| 00:52.30 | *** join/#brlcad packrat (~packrator@99-67-225-40.lightspeed.livnmi.sbcglobal.net) |
| 01:10.34 | *** join/#brlcad crazy_imp (~mj@a89-182-14-228.net-htp.de) |
| 02:33.28 | *** join/#brlcad starseeker (~starseeke@BZ.BZFLAG.BZ) |
| 02:58.06 | *** join/#brlcad IriX64 (~IriX64@bas2-sudbury98-1096601295.dsl.bell.ca) |
| 03:00.46 | *** join/#brlcad PrezKennedy (MK@whitecalf.net) |
| 03:12.43 | *** join/#brlcad dli_ (~dli@dsl-173-248-203-45.acanac.net) |
| 04:09.01 | *** join/#brlcad KimK (~Kim__@ip174-71-95-176.om.om.cox.net) |
| 06:02.25 | *** join/#brlcad Stattrav (~Stattrav@111.93.134.142) |
| 06:02.25 | *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav) |
| 08:02.19 | *** join/#brlcad d_rossberg (~rossberg@BZ.BZFLAG.BZ) |
| 10:10.31 | *** join/#brlcad KimK (~Kim__@ip174-71-95-176.om.om.cox.net) |
| 12:29.25 | CIA-52 | BRL-CAD: 03Rossberg 07http://brlcad.org * r2753 10/wiki/Google_Summer_of_Code/Project_Ideas: corrected my irc name |
| 13:26.07 | CIA-52 | BRL-CAD: 0368.34.98.23 07http://brlcad.org * r2754 10/wiki/Google_Summer_of_Code/Project_Ideas: group method of communicatio together |
| 13:41.06 | CIA-52 | BRL-CAD: 03Sean 07http://brlcad.org * r2755 10/wiki/Google_Summer_of_Code/Project_Ideas: extra info |
| 13:53.19 | *** join/#brlcad Stattrav (~Stattrav@111.93.134.142) |
| 13:53.19 | *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav) |
| 15:09.55 | *** join/#brlcad AbhijitKane7 (~Abhijit@111.93.5.194) |
| 15:22.52 | *** join/#brlcad AbhijitKane (~Abhijit@111.93.5.194) |
| 15:23.27 | AbhijitKane | Hi, I intend to apply to BRLCAD thru the GSOC program, and work for the "Materials Database" project. |
| 15:23.37 | AbhijitKane | Is there any specific mentor associated with the project? |
| 15:32.13 | *** join/#brlcad Zaebos (~irc@pd95b7f5e.dip0.t-ipconnect.de) |
| 15:38.33 | brlcad | hi AbhijitKane, what draws you to that particular project? |
| 15:38.47 | brlcad | hello Zaebos |
| 15:39.02 | brlcad | saw your message on the mailing list as well |
| 15:40.10 | AbhijitKane | I've always been interested in web-related projects, and I think that a web interface would help a lot of people access the data that you have. |
| 15:40.31 | Zaebos | hi |
| 15:40.32 | brlcad | AbhijitKane: what are your thoughts for that project? |
| 15:40.41 | brlcad | what do you envision it entailing? |
| 15:41.46 | AbhijitKane | actually, thats what I wanted to clear |
| 15:41.52 | *** part/#brlcad Zaebos (~irc@pd95b7f5e.dip0.t-ipconnect.de) |
| 15:42.11 | AbhijitKane | At first glance, i thought it was a simple front-end to query the database |
| 15:43.03 | AbhijitKane | but I'm not sure what else is required to be done |
| 15:43.42 | brlcad | how much do you know about materials? |
| 15:43.49 | brlcad | and material properties |
| 15:44.15 | brlcad | it is a simple front-end to the database, but it also entails establishing the database |
| 15:44.28 | brlcad | setting up the schema |
| 15:44.58 | AbhijitKane | Not much. I've just done one course related to material properties - I haven't got any other exposure to the subject |
| 15:45.02 | brlcad | there was someone that actually worked on this project five or so years ago, but they never finished |
| 15:45.15 | brlcad | made great progress, but never got to production status |
| 15:45.34 | AbhijitKane | Oh, is that the "proof-of-concept web work of relevance" mentioned on the website? |
| 15:45.40 | brlcad | yes |
| 15:46.19 | brlcad | so your project could either leverage all that previous work as a starting point or you could start from scratch |
| 15:47.24 | brlcad | if you start from scratch, you'll need to do some basic research on materials science, so our database captures common properties sufficiently |
| 15:47.24 | AbhijitKane | Is the database schema / website source code included in the BRLCAD code? |
| 15:47.29 | brlcad | it is not |
| 15:47.52 | brlcad | I'll have to dig around for it in our archives -- it was quite a while ago |
| 15:48.10 | brlcad | it's available (somewhere), it'll just take a little while to find it |
| 15:48.26 | brlcad | regardless -- it was a homegrown setup |
| 15:48.37 | brlcad | how were you envisioning the implementation? |
| 15:49.11 | brlcad | using a content management system, rapid dev toolkits, or also from scratch? something else? |
| 15:49.48 | AbhijitKane | I have the most experience with MySQL. I also have a fair idea about UML, so if there are UML diagrams that I could look at, it would be advantageous. |
| 15:50.24 | AbhijitKane | Not a CMS, but I could take a look at the old code or work from scratch |
| 15:50.34 | brlcad | heh, no UML |
| 15:51.41 | brlcad | personally (and others may have different opinions), I'd prefer that IFF you are not going to use any web toolkits or CMS that you then leverage the old code and pick up where they left off |
| 15:51.55 | brlcad | otherwise there'd be no difference if it was all just from scratch again, effort wasted |
| 15:52.09 | brlcad | rather, duplicate effort .. not necessarily wasted |
| 15:52.19 | AbhijitKane | i agree |
| 15:52.55 | brlcad | have you ever used a revision control system before for web development? |
| 15:53.10 | AbhijitKane | I used subversion for a project that I did last year |
| 15:53.26 | brlcad | it'll be required for this, as will creating a simple patch showing your ability |
| 15:53.33 | brlcad | oh? |
| 15:53.36 | brlcad | who'd you work with? |
| 15:53.56 | AbhijitKane | It wasn't a part of gsoc. I was interning with a start-up |
| 15:55.14 | brlcad | anything on-line that we can take a look at? |
| 15:56.33 | brlcad | AbhijitKane: also, what got you interested in BRL-CAD? |
| 15:56.41 | AbhijitKane | I helped with this: http://www.vindev.net/phototourv2/public/apps/html5 |
| 15:56.51 | AbhijitKane | its an HTML5 implementation of www.phototour.in |
| 15:56.59 | brlcad | be sure to put that in your application |
| 15:57.19 | AbhijitKane | ok |
| 15:58.16 | AbhijitKane | I was just looking through all the organisations that wanted back-end web development, and BRL-CAD was one of the few |
| 16:00.05 | brlcad | nods |
| 16:01.05 | AbhijitKane | Is there any reading material related to material properties that you can recommend? |
| 16:02.05 | brlcad | there are some references on the detail page |
| 16:02.34 | brlcad | there are several online commercial material database websites that you could look through for comparison/reference, see what all they include |
| 16:02.48 | AbhijitKane | ok |
| 16:02.52 | brlcad | few google searches will lead you to two or three of them |
| 16:03.56 | AbhijitKane | i'll take a look at them |
| 16:04.14 | AbhijitKane | I have to go - be back in an hour or so |
| 16:04.22 | brlcad | nods |
| 16:04.24 | brlcad | cya later! |
| 16:04.28 | AbhijitKane | bye! |
| 16:04.45 | brlcad | can't wait to see the proposal |
| 16:07.19 | *** join/#brlcad adityag (~ADITYA@182.237.144.88) |
| 16:09.59 | adityag | i am pretty fluent in C & web related projects. i would love to contribute if im guided by some one experienced |
| 16:10.44 | adityag | brlcad : i would love to work here |
| 16:11.11 | brlcad | adityag: howdy and welcome |
| 16:11.34 | brlcad | the C background is great, we have tons of C projects ;) |
| 16:11.49 | brlcad | suggest you scan down through our project ideas page and see if anything catches your interest |
| 16:11.59 | adityag | in fact, i won a national level C/C++ programming contest |
| 16:12.10 | brlcad | which nation? :) |
| 16:14.47 | *** join/#brlcad dli (~dli@dsl-173-248-203-45.acanac.net) |
| 16:15.04 | *** join/#brlcad adityag1 (~ADITYA@182.237.144.88) |
| 16:16.26 | brlcad | adityag1: connection problems? :) |
| 16:16.40 | brlcad | waves hello to dli |
| 16:17.02 | dli | brlcad, hi |
| 16:18.11 | dli | brlcad, I am trying to become a gentoo developer to update the brlcad package, but they are very slow there |
| 16:18.24 | brlcad | dli: awesome! |
| 16:18.31 | starseeker | dli: my sympathies |
| 16:18.33 | brlcad | I was just about to ask if you were you |
| 16:19.07 | brlcad | a Lin Di just messaged the mailing list, thought they might have used your irc nick :) |
| 16:20.06 | dli | brlcad, couldn't be me :( my family name is Li :( |
| 16:22.09 | dli | brlcad, the gentoo story, andreas Huettel told me he would be my mentor for gentoo-scicence, but I have to wait for him to start the recuiting process |
| 16:22.32 | *** join/#brlcad adityag (~ADITYA@182.237.144.88) |
| 16:22.52 | starseeker | adityag: welcome back :-) |
| 16:24.22 | brlcad | his peers don't like him |
| 16:24.48 | adityag | brlcad starseeker: connection problem + dinner time |
| 16:25.24 | brlcad | dli: that's great news either way, you're well on your way |
| 16:25.37 | *** join/#brlcad AbhijitKane (~Abhijit@111.93.5.194) |
| 16:25.43 | brlcad | adityag: understandable |
| 16:26.17 | brlcad | we're here all day -- we don't expect immediate and perpetual connectivity/responses, neither should you ;) |
| 16:26.57 | adityag | yes cool |
| 16:29.02 | dli | brlcad, for me to test the intersection program (from previous GSoC?), is it possible for me to skip building the whole brlcad package. my gentoo is on a slow core 2 thinkpad, got to speed up the process |
| 16:30.58 | brlcad | dli: you can build by hand, you just cd to the dir and build from there |
| 16:32.18 | brlcad | "make depends" and "make" in src/proc-db should get you built |
| 16:34.49 | dli | brlcad, maybe, I should write some cmake or autotools for the folder first |
| 16:35.45 | starseeker | dli: it shouldn't be necessary |
| 16:36.10 | brlcad | dli: what's the problem? |
| 16:36.24 | brlcad | cmake branch should work as should trunk's autotools build |
| 16:37.09 | brlcad | adityag: so if you didn't see my question, which nation? :) |
| 16:38.06 | dli | brlcad, want to limit the range of code base for me to work on |
| 16:39.56 | *** join/#brlcad piksi (piksi@pi-xi.net) |
| 16:40.40 | starseeker | dli: just cd into the directory with the code in question and type make there - it should only build what is needed |
| 16:40.41 | adityag | brlcad: India.... & u ? |
| 16:41.24 | adityag | brlcad: for more info about me http://aditya.ritzsoftec.com |
| 16:41.57 | dli | starseeker, let me play with it more |
| 16:42.11 | brlcad | adityag: maryland (usa) |
| 16:45.25 | brlcad | adityag: thanks (also good to include in your proposal) |
| 16:45.38 | brlcad | adityag: what platform do you work on? |
| 16:46.42 | adityag | <PROTECTED> |
| 16:48.02 | brlcad | k |
| 16:48.11 | brlcad | what is slmbk? |
| 16:50.01 | adityag | its a social network site which i had lauched in 2008 which revolves around filling scrap/slam books for students |
| 16:54.19 | *** join/#brlcad adityag1 (~ADITYA@182.237.144.88) |
| 16:54.46 | adityag1 | brlcad: i tried working on a few start ups.. i failed every time, so i want to contribute now |
| 16:56.15 | *** join/#brlcad adityag (~ADITYA@182.237.144.88) |
| 16:59.00 | *** join/#brlcad adityag (~ADITYA@182.237.144.88) |
| 17:01.27 | *** join/#brlcad adityag (~ADITYA@182.237.144.88) |
| 17:02.30 | brlcad | adityag: okay thanks! |
| 17:03.44 | AbhijitKane | brlcad: hi again |
| 17:04.09 | AbhijitKane | went thrrough a few material database sites - most of them focus on their querying abilities |
| 17:04.24 | AbhijitKane | ranges for value of various properties and things like that |
| 17:05.23 | AbhijitKane | :adityag: hi! where are you from? |
| 17:06.57 | brlcad | AbhijitKane: great |
| 17:07.13 | *** join/#brlcad adityag (~ADITYA@182.237.144.88) |
| 17:09.06 | brlcad | AbhijitKane: it'd be great to include a rough mock-up of the layout with your application -- wouldnt' have to be anything too complex or fancy but it would give the reviewers a sense of the project goal |
| 17:09.55 | AbhijitKane | ok sure |
| 17:10.21 | AbhijitKane | i don't know if im supposed to ask - but is adityag applying for the same project? |
| 17:10.35 | brlcad | no idea :) |
| 17:10.42 | brlcad | sounded like he was interested in a C project |
| 17:10.49 | AbhijitKane | ohk |
| 17:11.03 | adityag | im interested in any cool projects |
| 17:11.33 | adityag | brlcad : suggest me the coolest idea from brlcad's list |
| 17:12.17 | brlcad | helps to include the '-' when referring to the project and without when referring to me .. otherwise just gets confusing ;) |
| 17:12.36 | brlcad | hm, depends how you defined 'cool' |
| 17:13.24 | brlcad | the first one on the list, NURBS surface-surface intersection calculations is pretty freaking cool, but really frickin' hard too |
| 17:13.43 | adityag | :-D. Whats the best idea about brl-cad ? |
| 17:13.48 | brlcad | otherwise, what's cool to me isn't necessarily cool to you :) |
| 17:13.58 | adityag | let me check out |
| 17:14.39 | brlcad | spend an hour or so and go down the list |
| 17:14.50 | brlcad | if the short summary doesn't catch your attention, move on to the next one |
| 17:15.35 | *** join/#brlcad adityag (~ADITYA@182.237.144.88) |
| 17:16.09 | brlcad | might have missed me saying "spend an hour of so and go down the list, if the short summary doesn't catch your attention, move on to the next one.." |
| 17:16.22 | brlcad | should have a great idea by the end which two or three sound the most interesting |
| 17:16.34 | brlcad | rather you pick something YOU like, than it be something I like |
| 17:17.06 | brlcad | even if it's something simple sounding like cleanup or something complex like nurbs |
| 17:17.25 | adityag | ok cool |
| 17:18.08 | brlcad | we don't really care what you work on -- we want you to become a brl-cad developer merely enjoying what you work on, so maybe you'll keep working on it |
| 17:19.20 | CIA-52 | BRL-CAD: 03starseeker * r43964 10/geomcore/trunk/tests/svntest/main.c: OK, get as far as printing out what's in the toplevel directory... |
| 17:26.29 | *** join/#brlcad adityag1 (~ADITYA@182.237.144.88) |
| 17:37.38 | *** join/#brlcad Ralith (~ralith@d142-058-094-084.wireless.sfu.ca) |
| 17:38.28 | *** join/#brlcad adityag (~ADITYA@182.237.144.88) |
| 17:41.32 | adityag | brlcad : can i get my couple of my friends to work with me on Code Refactoring, GUI, Web Developments ? we can work on others too but im pretty sure that we can work on the mentioned topics. |
| 17:47.42 | *** join/#brlcad Ralith (~ralith@d142-058-094-084.wireless.sfu.ca) |
| 17:58.07 | *** join/#brlcad adityag (~ADITYA@182.237.144.88) |
| 18:31.53 | *** join/#brlcad Ralith (~ralith@d142-058-094-084.wireless.sfu.ca) |
| 18:45.45 | CIA-52 | BRL-CAD: 03erikgreenwald * r43965 10/geomcore/trunk/src/GS/testclient/gstestclient.c: try to send a disconnect request |
| 18:50.37 | CIA-52 | BRL-CAD: 03erikgreenwald * r43966 10/geomcore/trunk/src/GS/testclient/gstestclient.c: glue header magic together. make package display a little more readable |
| 19:03.32 | CIA-52 | BRL-CAD: 03bob1961 * r43967 10/brlcad/trunk/src/libged/ged.c: Initialize libged's _FreeSolid list if not already initialized. |
| 19:12.26 | brlcad | adityag: if they're gsoc students, their work has to be independent of yours, but otherwise they're welcome to apply too |
| 19:13.28 | brlcad | there's no guarantee that any student will be selected really, it really depends how each student is evaluated, how many proposals, what sort of interest there is, etc |
| 19:17.38 | adityag | brlcad : we could apply to 3 different projects as i mentioned you. |
| 19:27.53 | *** join/#brlcad kunigami (~kunigami@201.53.192.197) |
| 19:32.51 | brlcad | you could apply to 3 different projects each -- they'll all still be evaluated independently ;) |
| 19:33.25 | brlcad | do generally recommend that seriously interested students should apply to at least two topics in case there is a conflict that needs resolving |
| 19:35.12 | CIA-52 | BRL-CAD: 03erikgreenwald * r43968 10/geomcore/trunk/src/GS/testclient/gstestclient.c: read node name msg and test magics |
| 19:39.39 | kunigami | hi, I'm interested in the vector output from raytracing: http://brlcad.org/wiki/Vector_output_from_raytracing -- Question: do I have to know beforehand algorithms to fit curves to points or I could learn them during the project? |
| 19:40.16 | brlcad | kunigami: hello! saw your message to the mailing list |
| 19:41.05 | brlcad | you can learn them during the project, but you should have some basic references already researched |
| 19:41.39 | kunigami | Perfect! |
| 19:41.52 | brlcad | two or three research papers that do exactly that, understanding the tradeoffs so that you spend most of your time figuring out how to implement the algorithm |
| 19:42.56 | brlcad | then your proposal should reflect that time with baby milestone steps, more than pure coding projects, so that we can make sure you're making progress |
| 19:43.00 | kunigami | Ok! I'll research some papers on the subject and ask back if they are enough! |
| 19:43.13 | brlcad | like if you find a particular paper, break the algo up into a bunch of steps that could all be testable |
| 19:45.02 | kunigami | hmm ok. splitting the algorithm is also good for debugging purposes :) |
| 19:56.46 | brlcad | kunigami: how'd you come to our project ideas page? what's your interest? |
| 20:06.44 | kunigami | well, I found BRL-CAD through GSoC page - I didn't know BRL-CAD before. I just searched for "computer graphics" keyword. which is also my interest. |
| 20:09.35 | kunigami | I've also interest in computational geometry. Last year I tried CGAL, but I couldn't get in touch with them, so I sent my proposal without their review :( Obvisusly it was rejected. |
| 20:11.53 | *** part/#brlcad adityag (~ADITYA@182.237.144.88) |
| 20:13.25 | kunigami | I've a particular interest in raytracers. I was going to contribute to yafaray, but this year they didn't get accepted to gsoc. glad to see there's another project that involves ray-tracing \o/ |
| 20:18.00 | CIA-52 | BRL-CAD: 03erikgreenwald * r43969 10/geomcore/trunk/include/Portal.h: 5309!=0x5309, change this to match the spec. |
| 20:23.46 | CIA-52 | BRL-CAD: 03starseeker * r43970 10/geomcore/trunk/tests/svntest/main.c: not working, but checkpointing |
| 21:21.18 | CIA-52 | BRL-CAD: 03starseeker * r43971 10/geomcore/trunk/tests/svntest/main.c: Ah, right - path from root, not from model_name |
| 21:46.35 | Ralith | brlcad: would it be worthwhile for me to apply to work on the toolpath generator? I'd be happy to discuss the results of my previous participation, if it'd help. |
| 21:50.33 | CIA-52 | BRL-CAD: 03starseeker * r43972 10/geomcore/trunk/tests/svntest/main.c: Woot! re-assembles the file, althought we're losing some information at the moment. |
| 21:55.00 | starseeker | about one minute fifty seconds with that code to process havoc.g |
| 21:56.14 | starseeker | that's disassembly and insertion, committing, and reconstruction |
| 21:56.50 | starseeker | not handling units correctly yet, or _GLOBAL - probably some other issues |
| 21:58.25 | starseeker | this API communication may be below the "safe for multiple clients" level, but does give a decent baseline. There are no working copies used either for insertion or reassembly. A traditional svn checkout does succeed |
| 22:03.37 | CIA-52 | BRL-CAD: 03starseeker * r43973 10/geomcore/trunk/tests/svntest/main.c: Remove some debug statements, use 1 for unit conversion factors... |
| 22:08.42 | starseeker | humph. OK, not preserving title, units, color tables, or "region_id" attributes - otherwise g_diff reports nothing |
| 22:10.01 | starseeker | that's a good note to call it a week on |
| 22:38.18 | brlcad | starseeker: wow, so 4.5 min reduciton |
| 22:38.21 | brlcad | that's substantial. |
| 22:40.52 | brlcad | pretty awesome progress, so that's apparently as fast as it'll get for the svn server overhead |
| 22:41.43 | brlcad | presumably most of that time is spent inserting and committing |
| 22:44.55 | brlcad | kunigami: that's cool, those are sister projects we don't interact with very often, but are in related domains |
| 22:46.01 | brlcad | cgal mostly just because their license is incompatible |
| 22:46.18 | brlcad | kunigami: did you see our yafaray-related project? |
| 22:49.12 | kunigami | no, I didn't. Is it listed in the project ideas? |
| 22:49.20 | brlcad | not sure.. ;) |
| 22:50.27 | brlcad | there was an idea at one point to hook into yafaray using our raytracer to shoot the ray, find hit point, then pass off to yafray (I need to get used to using their new name...) for optics calculations |
| 22:50.59 | brlcad | basically so we can render our geometry using their global illumination model as a plugin to our tracer |
| 22:51.26 | kunigami | interesting! |
| 22:51.30 | brlcad | yeah very |
| 22:51.33 | brlcad | BUT ...probably wouldn't recommend it for gsoc unless you were already really familiar with |
| 22:51.38 | brlcad | brl-cad or yafray |
| 22:51.55 | brlcad | at the source code level, since it's undoubtedly a tricky integration of data management |
| 22:52.30 | brlcad | though that could be a great out-year project after becoming familiar with the code this summer |
| 22:52.42 | kunigami | hmm ok. I'll consider helping with it afterwards |
| 22:53.17 | brlcad | the first render project would be a good step towards becoming familiarized with the code |
| 22:53.24 | kunigami | yes! I hope to learn a lot about ray-tracers :) |
| 22:54.06 | kunigami | you mean the "shader enhancements" one? |
| 22:54.11 | brlcad | yeah |
| 22:54.26 | kunigami | hmmm, I'll take a look on it too |
| 22:54.29 | brlcad | hooking in something like OSL would cover related code |
| 22:55.27 | kunigami | ok! |
| 22:56.25 | brlcad | kunigami: what projects were you considering before I mentioned that? :) |
| 22:58.34 | kunigami | actually only the "vector output from raytracing", but when I joined the room, I got the end of a conversation about applying to more than one project |
| 22:58.39 | CIA-52 | BRL-CAD: 03Sean 07http://brlcad.org * r2756 10/wiki/Shader_Enhancements: link to main wikipedia article and yafaray |
| 22:58.51 | brlcad | ah excellent |
| 22:58.55 | kunigami | so I'm looking for two projects |
| 22:59.10 | brlcad | yeah, that's perfect |
| 22:59.47 | brlcad | because usually we narrow down to the individual first, then pray we've not narrowed down to two students that have proposed the same project |
| 22:59.59 | brlcad | submitting two makes that never an issue |
| 23:00.35 | kunigami | hmm ok. good to know that |
| 23:01.02 | brlcad | wow, wikipedia page on shaders doesn't mention OSL |
| 23:01.38 | brlcad | they're the hot cool new kid on the block for ray tracing |
| 23:02.40 | kunigami | OSL = open shading language? |
| 23:03.06 | brlcad | yeah |
| 23:03.21 | brlcad | sony imageworks released that project about two or three years ago |
| 23:03.37 | brlcad | it's a shader language specifically tuned to ray tracing |
| 23:03.46 | kunigami | hmm nice! didn't know it |
| 23:04.08 | brlcad | unlike renderman, which was designed for pixar-style micropolygon raster engines |
| 23:05.46 | brlcad | ahh, that's why it's not on our ideas list .. it was a discussion with the yafaray devs |
| 23:06.21 | CIA-52 | BRL-CAD: 03Sean 07http://brlcad.org * r2757 10/wiki/Shader_Enhancements: typo |
| 23:24.05 | starseeker | brlcad: I'll probably have to have a third go at it using the svn_ra layer (svn_fs most likely doesn't have the multi-client safty features) |
| 23:24.55 | starseeker | that requires a little more understanding of things like batons though, so I went for the svn_fs approach as faster (and the same split-and-recombine-without-using-files issues had to be addressed, so that will apply) |
| 23:26.45 | starseeker | I think the majority of the time was committing, but inserting was also significant |
| 23:33.08 | brlcad | ok |
| 23:33.49 | brlcad | locking wouldn't amount to 4 min of activity .. had to be lots of i/o, book-keeping, unnecessary allocations, etc |
| 23:34.02 | brlcad | hard to say, would need to profile |
| 23:34.14 | brlcad | the good news is that the raw server layer is very promising |
| 23:34.48 | brlcad | if we had to, we lock the write session or even lock per node |
| 23:35.05 | brlcad | or profile and optimize their code |
| 23:35.11 | brlcad | either way, very promising |
| 23:35.36 | brlcad | still waiting for that magical "holy crap" moment where it drops to sub 5sec :) |