run a monte carlo on possible part tolerances, the results may surprise you
[23:50:10] <SWPadnos> or just do tolerance sweeps
[23:50:27] <les_w> at least I don't need 800 microhenries of shunt induxtance for power factor.
[23:50:29] <skunkworksB> bbl
[23:50:41] <les_w> A tiny piece of foil did the trick.
[23:50:46] <SWPadnos> heh
[23:50:51] <SWPadnos> macgyverisms
[23:50:58] <SWPadnos> that could have been a gum wrapper
[23:51:03] <les_w> I guess
[23:51:31] <les_w> But I must be pretty close to a perfect impedance match
[23:51:46] <les_w> because I can't seem to eek more power out of it
[23:51:58] <les_w> but I need some
[23:52:43] <SWPadnos> have you done a physical simulation of the piezo itself, or are you doing the electrical equivalent instead?
[23:53:31] <les_w> I put the piezos on the analyzer and it spits out an equivalent circuit
[23:54:18] <les_w> daubs of epoxy here and there change the values
[23:55:04] <SWPadnos> right - changes in resonance modes (or just weights)
[23:55:17] <les_w> yes
[23:55:45] <les_w> well I can sure see when the 5 min epoxy trips....very obvious on the analyzer screen
[23:56:11] <les_w> after about 3 minutes things start moving fast
[23:56:22] <SWPadnos> heh
[23:56:29] <SWPadnos> it's a glue tester ;)
[23:56:33] <les_w> yeah
[23:56:55] <les_w> what I am really trying to see is radiation resistance
[23:57:23] <les_w> and I am
[23:57:28] <SWPadnos> what kind of radiation?
[23:57:37] <SWPadnos> (sound, RF, light, gamma rays ...)
[23:57:52] <les_w> put a reflector 1/2 wavelength in front of it and the ESR drops to nothing
[23:58:03] <les_w> ultrasonic
[23:58:05] <SWPadnos> yes
[23:58:07] <SWPadnos> ok
[23:58:30] <SWPadnos> reflections and the housing will be a major source of annoyance
[23:58:39] <les_w> oh sure
[23:58:46] <SWPadnos> test setups don't translate well to products
[23:59:05] <les_w> that's why they call me!
[23:59:16] <SWPadnos> heh