#emc | Logs for 2006-06-05

[00:01:14] <A-L-P-H-A> who was around earlier when I asked about the UPS issue?
[00:01:15] <A-L-P-H-A> http://www.electronic-circuits-diagrams.com/telephonesimages/10.gif
[00:01:19] <A-L-P-H-A> http://www.electronic-circuits-diagrams.com/telephonesimages/telephonesckt10.shtml
[00:01:24] <les_w> I'm just hoping the economy holds on so corporations continue to give me funding for such screwy things
[00:02:01] <A-L-P-H-A> les_w... invite the perfect woman.
[00:02:05] <A-L-P-H-A> invent
[00:02:06] <A-L-P-H-A> hahaha
[00:02:26] <les_w> I have not doe such a good job at that I guess
[00:02:31] <les_w> done
[00:02:55] <les_w> HEY
[00:03:00] <les_w> FENN
[00:03:03] <les_w> !!
[00:03:59] <les_w> You are a microbiologist by education right?
[00:04:03] <fenn> yes
[00:04:12] <A-L-P-H-A> Anyone good at critiques pour resume?
[00:04:25] <les_w> Want some work?
[00:04:50] <fenn> hmmm.. is it easy stuff?
[00:05:01] <A-L-P-H-A> fenn, wanna brew some ricin? ebola? hemeragic fever?
[00:05:05] <fenn> bleh i hate microbiology
[00:05:07] <les_w> of course.
[00:05:28] <jmkasunich> you hate it but you majored in it?
[00:05:31] <A-L-P-H-A> fenn, ever watch regensis?
[00:05:35] <A-L-P-H-A> regenesis
[00:05:58] <fenn> jmkasunich: i was into molecular biology and genetic engineering, but they only had undergraduate program for microbiology and biochemistry
[00:06:46] <les_w> Could you write a slick paper on UV DNA mutation?
[00:07:10] <les_w> A short one
[00:07:15] <fenn> no but i'd happily proofread one you wrote
[00:07:21] <les_w> targeted to a specific audience
[00:07:49] <les_w> It's marketing really
[00:08:00] <fenn> there's tons of stuff on the web
[00:08:19] <les_w> Like those camping gear UV lights that you put in water.....
[00:08:27] <les_w> that don't do a damn thing
[00:08:49] <les_w> I know I spent a whole day searching the web
[00:09:12] <les_w> But targeted propaganda is another thing.
[00:12:58] <A-L-P-H-A> I've got a bio background (honest). I could write a paper on that... but I'd need a grant to do so.
[00:13:13] <A-L-P-H-A> and then I would have to state, as I should for ethical reasons, who the grant was from.
[00:13:38] <les_w> I'm just an engineer that can read.
[00:14:09] <A-L-P-H-A> can you chew gum and walk at the same time?
[00:14:18] <les_w> uh...
[00:14:33] <les_w> haha
[00:15:05] <fenn> * fenn must be allergic to money or something
[00:15:16] <fenn> ah! i'm allergic to work, that's it
[00:15:49] <les_w> I do have to check my facts. I said the DNA bases were weakly hydrogen bonded to the ribose. Fenn says that it is a covalent bond.
[00:16:20] <les_w> I'll have to check that.
[00:16:57] <fenn> do a google image search for "dna hydrogen bonding"
[00:17:07] <les_w> yeah
[00:17:47] <les_w> There was some disagreement in papers I saw about UVC mutation
[00:18:14] <les_w> But the structure of DNA is known.
[00:18:43] <fenn> heh you bet it is
[00:18:46] <les_w> I don't want to misstate something in a paper.
[00:18:57] <fenn> alpha beta and zeta forms
[00:19:14] <fenn> and probably even more
[00:19:38] <les_w> Some say it is only thymine thymine dimerization
[00:19:54] <fenn> there are other mechanisms too i'm sure
[00:19:58] <les_w> others say guianine and cytosine are involved as well
[00:20:09] <fenn> that is more of an organic chemistry question
[00:20:15] <les_w> yeah
[00:22:20] <fenn> Although 185 nm is better absorbed by DNA, the quartz glass used in commercially-available lamps, as well as environmental media such as water, are more opaque to 185 nm than 254 nm
[00:22:30] <fenn> hmmm
[00:23:21] <les_w> unless it's doped....
[00:23:45] <fenn> ;)
[00:24:25] <les_w> 185 is nasty stuff anyway...it'll fry you crispy way fast
[00:25:54] <les_w> and make you grow two heads
[00:27:21] <fenn> there isnt a whole lot of uv-c in sunlight is there
[00:28:03] <les_w> I read some of DNA mutatin repair mechanisms....that involve fatty acid esters of all things
[00:28:14] <les_w> no uvc in sunlight
[00:28:41] <les_w> except at the south pole
[00:29:06] <les_w> ozone hole
[00:29:29] <fenn> fatty acid esters? i havent heard of that
[00:29:47] <les_w> I don't understand it really
[00:29:54] <les_w> just read about it
[00:31:07] <fenn> in rat colon cells perhaps? :)
[00:31:11] <les_w> I am more concerned about bugs hiding under an opaque film of grease on a food processing machine....where I can't blast 'em
[00:32:45] <fenn> this is the sort of thing you'd have to do field trials i think
[00:33:44] <les_w> Anyway I have to write up this paper. I have a one track mind, so I have to write and ship off the paper befor I continue messing with the impedance analyzer
[00:34:42] <les_w> We will check uvc transmissibility in fatty acids/esters....I have designed a test rig to do so
[00:35:25] <les_w> they will build it in chicago.
[00:36:31] <les_w> Have to watch out for flourecsence spoiling the measurement though
[00:37:00] <les_w> so must have a bandpass filter before the silicon carbide photodiode detector.
[00:37:59] <fenn> where do you get that?
[00:38:14] <les_w> what? the diode?
[00:38:22] <les_w> cree
[00:38:23] <fenn> a UV bandpass filter
[00:38:40] <les_w> it's just a quarter wave stack
[00:38:50] <les_w> can have it built up
[00:39:46] <fenn> ooo cool: http://www.naturfotograf.com/uvstart.html
[00:39:50] <les_w> like the antireflective coating on my glasses...just thinner and different materials
[00:40:36] <les_w> see like a bee!!!
[00:50:21] <giacus> G'night
[00:53:31] <les_w> Anyway, the targets are e.coli and fecal coliform. Yes, at the slaughterhouse lower intestines are often breeched, so your sunday BBQ meat is often soaked in $hit fora time.
[00:53:36] <les_w> gross, huh?
[00:55:33] <fenn> one of many reasons i dont eat dead animals
[00:56:02] <cradek> easy enough to solve that one
[00:56:12] <les_w> how?
[00:56:12] <fenn> "dont do that"
[00:56:53] <fenn> cradek: slaughterhouses are pretty crazy places, overworked dudes with 2 foot long knives goin like mad
[00:57:08] <fenn> or big gas powered saws
[00:57:10] <cradek> where not to be
[00:57:49] <fenn> if the statistics werent totally skewed it'd probably be _the_ most dangerous job
[00:58:04] <les_w> I guess
[00:58:15] <cradek> it sure has to be one of the worst jobs
[00:58:19] <cradek> danger aside
[01:08:37] <cradek> new EMC 2.0.1 packages in the ubuntu repository
[01:15:18] <CIA-8> 03cradek 07v2_0_branch * 10emc2/VERSION: release
[01:15:48] <CIA-8> 03cradek 07v2_0_branch * 10emc2/configs/stepper/ (stepper_inch.ini stepper_mm.ini):
[01:15:48] <CIA-8> more segments per second throughput by default (we were getting only about
[01:15:48] <CIA-8> 10 which is too low for a lot of programs).
[01:15:49] <CIA-8> 03cradek 07v2_0_branch * 10emc2/configs/stepper-xyza/inch.ini:
[01:15:49] <CIA-8> more segments per second throughput by default (we were getting only about
[01:15:51] <CIA-8> 10 which is too low for a lot of programs).
[01:16:27] <skunkworks> Nice work!
[01:16:40] <CIA-8> 03cradek 07v2_0_branch * 10emc2/debian/changelog: as released
[01:16:59] <CIA-8> 03cradek 07v2_0_branch * 10emc2/debian/ (postinst emc2.files): changes for exclusive parport access
[01:19:06] <CIA-8> 03cradek 07v2_0_branch * 10emc2/VERSION: bump version number after release
[01:20:22] <cradek> whew
[01:20:28] <A-L-P-H-A> :D
[01:20:41] <A-L-P-H-A> cradek... at least someone's maintaining stuff... as i've got no clue about this stuff.
[01:20:48] <A-L-P-H-A> I should get my hands dirty and do something.
[01:20:51] <cradek> who volunteered me for this work??
[01:20:56] <cradek> oh right, it was me
[01:21:33] <cradek> A-L-P-H-A: there's always stuff to do if you have time to give!
[01:21:54] <A-L-P-H-A> I've gotta relearn stuff...
[01:22:12] <A-L-P-H-A> But I do have some spare brain neurons to fry for EMC.
[01:24:44] <CIA-8> 03jmkasunich 07HEAD * 10emc2/src/hal/components/pwmgen.c: Some progress on pwm component. The 'pure PWM' and 'PDM' (pulse density modulation) modes work correctly now, but the 'dithered PWM' mode still needs some work
[01:34:59] <skunkworks> I am going to have to setup a computer to run head or testing I think :)
[01:41:10] <jepler> vmware server works better on my machine where the host is not a realtime kernel
[01:41:16] <jepler> * jepler installed dapper, and is now installing breezy
[01:41:38] <jmkasunich> jepler, what is the host?
[01:41:51] <jmkasunich> stock breezy?
[01:42:58] <jepler> jmkasunich: the host is an x86-64 running fedora core 4
[01:43:17] <jepler> after breezy is installed I'm going to install the realtime kernel and see if it can rebuild emc2
[01:43:26] <jepler> and how long it takes
[01:43:38] <SWPadnos> I found the RT kernels a lot slower as guests
[01:43:43] <jmkasunich> install the RT kernel on the breezy vm I asume
[01:43:47] <jepler> jmkasunich: right
[01:44:20] <jmkasunich> does FC support kernels in rpms?
[01:44:36] <cradek> yes
[01:45:10] <cradek> but trying to make a custom one makes debian's make-kpkg look really good
[01:45:20] <jmkasunich> ok, forget I asked
[01:45:43] <cradek> well I've tried to rpmize a new kernel version - maybe adding a patch wouldn't be so bad
[01:46:48] <jepler> the breezy installer kicks the ass of the dapper live cd installer
[01:46:57] <jepler> I'm sorry to say
[01:47:03] <cradek> is it no longer text mode?
[01:47:09] <jepler> cradek: no, it's an X application, you idiot
[01:47:29] <jepler> they copied the fedora core installer bug "you have to mouse out of the next button before you can click it again"
[01:47:47] <cradek> why the f are people afraid of text mode? the breezy installer worked *great*
[01:47:48] <jepler> they lost the feature where they show you a short list of time zones that are relevant based on the other information collected (locale & keyboard)
[01:48:29] <cradek> the only one I saw say it was "unfriendly" was bigAl
[01:48:29] <jmkasunich> cradek, they don't like text mode for the same reason they do stupid sh!t like bootsplash
[01:48:41] <jmkasunich> non-geeks are afraid of anything that isn't cute and cuddly
[01:49:00] <cradek> and ideally, the corners should be round so you don't get hurt
[01:49:04] <jepler> they ask you what language and keyboard to use, but this is after you've booted the OS in english with an american keyboard, and gotten far enough to double-click an icon
[01:49:08] <jmkasunich> cartoon windows
[01:50:16] <jepler> but hey, you can wait for openoffice to start while it's installing packages
[01:51:08] <cradek> I admit having one cd for live + install is cool.
[01:51:13] <jepler> yeah it's a great gimmick
[01:51:33] <jmkasunich> cuts the cost of the free CDs they mail out in half
[01:51:41] <jmkasunich> non-trivial when you intend to mail millions
[01:51:52] <cradek> yeah half is sure a lot
[01:51:57] <cradek> has anyone ordered cds yet?
[01:55:26] <jepler> vmware server is pretty cool. I can install the client on my laptop (which is confusingly called "vmware-server-console") and connect via ssh to the machine where vmware is actually hosting the server
[02:22:34] <CIA-8> 03jmkasunich 07HEAD * 10emc2/src/hal/components/pwmgen.c: Final (hopefully) version of PWM generator, all modes work correctly
[02:22:58] <jmkasunich> there you go cradek, get testing!
[02:23:10] <cradek> haha
[02:23:15] <cradek> ok and you get to bed
[02:23:25] <jmkasunich> have fun
[02:23:40] <cradek> hope you have a good trip
[02:23:44] <jmkasunich> goodnight all, gonna be away for a couple days
[02:24:01] <jepler> see you jmkasunich
[02:52:05] <A-L-P-H-A> omg this is just SEXY... http://osvids.com/files/page3-1036-pop.html
[02:52:08] <A-L-P-H-A> maybe I'm just a nerd.
[02:52:10] <A-L-P-H-A> :/
[02:52:18] <A-L-P-H-A> but tell me that ain't fuck'n COOL
[02:53:06] <jepler> "The Camtasia Studio video content presented here requires JavaScript to be enabled and the latest version of the Macromedia Flash Player"
[02:53:09] <jepler> no, that web page sucks
[02:53:36] <jepler> (I do have javascript enabled, and I thought I had the latest flash for linux, so whatever)
[02:58:11] <A-L-P-H-A> worked on my windoz machine
[02:58:17] <A-L-P-H-A> CRAP
[02:58:20] <A-L-P-H-A> I can't use it on my laptop.
[02:58:27] <A-L-P-H-A> I'm using the 'ati' drivers... not the 'radeon' drivers.
[02:58:46] <jepler> I've seen a presentation about "xgl" .. shows the desktops on 4 faces of the cube and does stuff like make transparent, jiggling windows
[02:58:52] <jepler> is that what this video is?
[02:58:59] <A-L-P-H-A> yup
[02:59:21] <jepler> ah
[03:00:07] <A-L-P-H-A> what's the best bang for the buck video card now these days?
[03:00:37] <A-L-P-H-A> need it to be AGP 3.0 x8.
[03:00:40] <A-L-P-H-A> not pciE
[03:03:22] <jepler> except for gaming, I'd buy the cheapest card that has linux drivers.
[03:03:49] <jepler> specifically for emc(2), you'll miss realtime deadlines as soon as you run opengl programs on nvidia's closed source drivers
[03:04:32] <jepler> but even on a 600MHz machine, AXIS is perfectly fast with software rendering for my typical 4kline gcode files
[03:06:50] <A-L-P-H-A> I want it for my workstation.
[03:10:47] <jepler> relatively inexpensive nvidia cards give good performance
[03:11:14] <jepler> I'd pay more attention to the outputs (e.g., do you need dual dvi) than the 3d texturing performance
[03:12:13] <jepler> at work we recently put dual monitors on many developers' desks, and where upgrades were needed it was "the cheapest 2-output nvidia card" that they put in
[03:17:21] <jepler> similar to: http://www.newegg.com/Product/Product.asp?Item=N82E16814130278
[03:17:23] <jepler> fanless == good
[03:18:03] <A-L-P-H-A> I have three monitors... I NEED three monitors...
[03:18:15] <A-L-P-H-A> I've got an ATI9200 PCI card, and ATI9200SE AGP card.
[03:18:19] <A-L-P-H-A> I don't play games.
[03:18:40] <A-L-P-H-A> but I want the eyecandy from XGL
[03:19:09] <jepler> it won't work with the ATI opengl drivers?
[03:20:14] <A-L-P-H-A> XGL won't work with the xorg.conf "ati" drivers... I will work with the ATI Binaries.
[03:20:20] <A-L-P-H-A> it will work with the "radeon" drivers.
[03:21:31] <cradek> I tried Xgl and was irritated by it within 10 minutes. I recommend saving your money (and saving yourself 10 minutes)
[03:21:45] <A-L-P-H-A> serious? :|
[03:21:48] <A-L-P-H-A> probably right.
[03:21:49] <cradek> it looked interesting but waiting for the windows to finish jiggling all the time was really irritating
[03:21:56] <cradek> yeah 100% serious
[03:21:57] <A-L-P-H-A> i disabled all the eye candy in winXP already too.
[03:22:22] <cradek> because you have to wait for it to stop doing its irritating thing before you can continue, right? Xgl is worse.
[03:22:51] <A-L-P-H-A> I need to figure out a way to convert 400 gig NTFS drive into a linux drive without backing up all the crap.
[03:23:10] <jepler> goodnight all
[03:23:13] <cradek> seems like you'll need 400 gig of free space somewhere
[03:23:19] <cradek> goodnight
[03:23:22] <A-L-P-H-A> night
[03:23:24] <jepler> cradek: did you try out halscope after I changed everything?
[03:23:38] <cradek> not yet but I'm anxious to see it
[03:23:43] <jepler> let me know how much you hate it
[03:23:44] <cradek> cursor?
[03:23:52] <A-L-P-H-A> okay... 167gig is porn... 82gigs is TV shows.
[03:23:54] <jepler> no, I never improved it after adding the "tooltip"
[03:24:09] <cradek> ok I'll play with it soon
[03:24:19] <jepler> that's ... about 82 gigs more TV and 166.8 gigs more porn than anyone actually needs!
[03:24:40] <A-L-P-H-A> jepler... have you watched Topgear?!?! it's awesome!
[03:24:59] <jepler> let's see .. 30 seconds .. times 4.8 megabytes per second .. yeah, that's about right
[03:25:10] <cradek> haha
[03:25:53] <cradek> that's easy to convert to ext3: just format it an fire up your IPT software again
[03:25:59] <cradek> d
[03:26:23] <A-L-P-H-A> IPT?
[03:27:12] <cradek> intellectual-property-theft (a generic acronym for all that file-sharing software)
[03:27:13] <A-L-P-H-A> I know that there is an NTFS mounting software.
[03:27:22] <A-L-P-H-A> cradek. I see. :P
[03:28:23] <cradek> I think linux can mount ntfs read-only and maybe even read-write if you're very brave
[03:28:30] <cradek> but I haven't looked into it for a long time
[03:28:49] <A-L-P-H-A> there is... but it's beta
[03:28:55] <cradek> not sure I've ever had an ntfs partition
[03:29:38] <A-L-P-H-A> WTF.
[03:29:44] <cradek> ?
[03:29:57] <A-L-P-H-A> Messenger has a 8.5 GIG "sharingMetadata" file.
[03:30:03] <A-L-P-H-A> MS is dying NOW.
[03:30:07] <cradek> haha
[03:30:17] <cradek> what's messenger?
[03:30:25] <A-L-P-H-A> MSN Messenger
[03:32:59] <A-L-P-H-A> it's a BUG.
[03:33:00] <A-L-P-H-A> wth.
[03:37:00] <A-L-P-H-A> Time: 23:40:37 -0500 GMT, Windows XP Professional, Service Pack 2 (5.1 - 2600), AMD XP 2800+ running at (1-AMD , 2123MHz, 512KB (100% Load)), DDR400 RAM Usage: 592/1024MB (57.81%), C: 33.18gb of 189.91gb free, E: 58.96gb of 372.62gb free, Current Uptime: 1hr 12mins 19secs, Record Uptime: 3wks 12hrs 48mins 50secs, 3 Viewsonic 17" LCDs.
[03:45:09] <cradek> ?
[03:45:51] <A-L-P-H-A> seeing how much I'm freeing up.
[03:46:11] <A-L-P-H-A> I want to keep all my videos and mp3s.
[03:46:16] <A-L-P-H-A> the rest... who cares.
[03:57:09] <A-L-P-H-A> hey... a 300GB SATA drive... $130CDN.
[07:56:00] <Lerneaen_Hydra> hi
[10:57:06] <les_w> morning
[11:06:02] <fenn> * fenn pecks as les_w's head
[11:06:48] <fenn> *peck peck peck* hellooo!
[11:18:10] <anonimasu> hello
[11:30:00] <les_w> hello!
[11:31:13] <les_w> I've got to go out and straighten up the shop. I really need to do that expansion. I don't have nearly enough room.
[11:59:55] <giacus> jmkasunich: was right
[12:00:04] <giacus> no mail from ML today
[12:00:25] <giacus> it should be down yet
[12:24:38] <jepler> it's happening to others in different projects too: http://sourceforge.net/tracker/index.php?func=detail&aid=1500818&group_id=1&atid=200001
[12:28:51] <giacus> yeah, right
[12:30:37] <giacus> Jeff, i'm compiling 2.6.16-magma modules right now on the vm
[12:30:54] <Lerneaen_Hydra> jepler: were you having issues with EMC on dapper?
[12:30:59] <giacus> hope to be lucky.. I'll tell you later if rtai work
[12:31:32] <jepler> Lerneaen_Hydra: I haven't tried installing it on dapper yet
[12:31:37] <Lerneaen_Hydra> oh
[12:31:39] <Lerneaen_Hydra> ok
[12:31:45] <jepler> Lerneaen_Hydra: I was installing breezy and dapper in vmware server, just to try out vmware
[12:32:21] <anonimasu> 1hm
[12:32:33] <anonimasu> les_w: are you here?
[12:33:17] <jepler> Lerneaen_Hydra: I wonder if vmware server running on one modern machine could replace the current emc "compile farm", which is 5 old computers
[12:33:18] <Lerneaen_Hydra> jepler: which is the current version of EMC as TESTING
[12:33:34] <Lerneaen_Hydra> possibly, IIRC the overhead was rather low
[12:34:01] <jepler> Lerneaen_Hydra: 2.0.1 was just released. I would guess that TESTING and 2.0.1 are the same thing right now except for the version number
[12:34:04] <anonimasu> jepler: vmware?
[12:34:08] <Lerneaen_Hydra> oh, ok
[12:34:10] <anonimasu> jepler: you can run xen :)
[12:34:40] <Lerneaen_Hydra> how often is testing updated?
[12:34:43] <Lerneaen_Hydra> (roughly)
[12:34:43] <jepler> anonimasu: does xen allow me to run an unmodified linux kernel as the guest?
[12:35:29] <anonimasu> yeah
[12:36:38] <jepler> Lerneaen_Hydra: I don't think it's being updated very often. but since the branch for the 2.0 series of releases is getting only bugfixes, it's always very stable. (TESTING refers to a CVS tag that is moved from time to time, not a branch)
[12:37:46] <Lerneaen_Hydra> jepler: oh, ok. if I am waiting for jogwheel support, should I go for testing or head?
[12:39:34] <jepler> Lerneaen_Hydra: that will not be added to the 2.0 branch, since it's not a bugfix. so you won't get it until we decide to put TESTING on the branch that will be 2.1.
[12:39:47] <Lerneaen_Hydra> oh, ok
[12:40:17] <Lerneaen_Hydra> jepler: so I guess going for head would be a good idea (it would give the functionality the quickest)
[12:41:02] <jepler> yes
[12:43:06] <anonimasu> hm, does anyone of you have a good servo sizing howto
[12:43:13] <anonimasu> or can point me to anything about it..
[12:43:17] <anonimasu> I've forgotten ;)
[12:47:16] <alex_joni> morning
[12:47:32] <alex_joni> jepler: made a release for 2.0.1 at sourceforge
[12:47:34] <jepler> hi alex_joni
[12:47:51] <cradek> actually the TESTING tag points to the pre-2.0.0 release
[12:47:52] <alex_joni> and updated the linuxcnc.org news section.. hope it's ok
[12:48:05] <alex_joni> cradek: this is an interesting point..
[12:48:09] <cradek> I'm not sure what should be done with it
[12:48:16] <alex_joni> do we want to keep TESTING on v2_0_branch?
[12:48:22] <alex_joni> or move it to stable HEAD releases?
[12:48:38] <alex_joni> I guess the later would be more usefull for users
[12:49:33] <alex_joni> oh, btw.. tried Xgl today (some LiveCD), and it works / looks stunning.. not sure it's usefull though ;)
[12:49:54] <cradek> I don't think there's any point to having it on the branch - any bugfix there will turn into a release
[12:50:00] <jepler> are we confident enough in HEAD to move TESTING there?
[12:50:09] <Lerneaen_Hydra> jepler: what's the easiest way to get head?
[12:50:23] <Lerneaen_Hydra> cvs? or is there a .deb for both emc and axis?
[12:50:34] <cradek> jepler: probably
[12:51:36] <jepler> Lerneaen_Hydra: cvs
[12:52:05] <Lerneaen_Hydra> jepler: any quick-guide available for doing that?
[12:52:18] <SkunkWorks> morning
[12:52:25] <Lerneaen_Hydra> especially getting emc and axis to work together
[12:52:36] <jepler> Lerneaen_Hydra: First, get emc2: cvs -z5 -d:ext:anon@cvs.linuxcnc.org:/cvs co emc2
[12:52:54] <jepler> Lerneaen_Hydra: then, get axis: http://axis.unpy.net/files/downloads/nightly/axis-latest.tar.bz2
[12:53:01] <jepler> untar axis-latest inside emc2/src
[12:53:27] <jepler> (so there will be a file like emc2/src/axis-ps999/setup.py)
[12:53:28] <alex_joni> cradek: what axis did you pack together with emc2.0.1 ?
[12:53:41] <alex_joni> * alex_joni will want to pack the same one lateron
[12:53:45] <jepler> alex_joni: CVS HEAD, I think
[12:53:50] <jepler> called '1.4a0'
[12:54:14] <cradek> yes
[12:54:14] <jepler> Lerneaen_Hydra: when you run configure it should say: checking for AXIS source... will build axis from <directory name>
[12:54:20] <alex_joni> jepler: ok, I might ask again later tonight when I'll get around to do it..
[12:54:21] <Lerneaen_Hydra> oh, ok
[12:54:40] <Lerneaen_Hydra> jepler: which directory will all this go into? and is it easy to update later on?
[12:55:22] <jepler> Lerneaen_Hydra: the "cvs ... co" will go in a directory called "emc2"
[12:55:32] <Lerneaen_Hydra> in my home dir?
[12:55:33] <jepler> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Installing_EMC2
[12:55:50] <jepler> it creates a subdirectory "emc2" in the directory where you execute the command
[12:56:08] <Lerneaen_Hydra> oh, ok
[12:56:31] <jepler> when you run configure, give it the flag --enable-run-in-place so it doesn't put any files outside the top directory
[12:56:43] <jepler> that way you can easily go back to the version installed as a .deb package
[12:59:03] <Lerneaen_Hydra> which kernel do I want?
[13:00:00] <jepler> the same one you've already been using for emc2
[13:00:11] <Lerneaen_Hydra> this is a new install
[13:00:14] <Lerneaen_Hydra> 6.06
[13:00:24] <jepler> oh
[13:00:58] <jepler> we haven't prepared a kernel for 6.06 yet, and the one from breezy doesn't work right (reports are that network and sound drivers don't load)
[13:01:12] <Lerneaen_Hydra> hmm, that doesn't sound so good ;)
[13:01:15] <Lerneaen_Hydra> I'll wait a while then
[13:04:00] <jepler> bbl
[13:30:32] <giacus> got no luck with latest compilation: http://www.giacus.org/files/shot6.png
[13:30:59] <giacus> I'm missing some fs module I suppose
[13:31:32] <cradek> you don't have devfs support in the kernel (or the devfs module in your initrd)
[13:31:51] <giacus> looking..
[13:31:53] <giacus> ty
[13:54:34] <jepler> http://thedailywtf.com/forums/thread/75718.aspx
[13:57:27] <giacus> lol
[14:30:20] <Lerneaen_Hydra> any idea of how much work is needed to get emc to work on 6.06?
[14:42:41] <jepler> someone has to build a realtime kernel that works properly
[14:42:50] <jepler> after that, emc2 will be very easy to build
[14:45:27] <cradek> Are you file update ? WUW.sad is already exist. [Yes] [No] [Cancel]
[14:46:00] <jepler> My favorite was [ _OK ] [ O_K ] [ _Cancel ] [ OK ]
[14:47:13] <cradek> yeah that's a nice one too
[14:50:53] <cradek> http://kflickr.sourceforge.net/wikka.php?wakka=Screen
[14:50:59] <cradek> btw, it's supposed to have three OK buttons
[14:56:35] <fenn> it should be [ OK ] [ OK ] [ Uh-huh ] [ OK ]
[15:05:17] <cradek> if you have 3 OK buttons, I suppose it's hard to pick keyboard accelerators for them all
[15:19:58] <A-L-P-H-A-2> hello
[15:24:02] <ValarQ> hiya
[15:27:30] <Lerneaen_Hydra> jepler: I guess a RT linux kernel for 6.06 is a high priority for the developers of the RT kernel?
[15:28:40] <jepler> Lerneaen_Hydra: the rtai people don't seem to produce kernel packages, just patches. cradek had to compile his own realtime-patched kernel for breezy, and that's the one most of us are using.
[15:29:16] <Lerneaen_Hydra> jepler: oh, why couldn't emc use the "standard" rt kernel?
[15:29:39] <SWPadnos> there isn't one
[15:29:48] <Lerneaen_Hydra> the one made by the rtai people
[15:29:52] <Lerneaen_Hydra> the patched one
[15:30:20] <SWPadnos> you end up with patched source, you still have to compile on your system, and optionally build packages that others can install
[15:30:26] <Lerneaen_Hydra> oh, maybe you meant that cradek used the patches and compiled a kernel, but didn't write any RT-kernel code
[15:30:29] <SWPadnos> that's what cradek does (as jepler said)
[15:30:33] <SWPadnos> right
[15:30:41] <alex_joni> Lerneaen_Hydra: and you still need to change a LOT of config options
[15:30:45] <alex_joni> till you get a proper kernel
[15:30:53] <alex_joni> it's not a trivial exercise
[15:30:56] <Lerneaen_Hydra> oh, like system capabilities and so on?
[15:31:08] <alex_joni> especially when you think about distributing it to a lot of platforms
[15:31:13] <alex_joni> Lerneaen_Hydra: right
[15:31:36] <cradek> building a rt kernel for your own machine isn't too hard (if you know how to build kernels) but building a package that lots of people can use is very difficult.
[15:31:53] <alex_joni> cradek: building a _good_ package ..
[15:32:07] <cradek> heh, that lots of people can use *successfully*
[15:32:21] <alex_joni> Lerneaen_Hydra: dapper is quite new, and it does have some infantility problems
[15:32:28] <Lerneaen_Hydra> because you need to know many of the intricacies between the differing processors and so on?
[15:32:31] <alex_joni> cradek: or that :)
[15:32:50] <alex_joni> Lerneaen_Hydra: right, and to figure out FB for the bootsplash, and modules, and filesystems
[15:32:50] <Lerneaen_Hydra> alex_joni: any serious problems?
[15:32:56] <alex_joni> Lerneaen_Hydra: nothing serious
[15:33:30] <alex_joni> Lerneaen_Hydra: doing a test kernel for dapper I encountered 2 bugs, one in the software used to build the kernel package (make-kpkg) and one in the kernel source itself
[15:33:56] <alex_joni> both aren't very serious, but if you don't nail them you can spend ages and still be in the same spot
[15:34:09] <Lerneaen_Hydra> oh, you need to change settings that would apply to the non-rt kernel too? (filesystems and so on)
[15:34:30] <alex_joni> no, but support for various filesystems is in the kernel
[15:34:31] <Lerneaen_Hydra> alex_joni: I take it you reported them?
[15:34:48] <alex_joni> Lerneaen_Hydra: was talking with the people who were responsible for them ;)
[15:34:55] <alex_joni> which was my luck actually..
[15:35:07] <alex_joni> otherwise I wouldn't have figured them out so quickly..
[15:35:28] <Lerneaen_Hydra> alex_joni: yes, but previously you wrote "right, and to figure out FB for the bootsplash, and modules, and filesystems", I read that as, you need to config everything, even things like that
[15:35:44] <giacus> Lerneaen_Hydra: i just compiled the kernel few days ago for the first time, not very hard
[15:35:46] <alex_joni> right, you do
[15:35:49] <giacus> and funny :P
[15:36:10] <alex_joni> Lerneaen_Hydra: you need to compile (actually enable the compiling) of all the possible modules users out there might want
[15:36:11] <cradek> giacus: does it boot yet? :-)
[15:36:21] <giacus> http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl/emcinfo.pl?RtaiSteps
[15:36:25] <giacus> cradek: yep
[15:36:30] <alex_joni> cradek: think giacus made it boot .. :)
[15:36:38] <giacus> as a stupid I was missing udev
[15:36:43] <alex_joni> but for one PC that is quite easy
[15:36:54] <giacus> pacthed with rtai now and compiling again
[15:37:03] <alex_joni> nowadays there are so many things .. so you can't expect to even know what most are
[15:37:05] <Lerneaen_Hydra> alex_joni: and there are no default (non-rt-kernel standards) settings?
[15:37:07] <alex_joni> or what is needed
[15:37:29] <alex_joni> Lerneaen_Hydra: there are the settings for the non-rt-kernel that comes with dapper
[15:37:38] <Lerneaen_Hydra> bloated or needed-complication?
[15:37:38] <alex_joni> but using those will break a RT kernel
[15:37:49] <Lerneaen_Hydra> alex_joni: oh. that sounds nasty
[15:37:56] <alex_joni> probably good enough.. but some stuff is enabled which can cause RT stuff to break
[15:38:03] <Lerneaen_Hydra> alex_joni: and I take it some of those settings are needed by some people?
[15:38:04] <alex_joni> for example Power Management & ACPI
[15:38:20] <alex_joni> Lerneaen_Hydra: there are some who whine when they don't have them ;)
[15:38:32] <alex_joni> like the PC automatically powering off after shutdown
[15:38:44] <Lerneaen_Hydra> oh, so that wasn't a bug
[15:38:47] <alex_joni> ACPI is a nasty thing
[15:38:56] <Lerneaen_Hydra> I was wondering why it didn't powerdown after a halt
[15:39:13] <alex_joni> it's terribly implemented (by the BIOS manufactureres) so the code to support it is pretty nasty
[15:39:43] <alex_joni> Lerneaen_Hydra: the problem with rtai & rtlinx & comedi is that they really lack some good documentation
[15:40:03] <alex_joni> not only the API's which kinda are there, but documentation on this kind of stuff
[15:40:22] <Lerneaen_Hydra> alex_joni: sounds like the same thing that plagues most OSS
[15:40:34] <alex_joni> otoh, kernel development rushes forward like a freighttrain, you probably find it hard to keep up if you don't have the same kind of manpower
[15:40:57] <alex_joni> Lerneaen_Hydra: I used to build my kernels a few years ago
[15:41:10] <alex_joni> and I just started again a couple of days ago
[15:41:22] <alex_joni> let me tell you there are worlds between the experiences
[15:41:43] <alex_joni> I couldn't even recognize the names of half of the stuff
[15:41:53] <alex_joni> like EFI, ABI, fbcon, ..
[15:43:05] <Lerneaen_Hydra> heh, nothing like ambiguous acronyms
[15:43:25] <alex_joni> Lerneaen_Hydra: I probably could fill two pages with stuff like that
[15:43:43] <alex_joni> so to sumarize it: you can build a good kernel for yourself easily
[15:43:54] <alex_joni> you can even get lucky and build one that works for most people
[15:44:25] <alex_joni> but to make one that will always work, requires someone with quite some skills (and time to follow all the kernel development stuff, and rtai devel stuff, ...)
[15:45:57] <jepler> and why "sometime before support for breezy is dropped" is the goal I'd set for having dapper packages, not "sometime in the first week after dapper is released".
[15:46:14] <cradek> I agree totally
[15:46:35] <cradek> when is breezy support ending?
[15:47:08] <Lerneaen_Hydra> IIRC they support for 1 year after it's outdated
[15:47:12] <SWPadnos> I think they had a 2-year policy before
[15:47:23] <cradek> ok so we have a year or two
[15:47:25] <SWPadnos> I think that would be from release data though
[15:47:28] <SWPadnos> date
[15:47:47] <cradek> ideally they should overlap quite a bit to allow people to upgrade at their convenience.
[15:48:13] <cradek> I mean, our emc support for both versions should overlap.
[15:50:10] <SWPadnos> that probably means that a version of the RT kernel based on 2.6.15 should be made for breezy
[15:50:14] <alex_joni> WartyWarthog - Version 4.10. The first release; supported until April 2006.
[15:50:21] <SWPadnos> so that people with dapper aren't surprised by a "downgrade"
[15:50:32] <alex_joni> HoaryHedgehog - Version 5.04. Previous stable release; supported until October 2006.
[15:50:48] <alex_joni> breezy doesn't even have a date set when support will stop for it..
[15:50:56] <SWPadnos> Tuessday ;)
[15:54:16] <Lerneaen_Hydra> alex_joni: IIRC they had long support, 5 years or so
[15:54:35] <Lerneaen_Hydra> probably so they can play with the newer versions more, maybe XGL and other such stuff
[15:55:12] <alex_joni> dapper will be LTS (Long Term Support)
[15:55:25] <cradek> 5 years without upgrading an OS sounds great to me
[15:55:34] <alex_joni> because edgy (the one after) will be quite aggressively reorganized / improved
[15:55:39] <cradek> I don't like the leading edge anymore
[15:55:48] <alex_joni> cradek: I can understand ;)
[15:55:56] <alex_joni> otoh, apt-get does it all
[15:56:26] <alex_joni> with proper dapper kernel & emc2 debs one should be able to apt-get distupgrade to dapper from breezy
[15:56:37] <alex_joni> but then again.. I see little reason to do that
[15:56:56] <alex_joni> Lerneaen_Hydra: I installed dapper a few days ago, and I can see very little difference
[15:57:09] <alex_joni> nothing that might make me suggest an user to upgrade
[15:57:42] <Lerneaen_Hydra> alex_joni: I installed it yesterday, and other then some graphical polish I didn't see much either
[15:57:48] <alex_joni> XGl is a "nifty" thing, but I bet it won't work with RTAI, so I don't see much reason to be joyfull about that
[15:58:12] <giacus> * giacus giacus use Debian/GNU Linux 2.6.16-flinstones
[15:58:29] <Lerneaen_Hydra> alex_joni: XGI is really something not for EMC, but it may be useful for dekstop systems
[15:58:49] <Lerneaen_Hydra> * Lerneaen_Hydra doesn't really know all that much about it
[15:58:57] <fenn> * fenn poops on XGl
[15:59:59] <cradek> be careful, you'll set off the jiggling again
[16:00:00] <giacus> really was GNU/Linux :P
[16:00:26] <giacus> Lerneaen_Hydra: what the machine should run emc ?
[16:00:35] <giacus> cpu, ram, video, etc ..
[16:00:56] <giacus> your*
[16:00:56] <Lerneaen_Hydra> giacus: you mean in my case?
[16:01:14] <giacus> yeah
[16:01:50] <Lerneaen_Hydra> intel PIII 800mhz, 512ram, integrated video, some 3.5gb harddisk I found lying around
[16:02:16] <giacus> similar to the mine, except that is celeron
[16:02:24] <giacus> is the video card nvidia ?
[16:02:37] <giacus> I suppose not
[16:02:44] <giacus> should be sys maybe...
[16:03:39] <giacus> SiS*
[16:04:54] <giacus> the only issue I found in the wiki seems related to nvidia cards
[16:05:03] <giacus> for the rest you should be happy
[16:05:27] <giacus> with that HW
[16:06:24] <Lerneaen_Hydra> yes, it's quite good actually
[16:07:01] <Lerneaen_Hydra> the integrated card is probably an intel graphics extreme (yeah, very extreme, can't even do 1280x1024 without artifacts)
[16:07:06] <jepler> onboard video can be trouble too
[16:07:20] <jepler> be sure to run the latency test and do stuff like focus, raise, and lower different windows
[16:07:30] <giacus> Lerneaen_Hydra: has no agp slot ?
[16:07:50] <Lerneaen_Hydra> it has one, but I havn't felt the need to use it
[16:08:00] <SkunkWorks> does anyone remember what video card was in the mazak?
[16:08:04] <Lerneaen_Hydra> jepler: how do I do the test?
[16:09:12] <giacus> cd /usr/realtime/testsuite/kern/latency/
[16:09:13] <giacus> ./run
[16:10:20] <jepler> I think you have to also manually create the /dev/rtf/3 device before running the latency test: mkdir /dev/rtf; mknod /dev/rtf/3 c 150 3
[16:10:48] <Lerneaen_Hydra> ok, I'll just install breezy first
[16:13:48] <alpha___> alpha___ is now known as A-L-P-H-A
[16:14:42] <A-L-P-H-A> k... need some help... http://ubuntuforums.org/showthread.php?p=1097290#post1097290
[16:14:55] <A-L-P-H-A> I can't get my third monitor to work in ubuntu.
[16:19:06] <giacus> A-L-P-H-A: should it depend by xorg or ati drivers ?
[16:19:18] <A-L-P-H-A> giacus: ?
[16:19:49] <A-L-P-H-A> giacus: I'm having the problem of the third monitor to even show anything. and if it doesn't work, it just locks up the whole computer
[16:20:19] <giacus> I meant could the cause be in the ati drivers ?
[16:21:07] <giacus> I seen you posted the xorg.conf ..
[16:21:28] <A-L-P-H-A> I don't know.
[16:21:38] <A-L-P-H-A> these are the OSS drivers.
[16:21:48] <A-L-P-H-A> not the ATI official binary drivers.
[16:22:18] <giacus> I'd check the drivers or the ati installer if you're using the closed drivers vers.
[16:22:43] <giacus> from there you should be able to set lot of options
[16:24:36] <giacus> i.e. how many monitors to use
[16:25:32] <A-L-P-H-A> 3...
[16:30:27] <cradek> I think you should post (or link to) the full X output too
[16:34:55] <giacus> * giacus cross fingers foot ..
[16:35:14] <jepler> new colors in halscope: http://emergent.unpy.net/files/sandbox/halscope.png
[16:35:15] <A-L-P-H-A> cradek: , how?
[16:35:34] <giacus> jepler running the rtai test on the vm
[16:35:43] <SWPadnos> pastebin with the contents of /var/log/Xorg.0.log (or something like that)
[16:35:57] <giacus> got lot of overruns
[16:37:49] <jepler> giacus: all the time, or just when you move windows or things like that?
[16:37:52] <CIA-8> 03jepler 07HEAD * 10emc2/src/hal/utils/ (scope_disp.c scope_usr.h scope_vert.c):
[16:37:52] <CIA-8> revise trace colors
[16:37:52] <CIA-8> show trace colors on the trace selection buttons
[16:37:52] <CIA-8> change readout to show next to the "Logging" button instead of in a tooltip,
[16:37:52] <CIA-8> and always come from selected trace instead of the trace near the cursor
[16:38:39] <fenn> jepler: you can get rid of that logging button if you want
[16:38:50] <fenn> it doesnt do anything yet anyway
[16:38:58] <giacus> jepler: just running the test, nothing other ..
[16:39:07] <jepler> giacus: that's unfortunate
[16:39:22] <giacus> give me a minut
[16:39:27] <fenn> the interface is kinda crammed
[16:40:27] <CIA-8> 03compile-farm 07Ubuntu 5.10 (2.6.12-magma) * 10emc2head/: build FAILED ; see http://linuxcnc.org/compile_farm/emc2head_slot7_log.txt
[16:40:42] <jepler> oh no!
[16:41:29] <CIA-8> 03jepler 07HEAD * 10emc2/src/hal/utils/scope_disp.c: oops
[16:42:29] <Roguish> hello, got a problem with emc2. "Starting emc...
[16:42:31] <Roguish> insmod: error inserting '/usr/realtime-2.6.12-magma/modules/rtai_hal.ko': -1 Invalid module format"
[16:42:47] <Roguish> waz up?
[16:43:38] <A-L-P-H-A> cradek: ???? still there?
[16:44:12] <jepler> Roguish: look in dmesg for more information about the error
[16:44:31] <Roguish> where is that located?
[16:44:40] <SWPadnos> run "dmesg"
[16:44:41] <giacus> jepler: http://www.giacus.org/files/rtai_vm-2.6.16-magma.avi
[16:44:43] <jepler> "dmesg" is a shell command, it will print the kernel message log to the screen
[16:45:25] <jepler> giacus: -ENOCODEC
[16:45:43] <SWPadnos> heh
[16:46:17] <Roguish> ok, what am i looking for?
[16:46:46] <jepler> Roguish: the last lines are the most recent. If you want to just put the last 20-30 lines on pastebin I'll look at them.
[16:46:49] <CIA-8> 03compile-farm 07Ubuntu 5.10 (2.6.12-magma) * 10emc2head/: build FAILED ; see http://linuxcnc.org/compile_farm/emc2head_slot7_log.txt
[16:46:56] <jepler> (http://pastebin.com/)
[16:47:06] <giacus> jepler: is xvid ok ?
[16:47:50] <jepler> giacus: what are you trying to show me?
[16:47:59] <giacus> the test running
[16:48:20] <giacus> :)
[16:48:21] <jepler> ummm how about a paste of the text?
[16:48:33] <giacus> sure
[16:48:39] <jepler> I'm very happy for you that you've found an exceedingly complicated way to do something that should be very simple
[16:48:40] <giacus> taking a shot ..
[16:48:50] <CIA-8> 03compile-farm 07BDI-4.20 (2.6.10-adeos) * 10emc2head/: build FAILED ; see http://linuxcnc.org/compile_farm/emc2head_slot5_log.txt
[16:48:50] <giacus> btw, the avi is very small
[16:49:02] <Roguish> ok, just put them there (I think)
[16:49:11] <jepler> Roguish: pastebin should give you a URL for me to go to
[16:49:29] <Roguish> http://pastebin.com/760110
[16:49:43] <CIA-8> 03jepler 07HEAD * 10emc2/src/hal/utils/scope_vert.c: get rid of unused logging button. #ifdefs for gtk1.2 to fix farm
[16:50:02] <jepler> Roguish: this is the important error: [ 3108.406002] rtai_hal: version magic '2.6.12-magma 586TSC gcc-3.4' should be '2.6.15-23-386 preempt 486 gcc-4.0'
[16:50:18] <jepler> Roguish: it indicates that you didn't boot the realtime kernel (2.6.12-magma) but are using the kernel '2.6.15-23-386' instead.
[16:50:36] <jepler> Roguish: after installing the realtime kernel, you must reboot, and select it from the boot menu if it is not the default (it should be the default)
[16:51:03] <Roguish> i just fired up the machine and did all the updates.
[16:51:28] <Roguish> updated ubunto and emc2
[16:52:09] <Roguish> how do i check to see if the realtime is default?
[16:52:29] <alex_joni> Roguish: try 'uname -a'
[16:52:33] <alex_joni> and look what the name is
[16:53:13] <CIA-8> 03compile-farm 07Ubuntu 5.10 (2.6.12-magma) * 10emc2head/: build FAILED ; see http://linuxcnc.org/compile_farm/emc2head_slot7_log.txt
[16:53:45] <Roguish> Linux xps400 2.6.15-23-386 #1 PREEMPT Tue May 23 13:49:40 UTC 2006 i686 GNU/Linux
[16:53:46] <CIA-8> 03jepler 07HEAD * 10emc2/src/hal/utils/scope_vert.c: try again to fix farm
[16:55:16] <jepler> Roguish: you have to boot the 2.6.12-magma kernel. When you reboot, you may see a message telling you to press escape to enter the boot menu. do so. Once you're at the menu, choose the line that says "2.6.12-magma" but not "recovery mode"
[16:55:42] <CIA-8> 03compile-farm 07BDI-TNG (2.4.18-rtai) * 10emc2head/: build FAILED ; see http://linuxcnc.org/compile_farm/emc2head_slot3_log.txt
[16:55:51] <Roguish> will try it
[16:56:01] <giacus> jepler: http://www.giacus.org/files/test.txt
[16:56:12] <giacus> I didn't install vmtools yet
[16:56:26] <giacus> I hope that could make the difference, right ?
[16:56:44] <SWPadnos> it helps a lot with mouse focis and the like
[16:56:47] <SWPadnos> focus
[16:57:04] <giacus> about overruns
[16:57:21] <SWPadnos> I don't think it has much effect on the actual VM execution
[16:57:39] <SWPadnos> I believe it's more about interoperability between the host and guest OSes
[16:57:42] <giacus> ok, lets try ..
[16:57:52] <giacus> only in that way we can know :)
[16:57:58] <SWPadnos> yep
[16:58:48] <giacus> SWPadnos: should I burn a cdrom for that ..
[16:58:52] <giacus> can't recall
[16:58:59] <SWPadnos> for the tools install?
[16:59:03] <giacus> yeah
[16:59:09] <CIA-8> 03compile-farm 07BDI-4.20 (2.6.10-adeos) * 10emc2head/: build FAILED ; see http://linuxcnc.org/compile_farm/emc2head_slot5_log.txt
[16:59:17] <SWPadnos> nope. sele3ct "install VMWare tools" on the host
[16:59:29] <SWPadnos> then browse to the CDROM in the guest, and install from there
[16:59:36] <SWPadnos> a fake ISO will be mounted in the guest
[16:59:39] <giacus> K, thanks
[16:59:40] <SWPadnos> (as the CDROM)
[16:59:44] <SWPadnos> np
[17:00:10] <SWPadnos> that threw me for a while - a Windows guest automatically runs the install because of "auto-run"
[17:00:39] <SWPadnos> I thought it was just vmware magic, and was surprised when it didn't work with the Linux guest ;)
[17:00:42] <CIA-8> 03compile-farm 07Ubuntu 5.10 (2.6.12-magma) * 10emc2head/: build PASSED
[17:02:17] <giacus> uhm, rpm or tar.gz, not deb
[17:02:38] <giacus> I's try alien
[17:02:42] <giacus> I'd
[17:06:58] <Roguish> gentlemen, still no luck: http://pastebin.com/760158
[17:07:57] <A-L-P-H-A> :(
[17:08:09] <A-L-P-H-A> I want ubuntu to just work with three monitors!!! :(
[17:09:08] <bill203> that's not too much to ask.
[17:11:10] <CIA-8> 03compile-farm 07BDI-TNG (2.4.18-rtai) * 10emc2head/: build PASSED
[17:13:44] <alex_joni> jepler: that 2.6.15-23 is the kernel on dapper
[17:14:02] <alex_joni> so it sounds like Roguish is up for a fight
[17:14:27] <SWPadnos> heh - I guess we have a dapper customer ;)
[17:15:13] <A-L-P-H-A> I am a dapper user... :(
[17:15:14] <alex_joni> jepler: I'm not always around here, but I noticed that the emc2 package fails on dapper because of udev
[17:15:23] <A-L-P-H-A> and I'm HATING this BLACK BLANK centre monitor!
[17:15:39] <alex_joni> A-L-P-H-A: that's the price for living on the edge
[17:16:15] <A-L-P-H-A> the price is too high
[17:16:22] <SWPadnos> did it work in breezy?
[17:16:27] <A-L-P-H-A> yeah
[17:16:30] <A-L-P-H-A> it did.
[17:16:32] <alex_joni> 2 out of 3 sounds like a good result
[17:16:33] <alex_joni> :D
[17:16:35] <A-L-P-H-A> actually, it maybe have been hoary
[17:16:39] <SWPadnos> do you have a copy of the XOrg config that worked?
[17:16:42] <fenn> you might try looking at the xorg.conf file from breezy
[17:16:46] <A-L-P-H-A> AJ... I feel like killing you. hahaha
[17:17:00] <alex_joni> A-L-P-H-A: I only have 1 most of the time..
[17:17:01] <A-L-P-H-A> SWPadnos: yeah... I did that... well, bits and pieces out of it.
[17:17:07] <alex_joni> so 2 sounds pretty much ok
[17:17:18] <A-L-P-H-A> alex_joni: I got the 2 working off the PCI card (dual head)
[17:17:22] <A-L-P-H-A> using xinerama.
[17:17:28] <A-L-P-H-A> but I can't activate the AGP card.
[17:17:31] <A-L-P-H-A> if I try, it dies.
[17:17:39] <A-L-P-H-A> like, locks up the computer.
[17:17:47] <alex_joni> what kernel?
[17:17:59] <SWPadnos> can you use only the AGP card? (ie, remove the PCI card and use AGP only)
[17:18:11] <A-L-P-H-A> SWPadnos: don't know.
[17:18:15] <A-L-P-H-A> alex_joni: how do I check?
[17:19:04] <alex_joni> uname -a
[17:19:41] <A-L-P-H-A> Linux alpha-desktop 2.6.15-23-386 #1 PREEMPT Tue May 23 13:49:40 UTC 2006 i686 GNU/Linux
[17:21:07] <CIA-8> 03compile-farm 07BDI-4.20 (2.6.10-adeos) * 10emc2head/: build PASSED
[17:21:22] <A-L-P-H-A> http://pastebin.com/760190 <-- My original xorg.conf from when paul_c was helping me out.
[17:21:47] <A-L-P-H-A> nothing should have changed.
[17:32:55] <giacus> the rtai test is going to be crazy here ..
[17:33:25] <Bo^Dick> the voltage on my parallel ports data pins is 3.34 volts when high
[17:33:28] <Bo^Dick> is this normal?
[17:35:27] <giacus> Bo^Dick: I think so
[17:35:40] <giacus> especially if is a laptop
[17:37:49] <giacus> any value up to 3V should be assumed as high I guess
[17:38:33] <giacus> sometime it could depend on motherboard model
[17:38:52] <giacus> but its hard to read 5V for what I know
[17:42:20] <Roguish> hello again. still no go, although it looks like realtime is running, i think. emc config screen comes up, i select demo_step_cl, then nothing........
[17:42:27] <Lerneaen_Hydra> isn't TTL wonderful...
[17:45:52] <A-L-P-H-A> giacus: could have a pullup resistor to it...
[17:48:08] <giacus> yes
[17:49:22] <Roguish> any more help here would be appreciated.
[17:50:19] <giacus> I meant as general rule any value < 0.8 V should be 0 logic, any value > 2V should be 1 logic
[17:50:28] <giacus> for what I remember ..
[17:51:36] <Lerneaen_Hydra> yes, with 0 and 5V as nominal voltage
[17:51:40] <fenn> Roguish: are you running from the terminal/
[17:51:52] <fenn> Roguish: run emc from the terminal (type "emc")
[17:52:02] <fenn> tell us what the error messages are
[17:52:05] <Roguish> no, but i can, ok
[17:53:03] <fenn> maybe i need pullup resistors on my parport too
[17:53:08] <fenn> that would explain a lot of things
[17:53:28] <Lerneaen_Hydra> fenn: is the best way to install emc (head version) on a fresh breezy install to first use cradek's script, then remove emc & axis, and then compile emc and axis (head)?
[17:53:55] <Lerneaen_Hydra> fenn: I don't have any, it depends on what they are connected to. Mine are connected to optocouplers
[17:54:01] <SWPadnos> you don't need to remove the installed version
[17:54:03] <fenn> if i were compiling emc i'd make sure to do ./configure --enable-run-in-place
[17:54:15] <Lerneaen_Hydra> SWPadnos: is there any reason to have it there?
[17:54:45] <SWPadnos> Lerneaen_Hydra, it's probably easiest to run apt-get build-dep emc2-axis, which will insure that all libraries and compilers are installed
[17:55:04] <SWPadnos> only for comparison between HEAD and released versions
[17:55:09] <Lerneaen_Hydra> oh, I've already run cradeks script
[17:55:15] <SWPadnos> which can be good for diagnosing hardware vs software problems
[17:55:34] <SWPadnos> did the script run build-dep? (I don't remember)
[17:55:41] <SWPadnos> apt-get build-dep, that is
[17:55:57] <Roguish> ok: Starting emc...
[17:55:58] <Roguish> .
[17:56:00] <Roguish> .
[17:56:01] <Roguish> .
[17:56:03] <Roguish> .
[17:56:04] <Roguish> .
[17:56:06] <Roguish> .
[17:56:07] <Roguish> .
[17:56:09] <Roguish> .
[17:56:10] <Roguish> .
[17:56:12] <Roguish> .
[17:56:13] <Roguish> Can't write to /dev/rtai_shm - aborting
[17:56:13] <fenn> yah i dont know what those dots are all about
[17:56:15] <Roguish> Realtime system did not load
[17:56:16] <Roguish> Shutting down and cleaning up EMC...
[17:56:18] <Roguish> RTAPI: ERROR: could not open shared memory
[17:56:19] <Roguish> HAL: ERROR: rtapi init failed
[17:56:21] <Roguish> halcmd: hal_init() failed
[17:56:22] <Roguish> NOTE: 'rtapi' kernel module must be loaded
[17:56:24] <Roguish> halcmd: hal_init() failed
[17:56:24] <fenn> ok
[17:56:26] <Roguish> NOTE: 'rtapi' kernel module must be loaded
[17:56:27] <Roguish> halcmd: hal_init() failed
[17:56:29] <Roguish> NOTE: 'rtapi' kernel module must be loaded
[17:56:31] <Roguish> Cleanup done
[17:56:44] <fenn> this is because the realtime device files arent in /dev
[17:56:54] <Roguish> that is a straight copy from the screen.
[17:57:02] <fenn> what distro is this again?
[17:57:15] <fenn> (not that i have a clue how any of them handle it)
[17:57:49] <fenn> anyway.. type this:
[17:57:59] <Roguish> this is the ubuntu
[17:58:01] <Lerneaen_Hydra> IIRC he was running dapper
[17:58:43] <fenn> mknod /dev/rtai_shm c 10 254 && mknod /dev/rtf3 c 150 3
[17:59:07] <Roguish> yes, it's dapper
[17:59:13] <fenn> then try it again and see what happens
[17:59:37] <Roguish> what, run emc from the command line?
[17:59:48] <fenn> oh, you might need to do some sudo'ing and chmodding
[18:00:03] <A-L-P-H-A-2> damn you ubuntu!
[18:00:06] <A-L-P-H-A-2> it crashed again!
[18:00:13] <A-L-P-H-A-2> well... from X fuck'n up
[18:00:14] <A-L-P-H-A-2> :(
[18:00:15] <fenn> actually, this is a realtime OS problem, you might get better answers by running the rtai latency tests instead
[18:00:18] <A-L-P-H-A-2> I'm angry now
[18:00:37] <SWPadnos> A-L-P-H-A-2, I had some problems with my ATI card and ubuntu (breezy as well)
[18:00:42] <giacus> cradek: why do you included bwidget package in the repository ? is it different compared to the same package provided by official repository ?
[18:00:51] <A-L-P-H-A-2> SWPadnos: this is just hurting me
[18:00:52] <fenn> Roguish: what did you do before it stopped working?
[18:00:52] <SWPadnos> I was just reminded of this because I upgraded, and the problems are back ;)
[18:02:02] <Roguish> i let ubuntu do it's automatic update of the emc package, then the update from breezy to dapper.
[18:03:11] <A-L-P-H-A-2> SWPadnos: what are you using?
[18:03:22] <SWPadnos> Radeon 8500DV
[18:03:25] <A-L-P-H-A-2> hardware, distro, kernal, drivers wise
[18:03:27] <SWPadnos> one monitor
[18:03:48] <A-L-P-H-A-2> ati binaries? or OSS DRI/ATI/Radeon drivers?
[18:03:48] <Roguish> guess i should have shut off all the automatic updating like i do with xp.
[18:03:53] <SWPadnos> well, the distro is now Dapper (upgraded from Breezy), default drivers
[18:04:03] <SWPadnos> the machine locks up when X loads
[18:04:17] <A-L-P-H-A-2> in your xorg.conf does it say driver "ati" or driver "radeon"
[18:04:18] <SWPadnos> I think it may have something to do with DRI. I'm looking for the solution I found before
[18:04:36] <A-L-P-H-A-2> yeah... that's what's happening with me... X locks up.
[18:04:40] <SWPadnos> ati
[18:04:57] <A-L-P-H-A-2> yeah, those are the DRI /mesa drivers.
[18:05:08] <A-L-P-H-A-2> frak'n stupid
[18:05:33] <fenn> Roguish: did it work after you did the mknod stuff?
[18:06:10] <fenn> if you havent done that yet, does ls /dev/rt* show anything?
[18:06:11] <SWPadnos> those are the only onesthat will work with an RT kernel
[18:07:41] <Roguish> fenn: rtc
[18:08:16] <fenn> ewird
[18:08:43] <fenn> Roguish: sorry, but i dont know that much about either breezy or dapper, so i cant really help, and i have to go.. bbl
[18:09:15] <fenn> Roguish: but try mknod'ing /dev/rtf0 to rtf9 and rtai_shm and see if you can run realtime
[18:09:50] <Roguish> feen: thanks for the help
[18:10:03] <fenn> this is what it looks like on my system: http://pastebin.com/760291
[18:10:19] <fenn> you might also have to copy the device files to /etc/udev/devices
[18:10:27] <fenn> or they will disappear on reboot
[18:21:22] <jepler> giacus: bwidget is in the emc2 repository because it's in "universe" and "universe" isn't turned on by default
[18:21:34] <jepler> giacus: the easiest fix seemed to be putting it in the emc2 repository
[18:25:27] <giacus> oh, ok. thanks
[18:25:57] <Bo^Dick> could someone recommend me a software for programming a pic with the "tait-style" hardware http://www.nomad.ee/PIC/parpic.gif
[18:30:40] <jepler> Bo^Dick: I don't know, but searching google for 'tait' 'linux' gives many results.
[18:30:44] <jepler> Bo^Dick: maybe you should investigate some of them
[18:30:47] <SWPadnos> A-L-P-H-A - have you tried starting up in single mode, then running X -configure?
[18:31:07] <A-L-P-H-A> SWPadnos: no. how would I do that?
[18:31:11] <SWPadnos> A-L-P-H-A this worked for me, though I had to manually edit the mouse device dile (/dev/input/mice)
[18:31:18] <A-L-P-H-A> "dile?
[18:31:20] <A-L-P-H-A> "
[18:31:29] <SWPadnos> err - file ;)
[18:31:30] <A-L-P-H-A> SWPadnos: got skype installed?
[18:31:33] <SWPadnos> nope
[18:31:40] <A-L-P-H-A> oh... :(
[18:31:51] <A-L-P-H-A> skype = poor mans great long distance.
[18:31:55] <Lerneaen_Hydra> Um, in the new (cvs) version of axis, where is the new offset functionality?
[18:31:59] <SWPadnos> hit ESC to get the GRUB menu, press 'e' on the default line, then 'e' again to edit the line
[18:32:24] <SWPadnos> add the word single to the end of the line, hit enter, then presss b to boot
[18:32:51] <SWPadnos> (to the end of the line that specifies the linux kernel to boot)
[18:33:10] <SWPadnos> that'll get you to single mode
[18:33:16] <A-L-P-H-A> what's single mode?
[18:33:28] <SWPadnos> single user mode - no X login
[18:33:38] <SWPadnos> (and no networking, among other things)
[18:33:45] <A-L-P-H-A> oh!
[18:33:57] <A-L-P-H-A> I've done that in like restore mode.
[18:34:04] <A-L-P-H-A> still crashes like a 'biaotch'
[18:34:06] <cradek> Lerneaen_Hydra: press 'End'
[18:34:16] <Lerneaen_Hydra> oh, there's no button for that?
[18:34:18] <cradek> Lerneaen_Hydra: it's not on a menu yet
[18:34:59] <A-L-P-H-A> SWPadnos: I'll try.
[18:35:06] <SWPadnos> A-L-P-H-A - apparently, restore mode still tries to use a graphical login, which is what you need to aboid
[18:35:07] <A-L-P-H-A> give me a sec to finish this freecell game.
[18:35:10] <SWPadnos> heh
[18:36:45] <A-L-P-H-A> k
[18:36:47] <A-L-P-H-A> won
[18:36:49] <A-L-P-H-A> :D
[18:36:54] <A-L-P-H-A> give me a minute to move my laptop.
[18:37:02] <A-L-P-H-A> and so I' can chat there, and do what is require here.
[18:37:20] <SWPadnos> ok by me
[18:37:47] <Lerneaen_Hydra> cradek: I don't have a visible Z axis in AXIS, even though it's configed in the ini
[18:38:00] <A-L-P-H-A> damn.
[18:38:09] <A-L-P-H-A> there's a DENT in my new computer case. :(
[18:38:22] <cradek> Lerneaen_Hydra: your ini must be wrong, check again
[18:38:42] <Lerneaen_Hydra> could it be becuase I called the two axes X and Z?
[18:38:45] <Lerneaen_Hydra> and not X and Y?
[18:38:50] <Lerneaen_Hydra> (non continuity)
[18:38:57] <cradek> oh this is your lathe?
[18:39:00] <Lerneaen_Hydra> yes
[18:39:09] <cradek> leave your ini with AXES=3 (XYZ) and set [DISPLAY]LATHE=1
[18:39:16] <Lerneaen_Hydra> I've set the lathe parameter to 1
[18:39:25] <Lerneaen_Hydra> oh, it should have 3 axes in the ini?
[18:39:34] <A-L-P-H-A> 'ight
[18:39:44] <Lerneaen_Hydra> what if I want the IO pins that the Y axis allocates?
[18:39:49] <A-L-P-H-A-2> SWPadnos: so hit "E" when grub loads?
[18:39:55] <cradek> you don't have to hook them to anything in hal
[18:40:07] <Lerneaen_Hydra> just as long as they're in the ini?
[18:40:11] <cradek> yes
[18:40:28] <SWPadnos> A-L-P-H-A-2, yes
[18:40:44] <SWPadnos> assuming you've managed to hit ESC fast enough to get the menu
[18:40:57] <A-L-P-H-A-2> I am in the menu
[18:41:07] <A-L-P-H-A-2> crap.
[18:41:12] <A-L-P-H-A-2> it's now change to "ls"
[18:41:15] <A-L-P-H-A-2> :(
[18:41:19] <A-L-P-H-A-2> what's it supposed to be again?
[18:41:53] <A-L-P-H-A-2> nm
[18:41:55] <A-L-P-H-A-2> I could undo!
[18:41:56] <A-L-P-H-A-2> :D
[18:41:58] <SWPadnos> whatever it was, plus "single" at the end (no quotes)
[18:42:12] <A-L-P-H-A-2> nono.
[18:42:19] <A-L-P-H-A-2> I have "root (hd0,1)
[18:42:33] <A-L-P-H-A-2> kernal /boot/vmlinux-2.6.15-23-386 root=/dev/hda2 ro quiet splash
[18:42:40] <Lerneaen_Hydra> cradek: it works now. thanks
[18:42:44] <A-L-P-H-A-2> initrd /boot/initrd.img-2.6.15-23.386
[18:42:45] <SWPadnos> the second line, beginning with "kernel"
[18:42:49] <A-L-P-H-A-2> savedefault
[18:42:54] <A-L-P-H-A-2> k
[18:42:56] <Lerneaen_Hydra> cradek: will the need for the Y axis be removed later?
[18:43:04] <A-L-P-H-A-2> at the end add 'single' without quotes?
[18:43:13] <cradek> Lerneaen_Hydra: hard to say. there's a lot to still be done for lathes.
[18:43:52] <Lerneaen_Hydra> And i've still got a spec to write... ;)
[18:44:00] <A-L-P-H-A-2> eeew... cold coffee sucks... it's kinda bitter...
[18:44:07] <A-L-P-H-A-2> should have added a pinch of salt
[18:44:13] <Lerneaen_Hydra> A-L-P-H-A-2: and hot coffee isn't?
[18:44:32] <A-L-P-H-A-2> Lerneaen_Hydra: no. cause the oils haven't oxidized to that bitter crap.
[18:45:09] <Lerneaen_Hydra> A-L-P-H-A-2: I find hot coffee to be too bitter, so I don't even want to try cold coffee then
[18:45:32] <A-L-P-H-A-2> iced coffee is different too!
[18:45:39] <A-L-P-H-A-2> this is some starbucks coffee bean stuff.
[18:45:45] <A-L-P-H-A-2> great / awesome when hot
[18:45:47] <A-L-P-H-A-2> crap when cold
[18:47:43] <A-L-P-H-A-2> SWPadnos: k... I'm editing the xconf.org file... I'll see if I have that issue still... trying new drivers and crap too
[18:47:57] <SWPadnos> did you run X -configure ?
[18:48:07] <A-L-P-H-A-2> no
[18:48:09] <A-L-P-H-A-2> should I?
[18:48:24] <SWPadnos> that's what helped me out. I haven't looked at the real differences between the files yet
[18:48:43] <SWPadnos> the only thing screwed up after that was the mouse file node
[18:50:21] <A-L-P-H-A-2> "cannot run in framebuffer mode"
[18:50:39] <Lerneaen_Hydra> what was the RT-test code?
[18:50:44] <A-L-P-H-A-2> argh
[18:50:59] <SWPadnos> hmm. I didn't get that
[18:51:23] <Lerneaen_Hydra> there was some script/line of bash code that ran a test of the RT-subsystem
[18:51:23] <SWPadnos> I'm not sure if I was in true text mode or FB mode
[18:57:13] <A-L-P-H-A-2> SWPadnos: ... k... it locked up... I'm gonna try it again...
[18:57:23] <SWPadnos> ok
[18:57:39] <A-L-P-H-A-2> you all know that skype allows FREE calling to within the USA and Canada right?
[18:57:46] <SWPadnos> yeah
[18:57:49] <A-L-P-H-A-2> like PC to Phone free. :D
[18:58:48] <A-L-P-H-A-2> why would this be such a pain in the butt?? :/
[18:58:56] <A-L-P-H-A-2> it doesn't even spit out an error to me, just LOCKS up
[18:59:46] <SWPadnos> I just got a message on the monitor that the refresh rate is out of range (12.5 Hz or something)
[19:00:06] <A-L-P-H-A-2> I had that set already
[19:00:31] <A-L-P-H-A-2> I'm gonna try to just make the "single" AGP monitor work.
[19:00:36] <A-L-P-H-A-2> and then ADD the PCI card back in.
[19:00:40] <A-L-P-H-A-2> think that'd be okay?
[19:00:53] <A-L-P-H-A-2> as I knwo have the config file to make the PCI mons work already
[19:00:58] <SWPadnos> probably.
[19:00:59] <A-L-P-H-A-2> seems logical to me... doesn't it?
[19:01:06] <SWPadnos> it's the PCI card that works now, right>
[19:01:08] <SWPadnos> ?
[19:01:14] <A-L-P-H-A-2> yeah.
[19:01:15] <A-L-P-H-A-2> PCI working
[19:01:17] <A-L-P-H-A-2> AGP no go
[19:01:45] <SWPadnos> ok. one day, I'll look at the substantive differences between the working and non-working conf files
[19:01:53] <SWPadnos> but until then ... good luck ;)
[19:02:11] <SWPadnos> I've got to do some work here. I'll be intermittent for a while ;)
[19:02:14] <CIA-8> 03jepler 07HEAD * 10emc2/src/hal/utils/scope_disp.c:
[19:02:14] <CIA-8> draw a small circle to mark the point shown on the readout.
[19:02:14] <CIA-8> update the point when redrawing traces. this also covers the case that the user clicked a new trace.
[19:02:17] <A-L-P-H-A-2> k
[19:02:47] <SWPadnos> is the CVS server sending out commit messages?
[19:03:04] <cradek> the sf mailing lists are all broken
[19:03:09] <SWPadnos> ah - ok.
[19:03:49] <Lerneaen_Hydra> did anyone know the command to do the RT-test?
[19:04:07] <alex_joni> ./run
[19:04:18] <cradek> sudo mkdir /dev/rtf; sudo mknod /dev/rtf/3 c 150 3; cd /usr/realtime*/tests/kern/latency; ./run
[19:04:22] <cradek> I think
[19:04:29] <alex_joni> google helps
[19:04:47] <alex_joni> check for a link in austria (.at), it has some short scripts to build the /dev/ missing stuff
[19:04:48] <Lerneaen_Hydra> What would a good keyowrd be for that?
[19:05:08] <SkunkWorks> is that on the wiki?
[19:05:23] <SkunkWorks> if not it should be added by someone :)
[19:05:27] <cradek> I agree
[19:05:39] <alex_joni> * alex_joni looks around for someone
[19:05:50] <SkunkWorks> where would be a good place?
[19:06:04] <alex_joni> that's the 100 point question
[19:06:21] <alex_joni> troubleshooting -> realtime system
[19:06:45] <alex_joni> but don't bother looking for those, I just made them up :-)
[19:07:26] <Lerneaen_Hydra> cradek: I dont have the directory realtime*
[19:07:56] <alex_joni> Lerneaen_Hydra: /usr/realtime-2.6.12-magma/
[19:08:03] <alex_joni> * stands for any available
[19:08:04] <Lerneaen_Hydra> oh, ok
[19:08:25] <Lerneaen_Hydra> yes I found the directory rather sctrange
[19:08:29] <Lerneaen_Hydra> *strange
[19:08:31] <giacus> jepler: the final report on the virtual machine here is: lot of overruns, I ran Emc and Axis but very very slow.
[19:09:20] <giacus> dinner.. bbl
[19:09:50] <Lerneaen_Hydra> cradek: I still get a no such file/dir
[19:10:08] <SkunkWorks> troubleshooting - maybe under advaceced topics?
[19:10:09] <Lerneaen_Hydra> and I am running 2.6.16-magma
[19:10:18] <Lerneaen_Hydra> err, 2.6.12
[19:11:21] <SkunkWorks> or maybe there should be a troubleshooting section
[19:11:24] <jepler> giacus: I guess I didn't understand that you were running on a virtual machine. I thought you were talking about running on actual hardware, since you mentioned onboard video.
[19:13:02] <giacus> jepler: oh.. no, that's vmware workstation, running 2.6.16-magma, emc2 testing, axis testing
[19:13:29] <giacus> the video card is emulated vmware, I got no 3d accell here..
[19:13:56] <giacus> real video card is ati radeon 9600 - 256
[19:14:21] <giacus> have to check vmtools again
[19:14:50] <giacus> the installer seems do not recognize correct linux headers
[19:14:51] <Lerneaen_Hydra> cradek: it's called testsuite, not tests
[19:15:45] <alex_joni> Lerneaen_Hydra: really? now that is something cradek probably never would have figured out
[19:16:12] <Lerneaen_Hydra> alex_joni: how do you mean?
[19:16:33] <alex_joni> * alex_joni swallows a remark
[19:16:53] <Lerneaen_Hydra> oh, somewhat, well, sarcasticly?
[19:17:01] <SkunkWorks> * SkunkWorks thinks alex is getting snippy :)
[19:17:08] <alex_joni> you think?
[19:17:35] <Lerneaen_Hydra> I was just commenting that, in case someone else wanted to run the test, and was reading this. I found it better at least than not saying anything
[19:17:54] <Lerneaen_Hydra> how long should it run?
[19:18:10] <SkunkWorks> alex - do you see a problem adding a troubleshooting section?
[19:18:23] <alex_joni> SkunkWorks: not at all
[19:18:28] <alex_joni> I would welcome it..
[19:18:34] <SkunkWorks> after the advanced topics?
[19:19:41] <alex_joni> more like under Using / Troubleshooting
[19:20:08] <SkunkWorks> ah - that makes sense.
[19:20:36] <alex_joni> but the whole wiki is lacking some sort of design :)
[19:20:46] <alex_joni> so it kinda grew without any directions
[19:21:36] <A-L-P-H-A-2> freak'n retarded!
[19:22:15] <A-L-P-H-A-2> wonder will this work?
[19:22:23] <A-L-P-H-A-2> ARGH sooo hate this.
[19:24:22] <SWPadnos> A-L-P-H-A-2, you may get better help from an ubuntu or ati forum :)
[19:25:51] <CIA-8> 03cradek 07HEAD * 10axis/setup.py: this fixes a build problem
[19:28:42] <jepler> cradek: is that the 'import os' thing again?
[19:28:46] <jepler> or something else?
[19:28:58] <cradek> yes
[19:29:01] <jepler> wtf wtf wtf
[19:29:17] <cradek> commit war!
[19:32:59] <alex_joni> * alex_joni raises a white flag and stays neutral
[19:33:23] <alex_joni> I'll open up some banks too, and start to produce chocolate
[19:33:33] <jepler> cradek: how come this import problem doesn't happen on my machine?
[19:34:16] <alex_joni> YAY.. AXIS sees the light of dapper
[19:34:56] <cradek> jepler: did you dpkg-buildpackage it?
[19:35:20] <Lerneaen_Hydra> when doing the RT test, I get an OVL MAX of 11k, and my base period is 20k, so things should be ok? How long do I need to run the test to make sure that nastystuff (R) doesn't happen?
[19:35:58] <cradek> Lerneaen_Hydra: start and exit a bunch of programs, move windows around, etc
[19:36:08] <cradek> Lerneaen_Hydra: if overruns stays 0, you're fine
[19:36:09] <jepler> oh, it does when I -buildpackage it
[19:36:37] <alex_joni> Lerneaen_Hydra: you shoudl try to stress your system while it tests
[19:36:57] <alex_joni> what cradek said
[19:38:25] <Lerneaen_Hydra> hmm, interesting. when starting lots of intensive apps (oo.org, gimp, and so on), and moving lots of windows around, I got a max of 26k
[19:38:30] <Lerneaen_Hydra> which is not that good
[19:38:33] <cradek> if it's a laptop, open and close the lid
[19:38:39] <Lerneaen_Hydra> its not a laptop
[19:38:46] <alex_joni> hmm.. I can't believe how small AXIS si..
[19:38:52] <alex_joni> s/si/is/
[19:38:58] <cradek> 26k is fine, let it run for a while while you do lots of different things
[19:40:42] <jepler> 26k is not fine for a base period of 20k, is it?
[19:41:10] <cradek> I don't know how/if those numbers are related
[19:41:49] <cradek> if the latency swings from 15-25k, it just means everything runs a little late? it doesn't necessarily mean it takes that long to finish
[19:42:05] <cradek> there's a calibration you can do ... somehow ... that makes the numbers smaller
[19:42:13] <cradek> it doesn't matter for our application
[19:42:42] <Lerneaen_Hydra> the max I get is 26k
[19:42:50] <Lerneaen_Hydra> but I don't need to change base_freq?
[19:42:59] <alex_joni> I think deviation matters most
[19:43:15] <cradek> I think it's just the overruns that are important
[19:43:18] <alex_joni> Lerneaen_Hydra: the base_freq in emc2 should work for ~3-400MHz processors
[19:43:21] <alex_joni> or even less
[19:43:50] <Lerneaen_Hydra> oh, 15µs isn't that much though (if the delay between emc and the parport is that little)
[19:43:55] <Lerneaen_Hydra> 0 overruns
[19:56:15] <SkunkWorks> and this is the reason why I don't write manuals http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?TroubleShooting
[19:56:45] <SkunkWorks> edit at will.
[20:01:29] <NickServ> This nickname is owned by someone else
[20:01:29] <NickServ> If this is your nickname, type /msg NickServ IDENTIFY <password>
[20:04:20] <lilo> [Global Notice] Hi all. We just had a fairly route loss to one of our largest sponsors. We're back, and so far everything appears to be stable.
[20:10:44] <lilo> [Global Notice] Whoops, that would be "fairly large route looss".... thanks :)
[20:52:41] <SkunkWorks> alex_joni: Thanks :)
[20:53:17] <alex_joni> SkunkWorks: heh, no problem
[20:53:22] <alex_joni> too tired to continue though :(
[20:53:38] <SkunkWorks> Get some sleep :)
[20:53:54] <alex_joni> yeah.. I should.. and I will :P
[20:54:03] <alex_joni> giacus: any movie for me to comment on?
[20:54:22] <anonimasu> hm, I might get the machine togther someday this week.
[20:56:00] <alex_joni> anonimasu: tried XGl yet?
[20:56:10] <anonimasu> xgl?
[20:56:24] <anonimasu> glxgears and stuff?
[20:56:43] <alex_joni> XGl = Gl enabled X server
[20:56:58] <alex_joni> I tried a kororaa Live CD todaz
[20:56:59] <anonimasu> no, I havent yet
[20:57:01] <alex_joni> today even
[20:57:01] <anonimasu> is it fast?
[20:57:05] <alex_joni> try it :)
[20:57:07] <alex_joni> it's neat
[20:57:32] <anonimasu> going to try it on the laptop
[20:57:43] <alex_joni> http://kororaa.org/static.php?page=static060318-181203
[20:57:53] <CIA-8> 03jepler 07HEAD * 10emc2/src/hal/utils/scope_disp.c:
[20:57:53] <CIA-8> reorganize handle_motion to remove a thinko (caused the readout label to not
[20:57:53] <CIA-8> work at at all) and do less work overall.
[20:57:53] <CIA-8> force the readout_label to refresh immediately, improves responsiveness
[20:57:53] <CIA-8> improve a comment
[20:59:08] <jepler> cradek: does that fix halscope so the readout works?
[21:00:25] <alex_joni> jepler: never heard of something like a "thinko"
[21:01:35] <jepler> alex_joni: it's like a typo
[21:01:53] <alex_joni> yeah, I figured
[21:02:02] <alex_joni> sounds neat though
[21:02:21] <SkunkWorks> * SkunkWorks wonders what it is supposed to be.
[21:02:43] <alex_joni> SkunkWorks: something like a typo but involving thinking not typing
[21:03:04] <alex_joni> so a rather borked idea (but close to the good one) I'd say..
[21:03:29] <jepler> right
[21:03:50] <jepler> in this case, I wrote if (...) { q = r; } when it should have been { r = q; }
[21:03:56] <SkunkWorks> that makes sense
[21:03:59] <jepler> not a slip of the fingers, but a failure to think about exactly what I meant
[21:04:12] <jepler> "I have to, umm, assign, q and r"
[21:04:29] <jepler> <confidently types the wrong thing>
[21:05:53] <cradek> jepler: coool
[21:06:00] <jepler> it works?
[21:07:25] <cradek> yep it looks great
[21:08:16] <cradek> (now I keep wanting to use the wheel to scroll in while centered on the pointer)
[21:08:59] <cradek> this is great - it sure has needed this feature
[21:09:05] <anonimasu> alex_joni: can you use any wm with xgl?
[21:09:06] <alex_joni> cradek: next you'll want a FFT analysis on the signal, displayed in 3D with GL
[21:09:18] <alex_joni> anonimasu: no, I think there is a special one..
[21:09:22] <anonimasu> :S
[21:09:24] <cradek> alex_joni: no, just derivatives
[21:09:27] <anonimasu> too bad
[21:09:46] <alex_joni> cradek: how about f(...) (user implementable functions)
[21:10:05] <cradek> just a couple derivatives would be all I care about
[21:10:07] <alex_joni> with a filter collection, and the user can drag the ones he likes
[21:10:21] <cradek> now you're just being silly
[21:11:17] <alex_joni> hmm.. thought you catched my joke earlier
[21:11:48] <SWPadnos> derivatives and integrals are pretty easy (as is scaling, if necessary)
[21:12:02] <SWPadnos> also trigger on slope instead of value
[21:12:14] <cradek> yeah that would be cool too
[21:12:31] <cradek> dragging the display and zooming with the wheel would be nice
[21:12:37] <cradek> (poor jepler)
[21:12:38] <SWPadnos> indeed it would
[21:12:57] <les_w> hi guys
[21:13:23] <SWPadnos> vertical scroll with the wheel, horizontal with shift+wheel, zoom with ctrl+wheel (optionally in Y with ctrl only, and in X with ctrl+shift)
[21:13:37] <jepler> if this was Tk I'd actually know how to do that
[21:13:46] <SWPadnos> and then drag around (or was that what was just added?)
[21:13:48] <SWPadnos> heh
[21:13:53] <cradek> no, I think dragging should scroll. wheel should zoom. that's it.
[21:14:01] <alex_joni> it's Tk + G
[21:14:04] <jepler> actually I figured the next most important thing was to give up a factor of 10 in speed, and compatibility with anything older than a year or so, by using cairo to draw the traces.
[21:14:06] <SWPadnos> zoom horizontally or vertically?
[21:14:25] <cradek> SWPadnos: I figured just manipulate the two top sliders
[21:14:30] <SWPadnos> (they're different here, whereas in most apps they're not)
[21:14:52] <SWPadnos> ok, so horizontal
[21:15:09] <cradek> yeah
[21:15:18] <cradek> ideally keeping the cursor location stationary on the screen
[21:16:24] <SWPadnos> right
[21:20:09] <A-L-P-H-A-2> SWPadnos: I can't figure it out... this sucks!
[21:20:13] <A-L-P-H-A-2> spent all day trying this out.
[21:20:21] <A-L-P-H-A-2> I ended up going back to windows.
[21:20:24] <SWPadnos> yeah - mine didn't survive the second reboot either
[21:20:39] <SWPadnos> I'll have to look into it a bit more sometime
[21:22:38] <alex_joni> A-L-P-H-A-2: figuring what out?
[21:26:20] <SWPadnos> how to get 3 screens working on two Radeons (under Dapper)
[21:26:28] <SWPadnos> I'm just trying to get one to work reliably
[21:36:58] <alex_joni> http://www.amazon.com/gp/product/B000CMKPDI/ref=dp_return_1/104-7138572-2267965?%5Fencoding=UTF8&n=284507&s=kitchen&v=glance
[21:51:06] <alex_joni> g'night guys..
[21:52:08] <Lerneaen_Hydra> 'night
[21:53:53] <K4ts> hello
[21:54:01] <K4ts> hi alex
[21:57:09] <les_w> hi k4ts
[21:58:53] <Lerneaen_Hydra> Lerneaen_Hydra is now known as Lerneaen_Hydra_a
[21:59:00] <Lerneaen_Hydra_a> Lerneaen_Hydra_a is now known as Hydra_away
[21:59:07] <Hydra_away> g'night all
[22:14:23] <giacus> hi les_w
[22:14:32] <les_w> hi jacky
[22:31:22] <dmessier> hi all..
[22:32:19] <dmessier> anyone familiar with a tune by the boom town rats.... "i dont like monday's"
[22:32:34] <bill203> sure.
[22:32:54] <bill203> my silicon chip is also switched to overload.
[22:33:00] <les_w> I had an ok monday.
[22:33:04] <les_w> went too fast
[22:33:05] <anonimasu> hey
[22:33:07] <anonimasu> les_w: got a sec?
[22:33:10] <dmessier> im gonna shooootttttt the whole lot down....
[22:33:11] <les_w> yeah
[22:33:44] <anonimasu> I'm trying to get motinoeering to work
[22:33:52] <anonimasu> motioneering but I cant remember how to use it :)
[22:33:53] <les_w> yes...
[22:34:12] <anonimasu> and I am wondering if you could help me with come calcs
[22:34:18] <les_w> sure
[22:34:33] <dmessier> sounds like fun
[22:34:40] <anonimasu> I'm just going to grab a copy on this comp first
[22:34:40] <les_w> hahaha
[22:34:47] <les_w> some typo
[22:35:31] <dmessier> mine dont count... i cant type
[22:35:40] <les_w> me either
[22:35:56] <anonimasu> hehe
[22:36:04] <dmessier> how you been les... been a bit??
[22:36:19] <les_w> busy. but good
[22:36:38] <dmessier> payin' busy???
[22:36:50] <les_w> I just hope the economy stays up
[22:36:57] <les_w> yes payin busy
[22:37:04] <les_w> I am booked solid
[22:37:10] <dmessier> good stock i see... ; )
[22:37:20] <anonimasu> done in a sec
[22:37:32] <anonimasu> here's the scenario:
[22:37:43] <anonimasu> direct drive on a 6mm 2mm/turn screw
[22:37:46] <les_w> gas price and other things are starting to take a toll
[22:37:54] <les_w> ok
[22:38:01] <anonimasu> with 10kg on the table
[22:38:14] <les_w> you fill in the numbers on the mechanism part
[22:38:31] <anonimasu> yeah
[22:38:44] <CIA-8> 03jepler 07HEAD * 10emc2/src/hal/utils/scope_disp.c:
[22:38:44] <CIA-8> drag the scope display with the middle button and zoom with the wheel
[22:38:44] <CIA-8> break out part of handle_motion into (new) update_readout
[22:39:29] <dmessier> whats the going jobing shop shop rate there fore lathe?? 3 ax mill etc???? i had a guy from up top trey to tell me he was bakin shite on 240 $ / hr i nearly went off
[22:39:40] <les_w> that is a very fine pitch
[22:39:55] <les_w> I charge $100/hr for everything
[22:40:31] <dmessier> sounds reasonable to me
[22:40:34] <anonimasu> yeah, but I am thinking I might get away with direct driving it
[22:40:53] <les_w> well, I am a graduate engineer...
[22:41:06] <anonimasu> pretty small motors with pretty high rpm
[22:41:18] <dmessier> im a pleab.. tecnician... sorry...
[22:41:19] <les_w> anon direct drive is easy with that fine pitch
[22:41:43] <anonimasu> 0.005549 N-m
[22:41:47] <anonimasu> is what it ends up at
[22:41:48] <les_w> It might be cheap...I can book 70 hrs a week easily
[22:41:56] <dmessier> with 24 yrs experience... cutting and fixin
[22:41:58] <les_w> but cannot keep that up for long
[22:41:58] <anonimasu> the "total other"
[22:42:17] <anonimasu> 2.022E-06 kg-m2
[22:42:29] <anonimasu> the inerta stuff..
[22:42:34] <CIA-8> 03compile-farm 07Ubuntu 5.10 (2.6.12-magma) * 10emc2head/: build FAILED ; see http://linuxcnc.org/compile_farm/emc2head_slot7_log.txt
[22:42:43] <les_w> yeah...load, screw, friction
[22:43:06] <anonimasu> hm, that sounds reasonalbe..
[22:43:11] <anonimasu> I've got plenty of overhead..
[22:43:15] <anonimasu> let me see
[22:43:15] <les_w> i'll do a run if you can give me the numbers
[22:43:25] <anonimasu> sure
[22:43:29] <anonimasu> what numbers do you need?
[22:44:09] <les_w> ok
[22:44:18] <les_w> no reducer...
[22:44:27] <les_w> length of leadscrew
[22:45:11] <anonimasu> 1m
[22:45:18] <les_w> k
[22:45:22] <les_w> ballscrew?
[22:45:26] <dmessier> gonna be LOTS a puleses
[22:45:34] <anonimasu> yeah
[22:45:42] <dmessier> machine??
[22:45:49] <les_w> ok 90 % efficiency
[22:45:54] <dmessier> cpu??? overhead???
[22:46:02] <anonimasu> yeah
[22:46:09] <dmessier> peace
[22:46:09] <anonimasu> steel on steel
[22:46:16] <anonimasu> lubed..
[22:46:26] <dmessier> timed??
[22:46:38] <anonimasu> timed?
[22:46:42] <les_w> ok axis mass?
[22:46:42] <anonimasu> I'm running a usc
[22:46:44] <dmessier> lube
[22:46:52] <anonimasu> mass?
[22:46:58] <les_w> yeah
[22:47:09] <les_w> how many kg is the driven axis
[22:47:14] <anonimasu> hm, say 25kg..
[22:47:17] <anonimasu> I have no exact idea
[22:47:19] <les_w> k
[22:47:26] <dmessier> whole mech unit
[22:47:41] <dmessier> sliding
[22:47:45] <anonimasu> yeah
[22:47:48] <les_w> k linear slides
[22:47:48] <anonimasu> or maybe 40..
[22:47:51] <les_w> ball?
[22:47:51] <anonimasu> no
[22:47:56] <anonimasu> no slides
[22:48:01] <anonimasu> converted manual mill
[22:48:04] <dmessier> ways??
[22:48:05] <les_w> ok
[22:48:07] <anonimasu> yeah
[22:48:35] <dmessier> new or recently checked GIBBS
[22:49:05] <dmessier> maybe new turkite>?
[22:49:08] <anonimasu> dmessier: there's a 0,01 deviation at one end but that might be a dimple in the table..
[22:49:10] <les_w> i'll put in about 100 N gib force
[22:49:27] <anonimasu> very very smooth..
[22:49:36] <dmessier> .01 " or .01 mm?
[22:49:42] <anonimasu> 0.01mm
[22:49:43] <anonimasu> that'
[22:50:01] <les_w> ok
[22:50:13] <les_w> how fast and what accel do you want?
[22:50:20] <dmessier> peace LOVE and joy... clean machine.. ; )
[22:50:49] <anonimasu> dmessier: this is one of the schaublin machines
[22:50:59] <anonimasu> 5m/min rapids
[22:51:05] <les_w> k
[22:51:09] <les_w> accel?
[22:51:15] <CIA-8> 03compile-farm 07BDI-TNG (2.4.18-rtai) * 10emc2head/: build FAILED ; see http://linuxcnc.org/compile_farm/emc2head_slot3_log.txt
[22:51:15] <anonimasu> what's a sensible one?
[22:51:27] <anonimasu> I dont have the largest working envelope either
[22:51:47] <anonimasu> about 0.6m
[22:51:48] <anonimasu> at most..
[22:51:52] <dmessier> ahhh super high precision to start with.... PLS dont sacrifice accuracy to speed ... ; ;(
[22:52:03] <les_w> oh, .25 meters/min ^2
[22:52:23] <anonimasu> dmessier: that's the rapids..
[22:52:46] <anonimasu> dmessier: never :)
[22:53:09] <les_w> let's back up. I can tell you right now that 6mm diam screw 1 meter length is too small diameter
[22:53:32] <dmessier> its jus a SMALL envelope...
[22:53:33] <les_w> don't exceed 50:1 aspect ratio on the screw
[22:54:06] <dmessier> it WILL whip
[22:54:11] <les_w> yes
[22:54:21] <anonimasu> it will be supported after the nut..
[22:54:21] <les_w> you need about 20mm screw
[22:54:30] <les_w> even with support
[22:54:35] <dmessier> its a length to dia i work to...
[22:55:02] <dmessier> you would need travelling blocks...
[22:55:09] <anonimasu> yeah, that's what I had in mind..
[22:55:14] <les_w> right on scissor things
[22:55:29] <les_w> easier to just use a larger screw
[22:55:35] <anonimasu> I have a screw already..
[22:55:42] <anonimasu> I can go down on the speed..
[22:56:00] <dmessier> ik... travelling support bloock add to friction in Les's equation
[22:56:02] <les_w> euler buckling problems though
[22:56:42] <les_w> you could prob make a 6 mm screw work some...
[22:57:00] <dmessier> slow her down BOYS
[22:57:01] <les_w> but it would cost far more than just getting a 20 mm screw
[22:57:22] <les_w> could I see apicture of the machine again?
[22:58:06] <dmessier> im thinkin large table small screw just like i could afford
[22:58:17] <anonimasu> http://www.technolada.ru/images/muller/3488_1.jpg
[22:58:24] <les_w> looking
[22:59:17] <les_w> nice
[22:59:30] <les_w> yes my opinion is that 6 mm is far too small
[22:59:32] <dmessier> VERY nice machine... DONT destroy it pls...
[22:59:41] <anonimasu> dmessier: destroy?
[22:59:48] <les_w> what was on it originally?
[23:00:04] <anonimasu> ni
[23:00:05] <anonimasu> err no
[23:00:29] <dmessier> slow it down on current h/w or beef it up to go as fast as i can... and it CAN go
[23:00:35] <les_w> does it now have an acme screw?
[23:00:43] <les_w> if so, what size?
[23:00:45] <anonimasu> it does have the standard screw
[23:00:57] <anonimasu> I cant remember
[23:01:01] <anonimasu> right now it's a manual..
[23:01:01] <les_w> size and pitch?
[23:01:08] <anonimasu> neither'
[23:01:12] <les_w> just a guess
[23:01:13] <anonimasu> it's a strange pitch..
[23:01:23] <anonimasu> 15 perhaps..
[23:01:31] <anonimasu> maybe 20..
[23:01:56] <dmessier> the Xaxis... 2 stage??? like a grinder??
[23:02:13] <anonimasu> what do you mean?
[23:02:54] <les_w> diameter and mm/revolution
[23:03:03] <les_w> roughly
[23:03:23] <anonimasu> 8mm/rev I think
[23:03:30] <les_w> bet it's about 20 mm and 5mm/rev
[23:03:34] <anonimasu> yeah
[23:03:58] <les_w> well that is the proper ballscrew size too
[23:04:08] <anonimasu> damn.
[23:04:16] <les_w> ssorry
[23:04:21] <anonimasu> ah, dont worry about it..
[23:04:26] <anonimasu> I'll use the screws for something else..
[23:04:31] <les_w> yeah
[23:05:02] <anonimasu> too bad on a pair of nice screws
[23:05:48] <anonimasu> with that kind of mm/Rev
[23:05:53] <anonimasu> i'd have to gear it :/
[23:05:55] <les_w> It's like buying wood...If I see a nice batch for a nice price...I buy it whether I have an immediate use or not
[23:06:41] <les_w> nah 20mmx5 is commonly direct drive with a servo
[23:06:48] <les_w> 2 or 3 to one with a stepper
[23:06:49] <anonimasu> I've got very tiny servos
[23:06:56] <anonimasu> let's see what it ends up at..
[23:07:02] <anonimasu> if I change the 20x5 in motioneering
[23:07:22] <anonimasu> 0.9nm
[23:07:53] <les_w> medium size
[23:08:11] <anonimasu> but that's with 100kg..
[23:08:12] <anonimasu> moving..
[23:08:15] <les_w> what is it in oz in?
[23:08:15] <anonimasu> 120 with load..
[23:08:45] <anonimasu> what is meant by "gib force"
[23:09:09] <les_w> resistance to motion caused by tight gib friction
[23:09:16] <anonimasu> ah, that's not 100kg ;)
[23:09:32] <anonimasu> more like 30 at most
[23:09:37] <les_w> no it's in newtons
[23:09:45] <les_w> kg is not a force unit
[23:09:55] <anonimasu> in motioneering it's in kg's
[23:10:12] <les_w> kgf...heh
[23:10:15] <les_w> bad unit
[23:10:18] <anonimasu> yeah..
[23:10:21] <les_w> I used 100 newtons
[23:10:33] <anonimasu> ah
[23:10:40] <anonimasu> that adds up well within range of my servos
[23:10:52] <les_w> that is about the force of 10 kg on a scale
[23:10:55] <anonimasu> 0.05202NM
[23:11:25] <anonimasu> and I have 0.16NM
[23:12:09] <anonimasu> err did the calc with 2mm again damn
[23:12:19] <les_w> heh
[23:12:21] <anonimasu> 0.13 with 5mm
[23:12:30] <les_w> fine
[23:12:35] <anonimasu> hm, that seems a bit close
[23:13:02] <les_w> well servos have peak torques many times higher than average
[23:13:21] <les_w> not like steppers
[23:13:22] <anonimasu> heh
[23:13:24] <anonimasu> typo
[23:13:31] <anonimasu> 0.374nm
[23:13:36] <anonimasu> 2.118nm peak
[23:13:59] <les_w> right peak is 5 times more
[23:14:02] <anonimasu> let's see what kind of precision that ends up at..
[23:14:03] <les_w> so you are fine
[23:14:32] <anonimasu> 1000ppr encoder
[23:14:44] <anonimasu> that ends up as 2000 in quadrature..
[23:14:55] <anonimasu> so 5/2000 0.0025
[23:15:06] <anonimasu> large :S
[23:15:30] <les_w> make sure the 20 mm diameter is there....about half the total effective mass will be ballscrew moment of inertia
[23:15:44] <les_w> and specify the density of steel...
[23:15:52] <anonimasu> it's there
[23:16:08] <les_w> k
[23:16:42] <anonimasu> is that too large?
[23:16:59] <anonimasu> I might gear it down 2:1
[23:17:02] <les_w> what?
[23:17:08] <anonimasu> the resolution
[23:17:43] <les_w> 1000 ppr usually ends up as 4000 after interpolation
[23:17:55] <anonimasu> I'm running with geckos
[23:18:03] <les_w> ti's ok at the low speeds you will be going
[23:18:07] <anonimasu> and a usc + the feedback board
[23:18:25] <les_w> well don't know the specs...prob ok
[23:19:06] <anonimasu> the motors goes up to 6500rpm..
[23:19:43] <les_w> gear as needed in motioneering to get a 1:1 inertia match
[23:19:46] <anonimasu> and at rapids with a 5mm screw I'll have 1200rpm with 6m/min rapids
[23:20:10] <anonimasu> hm, ok
[23:20:31] <les_w> you have to know the rotor inertia of your motors
[23:20:40] <anonimasu> I do :)
[23:20:44] <les_w> ok
[23:21:05] <anonimasu> Kg*m2 5.86 x 10-5
[23:21:14] <les_w> go for something reasonable...does not have to be exactly 1:1
[23:21:46] <les_w> needs to be better than 5:1 though
[23:21:54] <anonimasu> 0.0083 oz/in-s^2
[23:22:11] <les_w> ok
[23:22:15] <les_w> really low
[23:22:22] <les_w> you will be gearing.
[23:23:05] <les_w> 3 or 4 to one
[23:23:10] <les_w> wsounds like
[23:23:14] <A-L-P-H-A> 4321
[23:23:14] <A-L-P-H-A> :D
[23:23:19] <les_w> haha
[23:24:17] <anonimasu> hm ok
[23:24:18] <les_w> actually, timing belt of course.
[23:25:00] <les_w> reflected inertia is the square of the belt ratio
[23:25:21] <les_w> 2:1, motor inertia quadruples
[23:25:36] <anonimasu> les, how important is the inertia?
[23:25:42] <les_w> 3:1, 9 times
[23:25:45] <anonimasu> at the speeds i'll be going at?
[23:25:55] <les_w> pretty important
[23:25:58] <SWPadnos> anonimasu, you may have an issue with your encoders, unles sthey're very high end
[23:26:14] <anonimasu> SWPadnos: galil motors
[23:26:47] <SWPadnos> the encoders are the issue. unless they're AB or BEI or the like (the ones that are in the $350-500 each range), they probably won't be able to keep up
[23:26:51] <giacus> SWPadnos: wmtools install won't accept /usr/src/linux-2.6.16-magma/include , any idea ?
[23:27:01] <giacus> sounds strange to me
[23:27:02] <SWPadnos> no clue at this point
[23:27:12] <anonimasu> SWPadnos: that's the specs from the manufacturer..
[23:27:13] <A-L-P-H-A> SWPadnos, I hate ubuntu! stupid thing is borked!
[23:27:29] <giacus> that is the source tree for the kernel i'm using
[23:27:33] <anonimasu> SWPadnos: I doubt they sell a motor with encoders that is unable to keep up with their max rating
[23:27:38] <SWPadnos> A-L-P-H-A - ask on the ubuntu forums. this may not be a distro specific problem
[23:27:46] <SWPadnos> anonimasu, I would hope not
[23:27:46] <A-L-P-H-A> have... no go.
[23:27:55] <anonimasu> SWPadnos: so, I dont think that's the problem :)
[23:28:05] <anonimasu> that would be shitty
[23:28:27] <SWPadnos> ok. you're at about 1/3 the spec of the USC at top speed
[23:28:33] <anonimasu> yeah
[23:28:33] <SWPadnos> though it can probably go faster
[23:28:40] <A-L-P-H-A> http://ubuntuforums.org/showthread.php?t=189651
[23:28:46] <anonimasu> the usc will push my motors to the limit at max speed...
[23:29:02] <anonimasu> :)
[23:29:04] <anonimasu> easily
[23:29:10] <SWPadnos> hmmm - actually, you're past the spec limit of the USC
[23:29:20] <anonimasu> I cant remember any specs really
[23:29:28] <anonimasu> I'm not going to go that fast anyway..
[23:29:29] <les_w> you mentioned 5 meters/min rapids....
[23:29:35] <anonimasu> about half..
[23:29:36] <anonimasu> yeah
[23:29:36] <les_w> that is pretty slow
[23:29:37] <SWPadnos> 6500 RPM is ~110 RPS, * 4000 counts/rev = 440k steps/sec
[23:29:43] <SWPadnos> the USC is rated for 300k
[23:30:01] <SWPadnos> though the FPGA (depending on the revision) can do either 2.5 or 5 MHz
[23:30:05] <les_w> oh really
[23:30:17] <les_w> oh, that's slow.
[23:30:19] <anonimasu> SWPadnos: I've been thinking about half..
[23:30:29] <SWPadnos> you get worse granularity, and it depends on the pulse rate, etc.
[23:30:44] <SWPadnos> galil drives as well?
[23:30:49] <anonimasu> no, geckos
[23:30:56] <SWPadnos> then those will be the limitation
[23:30:58] <les_w> anon think about half more!
[23:31:04] <SWPadnos> they have a 250k step/sec max freq
[23:31:18] <anonimasu> works fine for m..
[23:31:19] <anonimasu> me
[23:31:23] <les_w> or
[23:31:31] <les_w> just don't do quadrature
[23:31:41] <SWPadnos> the geckos always do quadrature, AFAIK
[23:31:41] <anonimasu> eh?
[23:31:44] <anonimasu> yeah
[23:31:51] <anonimasu> that's true
[23:31:53] <les_w> ha
[23:31:58] <SWPadnos> ok. dinnertime. talk amongst yourselves :)
[23:32:05] <anonimasu> les_w: what does the inertia do+
[23:32:06] <les_w> haha
[23:32:13] <anonimasu> err if there's a mismatch?
[23:32:43] <les_w> well if geckos always do quad and you want to use what you have
[23:33:02] <les_w> you can always divide with a simple circuit
[23:33:10] <anonimasu> I have a multiplier..
[23:33:15] <anonimasu> on my geckos..
[23:33:22] <anonimasu> and I want that kind of resolution
[23:34:21] <anonimasu> the 5m/min rapids is something I just pulled from somewhere..
[23:34:29] <les_w> I am not that familiar with the geckos and emc stepper....as an industrial user we don't use stepper sytems on machine tools
[23:35:02] <les_w> only full servo
[23:35:03] <anonimasu> if this were industrially I could order a siemens for the control with matching drives..
[23:35:10] <les_w> heh
[23:35:12] <les_w> yeah
[23:35:13] <anonimasu> seriously
[23:35:36] <anonimasu> :)
[23:35:57] <les_w> For hobby use steppers are great
[23:36:10] <anonimasu> this isnt steppers it's servos with geckodrives..
[23:36:19] <les_w> I understand
[23:36:25] <anonimasu> I'm not putting steppers on a machine again..
[23:36:27] <les_w> but emc is outputing steps
[23:36:38] <anonimasu> the usc is..
[23:36:49] <les_w> k
[23:37:03] <anonimasu> dosent add nearly as much for emc to do
[23:37:23] <les_w> you need a motenc lite
[23:37:25] <les_w> haha
[23:37:38] <anonimasu> _if_ I want I can have the feedback to emc.. too..
[23:37:53] <les_w> perhaps I'll get to rich and just send you one someday
[23:37:59] <anonimasu> lol
[23:38:11] <anonimasu> the issue with such a card if just that I'd have to get new drives..
[23:38:49] <anonimasu> that's where the money goes off.
[23:38:50] <les_w> or break out the soldering iron
[23:38:59] <anonimasu> hm, modify the geckos?
[23:39:09] <anonimasu> or build own ones
[23:39:20] <anonimasu> somehow I dont think that's a good idea :)
[23:39:29] <anonimasu> I'll end up seeing magic smoke ;)
[23:39:29] <les_w> no, convert motenc analog out to whatever the drives need
[23:39:46] <anonimasu> ah well, that's really a non issue yet..
[23:40:13] <les_w> hey is your dcc functional?
[23:40:16] <anonimasu> yes
[23:40:19] <les_w> file transfer?
[23:40:24] <anonimasu> sure
[23:40:30] <les_w> let me send you a picture ok?
[23:40:35] <anonimasu> ok
[23:40:55] <anonimasu> I could buy a motenc lite + drives, but right now I want the machine togther :)
[23:42:58] <anonimasu> *yawns*
[23:43:07] <anonimasu> les_w: so how much does the inertia do?
[23:43:17] <anonimasu> I have no clue. :)
[23:45:09] <les_w> trying to send you a file
[23:45:34] <les_w> you must autorize
[23:45:48] <les_w> authorize
[23:46:01] <anonimasu> working on it
[23:48:13] <anonimasu> try it again
[23:48:28] <les_w> k
[23:48:54] <anonimasu> this is weird..
[23:49:27] <anonimasu> if you can mail it instead
[23:49:32] <les_w> oh well...just a picture of my new impedance analyzer
[23:49:39] <anonimasu> still I want to see it :)
[23:49:42] <les_w> ok email
[23:49:50] <les_w> post it here or pm
[23:49:54] <anonimasu> i'll pm you it
[23:49:58] <les_w> k
[23:50:47] <anonimasu> i'ölli'll be back in a bit going to move home
[23:50:55] <anonimasu> *types like a drunk*
[23:51:58] <les_w> sent
[23:52:09] <les_w> to answer your question...
[23:52:28] <les_w> in high speed machines like routers....
[23:52:53] <les_w> inertia specifies the motor as about an 80% factor
[23:53:19] <les_w> in low speed high friction systems like conventional ways...25 to 50%