| 00:17.41 | *** join/#brlcad poolio_ (n=poolio@c-69-251-3-107.hsd1.md.comcast.net) | |
| 00:24.01 | *** join/#brlcad Twingy (n=justin@74.92.144.217) | |
| 00:26.39 | poolio | oops. |
| 00:28.22 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/src/gtools/beset/.cvsignore: ignore the beset product |
| 00:28.32 | poolio | brlcad: good choice :) |
| 00:39.02 | poolio | brlcad: it seems like part of the problem / the problem has to do with rti_radius...If I create a sphere with radius of 4, the min/max is correctly computed, but the bounding sphere has a radius of 6.9282...? |
| 00:39.25 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/src/gtools/beset/ (beset.c beset.h fitness.h population.c population.h): petty cosmetics while browsing, mostly ws |
| 00:39.39 | poolio | err i've been updating stuff :\ |
| 00:40.11 | brlcad | no worries, nothing likely to conflict |
| 00:40.25 | brlcad | still reading what all you have going |
| 00:41.04 | poolio | alright, but that rti_radius vs bounding box thing is confusing me |
| 00:41.15 | poolio | and possibly part of the problem, or maybe it's my confusion with how the raytracer works |
| 00:41.39 | poolio | thanks for fixing my headers :) |
| 00:41.39 | brlcad | you'll have to back me up to some example so I can see |
| 00:41.46 | poolio | Yeah sure |
| 00:42.06 | poolio | It involves some effor on your end though, I could throw some code together to do it automatically if you want though |
| 00:42.36 | brlcad | nah, just say what I need to do |
| 00:42.41 | brlcad | make a sph in mged? |
| 00:42.45 | poolio | yes |
| 00:42.49 | brlcad | radius 4? |
| 00:42.51 | poolio | sure |
| 00:42.53 | poolio | 4.0 |
| 00:43.06 | poolio | the output of the program is god awful |
| 00:43.18 | poolio | you could also modify the program and do it that way, but i think the mged way is the easiest |
| 00:43.35 | poolio | so make a sphere, run the program with ./beset database.g 1 1 spherename.s aadf |
| 00:43.47 | poolio | the aadf is just me still not decrementing the argc check |
| 00:44.05 | brlcad | already done |
| 00:44.18 | brlcad | not sure what those numbers mean yet |
| 00:44.31 | poolio | so they are the in/out points of the shotline ray |
| 00:44.36 | poolio | 1 1 = 1x1 grid of rays |
| 00:44.38 | poolio | so just 1 ray |
| 00:44.47 | brlcad | shooting in what direction? |
| 00:44.48 | poolio | that should be centered ... i think |
| 00:44.52 | brlcad | ah, k |
| 00:45.02 | poolio | hopefully shooting along the z axis |
| 00:45.17 | poolio | but i think i've confused my axis somewhere, but it's shooting along some defined axis |
| 00:45.35 | poolio | and the numbers are wrong...that's another issue.. |
| 00:45.54 | brlcad | what are the bracketed values? |
| 00:45.58 | poolio | ok so |
| 00:46.08 | poolio | they're the in/out points of the ray |
| 00:46.13 | poolio | the left one is the model shape |
| 00:46.21 | poolio | model shape = stored shape in database |
| 00:46.35 | poolio | the right one is the GA one which should be set to some value in population.c |
| 00:46.37 | poolio | some whole numbered value |
| 00:46.46 | poolio | I've modified everything to try to isolate what is going on |
| 00:47.03 | poolio | so basically, with a radius of 4, the in/out hit points are those |
| 00:47.11 | poolio | (those are normalized to the rti_radius*2 (diameter)) |
| 00:47.27 | brlcad | ah, normalized |
| 00:47.28 | poolio | the reason it's not 1 and is instead .57735 has to do with rti_radius being completely off for some reason |
| 00:48.21 | brlcad | where do you normalize? |
| 00:48.30 | poolio | I normalize in fitness.c |
| 00:49.04 | poolio | stored rays are normalized as they are stored in capture_hit |
| 00:49.18 | poolio | candidates rays are normalized as they are compared in compare_hit |
| 00:49.43 | poolio | the printing is coming from compare_hit |
| 00:49.55 | poolio | If you wait one minute I think I'll make it a bit more clear and stop wasting your time :) |
| 00:59.07 | poolio | brlcad: alright cvs update and you should get output that means something |
| 01:00.10 | *** join/#brlcad SWPadnos_ (n=Me@dsl245.esjtvtli.sover.net) | |
| 01:00.23 | poolio | brlcad: it's beset.c and fitness.c |
| 01:00.51 | CIA-4 | BRL-CAD: 03poolio * 10brlcad/src/gtools/beset/ (beset.c fitness.c): added logical debugging output |
| 01:01.07 | poolio | brlcad: same instructions but don't need the junk 5th argument |
| 01:01.28 | poolio | ./beset db.g 1 1 sphere.s |
| 01:01.42 | poolio | try it with something like 4 then 4.01 |
| 01:02.53 | poolio | my output: http://rafb.net/p/JIkxZt54.html |
| 01:03.42 | poolio | whole.s is a sphere of radius 4, dec.s is a sphere of radius 4.01 |
| 01:07.26 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/src/gtools/beset/beset.c: error check fit_prep() return value |
| 01:08.23 | poolio | brlcad: I add in error checking after I get it working in known conditions, I wouldn't worry about it for now :P |
| 01:11.27 | brlcad | i figured half as much, though there's a reason |
| 01:11.45 | poolio | Yeah, it could be stuff like that that's failing and I don't check and that's why it's going crazy |
| 01:12.49 | poolio | I should probably do that now, I'm just not sure if you're doing it and didn't want to commit stuff that you might be changing |
| 01:13.00 | brlcad | it's a habit to *always* write your checks at the same time, sanity checking as you go along -- you need the code regardless so you might as well write it while it's in your locus of attention |
| 01:13.32 | brlcad | don't worry about committing over me.. that's what cvs is for ;) |
| 01:13.33 | poolio | brlcad: Yeah I know, my issue is that I hate having to scroll past it all the time |
| 01:13.49 | brlcad | you get used to it, becomes second nature |
| 01:14.39 | poolio | so any ideas off the top of your head with the weird raytrace shifting thing? |
| 01:14.46 | brlcad | still looking |
| 01:15.05 | brlcad | which is why I was adding checks while poking .. it's in my locus of attention ;) |
| 01:15.16 | brlcad | always better earlier than later.. |
| 01:15.18 | poolio | I appreciate it :) |
| 01:15.30 | brlcad | lots can happen between early and later |
| 01:15.52 | brlcad | the sort of bugs and unchecked values that can send you debugging for days |
| 01:16.10 | poolio | I've already been debugging for days :\ |
| 01:16.28 | brlcad | that's probably more a combination of reasons :) |
| 01:16.41 | poolio | stupidity weighing heavily |
| 01:18.23 | poolio | I also had a kind of general question about C coding. Is it considered a bad practice to have code that's not part of the main routine output info? Like I'm specifying certain return values and then checking the return values in main() and printing out error info. Before I had it just print the error and exit from the routine |
| 01:21.41 | brlcad | nah |
| 01:22.06 | brlcad | (re stupidity) |
| 01:22.06 | poolio | k |
| 01:22.06 | poolio | oh |
| 01:22.06 | poolio | :P |
| 01:22.18 | brlcad | a like your layout, not too tricky to follow |
| 01:22.26 | poolio | brlcad: hurray :) |
| 01:22.37 | poolio | by layout you just mean the file structure and routine hierarchy? |
| 01:23.16 | brlcad | and to your second question .. heck no, that's fine either way |
| 01:23.27 | poolio | alright cool. |
| 01:23.37 | brlcad | generally, error/failure recovery is just a big design decision.. just should try to be consistent on the approach |
| 01:24.20 | brlcad | whether you use return codes, or abort in place, or try to recover from failures, or use result structures, or throw exceptions, or set jump points, etc |
| 01:24.44 | poolio | I think I might just let the routines exit. I mean all the errors that can be errors are fatal errors |
| 01:24.46 | brlcad | for something like this -- if it's a fatal error, I'd just abort |
| 01:24.57 | brlcad | s/abort/terminate the application/ |
| 01:24.58 | poolio | Well not all the errors, but all the errors I'm printing error messages for |
| 01:25.05 | poolio | brlcad: ok cool |
| 01:25.51 | brlcad | it's generally only worth the overhead/complexity of return codes/structs/exceptions/etc if you're actually going to handle them under some conditions |
| 01:26.12 | poolio | yeah that's what I'm considering... it just adds more conditional checks into main() that do nothing |
| 01:26.18 | brlcad | or if there's some secondary benefit, like being able to print more informative messgaes by returning values higher up the chain |
| 01:26.41 | poolio | Well that's what I was thinking with error messages, I wanted to print the program name too but I don't think that's of much importance |
| 01:26.56 | brlcad | usually it's a balance, particularly if you have things that can return null, you check your nulls regardless |
| 01:27.04 | brlcad | s/your/for/ |
| 01:27.45 | poolio | well I feel like bu_malloc has cleaned up a lot of code pertaining to null pointers |
| 01:28.57 | brlcad | that was done that way primarily because it's such a common pattern, and it was an architecture decision to never have memory failures reach application code |
| 01:29.11 | brlcad | i wouldn't take that to an extreme for all call types ;) |
| 01:29.16 | brlcad | like I said, it's a balance |
| 01:29.26 | poolio | yeah, I like that decision. |
| 01:30.28 | brlcad | if you have a routine that has several types of possible errors, it might actually make sense to return an error code or null pointer or what have you, and have the one or two callers just check that value than have N print-error/release-memory/shut-down statements in that deep function |
| 01:32.13 | brlcad | the more frequent decision is usually whether to use return codes (0 good, !0 bad) or truthfull results (true succeed, false failure) |
| 01:32.55 | brlcad | or if you're c++, whether to use exceptions at all or not, other examples abound |
| 01:34.53 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/src/gtools/beset/fitness.c: quell warning |
| 01:41.52 | poolio | brlcad: eek, I changed that.... |
| 01:42.02 | poolio | brlcad: I just made it print the error and exit...oh well |
| 01:42.19 | brlcad | *ahem* commit early, commit often |
| 01:42.42 | poolio | I think I might actually write that on top of my monitor |
| 01:42.51 | poolio | DON'T FORGET KIDS: commit early, commit often |
| 01:42.57 | poolio | (that better not be relationship advice) |
| 01:43.12 | brlcad | you'll hear that in jusst about every open source project if you've not already |
| 01:43.44 | brlcad | only way to effectively coordinate distributed devs without halting progress -- first to commit "wins", those that follow get to potentially resolve conflicts ;) |
| 01:43.53 | poolio | grargh. |
| 01:44.28 | brlcad | so I have the answer to one question for you |
| 01:45.11 | brlcad | the radius it assigns is only guaranteed to be larger, not tight fitting |
| 01:45.38 | poolio | alright. so I think I'll stick to the bounding box |
| 01:45.55 | brlcad | it computes the bounding sphere by taking half the diameter of the bounding box |
| 01:45.56 | poolio | so is it a sphere that fits inside it the bounding box? |
| 01:46.03 | poolio | ah ok |
| 01:46.04 | brlcad | and if you remember your trig, then those numbers make more sense ;) |
| 01:46.38 | poolio | Yes, but the issue with numbers had to do when it wasn't a whole number |
| 01:46.43 | brlcad | s/diameter/length from the longest corner-to-corner span/ |
| 01:47.07 | brlcad | the radius is still fine to use, shouldn't matter |
| 01:47.48 | brlcad | only happens to be obvious with a sphere since the bounding box is significantly bigger than the sphere, then that box's bounding sphere is then larger |
| 01:47.58 | poolio | yeah true |
| 01:48.05 | poolio | so I can still leave that I guess |
| 01:48.50 | brlcad | btw, a good testing value (aside from a debugger) for shooting rays is nirt and/or rtshot |
| 01:51.33 | brlcad | e.g. rtshot -d 0 0 -1 -p 0 0 1000 test.g sph (shoots down the Z axis) |
| 01:55.45 | poolio | brlcad: alright, so there is something messed up with my raytracing of rational numbers |
| 01:56.30 | poolio | thanks for the time and effort |
| 01:59.49 | brlcad | which axes are your U,V axis along? |
| 02:00.02 | brlcad | xy? |
| 02:00.04 | poolio | I want to say X and Y, but I also want to say I didn't check that and they might not be |
| 02:20.40 | CIA-4 | BRL-CAD: 03poolio * 10brlcad/src/gtools/beset/ (fitness.c fitness.h): quell more warnings, exit instead of returning in fit_prep() |
| 02:23.45 | poolio | brlcad: any luck? |
| 02:32.23 | brlcad | well, the first shotline looked fine, adding some debug |
| 02:32.46 | poolio | which first shotline? |
| 02:34.00 | brlcad | you shoot up the Z axis at the R4 sphere, get a 8mm thick segment |
| 02:34.58 | poolio | Yep. you shoot up the Z axis at the R4.01 sphere and get an 8.02mm thick segment, which is correct, but for some reason it's shifted over 1 mm? |
| 02:51.43 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/src/gtools/beset/beset.c: add support for librt -x debugging (e.g. -x 1 for rt_shootray() shotline debugging, see raytrace.h for debug flags) |
| 02:54.00 | brlcad | you're right, it's shifted |
| 02:57.21 | poolio | brlcad: aha! the issue is that the bounding box is whole numbers it looks like |
| 02:59.05 | brlcad | yes, it kicks the box up to the next whole number |
| 02:59.11 | poolio | oh man... |
| 02:59.16 | poolio | anyway to get decimal bounding boxes? |
| 02:59.17 | brlcad | your second shotline against 4.01 looks right... |
| 02:59.28 | poolio | It is, the issue was I thought the bounding box would fit the shape |
| 02:59.46 | poolio | Somehow my brain missed the change in bounding box size |
| 03:00.08 | brlcad | 9.01 .99 = 8.02 (shotline thickness) |
| 03:00.26 | poolio | brlcad: yes I see the rt output is right, the issue is the bounding box, how can I get a more accurate one? |
| 03:01.13 | brlcad | hm, well that gets a bit more complicated |
| 03:01.29 | brlcad | you could compute them yourself.. but you shouldn't need a tight fitting box |
| 03:02.10 | poolio | Well it's neccesary to have a tight fitting box if I want it to be able to disregard scale along the z-axis (ray) |
| 03:02.33 | brlcad | firstoff, remember that these are not unitless values.. the box is within 1mm |
| 03:04.21 | poolio | I mean the option would be to make the units/size bigger, but it wouldn't be too difficult to calculate it on my own |
| 03:05.02 | SuperTaz | howdy |
| 03:05.05 | SuperTaz | I have returned |
| 03:05.08 | poolio | I also just need the min/max in the z directon |
| 03:05.20 | SuperTaz | the build process still hates me, though |
| 03:06.18 | poolio | brlcad: so shoot off the rays on each model, keeping track of the min/max (and all the individual rays and their parititons) then shift the rays so that the ray closest tot he point is 0 distance away and it should be good |
| 03:06.29 | poolio | normalize with max ray distance - min ray distance |
| 03:06.34 | poolio | that way you wouldn't ahve to re-raytrace |
| 03:07.32 | poolio | but it would require storing all the rays, and separating the ray-by-ray comparison |
| 03:08.12 | poolio | Thats my besat idea... |
| 03:08.15 | brlcad | that's pretty good actually |
| 03:08.31 | brlcad | should be faster to process too |
| 03:08.41 | poolio | faster than the current implmenetation? |
| 03:08.52 | brlcad | that said, you can get at the bounding box before it's clamped |
| 03:09.38 | poolio | what do you mean (i don't quite understand :P) ? |
| 03:10.13 | brlcad | sorry, doing too many things at once :) |
| 03:10.32 | brlcad | what I mean is that the clamping is done for ray-trace partitioning/optimization reasons |
| 03:10.35 | poolio | you mean in librt and prepping? I can get a closer bounding box? |
| 03:10.45 | poolio | brlcad: yeah understood, but wouldn't I have to hack up librt? |
| 03:10.46 | brlcad | before calling prep, it should be the tight-fit box |
| 03:11.20 | poolio | wait, so when I extract the rt_i from the database it's the actual bounding box? |
| 03:11.53 | SuperTaz | brlcad: here's another pastebin link, this time with a couple of extra lines above to give you better context: |
| 03:11.57 | SuperTaz | http://pastebin.ca/594796 |
| 03:12.29 | brlcad | poolio: the bounding sizes are computed during the rt_gettrees() |
| 03:12.48 | poolio | brlcad: cheers :) |
| 03:12.51 | brlcad | once you prepare for a ray-trace, the boxes are clamped so that the spatial partitioning is well-behaved |
| 03:13.03 | brlcad | (i.e. during prep) |
| 03:13.03 | poolio | brlcad: so just overwrite the clamped values after rt_prep() ? |
| 03:13.16 | brlcad | eep, no, don't do that |
| 03:13.26 | brlcad | but you can grab/stash them if you like |
| 03:13.33 | poolio | wait, why not? |
| 03:13.42 | brlcad | it clamps them for a reason :) |
| 03:13.54 | poolio | What's the reason? |
| 03:14.00 | poolio | is it a lot faster when using whole numbers? |
| 03:14.09 | poolio | (for the bounding box) |
| 03:14.39 | brlcad | that, and there are all sorts of tolerancing and scene iteration that occurs as the ray is marched through a scene |
| 03:15.01 | poolio | brlcad: so stash the value and use it to shift and normalize the rays? |
| 03:15.15 | SuperTaz | brlcad: it looks like a problem with whitespace in CFLAGS |
| 03:15.28 | brlcad | you don't have to perform the same equations when progressing a ray and can get away with just checking signs for example |
| 03:16.09 | poolio | brlcad: wait what? are you referencing the advantages of clamping the value? |
| 03:16.10 | brlcad | poolio: sure, you can do that .. though I like your other idea too of just keeping track of all of the rays and computing as a post-process -- either way |
| 03:16.46 | poolio | brlcad: Do you see an advantage to post-processing? To me that's just more data storage which seems pretty pointless. You can store the model easily enough and obtain that raytraced data if needed. That's the idea of the whole program ;) |
| 03:16.49 | brlcad | SuperTaz: I've seen that before, but it's been a long time |
| 03:18.01 | SuperTaz | brlcad: I'm trying to track it down, but so far I've had no luck in figuring out where it's introduced |
| 03:18.48 | brlcad | the wierdness is that it hasn't "run before" yet, I think it's a matter of the CFLAGS from the top-level configure not matching what configure is passing to the sub-configure |
| 03:19.08 | SuperTaz | brlcad: that's exactly what's happening |
| 03:19.10 | brlcad | poolio: not a really strong advantage |
| 03:19.19 | poolio | brlcad: I like the rt_shootray debbugging, thanks for that :) |
| 03:19.24 | brlcad | some advantages and disadvantages |
| 03:19.51 | poolio | mind iterating some of the advantages? |
| 03:20.18 | SuperTaz | brlcad: the question is whether there's an easy hack |
| 03:20.20 | SuperTaz | ls |
| 03:20.23 | SuperTaz | oops |
| 03:21.00 | brlcad | good question |
| 03:22.39 | brlcad | poolio: cache coherency -- mild performance difference (which is an entirely pointless statement without profiling of course) |
| 03:22.44 | SuperTaz | I think I found something, though...the generated configure script seems to set up some whitespace |
| 03:22.56 | poolio | brlcad: yeah, over my head. I have no clue what you mean by cache coherency :P |
| 03:23.05 | brlcad | pointless because it could be entirely countered by avoiding memory allocations and other management logic |
| 03:23.07 | poolio | brlcad: optimization comes after functionality |
| 03:23.21 | brlcad | yup |
| 03:23.29 | brlcad | and speculative optimization is bad |
| 03:23.33 | brlcad | shame on me for saying it ;) |
| 03:23.35 | poolio | brlcad: alright, I'll try to fix up this damn bug finally |
| 03:23.54 | SuperTaz | brlcad: I'm trying to fake it out...I just modified the cache file and re-ran configure |
| 03:24.02 | SuperTaz | don't think it'll work, but it's worth a shot |
| 03:25.14 | SuperTaz | poolio: it's the law of diminishing returns at work |
| 03:25.51 | brlcad | poolio: the only other reason would be if you ever changed your ray shooter to not shoot rays orthogonally down an axis |
| 03:26.15 | SuperTaz | poolio: in the performance realm, we have to analyze whether the potential gains of [further] optimization of a particular piece of functionality outweighs the investment |
| 03:26.25 | brlcad | that will undoubtedly cause aliasing issues down the road without jittering the rays, though whether that matters remains to be seen |
| 03:26.34 | poolio | brlcad: which I don't really see a reason to, but it'd be interesting to try to run the GA at different axis and see how it does |
| 03:26.54 | brlcad | poolio: even better, not along any axis at all |
| 03:26.56 | poolio | SuperTaz: yes yes. |
| 03:27.09 | poolio | brlcad: I fail to comprehend, but okay :) |
| 03:27.14 | brlcad | sample from all around the bounding sphere, for example, until you converge on a mass |
| 03:27.35 | poolio | oh, by axis i just meant arbitrary line through the object, not x,y,z |
| 03:28.13 | poolio | brlcad: wait, why wouldn't shooting it from just one angle converge on a mass? |
| 03:29.31 | SuperTaz | brlcad: wouldn't random sampling work out to be more costly in the long run? |
| 03:30.58 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/src/other/tk/Makefile.am: tk doesn't have a unix/dltest dir to prodclean |
| 03:31.14 | brlcad | SuperTaz: depends on the goal/purpose |
| 03:31.28 | SuperTaz | or are you talking about aiming inward from the boundary horizon, instead of outward? |
| 03:32.19 | brlcad | shooting rays at the object to effectively point-sample it through-and-through |
| 03:32.42 | SuperTaz | assuming that you're using the center of the bounding sphere as either the target or origin of the ray, firing inwards would be more efficient than firing outwards |
| 03:32.59 | poolio | errrrr |
| 03:33.10 | brlcad | you generally don't want to shoot rays from inside geometry |
| 03:33.20 | SuperTaz | yes |
| 03:33.43 | SuperTaz | but I mean as the anchorpoint of the trajectory if you're not firing along an axis |
| 03:34.16 | SuperTaz | i.e. using it as the fulcrum on which the arbitrary axis rests |
| 03:34.20 | brlcad | also wouldn't want to always shoot at the center |
| 03:34.43 | SuperTaz | yes, but I'm talking about shooting inwards from the horizon, instead of outwards from the object |
| 03:34.50 | brlcad | that will bias the ray density for various shapes, sampling more points at portions near the center |
| 03:34.56 | SuperTaz | if you're randomly sampling |
| 03:35.11 | brlcad | i don't think anyone ever suggested shooting "outwards" |
| 03:35.12 | SuperTaz | hrmmm...I suppose that's true |
| 03:35.14 | brlcad | :) |
| 03:35.32 | SuperTaz | no, but I misunderstood the topic, seeing as how I walked in late, I believe ;) |
| 03:36.46 | SuperTaz | and was largely thinking aloud and re-adjusting my assumptions, when I realized that random sampling is only inefficient when it's outward (the further away you get from the centroid, the less coverage area you have) |
| 03:38.01 | SuperTaz | i.e. if you have a 1 degree spread between 2 rays, it's pretty dense right near the object, and exponentially approaches completely useless the further away you get from the object (the distance between the two objects increases) |
| 03:38.12 | SuperTaz | err...rays, not objects at the end of that |
| 03:38.34 | SuperTaz | perhaps instead of using a centroid, there's some sort of path one could create, though |
| 03:38.39 | SuperTaz | BTW |
| 03:38.56 | SuperTaz | configure is re-starting itself with new CFLAGS, which is screwing it up |
| 03:39.28 | brlcad | which autoconf are you using? |
| 03:39.36 | SuperTaz | sec |
| 03:39.39 | brlcad | default mac version, or fink or ports or something? |
| 03:39.59 | SuperTaz | autoconf (GNU Autoconf) 2.59 |
| 03:40.08 | SuperTaz | ports, I believe |
| 03:40.25 | SuperTaz | might be stock, though |
| 03:40.39 | SuperTaz | looks like it's stock |
| 03:40.47 | SuperTaz | it's in /usr/bin |
| 03:40.57 | SuperTaz | ports is /opt/bin or /usr/local/bin, iirc |
| 03:42.18 | SuperTaz | I'm considering trying to specify the cflags, and seeing what happens |
| 03:42.58 | SuperTaz | re-extracting the src over what's existing, though, since I accidentally typed autoconf -v before autoconf --version |
| 03:43.08 | SuperTaz | and autoconf -v hoses stuff |
| 03:44.44 | SuperTaz | there is a spot where it does a check and sets CFLAGS=" " |
| 03:44.53 | SuperTaz | I suspect there's more than one place that it does that |
| 03:44.55 | SuperTaz | probably 2 |
| 03:45.01 | SuperTaz | or it hits that spot twice |
| 03:45.13 | SuperTaz | and the whitespace stays |
| 03:45.36 | SuperTaz | specifying the cflags on the command line will PROBABLY remove the problem |
| 03:45.41 | SuperTaz | we shall see |
| 03:46.15 | SuperTaz | of course, it would be best to remedy the bug |
| 03:46.22 | SuperTaz | thanks, yukon |
| 03:46.24 | brlcad | could just be a matter of trimming the whitespace in configure.ac before it gets to the sub-configures |
| 03:46.39 | SuperTaz | I've been doing this stuff for many years...I've learned to be patient |
| 03:47.01 | SuperTaz | port one application or write one network stack and you will learn enough patience to last a lifetime |
| 03:47.43 | SuperTaz | brlcad: that might work |
| 03:48.43 | SuperTaz | I'm happy to apply a patch if you want to experiment, even if I do get it to build successfully with my attempted trickery ;) |
| 03:49.08 | SuperTaz | I've never really messed with autoconf much |
| 03:49.20 | SuperTaz | so it's admittedly a little bit of a weakness |
| 03:49.32 | SuperTaz | (and a lot of smoke and mirrors, imho :P ) |
| 03:49.59 | brlcad | yeah, lemme test here |
| 03:50.03 | SuperTaz | k |
| 03:50.09 | yukonbob | SuperTaz: you running 7.10.0 |
| 03:50.10 | yukonbob | ? |
| 03:50.19 | yukonbob | (or rather, trying to run) |
| 03:50.34 | SuperTaz | yes |
| 03:50.42 | SuperTaz | on OS X 10.4.10 |
| 03:52.40 | SuperTaz | have something I am trying to design, and learned modelling in the mid-90's with Alias and Explore, so I'm used to both NURBS and solids modellers |
| 03:53.20 | SuperTaz | but I also had dynamation and kinemation expertise, and this project actually needs materials simulation |
| 03:53.42 | brlcad | yukonbob: most people are oblivious to deps, even 'most' devs |
| 03:54.01 | SuperTaz | which means I either have to find a way to get the now-extinct Explore and port it to OS X or Windows, or find something else |
| 03:54.02 | brlcad | they care about features, functionality, and a clean compile usually from my experience |
| 03:54.06 | SuperTaz | enter BRL-CAD ;) |
| 03:54.34 | brlcad | SuperTaz: we're open to new devs if you have an itch that you want to scratch ;) |
| 03:54.38 | SuperTaz | and, frankly, I don't use tcl/tk for much on this machine |
| 03:54.42 | SuperTaz | hehehe brl |
| 03:55.54 | yukonbob | ?is there currently any nurb support, or is that what poolio is working on atm |
| 03:56.01 | SuperTaz | heh |
| 03:56.51 | SuperTaz | you know...I like both types of modelling |
| 03:57.01 | SuperTaz | which probably makes me a rare breed |
| 03:57.20 | poolio | yukonbob: nope. i'm not really working on anything. |
| 03:57.22 | brlcad | the big tasks going on right (i.e. the major focus of this year) now are a new BREP/NURBS implementation, a STEP geometry converter, and a new modeling interface |
| 03:57.39 | poolio | yukonbob: a bizarre GA project :) |
| 03:57.47 | SuperTaz | nice |
| 03:57.50 | yukonbob | "GA"? |
| 03:57.56 | poolio | genetic algorithm |
| 03:58.01 | poolio | brlcad: who's goign to do the STEP converter? |
| 03:58.02 | yukonbob | ah, right. |
| 03:58.03 | SuperTaz | actually, the materials stuff does look interesting to me :) |
| 03:58.14 | brlcad | yukonbob: there is support for nurbs in brl-cad .. albeit slow and problematic, you can generally only create them programmaticly or via iges import |
| 03:58.28 | SuperTaz | and the physics engine support would be nice to have, too |
| 03:59.01 | brlcad | more importantly, though, is a completely new (re)implementation and integration of a brep/nurbs primitive |
| 03:59.08 | SuperTaz | of course, with nearly a million lines of code already in the project, getting acquainted with the guts of something I haven't even seen, yet should be a bit of a challenge ;) |
| 03:59.48 | SuperTaz | brlcad: essentially, you're going to add NURBS to the GUI modeller? |
| 03:59.49 | brlcad | poolio: I've worked on it some, but I'm trying to find someone else who can work on the step converter |
| 04:00.34 | SuperTaz | so you can do what they did with Explore in the mid 90's, when they added NURBS and blob modelling to the solids modelling |
| 04:00.39 | brlcad | SuperTaz: also fortunately, the package is fairly compartmentalized into various tools and libraries.. so you don't necessarily need to understand the majority of those millino lines before you can make something useful/interesting/improved |
| 04:01.13 | SuperTaz | brlcad: good to know |
| 04:01.20 | SuperTaz | I did notice that in the diagram |
| 04:01.37 | poolio | SuperTaz: I'm a fairly young amateur coder and if you hvae a specific project, you can probably learn most of what you need to know in at most a couple weeks |
| 04:01.39 | SuperTaz | of course, physics and materials are both fairly complicated |
| 04:01.44 | brlcad | yes, going to add nurbs modeling to the gui -- probably focusing more on a new gui instead of hooking it into mged .. though mged might get some functionality in that area .. still tbd |
| 04:02.21 | brlcad | SuperTaz: diagram? |
| 04:02.48 | SuperTaz | poolio: I used to do standards work and R&D at a fortune 50 manufacturer, so I know...but there's a different between finding where to put the shoehorn, and understand why the shoehorn works :) |
| 04:03.01 | SuperTaz | brl: flowchart of how data moves? |
| 04:03.03 | SuperTaz | on the web |
| 04:03.13 | poolio | SuperTaz: yep, I'm currently satisfied with just sticking the shoehorn in there. |
| 04:04.12 | SuperTaz | pool: that's good for smaller stuff, but for major undertakings, such as interfacing complex libraries like a physics engine, you really have to understand the mechanics of the shoe, shoehorn, sock, and foot |
| 04:04.43 | SuperTaz | just takes a little longer :) |
| 04:05.00 | SuperTaz | and is a little more frustrating when you get things wrong, until you're intimately familiar with it :) |
| 04:05.02 | poolio | SuperTaz: true, you're at a far differnet leve than I |
| 04:05.33 | brlcad | SuperTaz: ahh, that old thing .. yes ;) |
| 04:05.46 | SuperTaz | hehehe |
| 04:06.02 | poolio | W0000000000T. |
| 04:06.06 | poolio | working :D :D |
| 04:06.24 | SuperTaz | getting the physics stuff in isn't going to happen easily |
| 04:06.39 | SuperTaz | btw, brl...specifying the cflags didn't help |
| 04:06.44 | SuperTaz | it still re-introduced the whitespace |
| 04:07.18 | SuperTaz | shame I don't have contact info for Frances from Xaos, anymore |
| 04:07.44 | SuperTaz | she would have been really adept at getting you your 2D compositor in place in a few days of effort |
| 04:08.14 | SuperTaz | she programmed most of the effects in Xaos Tools |
| 04:09.34 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/misc/brlcad-config.in: |
| 04:09.34 | CIA-4 | BRL-CAD: refactor the entire library processing so that it uses the exact same list that |
| 04:09.34 | CIA-4 | BRL-CAD: is used to compile from configure without maintaining a separate list in here. |
| 04:09.34 | CIA-4 | BRL-CAD: add additional options for libdir, includedir, and ldflags too. still need to |
| 04:09.34 | CIA-4 | BRL-CAD: scan the FLAGS vars for automake vars |
| 04:09.40 | SuperTaz | back in the 90's, in the days when an Onyx was the size of a refigerator and was about as powerful as my PC, and when the notion of a z-buffer was so new that you couldn't move the render window, else you'd get artifacts from all over :P |
| 04:10.30 | SuperTaz | btw...it's only the CFLAGS that are seeing injected whitespace |
| 04:10.38 | SuperTaz | the CPPFLAGS and LDFLAGS are just fine |
| 04:12.09 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/src/gtools/beset/beset.c: remove the if, just make the EOF check part of the case |
| 04:12.16 | SuperTaz | wound up with the following when I tried specifying the same CFLAGS at the commandline that it added: |
| 04:12.17 | SuperTaz | http://pastebin.ca/594860 |
| 04:12.39 | brlcad | CFLAGS is special |
| 04:12.59 | SuperTaz | in the short bus sense? |
| 04:13.17 | SuperTaz | or as in you have oodles of special handling in there (which I seemed to notice...) |
| 04:13.34 | yukonbob | SuperTaz: lots of info on Frances Dose on the 'net, if she's the one you were thinking of that worked at Xaos |
| 04:13.42 | poolio | brlcad: stop touching up mah code ;) |
| 04:13.48 | poolio | well I guess that was yours but ... heh :) |
| 04:14.00 | brlcad | no, there's only one place where CFLAGS is treated special and it's done to override autoconf's default -g -O2 behavior |
| 04:14.34 | SuperTaz | yes, that's the one, yukon :) |
| 04:14.42 | SuperTaz | getting her to contribute would be interesting, though |
| 04:15.46 | SuperTaz | not sure that che could, though |
| 04:15.51 | SuperTaz | she's working for philips now |
| 04:16.21 | yukonbob | http://www.francesdose.com/Coder.html |
| 04:16.25 | SuperTaz | so she might be contractually prohibited from doing it |
| 04:16.37 | SuperTaz | yeah |
| 04:16.43 | brlcad | ah, I see what the issue is |
| 04:17.02 | SuperTaz | it was pretty cool to learn pandemonium from the person who programmed all of the routine |
| 04:17.16 | brlcad | it's retrying configure because it fails to find X11 |
| 04:17.18 | SuperTaz | and I also got to meet the GUI designer |
| 04:17.24 | SuperTaz | hrmmm |
| 04:17.32 | SuperTaz | that's what it did the first time |
| 04:17.41 | SuperTaz | so I specified the location of x11 |
| 04:17.43 | SuperTaz | and then it found it |
| 04:17.52 | brlcad | checking for X11 link functionality... no |
| 04:17.55 | SuperTaz | here's my configure line (before the CFLAGS attempt) |
| 04:17.57 | *** join/#brlcad PrezKennedy (n=Matthew@c-76-106-124-125.hsd1.md.comcast.net) | |
| 04:18.14 | SuperTaz | ./configure --with-x11=/usr/X11R6 --prefix=/usr/local/brlcad --disable-retry |
| 04:18.22 | SuperTaz | hrmmm |
| 04:18.28 | SuperTaz | any ideas? |
| 04:18.44 | brlcad | you used that and it gave you http://pastebin.ca/594860 ? |
| 04:23.48 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/configure.ac: finally rip out the retry code -- causes too many headaches and confusion when the retry also fails since it has the tendency to hide the original real problem. |
| 04:24.48 | SuperTaz | yup |
| 04:24.51 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/m4/ (Makefile.am retry.m4): retry macros are no longer needed, the configure retry code was ripped out |
| 04:25.18 | brlcad | for some reason, it's not obeying the --disable flag |
| 04:25.59 | brlcad | you will likely have better luck working off of cvs head now ;) |
| 04:26.06 | SuperTaz | hahaha |
| 04:26.19 | SuperTaz | I need to install cvs on this box, I suppose |
| 04:26.23 | SuperTaz | I only have svn ;) |
| 04:26.43 | SuperTaz | cause that's what I use for everything these days :) |
| 04:26.49 | brlcad | we're doing the conversion over to svn later this year |
| 04:26.57 | SuperTaz | ahhh, good :) |
| 04:27.09 | SuperTaz | ahhh |
| 04:27.15 | SuperTaz | apple provides cvs |
| 04:27.26 | brlcad | I've converted most of the other projects I work with a long while ago, but brl-cad's conversion will need to be rather careful |
| 04:27.33 | brlcad | ~cadcvs |
| 04:28.33 | ibot | To obtain BRL-CAD from anonymous CVS: cvs -d :pserver:anonymous@brlcad.cvs.sourceforge.net:/cvsroot/brlcad login && cvs -d :pserver:anonymous@brlcad.cvs.sourceforge.net:/cvsroot/brlcad checkout -P brlcad |
| 04:28.35 | SuperTaz | *nod* |
| 04:28.35 | SuperTaz | I have to go find the instructions for using cvs to suck down brlcad ;) |
| 04:28.36 | brlcad | we've got almost 25 years of comprehensive cvs history to retain ... |
| 04:28.36 | brlcad | ibot: wake up |
| 04:28.45 | ibot | ACTION throws a barrel-full of ice water on up and shouts "GOOD MORNING!!!!" |
| 04:28.46 | SuperTaz | well, you COULD keep the cvs repository around |
| 04:28.48 | SuperTaz | though I'm sure it's possible to keep it |
| 04:29.10 | brlcad | it is perfectly possible, but it's not automatic |
| 04:29.17 | SuperTaz | *nod* |
| 04:29.48 | poolio | brlcad: what the heck. why was I quicksorting the candidate individuals based on fitness? |
| 04:29.52 | brlcad | and from my perspective, it's absolutely necessary .. not going to import fresh and just leave the history in cvs |
| 04:30.19 | brlcad | poolio: for crossover selection and other advancement criteria? |
| 04:30.36 | poolio | brlcad: well they dont need to be sorted for any of those... oh well.. |
| 04:30.46 | poolio | ill just comment it out, i'm sure i'm going somewhere with it I just have forgotten where |
| 04:31.14 | SuperTaz | good...you SHOULD keep the history |
| 04:31.16 | brlcad | you have lots of knobs you can tweak, like migrating the top N unmodified/unmutated, droping the bottown M outright, etc |
| 04:32.08 | SuperTaz | if you didn't want to use subversion internally, you could just expose nightly builds via svn (script a cvs checkout and a svn checkin) |
| 04:32.31 | brlcad | SuperTaz: we're rather proud of our extensive history, we just may have the oldest repository histories that has remained intact |
| 04:32.35 | SuperTaz | I suspect this cvs checkout is gonna take a while |
| 04:32.38 | poolio | brlcad: well the way I do it is that I use the fitness as a weight, and a weighted random individual picker |
| 04:32.50 | poolio | brlcad: but I guess for other selection methods it makes sense |
| 04:33.01 | SuperTaz | brlcad: that's awesome :) I read up on the history of the project...it's part of what attracted me to it :) |
| 04:33.03 | brlcad | the fact that we started even before cvs existed |
| 04:33.23 | SuperTaz | yeah |
| 04:33.42 | SuperTaz | what did the original team use for version control? |
| 04:33.54 | PrezKennedy | a notepad |
| 04:33.56 | PrezKennedy | haha |
| 04:34.02 | brlcad | talking to the ohloh folks a few months back, our history does/did go back the farthest of all projects in ohloh |
| 04:34.17 | brlcad | SuperTaz: it was in rcs before moving to cvs |
| 04:34.19 | SuperTaz | heh...that's cool |
| 04:34.21 | SuperTaz | ahhh |
| 04:34.25 | SuperTaz | god...I remember rcs |
| 04:34.38 | SuperTaz | almost forgot that we used to use rcs |
| 04:34.53 | SuperTaz | heh |
| 04:34.56 | SuperTaz | scary, huh? |
| 04:35.00 | brlcad | we gave the ohloh folks quite a head turn when brl-cad was added to their index |
| 04:35.12 | SuperTaz | it's been a good decade plus since I used ci/co ;) |
| 04:35.27 | brlcad | they thought their import was catching "all commits to head" when in fact it wasn't |
| 04:35.37 | SuperTaz | hahahaha :) |
| 04:35.50 | brlcad | all of brl-cad stats were massively wrong, missing thousands of commits through the 80's |
| 04:36.10 | SuperTaz | yikes |
| 04:36.15 | brlcad | a little hidden tidbit of knowledge about how cvs works |
| 04:36.29 | poolio | brlcad: I'm off for the night, but thanks for fixing up my code |
| 04:36.32 | brlcad | they were only pulling the latest rcs revision from cvs |
| 04:36.37 | brlcad | on head |
| 04:36.41 | SuperTaz | *nod* |
| 04:36.53 | CIA-4 | BRL-CAD: 03poolio * 10brlcad/src/gtools/beset/ (beset.c fitness.c population.c fitness.h): fixed bug where spheres with non-whole radii produced odd results |
| 04:37.02 | brlcad | since we predated, the rcs revision numbers have continued to be tweaked/incremented over the years for various accounting purposes |
| 04:37.10 | SuperTaz | *nod* |
| 04:37.10 | brlcad | night poolio |
| 04:37.21 | SuperTaz | wow...checkout finally finished :) |
| 04:37.30 | brlcad | they had to rearchitect their database and reimport all CVS projects when the found that out |
| 04:38.35 | SuperTaz | hehehe...nice |
| 04:38.43 | SuperTaz | nothing like a paradigm shift :) |
| 04:38.58 | SuperTaz | hadn't they ever seen a project that had been ported from rcs to cvs before?!? |
| 04:39.05 | SuperTaz | or am I just old? |
| 04:39.10 | SuperTaz | (and I'm not that old!) |
| 04:39.11 | brlcad | you're just that old ;) |
| 04:39.14 | poolio | 480 person years. geeeeez. |
| 04:39.19 | poolio | gnite guys |
| 04:39.20 | SuperTaz | yikes |
| 04:39.22 | SuperTaz | nite poolio |
| 04:40.29 | SuperTaz | all I can say is...I wasn't around the last time the Cubs won the World Series ;) |
| 04:41.22 | brlcad | heh, nor was I by a long shot :) |
| 04:41.53 | SuperTaz | how long have you been monkeying around with brl-cad? ;) |
| 04:42.07 | *** join/#brlcad cad54 (n=3d588390@bz.bzflag.bz) | |
| 04:43.44 | brlcad | since just a couple years before mike passed |
| 04:43.50 | SuperTaz | *nod* |
| 04:44.04 | SuperTaz | he passed in 2000, correct? |
| 04:44.04 | brlcad | about 9 years now |
| 04:44.27 | brlcad | yeah, Nov 2000 |
| 04:44.34 | brlcad | did you know him? |
| 04:44.45 | SuperTaz | no, I don't think we ever crossed paths |
| 04:44.56 | SuperTaz | unless we met at Siggraph or some such |
| 04:45.55 | SuperTaz | but back in the days of Siggraph (for me), I was mainly in the animation, mastering, and film areas |
| 04:46.03 | brlcad | hard to miss him at siggraph, he was very much in the "elite elders" crowd |
| 04:46.04 | SuperTaz | not as much in the cad area |
| 04:46.10 | SuperTaz | hehehe |
| 04:46.28 | SuperTaz | yeah, I may have met him in passing in the 90's |
| 04:46.44 | SuperTaz | I would remember if I'd had the opportunity to talk in depth |
| 04:47.20 | brlcad | he really was a brilliant guy (in many ways) |
| 04:47.33 | brlcad | and exceptionally charismatic |
| 04:47.41 | SuperTaz | I remember that's where I got my introduction to the guys from MRL at NYU |
| 04:48.06 | brlcad | one of the main reasons I was attracted towards working on brl-cad |
| 04:48.07 | SuperTaz | got to check out their robotics and fab lab there when I was next in NY...that was great |
| 04:48.10 | SuperTaz | *nod* |
| 04:48.18 | SuperTaz | it's a shame I never got to know him |
| 04:48.34 | SuperTaz | those are my favorite types of people :) |
| 04:49.04 | SuperTaz | at the IETF meetings, that was the crowd I was always around |
| 04:49.14 | brlcad | aha |
| 04:49.16 | SuperTaz | used to play nuclear war with the elders :) |
| 04:49.52 | SuperTaz | now that was some fun |
| 04:50.14 | brlcad | speaking of the ietf.. do you know don merrit? |
| 04:51.14 | SuperTaz | not really |
| 04:51.41 | SuperTaz | it's been a few years since I was really involved with the IETF |
| 04:51.44 | brlcad | he was the long-time ietf representative from BRL / ARL |
| 04:51.48 | SuperTaz | *nod* |
| 04:52.10 | SuperTaz | I didn't really know him |
| 04:52.40 | brlcad | just came to mind because I ran into him at the ball field just last week and some comment about the ietf came up :) |
| 04:52.47 | SuperTaz | hehehe |
| 04:53.11 | SuperTaz | I knew Scott Bradner and Allison Mankin |
| 04:53.19 | SuperTaz | I was in transport |
| 04:53.26 | SuperTaz | sigtran/TSVWG |
| 04:53.29 | SuperTaz | seamoby |
| 04:53.33 | SuperTaz | rohc |
| 04:53.53 | SuperTaz | knew Randy Bush |
| 04:54.03 | SuperTaz | used to run into him a bunch |
| 04:54.14 | SuperTaz | a lot of the guys I worked with were from europe, too |
| 04:54.26 | brlcad | foreign names to me :) |
| 04:54.35 | SuperTaz | yeah, different area |
| 04:56.55 | brlcad | so what caught your eye with brl-cad? |
| 04:56.57 | SuperTaz | actually, I think he did a bunch of work in the transport area |
| 04:56.59 | brlcad | presume you haven't used it before? |
| 04:57.06 | SuperTaz | no, I still haven't used it ;) |
| 04:57.54 | SuperTaz | it was that it's a solids modeller |
| 04:58.06 | SuperTaz | and that it can import and export several formats |
| 04:58.39 | brlcad | been a bit of a rocky start, from adapting to the conversion to open source, setting up an infrastructure that can grow collaboratively |
| 04:59.00 | SuperTaz | essentially, I have a design in my head that I need model and ultimately produce drawings of |
| 04:59.02 | brlcad | fighting the old-age bagged (the modeler's gui) which maintaining active development ;) |
| 04:59.12 | SuperTaz | heh :) |
| 04:59.19 | brlcad | s/which/while/ |
| 04:59.37 | SuperTaz | yeah, I figured ;) |
| 04:59.54 | SuperTaz | (btw...configure is running with ./configure --with-x11=/usr/X11R6 --prefix=/usr/local/brlcad) |
| 05:00.09 | SuperTaz | (we'll see if it finds X11 and/or completes this time |
| 05:00.10 | SuperTaz | ) |
| 05:00.27 | SuperTaz | anyway, I need to get this idea out of my head and figure out whether it'll work |
| 05:00.42 | SuperTaz | it'd be great if I could simulate what happens when it's dropped |
| 05:00.45 | brlcad | if you get interested in running a performance test, you'll probably also want to add --enable-optimized for faster performance |
| 05:00.50 | SuperTaz | but I realize that won't really happen |
| 05:00.57 | SuperTaz | oh? |
| 05:01.02 | SuperTaz | I can stop it |
| 05:01.06 | SuperTaz | and add that |
| 05:01.14 | brlcad | either way, it's perfectly fine without it too |
| 05:01.44 | brlcad | brl-cad also includes a benchmark performance suite that most like running, gives a really historic perspective of your system's performance |
| 05:01.48 | SuperTaz | stopped it, added the flag, re-ran it |
| 05:02.20 | brlcad | don't forget to make clean to clear out the unoptimized object files |
| 05:02.21 | SuperTaz | *nod* yeah, that's pretty cool |
| 05:02.31 | SuperTaz | it was in configure, still |
| 05:02.39 | SuperTaz | didn't get to make |
| 05:02.58 | SuperTaz | unless you broke the cardinal rule, and configure compiles more than tests?? |
| 05:03.03 | brlcad | you can run the benchmark either via "make benchmark" after the regular build completes or via the "benchmark" command after installation completes |
| 05:03.11 | SuperTaz | *nod* |
| 05:03.17 | SuperTaz | I'll probably install first |
| 05:04.09 | PrezKennedy | alright you old folks time for me to hit the hay |
| 05:04.09 | SuperTaz | this machine is an old PowerMac G4 |
| 05:04.17 | brlcad | PrezKennedy: heh |
| 05:04.25 | PrezKennedy | crazy people expect me to care about their hardware problems on a friday morning |
| 05:04.29 | SuperTaz | heh |
| 05:04.39 | SuperTaz | neither am I, neither am I! |
| 05:04.55 | SuperTaz | relative to some people, I'm quite young |
| 05:05.32 | SuperTaz | I haven't even lived half of my life (I hope) |
| 05:05.39 | brlcad | and with the upcoming advancements in medicine, one might live to be 300 |
| 05:05.56 | brlcad | *ahem* |
| 05:06.00 | SuperTaz | well, I CERTAINLY haven't lived half of my life if I break 150 :) |
| 05:06.16 | SuperTaz | or 120 |
| 05:06.18 | SuperTaz | or 100 |
| 05:06.21 | brlcad | you don't look a day over 130 |
| 05:06.21 | PrezKennedy | i havent even hit a quarter of mine |
| 05:06.23 | SuperTaz | . . . |
| 05:06.26 | PrezKennedy | i win! |
| 05:06.35 | SuperTaz | well |
| 05:06.38 | SuperTaz | yes, you win |
| 05:06.41 | SuperTaz | I suppose |
| 05:06.48 | SuperTaz | if it's a contest |
| 05:06.50 | brlcad | dunno |
| 05:06.54 | brlcad | with a name like PrezKennedy |
| 05:07.04 | brlcad | sounds like a recipe for "early retirement" |
| 05:07.08 | SuperTaz | hahahaha |
| 05:07.20 | SuperTaz | he may have lived half of his life already, and he just doesn't know it |
| 05:07.40 | SuperTaz | assuming he's at least in the second half of his teens |
| 05:07.51 | PrezKennedy | early 20's |
| 05:08.02 | SuperTaz | ahhh |
| 05:08.14 | brlcad | that's approaching mid-life crisis then |
| 05:08.14 | SuperTaz | <==== not early 20's ;) |
| 05:09.33 | brlcad | age is but a state of mind and I'm getting younger every day |
| 05:09.41 | SuperTaz | heh |
| 05:09.51 | SuperTaz | kennedy was 46 when he died |
| 05:10.08 | SuperTaz | so if you're 23 or older, you're screwed, son ;) |
| 05:10.31 | SuperTaz | (yes, I had to look that one up) |
| 05:10.53 | PrezKennedy | no relation |
| 05:10.56 | PrezKennedy | so im not worried |
| 05:11.04 | SuperTaz | wow...he won't even know where he was when the challenger exploded |
| 05:11.12 | SuperTaz | geez |
| 05:11.35 | PrezKennedy | give it a few years and neither will you ;-) |
| 05:11.58 | SuperTaz | you know...that's something that will stick in my mind for the rest of my life |
| 05:12.32 | SuperTaz | I remember exactly where I was |
| 05:12.36 | SuperTaz | and how stunned I was |
| 05:12.49 | SuperTaz | and we interrupted our lunch to have a minute of silence |
| 05:13.48 | brlcad | had a really interesting lecture about the name for those "life anchor" events in psychology in college |
| 05:13.52 | SuperTaz | I felt so awful for her students |
| 05:13.58 | brlcad | how you can date generations by those events |
| 05:13.59 | SuperTaz | yeah? |
| 05:14.05 | SuperTaz | it's true |
| 05:14.11 | brlcad | as one tends to stick out moreso than others |
| 05:14.17 | brlcad | even as other major events occur |
| 05:14.21 | SuperTaz | I've had more than one of them, unfortunately |
| 05:14.54 | SuperTaz | that, the world trade center bombing, the oklahoma city bombing, the final attack on the world trade center |
| 05:15.17 | brlcad | yep |
| 05:15.34 | SuperTaz | okay...configure completed that time |
| 05:15.42 | SuperTaz | I'm going to scroll back to see if it caught X11 |
| 05:16.07 | SuperTaz | looks like it |
| 05:16.14 | SuperTaz | 13 minutes, 37 seconds to configure |
| 05:16.21 | SuperTaz | the joy of a G4 400 |
| 05:16.34 | SuperTaz | make will take hours, I'm sure |
| 05:17.52 | SuperTaz | probably let it grind away while I sleep ;) |
| 05:18.47 | SuperTaz | take it psych 101 was after the challenger exploded for you? ;) |
| 05:19.25 | *** join/#brlcad tofu (n=sean@bz.bzflag.bz) | |
| 05:20.04 | tofu | others being landing on the moon, kennedy being shot, the tsunami, cuban missile crisis, pearl harbor, ... |
| 05:20.09 | SuperTaz | heh |
| 05:20.23 | SuperTaz | you were certainly not alive for all of those? |
| 05:20.24 | *** mode/#brlcad [+o brlcad] by ChanServ | |
| 05:20.45 | brlcad | heck no |
| 05:20.47 | SuperTaz | katrina |
| 05:21.14 | brlcad | doubt katrina was "big enough" |
| 05:21.24 | PrezKennedy | gnight folks |
| 05:21.27 | SuperTaz | I'm guessing psych 101 for you was somewhere between the challenger exploding and the first world trade center bombing? |
| 05:21.28 | brlcad | generally has to be "shocking" |
| 05:21.33 | brlcad | not just tragic |
| 05:21.34 | *** join/#brlcad poolio (n=poolio@c-69-251-3-107.hsd1.md.comcast.net) | |
| 05:21.53 | SuperTaz | I dunno...katrina was pretty profound for a lot of people |
| 05:22.06 | SuperTaz | and pretty shocking |
| 05:22.38 | SuperTaz | that the government couldn't handle it, that they let the levees fail, that a HUGE piece of american history was wiped out (NO had a LOT of history that was destroyed) |
| 05:22.39 | brlcad | no, I experienced challenger up front and center -- I was one of millions of kids in grade school that tuned in from the classroom attently watching the launch |
| 05:22.47 | SuperTaz | *nod* |
| 05:23.00 | SuperTaz | that puts you in your late 20's to early 30's |
| 05:23.14 | brlcad | thereabouts |
| 05:23.25 | SuperTaz | we weren't watching the launch |
| 05:23.30 | SuperTaz | we were having lunch |
| 05:24.02 | SuperTaz | the kitchen staff were listening to the launch on the radio |
| 05:24.09 | SuperTaz | and came out into the lunchroom to tell the asst. headmaster, whose table I was sitting at |
| 05:24.41 | SuperTaz | at the time, I still had dreams of being an astronaut |
| 05:24.49 | SuperTaz | hadn't quite broken the height limit, yet |
| 05:26.06 | SuperTaz | btw...I'd be a fan of seeing the site redone in rails |
| 05:26.40 | SuperTaz | as opposed to drupal/mediawiki |
| 05:26.48 | SuperTaz | just a personal bias ;) |
| 05:27.22 | *** join/#brlcad zapp (n=deltazap@pool-71-251-104-121.tampfl.fios.verizon.net) | |
| 05:29.27 | brlcad | :) |
| 05:30.04 | brlcad | or lack thereof ;) |
| 05:30.07 | SuperTaz | hahahaha |
| 05:30.11 | SuperTaz | complete lack thereof |
| 05:30.15 | SuperTaz | I develop that way, too ;) |
| 05:30.19 | SuperTaz | it makes me more efficient |
| 05:32.46 | SuperTaz | there don't seem to be any open source precision physics engines, btw |
| 05:33.27 | SuperTaz | and the real-time physics engines are pretty much useless for simulation, unless you want them for basic animation purposes only, instead of for simulation |
| 05:35.05 | brlcad | looked at ODE? |
| 05:35.12 | SuperTaz | yeah |
| 05:35.29 | SuperTaz | it's realtime rigid body |
| 05:35.47 | SuperTaz | <PROTECTED> |
| 05:35.53 | SuperTaz | i.e. games engine |
| 05:36.05 | brlcad | part of the puzzle for a games engine |
| 05:36.09 | SuperTaz | or so it seems...I'm reading more |
| 05:36.21 | brlcad | though useful for basic physics interactions too |
| 05:36.28 | SuperTaz | yes |
| 05:37.32 | SuperTaz | but if I put a physics engine into BRL-CAD, I'd want it to tell me how much force was applied at impact, and whether it exceeds deformation and breakage characteristics of a material |
| 05:38.22 | brlcad | there are only a couple open source libs, ode being one of the best |
| 05:38.24 | SuperTaz | though that may all be possible, in which case it'd be quite adequate |
| 05:38.37 | brlcad | it could possibly do that, but you'd have to break up your problem domain some, I'd imagine |
| 05:38.51 | brlcad | doubt the Bullet library could handle that either |
| 05:38.55 | SuperTaz | yeah, and that could defeat the purpose |
| 05:39.03 | SuperTaz | no...bullet is lower-end than ODE |
| 05:39.04 | brlcad | though I've seen bullet do deformation dynamics too |
| 05:39.09 | SuperTaz | can it? |
| 05:39.16 | SuperTaz | it seems to be lower-end than ODE |
| 05:39.21 | brlcad | it is |
| 05:39.26 | SuperTaz | deformation is pretty high-end |
| 05:39.37 | brlcad | it doesn't do all the work for you |
| 05:40.09 | brlcad | effectively hacked together, doing some of the interactions yourself letting bullet handle the first order physics |
| 05:40.20 | SuperTaz | ahhh |
| 05:40.58 | SuperTaz | I don't really care about lots and lots of deformation...I'm more concerned with impact forces and material ratings |
| 05:41.15 | SuperTaz | which are pretty exact |
| 05:41.46 | brlcad | ODE's probably the farthest along, but they're also still mostly looking at first order rigid body dynamics |
| 05:41.57 | brlcad | low energy interations, inelastic deformations |
| 05:42.14 | SuperTaz | *nod* |
| 05:42.25 | SuperTaz | it's just not there, yet |
| 05:44.30 | brlcad | they are what I'd planned on starting with for the new modeling interface simply because they're still the farthest along and can help set up basic interactions (even just for "good" collision detection) |
| 05:44.44 | brlcad | and adding on material interactions down the road |
| 05:44.47 | SuperTaz | *nod* |
| 05:45.22 | SuperTaz | the real question should be what you're planning on implementing it for |
| 05:45.41 | SuperTaz | you shouldn't be implementing a physics engine just to implement a physics engine |
| 05:45.47 | brlcad | actually nearly the same purpose you mentioned would be the driver |
| 05:45.54 | SuperTaz | you should be doing it with a specific application in mind |
| 05:45.56 | SuperTaz | *nod* |
| 05:46.09 | SuperTaz | well, it'd be a sensible part of a materials system :) |
| 05:47.05 | brlcad | it would be amusing to out-perform the (closed) physics interaction codes that hook up with brl-cad for performing V/L analyses |
| 05:48.03 | SuperTaz | don't know that that would happen |
| 05:48.15 | SuperTaz | i'd imagine they're pretty well optimized |
| 05:48.21 | brlcad | heh |
| 05:48.29 | brlcad | you'd think that wouldn't you.. :) |
| 05:48.47 | SuperTaz | you would |
| 05:48.54 | SuperTaz | doesn't mean it's true |
| 05:48.57 | SuperTaz | but you'd hope ;) |
| 05:49.47 | brlcad | the premiere code that does the interaction is far from real-time, also takes extensive prepartion to set up |
| 05:49.49 | SuperTaz | they used ODE for eternal sunshine of the spotless mind?? |
| 05:50.06 | SuperTaz | well, real-time = less accurate |
| 05:50.17 | SuperTaz | you're talking about precision vs. speed |
| 05:50.35 | SuperTaz | precision is definitely better in materials simulations |
| 05:52.05 | brlcad | yep |
| 05:53.29 | brlcad | when you care that it's "rolled aluminum 4122" vs "hardened aluminum 5293" compared to the sims that just care that it was "aluminum" or even worse "some metal" material |
| 05:53.55 | SuperTaz | well, for materials simulations, you always care |
| 05:53.56 | SuperTaz | a lot |
| 05:54.33 | brlcad | depends on what you mean by a materials simulation I'd think |
| 05:54.51 | SuperTaz | if you're trying to determine what's going to happen when you drop something from 50 feet, you care about the properties of the materials, how they dampen force, shock, and momentum, etc. |
| 05:55.43 | brlcad | well sure, that's a very specific relatively high-energy interaction |
| 05:55.57 | SuperTaz | that's what I need to simulate :) |
| 05:56.24 | brlcad | simulate dropping that material from 1 inch, and the simulation becomes very different |
| 05:56.48 | SuperTaz | actually, it's harder at 1 inch than at 50 feet |
| 05:56.57 | SuperTaz | because you have to be far more precise |
| 05:57.13 | SuperTaz | calculations require high precision at 1" |
| 05:57.22 | brlcad | still, depending on the purpose |
| 05:57.25 | SuperTaz | at 50', they just have to be accurate |
| 05:57.49 | brlcad | there was an implicit constraint of looking at the same order of magnitude change |
| 05:57.59 | SuperTaz | heh :) |
| 05:58.00 | SuperTaz | okay |
| 05:58.00 | brlcad | e.g. "how much mass is retained" |
| 05:58.12 | brlcad | very different between the two |
| 05:59.15 | brlcad | but I agree that you can ask just as hard if not harder questions for any interaction |
| 05:59.50 | SuperTaz | *nod* |
| 06:00.38 | SuperTaz | it should be sufficient to estimate how much force is applied to what areas, and whether deformation/breakage/absorption/damping threshholds are crossed |
| 06:00.59 | SuperTaz | and then what amount of energy is translated, in what direction, and so on |
| 06:01.48 | SuperTaz | basically, if I have 4 pieces of steel and dampers between them, and hit them with something, how much will each move/rebound if they're on the floor |
| 06:03.50 | SuperTaz | well, I used the example because the ground doesn't move |
| 06:04.00 | SuperTaz | it's a simpler calculation |
| 06:04.09 | brlcad | neither does a 70 ton tank generally speaking ;) |
| 06:04.15 | SuperTaz | right |
| 06:04.16 | brlcad | at least not significantly |
| 06:04.59 | SuperTaz | actually, they move about the same amount, when it's a relatively low energy application (i.e. dropping 10 lbs 50 feet onto the stack) |
| 06:05.45 | SuperTaz | they absorb/dissipate with similar characteristics |
| 06:05.56 | SuperTaz | it's stuff like the human body that gets harder |
| 06:05.59 | SuperTaz | but you can cheat |
| 06:06.04 | brlcad | yup |
| 06:06.15 | SuperTaz | essentially, you can similate a ballistics gel primitive |
| 06:06.20 | SuperTaz | and then you get close enough |
| 06:06.27 | brlcad | the velocities are generally much different too |
| 06:06.36 | SuperTaz | yes |
| 06:06.38 | SuperTaz | oops |
| 06:06.54 | SuperTaz | hrmmm...it's on opennurbs |
| 06:07.07 | brlcad | hah, wow |
| 06:07.12 | SuperTaz | slow, huh? :) |
| 06:07.17 | brlcad | quite |
| 06:07.22 | SuperTaz | G4 400 :) |
| 06:07.23 | brlcad | that a g4? |
| 06:07.25 | brlcad | heh |
| 06:07.28 | SuperTaz | yup |
| 06:07.31 | SuperTaz | it's archaic |
| 06:07.37 | SuperTaz | no $$ to upgrade it |
| 06:07.48 | SuperTaz | would be nice to bump it up to a faster G4 |
| 06:07.53 | SuperTaz | someday... |
| 06:07.53 | brlcad | that's probably going to take about 2 hours from start to finish |
| 06:08.01 | SuperTaz | 2 hours isn't too bad |
| 06:08.59 | SuperTaz | heh |
| 06:09.00 | SuperTaz | nice |
| 06:09.47 | SuperTaz | you working for DoD? |
| 06:09.48 | brlcad | used to take 2, but opennurbs and other developments over the past year have slowed things down |
| 06:09.59 | SuperTaz | yeah, I'm sure they'd add to it |
| 06:10.39 | brlcad | aside from g++ just taking way longer than gcc, in general |
| 06:11.02 | SuperTaz | SGI tried to hire me when they bought Cray, to figure out how to internetwork the two platforms |
| 06:11.04 | brlcad | though opennurbs is the only bit that is affected by that |
| 06:11.31 | SuperTaz | it's in the d's, now |
| 06:11.36 | SuperTaz | openurbs_defines.cpp |
| 06:11.46 | SuperTaz | detail... |
| 06:11.49 | brlcad | the one technology sgi did fortunately retain/leverage successfully |
| 06:12.14 | SuperTaz | yes, I didn't go work for them |
| 06:12.19 | brlcad | the craylink fiber that become numalink |
| 06:12.23 | SuperTaz | if I had, my life would have been different |
| 06:12.24 | SuperTaz | yup |
| 06:12.32 | SuperTaz | that's what I was supposed to work on, too ;) |
| 06:12.53 | SuperTaz | they were all like, WTF are we supposed to do with this?!? |
| 06:13.09 | brlcad | =) |
| 06:13.11 | SuperTaz | meantime, I'd just finished up a contract at Sun, managing a performance lab ;) |
| 06:13.32 | SuperTaz | dealing with the (at the time) super-fast OC-3 and OC-12 stuff |
| 06:13.41 | SuperTaz | and OMG!!! FIBER TO THE DESKTOP!!! |
| 06:14.23 | SuperTaz | nothing like trying to determine why a driver is caught in a lock-spin-lock pattern, instead of spin-lock-spin |
| 06:14.49 | SuperTaz | damned drivers were slower than 100M until that got fixed |
| 06:14.56 | brlcad | i still often wish apple would have bought out sgi to recover the high-end graphics niche and supercomputing technologies sgi still held |
| 06:15.16 | SuperTaz | yeah, woulda been nice |
| 06:15.25 | SuperTaz | but it didn't happen |
| 06:18.16 | SuperTaz | my G4 would have a 64 meg framebuffer with a z-buffer in it |
| 06:18.24 | SuperTaz | and wouldn't be a 16 meg ATI Rage PRo |
| 06:18.26 | SuperTaz | :P |
| 06:18.29 | brlcad | hehe |
| 06:18.35 | brlcad | is that a single? |
| 06:18.44 | SuperTaz | actually, though, I know a guy who designs the ATI GPUs :) |
| 06:18.50 | *** join/#brlcad elite01 (n=elite01@dslc-082-082-081-092.pools.arcor-ip.net) | |
| 06:19.01 | SuperTaz | it's dual |
| 06:19.05 | SuperTaz | the machine is SP, though |
| 06:19.13 | SuperTaz | but I've got a VGA port and a DVI port |
| 06:19.30 | SuperTaz | and the damned DVI port won't do 1650 |
| 06:19.35 | SuperTaz | which SUCKS |
| 06:19.50 | brlcad | ah, then you could speed up your compile |
| 06:19.51 | SuperTaz | because I have a Samsung 226BW |
| 06:19.56 | brlcad | you using -j3 ? |
| 06:20.02 | SuperTaz | no |
| 06:20.08 | SuperTaz | I normally would have |
| 06:20.20 | SuperTaz | but I was just happy to get it to compile |
| 06:20.26 | SuperTaz | it's single proc, single core |
| 06:20.44 | SuperTaz | so -j2 or -j3 at the outside are about all you can do to maximize |
| 06:21.02 | SuperTaz | next build I'll do -j3 |
| 06:21.07 | brlcad | erm .. "it's dual" but the machine is SP? |
| 06:21.28 | brlcad | it's either got one cpu or two, which is it? :) |
| 06:23.19 | elite01 | HT - also known as 1.2 CPUs :) |
| 06:23.46 | brlcad | the g4 didn't have ht.. |
| 06:24.13 | elite01 | no idea about macs |
| 06:25.44 | SuperTaz | heh |
| 06:25.49 | SuperTaz | no |
| 06:25.55 | SuperTaz | it's SP |
| 06:25.58 | SuperTaz | single proc |
| 06:26.03 | SuperTaz | it's dual capable |
| 06:26.19 | SuperTaz | but I can't afford a daughter card with two procs |
| 06:26.26 | SuperTaz | it's dual headed |
| 06:26.35 | brlcad | ah |
| 06:26.35 | elite01 | ok |
| 06:26.37 | SuperTaz | (If I want) |
| 06:26.55 | brlcad | then yeah, -j wouldn't really do much for ya |
| 06:27.03 | SuperTaz | -j2 helps |
| 06:27.13 | SuperTaz | -j3 even, sometimes |
| 06:28.10 | SuperTaz | because even on a SP machine you can squeeze more efficiency out when one process is io bound and the other is processor bound |
| 06:28.21 | SuperTaz | -j3 can help or hinder |
| 06:28.28 | SuperTaz | -j4 and above only hinder |
| 06:28.38 | elite01 | makes sense |
| 06:29.07 | brlcad | usually use something 1.5 to 2x whatever is available for that same reason |
| 06:29.16 | SuperTaz | yeah |
| 06:29.29 | SuperTaz | as I said before, though |
| 06:29.34 | brlcad | particularly for autocruft apps that have big i/o interlacing phases |
| 06:29.41 | SuperTaz | I just was happy enough to get the compile even STARTED ;) |
| 06:30.03 | SuperTaz | opennurbs...STILL |
| 06:30.11 | SuperTaz | planesurface |
| 06:30.37 | *** join/#brlcad MinuteElectron (n=chatzill@unaffiliated/MinuteElectron) | |
| 06:30.50 | SuperTaz | I must have forgotten to feed the hamster before I plugged it in to the accelerator port... |
| 06:30.58 | SuperTaz | too late now |
| 06:31.09 | SuperTaz | brb...need to take the dog out :) |
| 06:33.46 | MinuteElectron | You mean... That I have been to bed and woken up and you were at work all that time..... |
| 06:41.20 | SuperTaz | I think he does |
| 06:41.42 | SuperTaz | true dedication (or lack of home life) |
| 06:41.50 | SuperTaz | oh, FINALLY |
| 06:41.59 | SuperTaz | linking opennurbs |
| 06:50.50 | SuperTaz | ok...I'm off ot bed |
| 06:50.54 | SuperTaz | night night all :) |
| 06:51.05 | MinuteElectron | goodnight |
| 07:16.17 | brlcad | MinuteElectron: something like that |
| 07:16.26 | MinuteElectron | :P |
| 07:17.00 | MinuteElectron | having a look at some of them webpages in that text document, the design I am making at the moment seams a bit... bland. I am making some changes to it. |
| 07:20.21 | brlcad | sounds good |
| 07:20.41 | brlcad | like I said, i really do expect that will be the hardest and probably the most time consuming part to get doing |
| 07:20.50 | brlcad | s/doing/going/ |
| 07:50.42 | MinuteElectron | brlcad: Does the logo need alpha transparency or can I downgrade it to single-color transparency? |
| 08:05.23 | MinuteElectron | brlcad: If you would like to take a look and give your views on a preliminary outline I have done, not much yet, but I would rather restyle it now than once I finished the entire thing - http://localhost/drupal-5.1/ |
| 08:05.33 | MinuteElectron | oops http://82.7.33.28/drupal-5.1/ |
| 08:19.26 | *** join/#brlcad MinuteElectron (n=chatzill@unaffiliated/MinuteElectron) | |
| 08:49.33 | *** join/#brlcad MinuteElectron (n=chatzill@unaffiliated/MinuteElectron) | |
| 09:24.34 | *** join/#brlcad MinuteElectron (n=chatzill@unaffiliated/MinuteElectron) | |
| 09:58.30 | *** join/#brlcad docelic (n=docelic@212.91.113.221) | |
| 10:33.46 | *** join/#brlcad MinuteElectron (n=chatzill@unaffiliated/MinuteElectron) | |
| 10:53.36 | *** join/#brlcad AchiestDragon (n=david@whipy.demon.co.uk) | |
| 11:36.38 | *** join/#brlcad elite01 (n=elite01@dslc-082-082-081-092.pools.arcor-ip.net) | |
| 11:52.27 | *** join/#brlcad MinuteElectron (n=chatzill@unaffiliated/MinuteElectron) | |
| 12:40.24 | poolio | mornin' |
| 13:17.43 | *** join/#brlcad akreal (n=ak@ll-81-222-164-251.awanti.ru) | |
| 13:39.43 | poolio | brlcad: sometime today it'd be sweet if you could explain a bit to me how to get a binary tree from the internal stored CSG tree :) |
| 14:08.34 | *** join/#brlcad MinuteElectron (n=chatzill@unaffiliated/MinuteElectron) | |
| 14:09.47 | brlcad | poolio: okay, but it'll be a while |
| 14:10.05 | poolio | brlcad: alright. I'm just having issues understanding how everything is stored and how to interact with the stored trees |
| 14:10.52 | poolio | any source code examples anywhere? I see the red black tree stuff, is that what I'm looking for? |
| 14:10.56 | MinuteElectron | brlcad: busy? |
| 14:11.35 | brlcad | poolio: no, but you could certainly store your tree in a rb treee |
| 14:11.38 | brlcad | MinuteElectron: a bit |
| 14:12.04 | MinuteElectron | Okay, whe you get a chance your opinon on http://localhost/drupal-5.1/ would be appreciated. I have done a lot of work on it this morning, but there is no rush. |
| 14:12.12 | brlcad | poolio: probably first want to get familiar with creating csg trees (man libwdb, look in src/proc-db and/or src/mk) |
| 14:12.21 | brlcad | MinuteElectron: localhost doesn't help me :) |
| 14:12.34 | MinuteElectron | Oops, http://82.7.33.28/drupal-5.1/ |
| 14:12.37 | poolio | brlcad: thanks |
| 14:13.22 | poolio | brlcad: is there a way to work online with the internally stored tree? like duplicate an entire CSG tree and the modify it internally and save it to a new database? |
| 14:13.24 | brlcad | poolio: then after you know how to create them, you can look at the code that traverses them -- the converters are pretty boilerplate src/conv/ |
| 14:13.40 | poolio | alright cool. that should be plenty to start me off with, thanks for the pointers |
| 14:13.40 | brlcad | there are at least three different ways to traverse too |
| 14:13.51 | MinuteElectron | The logos on the side are there to make the sidebar bigger for the time being, but will be removed once I add more stuff to the sidebar. |
| 14:13.51 | poolio | eek. and i'm sure advantages and disadvantages for reach |
| 14:14.05 | brlcad | db_walk_tree() being one, manual iteration, and another whose name escapes me |
| 14:15.15 | brlcad | MinuteElectron: looking like progress, though I'd remove all the powered by spam on the panel :) .. there should be a page dedicated that talks about the site, gives credits |
| 14:15.39 | MinuteElectron | Yeah, as I said - that was just to make the sidebar longer. |
| 14:15.53 | brlcad | ah, missed that |
| 14:16.01 | MinuteElectron | dw |
| 14:16.43 | MinuteElectron | brlcad: I wasn't keeping logs, can you remind me what is going into MediaWiki and what is going into Drupal? |
| 14:17.36 | brlcad | mediawiki is all of the dynamic content, drupal is for the stuff that hooks into a database or stuff that is fairly static |
| 14:18.24 | brlcad | e.g. there's going to be a knowledge base "faq" of sorts -- that's entirely mediawiki |
| 14:19.01 | brlcad | but a profile page of premiere brl-cad users would be in a drupal page |
| 14:19.04 | MinuteElectron | Okay, that gives me a good starting point. |
| 14:20.11 | MinuteElectron | So as a brainstorm: |
| 14:20.28 | MinuteElectron | What are the items going to be listed on the navigation? |
| 14:20.35 | CIA-4 | BRL-CAD: 03brlcad * 10brlcad/misc/brlcad-config.in: strip out flags that have ${variables} in them that refer to the src path |
| 14:21.00 | MinuteElectron | I presume you want 'Home', 'About', 'News' |
| 14:22.29 | MinuteElectron | I guess that they will all be on Drupal. |
| 14:22.37 | MinuteElectron | Download would also be on Drupal |
| 14:23.31 | MinuteElectron | and probably Contact (although I am not sure whether or not it would be best to use a MediaWiki extension, Drupal extension or custom form for that). |
| 14:23.51 | MinuteElectron | Then the documentation and FAQ would be on MediaWiki. |
| 14:23.52 | brlcad | MinuteElectron: http://my.brlcad.org/sitemap.txt |
| 14:24.01 | MinuteElectron | Oh, cool. |
| 14:24.06 | brlcad | subject to change and rearrangmenet of course, but that's a start |
| 14:24.13 | MinuteElectron | You always seem to be one step ahead of me :P |
| 14:24.35 | brlcad | been ready for this for quite a while -- just need someone to actually do it and take all the glory ;) |
| 14:26.03 | brlcad | also, be sure to drop the register mark -- I'm not convinced the phrase itself is registered any more, at most it's a TM |
| 14:26.28 | MinuteElectron | ok |
| 14:45.21 | poolio | |10:18| <@ brlcad> mediawiki is all of the dynamic content, drupal is for the stuff that hooks into a database or stuff that is fairly static |
| 14:45.24 | poolio | |10:18| <@ brlcad> e.g. there's going to be a knowledge base "faq" of sorts -- that's entirely mediawiki |
| 14:45.27 | poolio | |10:19| <@ brlcad> but a profile page of premiere brl-cad users would be in a drupal page |
| 14:45.32 | poolio | ah sorry guys |
| 14:45.38 | poolio | Middle mouse button in dismay :\ |
| 15:29.18 | SuperTaz | Elapsed compilation time: 3 hours, 33 minutes, 42 seconds |
| 15:29.29 | SuperTaz | that's how long it takes on a G4 400 |
| 15:30.02 | brlcad | heh SuperTaz .. |
| 15:30.29 | SuperTaz | sad, huh? |
| 15:31.03 | brlcad | i thought my dual 500 g4 was slow :) |
| 15:31.17 | SuperTaz | hahaha |
| 15:31.31 | SuperTaz | this thing crawls :) |
| 15:32.19 | poolio | makes my aging laptop feel speedy :) |
| 15:32.29 | SuperTaz | heh :) |
| 15:33.05 | SuperTaz | yeah...I'm doing make install, first |
| 15:33.47 | poolio | brlcad: I'm taking g2asc.c and stripping it to my needs :) |
| 15:46.47 | MinuteElectron | brlcad: http://82.7.33.28/drupal-5.1/?q=user/register and http://82.7.33.28/drupal-5.1/ |
| 15:47.50 | MinuteElectron | Please suggest improvements, everything is working (fully tested in Safari, IE, Firefox and Safari) except the list in the left box when you are logged in and the tabs (no longer tabs) in IE. |
| 15:49.04 | SuperTaz | install just finished |
| 15:50.00 | poolio | is the color scheme permanent? like the "solid modeling for a strong defense" banner doesn't really fit well with the left nav bar |
| 15:50.22 | MinuteElectron | It can be changed, what color do you suggest? |
| 15:50.54 | poolio | dunno, i'm not a good design person, I just don't think it works :P |
| 15:51.02 | MinuteElectron | hmm |
| 15:51.07 | poolio | You could try just matching it to the left nav green |
| 15:51.17 | MinuteElectron | perhaps, one sec |
| 15:51.53 | poolio | http://82.7.33.28/drupal-5.1/themes/brlcad/gradient.jpg <-- scale it down pleeeeease :) |
| 15:52.31 | SuperTaz | Minimum run time is 3 minutes, 12 seconds |
| 15:52.31 | SuperTaz | Maximum run time is 30 minutes |
| 15:52.31 | SuperTaz | Estimated time is 9 minutes, 36 seconds |
| 15:52.59 | MinuteElectron | poolio: Will do once I am finished. |
| 15:53.53 | MinuteElectron | poolio: look now |
| 15:54.07 | MinuteElectron | I think the colors do need to be different. |
| 15:54.18 | MinuteElectron | Let me try a different one |
| 15:55.25 | SuperTaz | interesting watching this thing run |
| 15:55.25 | poolio | MinuteElectron: I don't think it's just the colors, I think my main gripe is I want more of a banner at the top of a page |
| 15:55.34 | SuperTaz | probably doesn't help that I'm doing other things at the same time, though |
| 15:55.42 | poolio | note: my opinion doesn't matter and it is only my opinion, so don't just change it cause I don't like it, other people may think it's awesome :) |
| 15:56.00 | MinuteElectron | no, I need a second opinon |
| 15:56.09 | MinuteElectron | poolio: so you want the banner to be alightly taller |
| 15:56.46 | SuperTaz | heh |
| 15:56.47 | MinuteElectron | *slightly taller? |
| 15:58.38 | MinuteElectron | poolio: look now |
| 16:00.41 | SuperTaz | Abs Fruit-Punch.local 47236.61 22868.08 27491.27 22049.86 25273.94 31281.82 29366.93 Fri Jun 29 10:59:04 CDT 2007 |
| 16:00.41 | SuperTaz | *vgr Fruit-Punch.local 344.76 341.00 490.30 413.22 357.53 2.11 324.82 |
| 16:00.41 | SuperTaz | #*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*# |
| 16:00.41 | SuperTaz | Benchmark results indicate an approximate VGR performance metric of 325 |
| 16:00.41 | SuperTaz | Logarithmic VGR metric is 2.51 (natural logarithm is 5.78) |
| 16:00.43 | SuperTaz | #*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*# |
| 16:01.57 | SuperTaz | that's a G4 400 |
| 16:02.02 | SuperTaz | with other stuff going on |
| 16:02.21 | brlcad | heh, nice |
| 16:03.15 | SuperTaz | yeah |
| 16:03.28 | brlcad | SuperTaz: it won't be incredibly affected by other activities on the machine as it is measuring cpu time, wallclock measurements are stored elsewhere |
| 16:03.36 | brlcad | at most, a 10% variance |
| 16:03.45 | SuperTaz | less than half the speed of your dual 500 unde 10.2 |
| 16:04.01 | SuperTaz | but 10.4 is known to have higher overhead in exchange for more stability |
| 16:04.03 | brlcad | which is probably about right given the lineage of those machines |
| 16:04.12 | SuperTaz | could be |
| 16:04.38 | MinuteElectron | brlcad: busy? |
| 16:04.44 | SuperTaz | is your dual 500 a sawtooth, gigabit ethernet, digital audio, or quicksilver? |
| 16:15.17 | SuperTaz | well, mged appears to actually run :) |
| 16:27.55 | MinuteElectron | brlcad: Please can you take a look at the website? |
| 16:29.49 | *** join/#brlcad tarzeau (i=sengun@kneazle.ethz.ch) | |
| 16:54.36 | AchiestDragon | Benchmark results indicate an approximate VGR performance metric of 3421 |
| 16:54.38 | AchiestDragon | Logarithmic VGR metric is 3.53 (natural logarithm is 8.14) |
| 16:54.49 | SuperTaz | what is that? |
| 16:54.54 | SuperTaz | (arch) |
| 16:55.31 | SuperTaz | not bad |
| 16:58.03 | elite01 | quad xeon and just a gig of ram? pah! :) |
| 16:58.48 | AchiestDragon | for £200 GBP im not grumbling |
| 17:00.15 | elite01 | uh |
| 17:00.15 | elite01 | ook |
| 17:00.31 | brlcad | SuperTaz: I suspect with the right CFLAGS, you'd actually come out at around 400MHz .. the G4 is almost right on 100 vgr's per MHz |
| 17:00.44 | brlcad | s/400MHz/400 vgrs/ |
| 17:02.32 | elite01 | where's thy benchmark result stored? |
| 17:02.48 | brlcad | elite01: hm? |
| 17:02.53 | elite01 | in brlcad? |
| 17:02.59 | elite01 | some file? |
| 17:03.08 | brlcad | don't understand the question |
| 17:03.27 | brlcad | when you run the benchmark, it outputs the results right then and there as well as writing out to log files |
| 17:03.29 | elite01 | if i run the brlcad benchmark, is the result stored in some file or just printed to the console? |
| 17:03.30 | SuperTaz | you think I could get it that high? |
| 17:04.22 | elite01 | can't find the log file, just individual ones for the pictures |
| 17:04.39 | brlcad | elite01: the file named "summary" has the important details |
| 17:04.41 | brlcad | two lines per run |
| 17:05.19 | brlcad | if you are using the latest cvs head sources, then there's also a run-####-benchmark.log file that has the entire output that you saw during the run written to a file |
| 17:05.43 | elite01 | hmm i can't find either |
| 17:06.35 | elite01 | anyway, this centrino duo seems to be pretty decent |
| 17:08.21 | *** join/#brlcad MinuteElectron (n=chatzill@unaffiliated/MinuteElectron) | |
| 17:08.28 | AchiestDragon | its ony a dual cpu xeon it shows up as 4 because of hyperthreading |
| 17:08.37 | elite01 | i see |
| 17:08.53 | SuperTaz | HT is blech |
| 17:09.03 | SuperTaz | get one with dual cores :) |
| 17:09.14 | elite01 | still, i just have 3324.60 bogomips on each cpu (which of course precisely describes the performance, as the name implies) |
| 17:10.45 | brlcad | elite01: are you running from cvs head? |
| 17:10.52 | brlcad | if so, the results are in the bench/ dir |
| 17:10.54 | elite01 | hmm no, just 7.10 |
| 17:13.59 | elite01 | Benchmark results indicate an approximate VGR performance metric of 3544 |
| 17:13.59 | elite01 | Logarithmic VGR metric is 3,55 (natural logarithm is 8,17) |
| 17:14.03 | elite01 | look at that, AchiestDragon :P |
| 17:14.12 | elite01 | on a T2300 |
| 17:14.13 | elite01 | gtg, cu |
| 18:09.21 | *** join/#brlcad tarzeau (i=sengun@berlin.ethz.ch) | |
| 18:15.45 | AchiestDragon | not bad |
| 18:16.59 | AchiestDragon | but as you can see from this pic http://www.whipy.demon.co.uk/desk.jpg i have 3 dual xeon machines atm |
| 18:22.26 | AchiestDragon | i have been looking at getting a dual quad core xeon mobo and chips but there a bit overpriced atm |
| 18:23.04 | SuperTaz | heh |
| 18:27.00 | brlcad | AchiestDragon: hehe, a rack in your room.. nice |
| 18:27.08 | brlcad | that's gotta be a bit noisy |
| 18:27.22 | brlcad | that wood paneling? |
| 18:27.54 | *** join/#brlcad dtidrow (n=dtidrow@host131.objectsciences.com) | |
| 18:28.04 | AchiestDragon | its got a half door it makes the systems a bit quieter , no its cardboard on the sides the side pannels are a bit tatty |
| 18:28.21 | brlcad | ah |
| 18:29.26 | dtidrow | just ran 'make benchmark' on my Dell M90 here: 5620 |
| 18:29.54 | *** join/#brlcad MinuteElectron (n=chatzill@unaffiliated/MinuteElectron) | |
| 18:29.58 | dtidrow | not too shabby for a latop :-) |
| 18:30.06 | AchiestDragon | :) |
| 18:30.31 | dtidrow | of course, work owns it ;-) |
| 18:31.02 | dtidrow | no way I could afford a $7k laptop... |
| 18:31.17 | brlcad | that's pretty sweet for a laptop |
| 18:31.30 | dtidrow | mbp? |
| 18:32.05 | dtidrow | you want system specs? |
| 18:32.09 | dtidrow | 10os[Linux 2.6.20-1.2948.fc6 i686] 10cpu[2 x Intel(R) Core(TM)2 CPU T7600 @ 2.33GHz @ 2.33GHz] 10mem[Physical : 2802MB/3288MB Free] 10disk[Total : 3.28GB/48.48GB Free] 10video[Quadro FX 2500M] 10sound[] |
| 18:34.03 | *** join/#brlcad cadguy (n=cadguy@c-67-166-125-250.hsd1.co.comcast.net) | |
| 18:34.09 | dtidrow | there's actually 4GB of system memory, but apparently it can only access ~3.3GB of it in 32-bit mode - I'd have to switch to a 64-bit install to get at the rest |
| 18:35.22 | poolio | man. i want a new laptop now :P |
| 18:35.46 | brlcad | mac book pro |
| 18:36.01 | dtidrow | oh, yeah - 1920x1200 flatpanel :-) |
| 18:36.15 | brlcad | ah, not that one |
| 18:36.23 | brlcad | i won't get that one for a few more weeks |
| 18:36.32 | poolio | ...bastard |
| 18:36.35 | brlcad | i have the one right before it now |
| 18:36.45 | poolio | brlcad: send me the old one ;) |
| 18:36.49 | dtidrow | lol |
| 18:36.50 | brlcad | hehe |
| 18:37.07 | brlcad | that new display on those is amazing |
| 18:37.34 | poolio | you're gonna kill your eyes looking at that flat panel. 1600 is bad enough |
| 18:37.35 | dtidrow | wish it had one of the mobile G80-based gfx cards, though... |
| 18:37.54 | dtidrow | pixels? what pixels? ;-) |
| 18:38.53 | dtidrow | yeah, I've had to up the font size on most of my apps - turning forty sucks... |
| 18:38.56 | AchiestDragon | i like the dell 2405fpw 24" lcd i have , the new 32" is even better aparantly |
| 18:39.30 | dtidrow | and the prices for those have really come down, too |
| 18:39.53 | AchiestDragon | they seem to be droping each month |
| 18:40.10 | dtidrow | I have one at home, but those new 24" LCD's for around $650 are really tempting... |
| 18:40.51 | AchiestDragon | btw if you still have a crt a new 19" lcd will pay for itself 2 years just by what you save in power over a 20" crt |
| 18:41.25 | poolio | yeah yeah. once I get my paycheck I might think about investing in one to take with me to college |
| 18:43.38 | MinuteElectron | I sit here with a feeling of awe. I thought getting a 19" LCD flat panel monitor was good, yet you guys talk about 24" LCD's as if they are mid-range. |
| 18:44.14 | dtidrow | they are now :-) |
| 18:44.20 | AchiestDragon | yes |
| 18:44.36 | dtidrow | just a couple of years ago they were $2000 or more |
| 18:44.50 | MinuteElectron | Heh, my 2.4Ghz processor could never compete with what you guys are benchmarking at. |
| 18:45.28 | brlcad | 24"? pffft. |
| 18:45.35 | dtidrow | I need to try my Athlon X2 at home and see how it does - still haven't finished updating it |
| 18:45.52 | dtidrow | Mike would be proud ;-) |
| 18:45.58 | brlcad | :-) |
| 18:45.58 | AchiestDragon | http://cgi.ebay.co.uk/New-Dell-Ultrasharp-2407WFP-24-LCD-Flat-Panel-Monitor_W0QQitemZ120134332519QQihZ002QQcategoryZ174QQssPageNameZWDVWQQrdZ1QQcmdZViewItem |
| 18:47.50 | dtidrow | http://www.tigerdirect.com/applications/SearchTools/item-details.asp?EdpNo=3066352&CatId=2775 - holy crap |
| 18:47.57 | dtidrow | wonder if it's any good.... |
| 18:48.23 | dtidrow | $440 for a 24" 1920x1200??? damn! |
| 18:51.19 | brlcad | nice |
| 18:51.46 | dtidrow | yeah - wonder what's wrong with it... ;-) |
| 18:52.02 | poolio | brlcad: if my monitor "breaks" and I suddenly cannot work.... hmmm.... |
| 18:52.30 | dtidrow | lol |
| 18:52.34 | brlcad | i probably have an old 12" b&w crt I could loan ya |
| 18:52.41 | dtidrow | rofl |
| 18:53.02 | poolio | brlcad: damn you. :P |
| 18:53.06 | brlcad | g'evening cadguy |
| 18:53.21 | cadguy | Top of the day brlcad |
| 18:53.28 | brlcad | or afternoon or whatever it is |
| 18:54.13 | dtidrow | yeah, that was nice |
| 18:54.22 | cadguy | Anybody running out to get their iPhone today? |
| 18:54.31 | dtidrow | pfft! |
| 18:54.50 | AchiestDragon | the 30" dell ones are £815 GBP here atm |
| 18:55.44 | SuperTaz | I have a 22" Samsung |
| 18:55.44 | SuperTaz | 2ms |
| 18:55.54 | SuperTaz | nicest damned monitor I've seen |
| 18:56.01 | SuperTaz | still gives me wood |
| 18:56.11 | brlcad | cadguy: probably tomorrow |
| 18:56.31 | MinuteElectron | SuperTaz: What model? |
| 18:56.35 | dtidrow | SuperTaz: which model? |
| 18:56.37 | brlcad | maybe later today, but pretty busy afternoon |
| 18:56.39 | dtidrow | gmta |
| 18:57.02 | SuperTaz | it's a 226BW |
| 18:57.17 | MinuteElectron | I have a 940N |
| 18:57.23 | poolio | You guys and your toys, share the weatlh. |
| 18:57.38 | *** join/#brlcad IriX64 (n=mario_du@bas2-sudbury98-1177726501.dsl.bell.ca) | |
| 18:57.44 | MinuteElectron | SuperTaz: It is the same, but only 19" and only has analog input and is not widescreen. |
| 18:58.06 | dtidrow | http://www.tigerdirect.com/applications/SearchTools/item-details.asp?EdpNo=2881203&CatId=2775 - not bad... |
| 18:58.21 | dtidrow | $320 before the rebate |
| 18:58.39 | MinuteElectron | <-- mine http://www.tigerdirect.com/applications/searchtools/item-Details.asp?EdpNo=2487377&sku=S203-1942 |
| 18:59.07 | SuperTaz | so, basically, it's a different monitor ;) |
| 18:59.22 | MinuteElectron | yeah |
| 18:59.45 | MinuteElectron | *space |
| 18:59.59 | IriX64 | http://irix64.spaces.live.com/photos (stuff albumn) 1st picture |
| 19:01.42 | poolio | Wow guys |
| 19:01.56 | poolio | this is my monitor: http://www.notebookreview.com/default.asp?newsID=2995&review=Asus+W3J |
| 19:02.02 | dtidrow | IriX64: isn't working here |
| 19:02.12 | IriX64 | that url? |
| 19:02.17 | IriX64 | just a sec |
| 19:02.25 | poolio | IriX64: it never works for me |
| 19:02.42 | dtidrow | poolio: nice compact laptop :-) |
| 19:02.44 | poolio | dtidrow: xml parsing error? |
| 19:02.49 | IriX64 | try now |
| 19:02.59 | poolio | dtidrow: Yes it is, the issue I have is it's a bit small and uncomfortable for 9 hour work days. |
| 19:03.07 | dtidrow | XML Parsing Error: syntax error |
| 19:03.07 | dtidrow | Location: http://irix64.spaces.live.com/photos/ |
| 19:03.07 | dtidrow | Line Number 3, Column 49:<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0//EN"> |
| 19:03.07 | dtidrow | ------------------------------------------------^ |
| 19:03.09 | poolio | I'm looking to get an ultraportable for college, like a 12 or 13" |
| 19:03.26 | IriX64 | what are you using? |
| 19:03.30 | *** join/#brlcad jimmyz (n=asd@host86-133-245-247.range86-133.btcentralplus.com) | |
| 19:03.38 | *** join/#brlcad jimmybrb (n=asd@host86-133-245-247.range86-133.btcentralplus.com) | |
| 19:04.01 | dtidrow | firefox v.1.5.0.12 |
| 19:10.12 | IriX64 | ftp://www3.sympatico.ca/ try that, havoc.png |
| 19:11.16 | cadguy | poolio: I've been impressed with the MacBook. Nice CPU. The only thing it lacks is a really high performance graphics chip (they save that for the Pro) |
| 19:11.16 | dtidrow | ugh - not loading :-\ |
| 19:11.25 | IriX64 | :) |
| 19:11.57 | poolio | cadguy: Yeah, I'm not sure I want a macbook. They're a bit overpriced for what you get and everyone has them. I'd also want something smaller and with better battery life I think |
| 19:12.20 | poolio | This 14" is really nice, and a nice GPU, the issue is just that the battery life is short cause of the GPU and it's a bit too heavy to trek around with. |
| 19:13.33 | cadguy | I've also had good luck with Lenovo. My issues is I want something *reliable*. The Macs and Lenovo's have been good to me. |
| 19:13.34 | poolio | cadguy: and as corny as it is, I need my right mouse button, mainly for key combos in my WM |
| 19:13.52 | poolio | is there a 12 or 13" lenovo? |
| 19:14.37 | poolio | I think with the new macbook pros they should have just gone ahead and divided it in two. I mean how often do you find yourself Ctrl+Clicking...there's a reason the mighty mouse exists for the desktop, now give us right click on the laptops |
| 19:15.25 | dtidrow | dang, this laptop isn't as expensive as I thought - ~$4200 for this configuration |
| 19:15.35 | cadguy | Check the lenovo X series. |
| 19:15.38 | poolio | are you kidding me? |
| 19:15.44 | poolio | dtidrow: that sounds hella expensive to me |
| 19:15.54 | dtidrow | less than what I thought it was |
| 19:15.57 | cadguy | Again, my criteria is *reliable* |
| 19:16.04 | dtidrow | I was guessing it was around $7000 |
| 19:16.27 | poolio | cadguy: as opposed to? bleeding edge? |
| 19:16.32 | cadguy | 61's start at 1215.00 US |
| 19:17.09 | poolio | cadguy: ugh, they are only XGA? |
| 19:17.09 | cadguy | gotta go get lunch |
| 19:17.21 | poolio | cadguy: enjoy |
| 19:18.29 | brlcad | cadguy: is that the same chris johnson?? |
| 19:18.39 | brlcad | (from the article) |
| 19:18.41 | MinuteElectron | cadguy: 61" monitor? |
| 19:18.42 | IriX64 | dtidrow: check that site for rotated :) |
| 19:19.33 | MinuteElectron | http://82.7.33.28/public/100_1028.jpg <-- my monitor |
| 19:19.36 | MinuteElectron | *computer |
| 19:20.33 | IriX64 | back to my compile :) |
| 19:20.45 | dtidrow | IriX64: which site? |
| 19:20.56 | poolio | MinuteElectron: Do I spy mIRC? |
| 19:21.13 | MinuteElectron | No, that would be ChatZilla. |
| 19:21.54 | MinuteElectron | poolio: lol, what is bad about mIRC? |
| 19:22.11 | SuperTaz | wrong question |
| 19:22.15 | poolio | ...nothing |
| 19:22.20 | MinuteElectron | oh, |
| 19:22.22 | MinuteElectron | I see, |
| 19:22.25 | MinuteElectron | mIRC is good. |
| 19:25.21 | MinuteElectron | brlcad: Are you busy? |
| 19:25.34 | brlcad | MinuteElectron: always |
| 19:25.34 | brlcad | what's up? |
| 19:26.50 | MinuteElectron | I altered the colors, added the tabs and added the navigation. |
| 19:27.06 | brlcad | I see I see |
| 19:27.10 | brlcad | hmmm |
| 19:28.10 | brlcad | don't like the tagline up top, but the bar color is interesting |
| 19:28.12 | MinuteElectron | The tabs (although they aren't really tabs) don't work properly in IE yet (still trying to diagnose the problem) and the box on the left has this freaky navigation thing which I can't get to go white (when you are logged in). |
| 19:28.35 | MinuteElectron | So keep the bar, but remove the tagline? |
| 19:28.55 | MinuteElectron | and perhaps make the bar a solid color? |
| 19:29.21 | IriX64 | it's beautiful MinuteElectron |
| 19:29.43 | MinuteElectron | IriX64: you really think so? wow, thanks. |
| 19:30.30 | MinuteElectron | brlcad: Are you still thinking? |
| 19:30.36 | IriX64 | first time i've seen it |
| 19:31.09 | IriX64 | login username a bit large but that could be a good thing :) |
| 19:31.10 | brlcad | dtidrow and whomever else was following, MBP benchmark results: |
| 19:31.12 | brlcad | Benchmark results indicate an approximate VGR performance metric of 5937 |
| 19:31.28 | poolio | brlcad: are you kidding me!?!?!? |
| 19:31.39 | poolio | brlcad: what kind of crazy optimizations did you do... I've got dual 1.83s and I only got 3k |
| 19:31.40 | IriX64 | what else was going on? |
| 19:32.44 | dtidrow | brlcad: MBP? |
| 19:32.49 | brlcad | poolio: did you --enable-optimized? |
| 19:32.58 | brlcad | dtidrow: mac book pro |
| 19:33.02 | dtidrow | ah |
| 19:33.05 | IriX64 | when mine builds i'll try one :) |
| 19:33.18 | dtidrow | which cpu? |
| 19:33.33 | IriX64 | me? opteron |
| 19:33.57 | brlcad | Intel(R) Core(TM)2 CPU T7600 @ 2.33GHz |
| 19:34.09 | dtidrow | 10cpu[2 x Intel(R) Core(TM)2 CPU T7600 @ 2.33GHz @ 2.33GHz w/ 4096 KB L2 Cache] |
| 19:34.17 | dtidrow | that one, iow :-) |
| 19:34.18 | CIA-4 | BRL-CAD: 03jlowenz * 10brlcad/src/librt/g_brep.cpp: update CloseTo signature (takes a Sample (rename to ON_CurveSample or something) to return the closes point) |
| 19:34.36 | brlcad | er, you have a 2x2cpu? |
| 19:35.21 | dtidrow | dual core, yes |
| 19:35.28 | brlcad | i.e. two dual cores |
| 19:35.52 | brlcad | or a single dual core |
| 19:36.00 | brlcad | the 2 x is throwing me off |
| 19:36.02 | dtidrow | just the one dual-core cpu |
| 19:36.07 | brlcad | okay, so same chip |
| 19:36.54 | brlcad | difference is probably in the OS and optimization options |
| 19:36.56 | *** join/#brlcad yukonbob (n=bch@whthyt224-180.northwestel.net) | |
| 19:36.58 | dtidrow | the cpu detection logic doesn't make the distinction between cores and cpu packages |
| 19:38.17 | CIA-4 | BRL-CAD: 03jlowenz * 10brlcad/src/other/openNURBS/ (opennurbs_curve.h opennurbs_curve.cpp): flesh out implementation of CloseTo (generic curve version samples the curve and uses a binary search to try and find the closest point) |
| 19:40.18 | dtidrow | well, I ran configure with --enable-optimized, did you do anything else in addition? |
| 19:43.37 | *** join/#brlcad PrezKennedy (n=Matthew@c-76-106-124-125.hsd1.md.comcast.net) | |
| 19:47.16 | brlcad | yup, a slew |
| 19:47.22 | brlcad | bbl |
| 19:52.55 | SuperTaz | brb |
| 20:00.14 | cadguy | MinuteElectron: I like the faded bar on the top. Looks cool. The green background on the left I think is a bit shocking or out of place. Pick complementary colors? |
| 20:00.55 | MinuteElectron | Hmm, I have one person saying the green on the left was bad and another saying that the green at the top (which I removed) was bad. It seams everyone hates green. |
| 20:01.29 | IriX64 | i like green reminds me i should commune with nature :) |
| 20:01.29 | *** join/#brlcad SuperTaz (n=taz@adsl-69-211-3-171.dsl.chcgil.ameritech.net) | |
| 20:01.31 | cadguy | The key is there is nothing else in the color scheme which works with the green. |
| 20:01.51 | MinuteElectron | There isn't? I thought that the gold went well... |
| 20:02.04 | MinuteElectron | Anyway, I need to think of a color to replace the green. |
| 20:02.14 | SuperTaz | there we go |
| 20:02.16 | cadguy | Everything else is softer. The green stands out as a saturated, in-your-face effect compared to the gradual muted tones elsewhere |
| 20:02.23 | IriX64 | ftp://www3.sympatico.ca (nowthisisasystem.png) shot of mine in action :) |
| 20:02.24 | SuperTaz | just added an extra 512 meg stick of RAM |
| 20:02.30 | SuperTaz | that helps a bit :) |
| 20:03.13 | cadguy | Desaturate the green and try it. |
| 20:03.39 | MinuteElectron | cadguy: ok |
| 20:03.43 | cadguy | Go for a slightly more olive look |
| 20:04.12 | cadguy | there's also the blue in the logo that you could work with. |
| 20:05.23 | cadguy | Perhaps the silver-grey of the right side of the eagle's head? Maybe a brown or a blue? |
| 20:06.25 | cadguy | The logo already gives a bit of the color scheme, unless you're willing to be modifying the logo somewhat. |
| 20:06.38 | MinuteElectron | look now |
| 20:07.25 | cadguy | Much less jarring. Need to mod the logo so the lower green area is "transparent" and matches the background I suppose. |
| 20:07.37 | cadguy | I like it much better now. |
| 20:07.45 | MinuteElectron | purge your cache please |
| 20:09.32 | IriX64 | forgot the faceplate, look for faceplate.png :) |
| 20:10.20 | cadguy | I'd still mod the green of the logo a little. Other than that, I think you've got something. Still need to drop the Tag line at the top? |
| 20:10.42 | MinuteElectron | brb |
| 20:12.10 | dtidrow | http://www.extremetech.com/article2/0,1558,2151961,00.asp?kc=ETRSS02129TX1K0000532 - rofl at reason #1 |
| 20:12.47 | dtidrow | ah - no wonder it' |
| 20:13.20 | dtidrow | it's so deserted around here - it's "iPhone Day" ;-) |
| 20:17.48 | IriX64 | wheres the uPhone :) |
| 20:22.44 | MinuteElectron | cadguy: removed the tagline and topbar gradient |
| 20:24.08 | MinuteElectron | cadguy: I am not going to edit the logo because I didn't make it. If it is really important I will ask brlcad - he might have the original files. |
| 20:24.11 | IriX64 | should break down and understand that mug tutorial, instead of just playinf with the example geometry sigh. |
| 20:24.21 | IriX64 | errr playing |
| 20:25.27 | cadguy | MunuteElectron: Looking good now. |
| 20:25.36 | MinuteElectron | cool |
| 20:25.38 | MinuteElectron | thxs |
| 20:26.07 | MinuteElectron | If only it didn't break in IE. |
| 20:27.07 | cadguy | There's a certain poetic justice there |
| 20:27.26 | IriX64 | how does it break in ie i just loaded it. |
| 20:27.49 | MinuteElectron | IE6 |
| 20:27.59 | IriX64 | thats me |
| 20:28.06 | MinuteElectron | hmm, which page are you on? |
| 20:28.06 | IriX64 | looks good |
| 20:28.14 | IriX64 | the test page |
| 20:28.18 | MinuteElectron | IriX64: do you have a different browseR? |
| 20:28.28 | IriX64 | i have firfox too |
| 20:28.33 | IriX64 | firefox |
| 20:29.05 | MinuteElectron | try it in there |
| 20:29.05 | IriX64 | just a sec. |
| 20:29.05 | MinuteElectron | it looks completely different. |
| 20:29.08 | MinuteElectron | then go to the create account page in IE, the yellow bar goes smaller |
| 20:30.30 | IriX64 | works here looks just as good :) |
| 20:31.04 | MinuteElectron | hmm |
| 20:31.11 | MinuteElectron | odd |
| 20:31.14 | MinuteElectron | very odd |
| 20:31.19 | MinuteElectron | what screen resolution are you on? |
| 20:31.28 | IriX64 | create account, just a sec ill reload ie |
| 20:31.40 | IriX64 | 1024x768 |
| 20:32.10 | MinuteElectron | Hmm, maybe that is the reason. |
| 20:32.30 | MinuteElectron | I am on 1280x1024 |
| 20:32.42 | MinuteElectron | Ahh well, if no one else can replicate it I needent worry. |
| 20:33.23 | IriX64 | ftp://www3.sympatico.ca ie.png |
| 20:34.37 | MinuteElectron | I cannot access that server for some reason. |
| 20:35.39 | MinuteElectron | I get a connection failed error. |
| 20:35.44 | IriX64 | really? |
| 20:35.46 | MinuteElectron | yeah |
| 20:35.54 | IriX64 | i was updating it sorry |
| 20:36.01 | IriX64 | try now |
| 20:36.10 | IriX64 | same server mapped.png :) |
| 20:37.00 | MinuteElectron | I still can't access it - it asks for a username, then a password then aborts |
| 20:40.19 | MinuteElectron | IriX64: You still there? |
| 20:42.58 | IriX64 | anonymous ftp |
| 20:43.19 | IriX64 | use any e-mail address i care little |
| 20:43.47 | IriX64 | who@who.com:) |
| 20:44.05 | IriX64 | waddayoucare@howdareyou.com:) |
| 20:44.12 | dtidrow | doesn't work here either |
| 20:44.33 | dtidrow | could be a filter of some sort on the domain, though |
| 20:44.41 | IriX64 | what are you using? |
| 20:44.44 | dtidrow | would commandline ftp work? |
| 20:44.49 | IriX64 | should |
| 20:44.54 | IriX64 | let me try |
| 20:44.58 | MinuteElectron | I am using WinSCP4 |
| 20:46.44 | dtidrow | [dtidrow@localhost ~]$ ftp www3.sympatico.ca |
| 20:46.44 | dtidrow | Connected to www3.sympatico.ca. |
| 20:46.44 | dtidrow | 220- Access to the Sympatico Personal Webserver |
| 20:46.44 | dtidrow | 220- is only available to Sympatico subscribers |
| 20:46.44 | dtidrow | 220- |
| 20:46.45 | dtidrow | 220- You must be connected via Bell Sympatico |
| 20:46.47 | dtidrow | 421 Service not available, remote server has closed connection |
| 20:46.49 | dtidrow | ftp> |
| 20:46.53 | MinuteElectron | same here |
| 20:47.04 | IriX64 | lemme call sympatico, ill be back |
| 21:01.43 | IriX64 | sigh they're looking at it sorry |
| 21:02.48 | MinuteElectron | dw |
| 21:04.25 | IriX64 | could give you my username and password i guess (*not*) :) |
| 21:06.04 | MinuteElectron | :P |
| 21:08.37 | IriX64 | sorry should have asked, do you want it? |
| 21:12.43 | MinuteElectron | want what? |
| 21:12.52 | IriX64 | a dcc of ie.png |
| 21:12.59 | IriX64 | dcc send |
| 21:14.03 | MinuteElectron | sure |
| 21:29.39 | IriX64 | sending |
| 21:29.53 | MinuteElectron | thanks |
| 21:30.28 | MinuteElectron | It failed. |
| 21:30.28 | IriX64 | i have no idea :) |
| 21:30.35 | IriX64 | drat |
| 21:30.51 | IriX64 | doh, im behind a wall :) |
| 21:31.03 | MinuteElectron | bah, one sec |
| 21:31.12 | MinuteElectron | IriX64: You have an SCP client? |
| 21:31.20 | MinuteElectron | *SFTP |
| 21:33.51 | MinuteElectron | IriX64: PM --> |
| 21:45.03 | IriX64 | alas no |
| 21:45.10 | IriX64 | i can mail it to yoy |
| 21:45.13 | IriX64 | you |
| 21:57.02 | IriX64 | finally a chance to do a make benchmark :) |
| 22:02.51 | MinuteElectron | afk |
| 22:03.22 | IriX64 | :) |
| 22:03.44 | IriX64 | prefer ask |
| 22:03.46 | IriX64 | ? |
| 22:03.53 | IriX64 | a sad kind :) |
| 22:05.32 | MinuteElectron | back |
| 22:09.08 | IriX64 | http://rafb.net/p/AEStPI53.html <--- benchmark on an opteron |
| 22:27.01 | IriX64 | thats an optimized build with compiler optimization enabled too |
| 22:28.56 | dtidrow | hmmmm |
| 22:29.02 | dtidrow | brlcad: you still around? |
| 22:31.17 | IriX64 | he answers "frayed knot" :P |
| 22:33.11 | IriX64 | minuteelectron, have you considered the stryker jpg for background on that page? |
| 22:33.21 | MinuteElectron | what is that? |
| 22:33.46 | IriX64 | look in screen shots and images on http://brlcad.org |
| 22:34.33 | IriX64 | it's my current wallpaper |
| 22:34.44 | MinuteElectron | hmm, I personally I don't think background images work very well - I can try but don't hope for much. |
| 22:34.57 | IriX64 | your choice :) |
| 22:34.58 | MinuteElectron | Also I would need it in a much highe resolution. |
| 22:35.12 | IriX64 | ill e-mail you one |
| 22:35.13 | MinuteElectron | and also it would much up depending on your resolution. |
| 22:35.18 | IriX64 | if you'll permit |
| 22:35.23 | SuperTaz | ermmm |
| 22:35.24 | MinuteElectron | yeah, ok |
| 22:35.31 | SuperTaz | background images are bad, mmkay? |
| 22:35.44 | MinuteElectron | background images are best if the are a simple repeating pattern. |
| 22:35.50 | SuperTaz | NO! |
| 22:35.59 | SuperTaz | background images are bad |
| 22:36.08 | MinuteElectron | ok |
| 22:36.16 | SuperTaz | the ONLY case where they're marginally okay |
| 22:36.32 | SuperTaz | is when they're a watermark |
| 22:37.22 | SuperTaz | i.e. they're in a static spot on the page and don't repeat |
| 22:37.25 | SuperTaz | and are faint |
| 22:37.31 | IriX64 | sent |
| 22:37.35 | SuperTaz | and even that isn't reliable cross-browser |
| 22:37.38 | MinuteElectron | recieved |
| 22:37.43 | IriX64 | sorry i don't do pages well |
| 22:37.52 | SuperTaz | it's okay |
| 22:38.02 | IriX64 | thanks |
| 22:38.04 | SuperTaz | it's really a question of several elements |
| 22:38.18 | SuperTaz | good design must have the following elements: |
| 22:38.29 | SuperTaz | 1) it should be easy to use |
| 22:38.43 | SuperTaz | 2) it should incorporate natural usage paradigms |
| 22:38.52 | SuperTaz | 3) it should be aesthetically pleasing |
| 22:39.04 | SuperTaz | 4) it should make the users feel good |
| 22:39.09 | SuperTaz | that's a good design |
| 22:39.22 | SuperTaz | well |
| 22:39.27 | IriX64 | woot ogl came up :) |
| 22:39.42 | SuperTaz | 1.5) it should do what the users need it to do |
| 22:40.08 | MinuteElectron | Does BRL-CAD compile and run on Windows? |
| 22:40.32 | IriX64 | depends who you talk too ;) |
| 22:40.52 | MinuteElectron | Oh o.O |
| 22:41.37 | MinuteElectron | how do you mean? |
| 22:42.24 | SuperTaz | I think he means "not reliably" |
| 22:42.43 | IriX64 | don't put lies in my mouth please :) |
| 22:42.46 | SuperTaz | I suspect you could probably get it to work with cygwin and XFree86 |
| 22:42.49 | MinuteElectron | Oh, ok. |
| 22:43.04 | IriX64 | how about X11R6 |
| 22:43.07 | IriX64 | :) |
| 22:43.07 | SuperTaz | but i haven't tried |
| 22:43.15 | AchiestDragon | well i can understand the 1 and 2 bit |
| 22:43.35 | SuperTaz | XFree86 IS X11R6 |
| 22:43.56 | IriX64 | adsl you have to remnt :) |
| 22:43.59 | SuperTaz | it's a specific distribution and port of X11R6 |
| 22:44.02 | IriX64 | rent too :) |
| 22:44.09 | MinuteElectron | dsl = damn small linux |
| 22:44.22 | IriX64 | pfu=pretty fine unix |
| 22:44.33 | MinuteElectron | it would appear not |
| 22:44.39 | AchiestDragon | think the best example i can think of is blender , it takes a while to learn how to use the gui , against well the 3d editor in autotrax that you can pickup on in minuts |
| 22:45.13 | MinuteElectron | How many GBs is BRL-CAD uncompiled? |
| 22:45.38 | IriX64 | whos brlcad? |
| 22:45.42 | AchiestDragon | 28Gb on the source download but that compressed |
| 22:45.51 | MinuteElectron | wtf? |
| 22:45.52 | AchiestDragon | mb not gb |
| 22:45.54 | dtidrow | MB, you mean |
| 22:45.55 | MinuteElectron | Holy crap. |
| 22:46.05 | MinuteElectron | oh 28MB |
| 22:46.08 | MinuteElectron | that is not much |
| 22:46.19 | MinuteElectron | I was expecting something like 5GBs. |
| 22:46.23 | IriX64 | it is when you're waiting for the transfer |
| 22:46.38 | IriX64 | use cvs much better |
| 22:46.40 | dtidrow | and about 176MB unpacked |
| 22:46.49 | AchiestDragon | depends how you look at it , i remeber working on systems when 4k was alot of ram |
| 22:46.50 | MinuteElectron | hmm, cool |
| 22:47.18 | IriX64 | my s-100 had a whopping 64K |
| 22:47.58 | AchiestDragon | my nascom1 had 1k and the pdp8 i had had 4k |
| 22:48.24 | IriX64 | how much did the sinclair have? |
| 22:48.25 | SuperTaz | I remember when your program had to fit in 4k... |
| 22:48.43 | IriX64 | brlcad wouldn't do well there :) |
| 22:48.44 | SuperTaz | ...and just how much people could do with 4k back then. |
| 22:48.53 | AchiestDragon | 1k on the zx80 and zx81 the spectrum had 16 or 48k |
| 22:48.59 | IriX64 | then microsoft came along :P |
| 22:49.17 | dtidrow | and massively fscked everything up.... |
| 22:49.44 | IriX64 | look at it this way you get much more bytes for your money now :P |
| 22:50.24 | AchiestDragon | i tend to do programming in asm and i used a lot of microcontrolers with 1 to 8k of ram doing quite a lot for the ram size , if i programed the same in c i would be looking at needing a chip with more than 4 times the ram |
| 22:51.11 | IriX64 | not if you have a good optimizing compiler |
| 22:51.23 | IriX64 | and you can do asm in c |
| 22:52.24 | IriX64 | ahh optimize for size instead of speed :) |
| 22:52.40 | SuperTaz | you can make scrambled eggs in a blender |
| 22:52.52 | SuperTaz | does that mean it's a good idea/ |
| 22:52.52 | SuperTaz | ? |
| 22:53.03 | IriX64 | it's all experience |
| 22:53.16 | AchiestDragon | yes but some code even optimized is not as quick as if you wrote the same task in native asm |
| 22:53.18 | SuperTaz | if you're trying to work in 1-4k, C is probably not the language you REALLY want |
| 22:53.43 | IriX64 | i know i'm just yanking your chain and it's working :) |
| 22:54.01 | SuperTaz | with microcontrollers, you often want ASM, or architecture-specific development tools |
| 22:54.27 | IriX64 | see above, back to testing this thing :) |
| 22:54.35 | AchiestDragon | like r=a+b then call function if r=1 usualy leads to some complex c maths lib call rather than a add forlowed by a call on condition |
| 22:55.56 | MinuteElectron | lol, my code had no validity errors whatsoever (except 5 irrelevant css warnings) - I am amazed. |
| 22:56.32 | AchiestDragon | most of the asm tools are avalabe free for most micros , the c compilers though are usualy 3rd party but gcc will do most |
| 22:56.41 | MinuteElectron | Oh dear, goodnight. |
| 23:02.18 | AchiestDragon | it was not so mutch having limited ram space that was a problem but the total lack of having any sutable storage for swap space , using papertape as swap space just does not work |
| 23:03.03 | IriX64 | http://rafb.net/p/vG1mWq75.html <--- if i can't show you the pretty pictures at least I can show you this much :) |
| 23:03.46 | IriX64 | does if you feed the output of the punch to the input of the reader :) |
| 23:04.53 | AchiestDragon | well not realy , that way ends up with a fifo ,, |
| 23:05.09 | IriX64 | heh just speculating |
| 23:05.26 | IriX64 | works quite well for testing a model 28 |
| 23:05.34 | IriX64 | or 35 and up |
| 23:05.53 | AchiestDragon | and you end up with miles of used papertape |
| 23:06.22 | IriX64 | sometimes necessary and they didn't recycle back then either, mores the pity |
| 23:06.58 | IriX64 | sometimes intermittent clutches were hard to find |
| 23:07.20 | IriX64 | break time |
| 23:09.03 | AchiestDragon | at one time the souce would come on papertape with brlcad having 176mb in the source that would take 120,000 miles of papertape for the pdp8 i had |
| 23:09.16 | dtidrow | heh |
| 23:12.37 | AchiestDragon | and take 54 days to read in on a asr33 teletype |
| 23:13.20 | AchiestDragon | Timeout on server |
| 23:13.21 | AchiestDragon | Connection was to 82.7.33.28 at port 80 |
| 23:13.21 | SuperTaz | I think he took it down when he went offline |
| 23:15.41 | brlcad | yeah, figured.. thanks |
| 23:28.38 | IriX64 | I can't get there either |
| 23:30.48 | IriX64 | hah a 1024 frame buffer |
| 23:30.53 | IriX64 | :) |
| 23:32.47 | IriX64 | and bonus it works, cyall l8r |
