| 00:18.21 | Notify | 03BRL-CAD:n_reed * 63960 brlcad/trunk/src/libtclcad/tclcad_obj.c: Improve interactive translation in Archer. Reduce accumlation errors in the edit matrix by deriving it from summed deltas instead of multiplying against the old matrix. Also, clear the edit matrix befor redrawing the updated solid to prevent it being momentarily drawn in the wrong position. |
| 00:25.07 | ``Erik | this'll be a fun upgrade O.o java, emacs, mediawiki, ... |
| 00:57.03 | Notify | 02GCI:mrajai2018 * 4639899508539392 : Task Claimed - I would like to work on this task. |
| 00:57.05 | Notify | 02GCI:mrajai2018 * 4639899508539392 : Claim Removed - The claim on this task has been removed, someone else can claim it now. |
| 01:19.57 | *** join/#brlcad darshpreets (~darshpree@202.164.53.117) |
| 03:31.04 | Notify | 02GCI:o7p9bxbnyj * 5250537473703936 : Ready for review - The work on this task is ready to be reviewed. |
| 03:43.29 | Notify | 02GCI:Melange * 4635701832318976 : No more Work can be submitted - Melange has detected that the deadline has passed and no more work can be submitted. The submitted work should be reviewed. |
| 03:55.20 | Notify | 03BRL-CAD:starseeker * 63962 (brlcad/trunk/src/libbrep/shape_recognition.cpp brlcad/trunk/src/libbrep/shape_recognition.h): Still problems with the child recognition logic... |
| 04:48.02 | *** join/#brlcad MarcTannous (4f733b36@gateway/web/cgi-irc/kiwiirc.com/ip.79.115.59.54) |
| 04:48.30 | MarcTannous | brlcad: thanks for all the follow up tasks, but as I said, I need someone to first check https://www.google-melange.com/gci/task/view/google/gci2014/6103408255696896 before I can claim other tasks |
| 04:48.30 | gcibot | [[ Create a web gallery of GCI sticker designs #3 || BRL-CAD || NeedsReview (2 days 8 hrs 51 min) || Ishwerdas, Ch3ck ]] |
| 04:48.34 | *** join/#brlcad dpmobile (~androirc@117.214.54.31) |
| 05:10.15 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 05:13.02 | *** join/#brlcad andromed` (~andromeda@108-225-17-54.lightspeed.sntcca.sbcglobal.net) |
| 05:13.46 | *** join/#brlcad andromed` (~andromeda@108-225-17-54.lightspeed.sntcca.sbcglobal.net) |
| 05:36.10 | *** join/#brlcad dpmobile (~androirc@117.200.83.193) |
| 05:45.09 | Notify | 02GCI:brlcad * 5798062521843712 : Task Closed - Congratulations, this task has been completed successfully. |
| 05:47.05 | *** join/#brlcad andromeda_galaxy (~andromeda@108.225.17.54) |
| 05:48.20 | Notify | 02GCI:brlcad * 4596409726664704 : Deadline extended - The deadline of the task has been extended with 2 days and 2 hours. |
| 06:00.52 | Notify | 02GCI:brlcad * 4596409726664704 : no rar files - Also, please do not submit rar files for your updates. They are not commonplace amongst open source software development for a variety of reasons. A... |
| 06:03.48 | Notify | 02GCI:brlcad * 5340178071683072 : Task Closed - Congratulations, this task has been completed successfully. |
| 06:11.47 | *** join/#brlcad andrei_ (c35a6e7d@gateway/web/freenode/ip.195.90.110.125) |
| 06:19.00 | Notify | 02GCI:brlcad * 5235927756570624 Design a BRL-CAD Sticker #12 https://www.google-melange.com/gci/task/view/google/gci2014/5235927756570624: very disappointing Ekamjot, It is not at all... |
| 06:19.00 | gcibot | [[ Design a BRL-CAD Sticker #12 || BRL-CAD || Closed || Deepak, Harmanpreet ]] |
| 06:22.00 | andrei_ | brlcad: you've found more forges? I seem to be the one closing that task |
| 06:22.02 | andrei_ | Sorry! |
| 06:22.41 | MarcTannous | I posted that one yesterday here in IRC |
| 06:22.46 | MarcTannous | when I was doing my web gallery |
| 06:23.00 | Notify | 02GCI:brlcad * 6143905720434688 Design scale/rotate/translate/shear icons #5 https://www.google-melange.com/gci/task/view/google/gci2014/6143905720434688: do not bother Ekamjot, please do not... |
| 06:23.01 | MarcTannous | not exactly identical, but it's clearly inspired after it, even the same colors |
| 06:23.01 | gcibot | [[ Design scale/rotate/translate/shear icons #5 || BRL-CAD || Closed || Jacob B, Mihai Neacsu ]] |
| 06:23.33 | MarcTannous | well I'm off to classes |
| 06:23.35 | MarcTannous | have a good day |
| 06:24.03 | andrei_ | you too :) |
| 06:24.59 | *** join/#brlcad dpmittal (~androirc@117.214.44.161) |
| 06:31.29 | *** join/#brlcad YashM (~YashM@59.88.183.90) |
| 06:50.05 | Notify | 02GCI:lin_huiqing * 5864524053741568 : Task Claimed - I would like to work on this task. |
| 06:50.45 | Notify | 02GCI:ch3ck * 5864524053741568 : Task Assigned - This task has been assigned to Lin_Huiqing. You have 100 hours to complete this task, good luck! |
| 06:59.04 | Notify | 02GCI:brlcad * 5860136425881600 : Deadline extended - The deadline of the task has been extended with 2 days and 2 hours. |
| 07:00.02 | Notify | 02GCI:brlcad * 5860136425881600 : time? - Jacob, do you need more time to work on this, or other assistance? The file you uploaded seems to be a compressed file containing simply an empty... |
| 07:05.22 | Notify | 02GCI:brlcad * 4635701832318976 : Deadline extended - The deadline of the task has been extended with 1 days and 6 hours. |
| 07:05.37 | Notify | 02GCI:o7p9bxbnyj * 5275647798673408 : Task Claimed - I would like to work on this task. |
| 07:05.38 | Notify | 02GCI:brlcad * 4635701832318976 : Task Needs More Work - One of the mentors has sent this task back for more work. Talk to the mentor(s) assigned to this task to satisfy the requirements needed to... |
| 07:08.54 | *** join/#brlcad raptor (~raptor@194.28.75.245) |
| 07:09.00 | raptor | hello |
| 07:09.18 | andromeda_galaxy | brlcad: what are the next steps for the stencil pattern other than eliminating the backout? Also, did you see my earlier comment about the problems with the bundle per-ray hit functions? |
| 07:10.08 | Notify | 02GCI:brlcad * 5275647798673408 : Task Assigned - This task has been assigned to Andromeda Galaxy. You have 100 hours to complete this task, good luck! |
| 07:10.38 | Notify | 02GCI:brlcad * 6103408255696896 : Task Closed - Congratulations, this task has been completed successfully. |
| 07:12.28 | Notify | 02GCI:tannousmarc * 5844897932247040 : Task Claimed - I would like to work on this task. |
| 07:12.58 | Notify | 02GCI:popescuandrei * 5844897932247040 : Task Assigned - This task has been assigned to Marc Tannous. You have 120 hours to complete this task, good luck! |
| 07:14.04 | Notify | 02GCI:ralpt3748752_qejwhe * 5050736735944704 : Task Claimed - I would like to work on this task. |
| 07:14.21 | andromeda_galaxy | for https://www.google-melange.com/gci/task/view/google/gci2014/5275647798673408, I know that the interface is somewhat bare-bones, but getting all the js & reasonable css for even this took ~5hrs, and styling should be extendable as long as the required information is present (which I believe it is) |
| 07:14.23 | gcibot | [[ PROJECT: Implement web application for testing commit ranges || BRL-CAD || NeedsReview (4 days 3 hrs 55 min) || Mihai Neacsu, Dishank ]] |
| 07:20.50 | Notify | 02GCI:wei1006 * 5869143559503872 : Task Claimed - I would like to work on this task. |
| 07:28.23 | *** join/#brlcad YashM (~YashM@59.88.183.90) |
| 07:37.49 | Notify | 02GCI:vanessalfs * 5859273187786752 : Task Claimed - I would like to work on this task. |
| 07:40.59 | Notify | 02GCI:rossberg * 5050736735944704 : Task Assigned - This task has been assigned to Raptor. You have 100 hours to complete this task, good luck! |
| 07:41.04 | Notify | 02GCI:rossberg * 6113445460049920 : Task Assigned - This task has been assigned to Vladimir Kuznetsov. You have 100 hours to complete this task, good luck! |
| 07:42.10 | andrei_ | hmm, Notify woke up |
| 07:42.18 | Notify | 02GCI:thebrokentree * 5002699137351680 : Task Claimed - I would like to work on this task. |
| 07:42.48 | Notify | 02GCI:popescuandrei * 5002699137351680 : Task Assigned - This task has been assigned to Choo Ming Jing. You have 100 hours to complete this task, good luck! |
| 07:47.08 | Notify | 02GCI:soh_xin_yi * 5806174775541760 : Task Claimed - I would like to work on this task. |
| 07:47.18 | Notify | 02GCI:ralpt3748752_qejwhe * 5050736735944704 : Ready for review - The work on this task is ready to be reviewed. |
| 07:49.29 | Notify | 02GCI:lin_huiqing * 5864524053741568 : Ready for review - The work on this task is ready to be reviewed. |
| 07:51.49 | andromedagalaxy | Andrei_: Notify has been behaving oddly with Melange for the last couple of days... It doesn't send all events, but it does send some, and there isn't a discernable pattern. It might just be Melanges fault, though... |
| 07:52.17 | andrei_ | yeah, at first I thought it failed to update, since I ve assigned some tasks |
| 07:52.27 | andrei_ | but then I saw the latest ones displaying |
| 07:52.45 | andrei_ | I've looked over your task |
| 07:52.46 | *** join/#brlcad bugport (~bugport@91.103.85.100) |
| 07:53.16 | andromedagalaxy | its kind of weird, but last time I noticed it, someone checked and said the mail queue was empty, so it mu |
| 07:53.31 | andromedagalaxy | *is likely Melange |
| 07:53.44 | Notify | 02GCI:ch3ck * 5864524053741568 : Task Needs More Work - One of the mentors has sent this task back for more work. Talk to the mentor(s) assigned to this task to satisfy the requirements needed to... |
| 07:54.04 | andromedagalaxy | anderei_: what did you think? |
| 07:54.42 | andrei_ | to be honest, you did complete what the task said, as there's no requirement of style |
| 07:54.50 | andrei_ | and by that I mean design |
| 07:55.15 | andromedagalaxy | Andrei_: ye |
| 07:55.20 | andrei_ | the only reason I haven't marked it as completed is that given your situation |
| 07:55.29 | andrei_ | brlcad might want you to improve |
| 07:55.36 | andrei_ | since you re not exactly in a rush |
| 07:55.47 | andrei_ | on the other hand |
| 07:55.58 | andrei_ | the follow-up task itself might involve doing the design |
| 07:56.14 | andrei_ | I m more inclined to marking it as complete, I don't think it will be an issue with that. |
| 07:56.39 | Notify | 02GCI:ariessa * 5888591171420160 : Task Claimed - I would like to work on this task. |
| 07:56.46 | andromedagalaxy | I was planning on doing some more design work on it, but getting all the js stuff working for a server integration demo room ~5hrs, so I decided not too... |
| 07:57.14 | Notify | 02GCI:popescuandrei * 5888591171420160 : Task Assigned - This task has been assigned to ariessa. You have 100 hours to complete this task, good luck! |
| 07:57.37 | andrei_ | I assume you've noticed I closed it |
| 07:58.05 | andromedagalaxy | Sorry, I didn't see that yet... |
| 07:58.23 | andrei_ | no worries, I said that because Notify skipped the message |
| 07:58.27 | andromedagalaxy | Just wanted to let everyone know why I didn't do more design... |
| 07:58.57 | andromedagalaxy | Notify is being kind of weird about melange... |
| 07:59.07 | andrei_ | well, you could ask brlcad, if he has a different opinion than mine. As I said, to me it seem reasonably completed |
| 07:59.12 | andrei_ | seems* |
| 08:00.11 | andromedagalaxy | andrei_: sound good... I'm most interested in doing a nice backend server in Haskell with web sockets |
| 08:01.10 | andrei_ | you'll have to speak to brlcad for that, at least to me, Haskell doesn't sound easily maintainable |
| 08:01.54 | andrei_ | mainly because the amount of people that know Haskell is rather limited |
| 08:02.05 | andromedagalaxy | Mentioned it to maths22 the other day, he said that he thought it would be alright... |
| 08:02.25 | andromedagalaxy | (And it does say any language...) |
| 08:02.25 | Notify | 02GCI:wangxiaotong * 4639899508539392 : Task Claimed - I would like to work on this task. |
| 08:02.56 | andrei_ | I wonder what's with the |
| 08:03.04 | *** join/#brlcad lin_huiqing_ (cab0ca8a@gateway/web/freenode/ip.202.176.202.138) |
| 08:03.07 | andrei_ | increasing number of claims these days |
| 08:03.55 | *** join/#brlcad Xin_ (cab0ca89@gateway/web/freenode/ip.202.176.202.137) |
| 08:04.05 | Xin_ | hi |
| 08:04.16 | andrei_ | Hello! |
| 08:05.15 | raptor | hello |
| 08:05.56 | andrei_ | I'm having a hard time figuring out |
| 08:06.11 | lin_huiqing_ | hi |
| 08:06.25 | andrei_ | which "Install and run BRL-CAD!" are genuine since they all use vm |
| 08:08.20 | lin_huiqing_ | https://www.google-melange.com/gci/task/view/google/gci2014/5864524053741568 |
| 08:08.22 | gcibot | [[ Beginner: Install and run BRL-CAD! #58 || BRL-CAD || NeedsWork || Harmanpreet, Ch3ck ]] |
| 08:09.05 | andromedagalaxy | Andrei_: true... By the way, on that task, ca |
| 08:09.33 | andromedagalaxy | Did you have specific better methods of data visualization in mind? |
| 08:09.40 | Notify | 02GCI:lin_huiqing * 5864524053741568 : #brlcad - freenode Web IRC - username: lin_huiqing |
| 08:10.04 | andromedagalaxy | andrei_: Or was it just the general lack of theming? |
| 08:10.18 | andrei_ | data visualisation is fine |
| 08:10.21 | Notify | 02GCI:lin_huiqing * 5864524053741568 : Ready for review - The work on this task is ready to be reviewed. |
| 08:10.26 | andrei_ | I was just reffering at theming |
| 08:10.50 | Ch3ck | lin_huiqing_: alright |
| 08:11.17 | andrei_ | erm, Ch3ck |
| 08:11.22 | andrei_ | you ve asked him to provide a summary |
| 08:11.50 | Ch3ck | andrei_: yeah accoding to the first part of the task |
| 08:11.53 | *** join/#brlcad lin_huiqing (cab0ca89@gateway/web/freenode/ip.202.176.202.137) |
| 08:12.17 | andrei_ | yeah, I just fail to see where he's put it |
| 08:12.17 | Ch3ck | lin_huiqing_: could you comment on the first part of the task on melange? |
| 08:12.17 | Notify | 02GCI:soh_xin_yi * 5806174775541760 : #BRL-CAD - Username: xin_ |
| 08:12.28 | andromedagalaxy | andrei_: okay, that makes sense.. |
| 08:12.36 | Ch3ck | andrei_: me too |
| 08:12.54 | andrei_ | andromedagalaxy: what you did is usable so far, and can be extended. The bottom line is that it's enough for that task :) |
| 08:12.59 | Ch3ck | heads to class, will look at it later, |
| 08:13.09 | andrei_ | what you did so far is usable*, sorry |
| 08:13.26 | Ch3ck | andrei_: could even close the taks if he adds the comments |
| 08:13.31 | andrei_ | I will |
| 08:13.43 | Ch3ck | andrei_: great |
| 08:13.57 | andromedagalaxy | Andrei_: that was my thought as well, I just wanted to make sure it wasn't functionally missing anything |
| 08:14.58 | andrei_ | I've seen that brlcad usually follow-up on tasks that are closed, where he has a different opinion |
| 08:15.33 | lin_huiqing | uh I don't really get what I'm supposed to do :/ |
| 08:15.49 | andrei_ | lin_huiqing, Ch3ck just asked you to explain what you've done |
| 08:15.52 | andrei_ | and how |
| 08:15.52 | *** join/#brlcad xinyi (cab0ca86@gateway/web/freenode/ip.202.176.202.134) |
| 08:15.58 | andrei_ | either here or on melange will do |
| 08:16.40 | andrei_ | like: what steps you took to get that drawing |
| 08:18.19 | *** join/#brlcad lin_huiqing_ (cab0ca86@gateway/web/freenode/ip.202.176.202.134) |
| 08:18.24 | andrei_ | lin_huiqing: there's nothing wrong with what you did |
| 08:18.28 | andrei_ | you just have to answer to this " Once you submit your screenshot, remember to leave a comment with details about which communication checklist item(s) you performed along with your evidence." |
| 08:19.11 | lin_huiqing_ | ok thanks! |
| 08:22.30 | lin_huiqing_ | First, I used gimp to create a png file of my name. Then, I converted the png file to a bw file. mged was then used to extrude the image and then ray trace it. |
| 08:23.05 | andrei_ | lin_huiqing: do you know what's the command for raytrace? |
| 08:23.06 | lin_huiqing_ | something like this? ^ |
| 08:23.08 | andrei_ | yep |
| 08:23.35 | lin_huiqing_ | rt |
| 08:23.39 | andrei_ | yeah |
| 08:24.38 | lin_huiqing_ | am i supposed to like paste my link here again? |
| 08:24.40 | Notify | 02GCI:popescuandrei * 5864524053741568 : Task Closed - Congratulations, this task has been completed successfully. |
| 08:24.47 | andrei_ | nah |
| 08:25.29 | andrei_ | lin_huiqing: I closed it, looks ok :) |
| 08:25.36 | lin_huiqing_ | Thanks! |
| 08:27.14 | *** join/#brlcad wei1006 (cab0ca8a@gateway/web/freenode/ip.202.176.202.138) |
| 08:28.53 | Notify | 02GCI:lim_yong_jun * 5821933278986240 : Task Claimed - I would like to work on this task. |
| 08:29.32 | Notify | 02GCI:wei1006 * 5869143559503872 : None - wei1006 |
| 08:30.15 | xinyi | hello, may I know how I would know if I have completed the task successfully? |
| 08:30.27 | Notify | 02GCI:popescuandrei * 5821933278986240 : Task Assigned - This task has been assigned to Lim Yong Jun. You have 100 hours to complete this task, good luck! |
| 08:30.44 | andrei_ | xinyi: the task would be marked as complete and you would receive an email from melange |
| 08:30.58 | YashM | http://www.google-melange.com/gci/dashboard/google/gci2014 |
| 08:31.02 | YashM | check completed tasks here |
| 08:31.33 | Notify | 02GCI:desiree_lim * 5876507767472128 : Task Claimed - I would like to work on this task. |
| 08:32.42 | andrei_ | how's it going, YashM ? :) |
| 08:32.45 | *** join/#brlcad xinyi_ (cab0ca85@gateway/web/freenode/ip.202.176.202.133) |
| 08:33.32 | YashM | Exams ended today, gonna do those web galleries soon now |
| 08:36.19 | xinyi_ | Hi, I opened BRL-CAD and typed in gimp and then converted the png file into bw, before using mged to extrude the image, and then taking a screenshot of it |
| 08:36.40 | *** join/#brlcad vanessalfs (cab0ca8a@gateway/web/freenode/ip.202.176.202.138) |
| 08:36.51 | vanessalfs | hi |
| 08:37.01 | andrei_ | xinyi: how did you get brlcad? |
| 08:37.04 | andrei_ | hello, vanessalfs |
| 08:37.20 | xinyi_ | https://www.google-melange.com/gci/task/view/google/gci2014/5806174775541760 |
| 08:37.21 | gcibot | [[ Beginner: Install and run BRL-CAD! #46 || BRL-CAD || NeedsReview (4 days 3 hrs 10 min) || Sean, Jacob B ]] |
| 08:37.58 | andrei_ | xinyi_ : I know which task is yours, I've asked you how did you get BRL-CAD? :) |
| 08:38.12 | vanessalfs | i have completed this task: https://www.google-melange.com/gci/task/view/google/gci2014/5002699137351680 |
| 08:38.13 | gcibot | [[ Beginner: Install and run BRL-CAD! #54 || BRL-CAD || Claimed (4 days 3 hrs 4 min) || Mandeep Kaur, Ch3ck ]] |
| 08:38.57 | *** join/#brlcad mingjing123 (cb748202@gateway/web/freenode/ip.203.116.130.2) |
| 08:39.06 | andrei_ | vanessalfs: okay, then you should upload the screenshot, add a comment on melange with how did you get BRL-CAD, and what did you do to obtain the result and mark it as needing review |
| 08:39.17 | wei1006 | I created the black and white image using brlcad. I created the image with my name with gimp. It was a png file, set the size of the image and converted it into bw file. I used the mged and extrude the image. I rotated the image. Lastly, I took a screenshot of it. |
| 08:39.42 | andrei_ | weil1006: I'm confused, which task is yours? |
| 08:39.54 | *** join/#brlcad xinyi (cab0ca85@gateway/web/freenode/ip.202.176.202.133) |
| 08:40.09 | xinyi | I downloaded the programme from http://brlcad.org/wiki/Deuces#Getting_Started |
| 08:40.53 | Notify | 02GCI:soh_xin_yi * 5806174775541760 : None - Hi sorry, my username should be xinyi |
| 08:42.28 | Notify | 02GCI:popescuandrei * 5806174775541760 : Task Closed - Congratulations, this task has been completed successfully. |
| 08:43.22 | mingjing123 | Hi Guys!! Ive finished the https://www.google-melange.com/gci/task/view/google/gci2014/5828709932072960 |
| 08:43.23 | gcibot | [[ Beginner: Install and run BRL-CAD! #45 || BRL-CAD || Open || Deepak, Dishank ]] |
| 08:43.48 | Notify | 02GCI:wei1006 * 5869143559503872 : Ready for review - The work on this task is ready to be reviewed. |
| 08:44.04 | mingjing123 | I was directed to chat here and what am i supposed to do now? |
| 08:44.07 | wei1006 | https://www.google-melange.com/gci/task/view/google/gci2014/5869143559503872 |
| 08:44.08 | gcibot | [[ Beginner: Install and run BRL-CAD! #59 || BRL-CAD || NeedsReview (4 days 2 hrs 38 min) || Harmanpreet, Ishwerdas ]] |
| 08:46.14 | Notify | 02GCI:desiree_lim * 5876507767472128 : Claim Removed - The claim on this task has been removed, someone else can claim it now. |
| 08:47.34 | Notify | 02GCI:desiree_lim * 5828709932072960 : Task Claimed - I would like to work on this task. |
| 08:51.49 | Notify | 02GCI:vs37nx * 5860136425881600 : Ready for review - The work on this task is ready to be reviewed. |
| 08:51.52 | wei1006 | hi I have uploaded the file |
| 08:53.05 | mingjing123 | hi i have uploaded the file too! |
| 08:53.11 | mingjing123 | hi wei1006 :D |
| 08:53.43 | wei1006 | hi :D |
| 08:58.39 | Notify | 02GCI:vanessalfs * 5859273187786752 Beginner: Install and run BRL-CAD! #55 https://www.google-melange.com/gci/task/view/google/gci2014/5859273187786752: None Hi i downloaded the virtual machine... |
| 08:58.40 | gcibot | [[ Beginner: Install and run BRL-CAD! #55 || BRL-CAD || NeedsReview (4 days 2 hrs 41 min) || Popescu Andrei, Dishank ]] |
| 08:58.54 | Notify | 02GCI:vanessalfs * 5859273187786752 : Ready for review - The work on this task is ready to be reviewed. |
| 09:00.15 | YashM | so many people installing BRL-CAD o.o |
| 09:00.20 | Notify | 02GCI:thebrokentree * 5002699137351680 : None - I have completed the task! I downloaded the virtual machine and created an image using GIMP with my ame on it. Then, i used the png-bw tool from the... |
| 09:00.43 | mingjing123 | hahah yeah |
| 09:12.18 | Notify | 02GCI:wangxiaotong * 4639899508539392 : Claim Removed - The claim on this task has been removed, someone else can claim it now. |
| 09:13.25 | Notify | 02GCI:wangxiaotong * 5859130111688704 : Task Claimed - I would like to work on this task. |
| 09:31.41 | *** join/#brlcad xiaotong (cab0ca88@gateway/web/freenode/ip.202.176.202.136) |
| 09:33.04 | Notify | 02GCI:popescuandrei * 5859130111688704 : Task Assigned - This task has been assigned to Wang Xiaotong. You have 100 hours to complete this task, good luck! |
| 09:33.14 | Notify | 02GCI:popescuandrei * 5831779759947776 : Task Assigned - This task has been assigned to Etzernal. You have 100 hours to complete this task, good luck! |
| 09:35.20 | xiaotong | hi, ive completed a task BEGINNER: INSTALL AND RUN BRL-CAD! #49 but im not sure what to do in part 1 |
| 09:40.16 | Notify | 02GCI:wangxiaotong * 5859130111688704 : None - i typed out my name in black and white and extruded it in mged. |
| 09:40.30 | *** join/#brlcad lin_huiqing (cab0ca85@gateway/web/freenode/ip.202.176.202.133) |
| 09:48.01 | Notify | 02GCI:utkarshsingh341 * 5020355882123264 : Ready for review - The work on this task is ready to be reviewed. |
| 09:56.53 | Notify | 02GCI:wei1006 * 5869143559503872 : None - I created the black and white image using brlcad. I created the image with my name with gimp. It was a png file, set the size of the image and converted it... |
| 09:58.34 | *** join/#brlcad mini_monstar (cab0ca85@gateway/web/freenode/ip.202.176.202.133) |
| 09:58.40 | mini_monstar | hello |
| 09:58.44 | mini_monstar | help |
| 09:58.54 | *** join/#brlcad merzo (~merzo@user-94-45-58-141.skif.com.ua) |
| 10:02.28 | *** join/#brlcad Pulse (cab0ca88@gateway/web/freenode/ip.202.176.202.136) |
| 10:02.31 | *** join/#brlcad xinjie (cab0ca87@gateway/web/freenode/ip.202.176.202.135) |
| 10:02.43 | *** part/#brlcad mini_monstar (cab0ca85@gateway/web/freenode/ip.202.176.202.133) |
| 10:04.31 | *** join/#brlcad mini_monstar (cb748202@gateway/web/freenode/ip.203.116.130.2) |
| 10:04.38 | mini_monstar | hello |
| 10:07.11 | andrei_ | is scared |
| 10:07.15 | andrei_ | Hello! |
| 10:07.54 | *** join/#brlcad Yong__ (cab0ca86@gateway/web/freenode/ip.202.176.202.134) |
| 10:08.15 | Yong__ | hi |
| 10:08.56 | mini_monstar | WASSUP |
| 10:10.04 | Yong__ | i have submitted the work |
| 10:10.08 | Yong__ | pls review. Thanks. |
| 10:10.10 | Notify | 02GCI:lim_yong_jun * 5821933278986240 : Ready for review - The work on this task is ready to be reviewed. |
| 10:11.05 | Notify | 02GCI:mini_monstar * 5831970483339264 : Task Claimed - I would like to work on this task. |
| 10:11.10 | Notify | 02GCI:pulse * 5886669710426112 : Task Claimed - I would like to work on this task. |
| 10:11.55 | Notify | 02GCI:popescuandrei * 5886669710426112 : Task Assigned - This task has been assigned to Pulse. You have 100 hours to complete this task, good luck! |
| 10:12.05 | Notify | 02GCI:xinjie * 5895161666076672 : Task Claimed - I would like to work on this task. |
| 10:12.07 | mini_monstar | I created a png of my name. I converted the png file into a bw file. Then I extruded it and screenshot it. |
| 10:12.10 | Notify | 02GCI:popescuandrei * 5831970483339264 : Task Assigned - This task has been assigned to Jovan Teo. You have 100 hours to complete this task, good luck! |
| 10:12.55 | Notify | 02GCI:popescuandrei * 5859273187786752 : Task Closed - Congratulations, this task has been completed successfully. |
| 10:13.02 | andrei_ | you're all from the same class, or school, aren't you? :) |
| 10:13.12 | mini_monstar | nope <3 |
| 10:13.23 | mini_monstar | i did it myself |
| 10:13.26 | mini_monstar | at home |
| 10:13.28 | mini_monstar | :( |
| 10:13.35 | mini_monstar | why dont you trust me |
| 10:13.40 | mini_monstar | :'( |
| 10:14.38 | andrei_ | I never said you didn't do it yourself. It's just there have been 10 claims in a couple of hours |
| 10:14.39 | mini_monstar | no im joking |
| 10:14.58 | mini_monstar | yea we from the same school |
| 10:15.10 | mini_monstar | we are very interested in learning this task |
| 10:15.19 | mini_monstar | WE ARE INTERESTED IN COMNPUTING |
| 10:15.22 | Pulse | Excuse my friend for his rude behaviour |
| 10:15.24 | mini_monstar | I LOVE COMPUTING |
| 10:16.27 | andrei_ | MarcTannous: isn't this design similar to yours? https://www.google-melange.com/gci/task/view/google/gci2014/5020355882123264 |
| 10:16.29 | gcibot | [[ Create a book layout for Doc Camp Book #6 || BRL-CAD || NeedsReview (8 hrs 26 min) || Kesha Shah, Sean ]] |
| 10:20.02 | *** join/#brlcad xinjie (cab0ca88@gateway/web/freenode/ip.202.176.202.136) |
| 10:21.19 | Pulse | I used the easy method in order to obtain your software, by downloading from the Virtual Machine http://brlcad.org/wiki/Deuces#Getting_Started. Installing debian, I was able to perform the following task. Downloading the Virtual Machine using GIMP, I used a 512 by 512 image with a black background, which I typed my name with it, saving it as a png file. After converting it into a bw image by running the png-bw command, I extruded the |
| 10:21.22 | xinjie | I made an image with a black background, adding my name in white text using GIMP. After that, I used mged to render the image and took a screenshot under the applications and accessories window. |
| 10:22.01 | Notify | 02GCI:mini_monstar * 5831970483339264 : None - I wrote on the irc chat. I am mini_monstar |
| 10:22.12 | Notify | 02GCI:tannousmarc * 5844897932247040 : Ready for review - The work on this task is ready to be reviewed. |
| 10:22.15 | mini_monstar | I created a png of my name. I converted the png file into a bw file. Then I extruded it and screenshot it. |
| 10:22.22 | Notify | 02GCI:pulse * 5886669710426112 : Upon Completion - I wrote on the IRC chat, my user name is Pulse. |
| 10:22.27 | Notify | 02GCI:xinjie * 5895161666076672 : Done!!!! :) - I wrote in the irc chat. My username is xinjie. |
| 10:22.31 | *** join/#brlcad MarcSchool (59eef606@gateway/web/cgi-irc/kiwiirc.com/ip.89.238.246.6) |
| 10:24.04 | mini_monstar | Yea the computer club in my school introduced us to this program if you're wondering. |
| 10:28.53 | Notify | 02GCI:pulse * 5886669710426112 : Ready for review - The work on this task is ready to be reviewed. |
| 10:29.04 | MarcSchool | mentors, this layout is very very similar to mine : https://www.google-melange.com/gci/task/view/google/gci2014/5020355882123264 |
| 10:29.05 | gcibot | [[ Create a book layout for Doc Camp Book #6 || BRL-CAD || NeedsReview (8 hrs 13 min) || Kesha Shah, Sean ]] |
| 10:29.27 | MarcSchool | Mine, submitted 1 and a half months ago : https://www.google-melange.com/gci/task/view/google/gci2014/5591216443883520 |
| 10:29.28 | gcibot | [[ Create a book layout for Doc Camp Book || BRL-CAD || Closed || Deepak, Harmanpreet ]] |
| 10:29.57 | MarcSchool | Same pages ( that I came up with ), same layout, same footer, even the same image I inserted |
| 10:30.06 | andrei_ | as I said, it seems similar to me as well, but I need a second or third opinion |
| 10:30.11 | andrei_ | before marking it as stolen |
| 10:30.12 | andrei_ | :) |
| 10:30.31 | andrei_ | but we probably will |
| 10:30.52 | MarcSchool | Don't know, loads of stuff Works are "inspired" from others |
| 10:30.55 | MarcSchool | as in heavily inspired |
| 10:31.04 | andrei_ | what do you mean? |
| 10:31.42 | Notify | 02GCI:tannousmarc * 5778240039813120 : Task Claimed - I would like to work on this task. |
| 10:32.10 | MarcSchool | Just an innocent example would be the fact that Sean made me come up with a byline for Archer in my very first task, on the 1st of December, and loads of designs I've seen use the same byline, "Think.Invent.Create" |
| 10:32.17 | Notify | 02GCI:maitreyav * 4635701832318976 : ZIP file - Hi Sean, I've re-uploaded my work as zip file, please look it so that I can make changes to it. |
| 10:32.23 | Notify | 02GCI:popescuandrei * 5778240039813120 : Task Assigned - This task has been assigned to Marc Tannous. You have 120 hours to complete this task, good luck! |
| 10:33.18 | Notify | 02GCI:tannousmarc * 5778240039813120 : Ready for review - The work on this task is ready to be reviewed. |
| 10:33.48 | Notify | 02GCI:wangxiaotong * 5859130111688704 : Ready for review - The work on this task is ready to be reviewed. |
| 10:34.36 | *** join/#brlcad xiaotong (cab0ca89@gateway/web/freenode/ip.202.176.202.137) |
| 10:36.45 | *** join/#brlcad merzo (~merzo@user-94-45-58-141.skif.com.ua) |
| 10:40.07 | *** join/#brlcad YashM (~YashM@59.88.183.90) |
| 10:40.41 | *** join/#brlcad Pulse (cab0ca8a@gateway/web/freenode/ip.202.176.202.138) |
| 10:57.42 | *** join/#brlcad FreezingCold (~FreezingC@135.0.41.14) |
| 11:28.27 | *** join/#brlcad YashM (~YashM@59.88.183.90) |
| 11:38.26 | Notify | 02GCI:zynatrix * 5571450517848064 : Task Claimed - I would like to work on this task. |
| 11:46.34 | *** join/#brlcad YashM (~YashM@59.88.183.90) |
| 11:47.04 | *** join/#brlcad merzo (~merzo@user-94-45-58-141.skif.com.ua) |
| 11:49.30 | Notify | 02GCI:mihaineacsu29 * 5895161666076672 : Task Assigned - This task has been assigned to Xinjie. You have 99 hours to complete this task, good luck! |
| 12:14.36 | gcibot | ning. [before amsg..] |
| 12:15.13 | ignacio | Good morning |
| 12:16.37 | *** join/#brlcad YashM (~YashM@59.88.183.90) |
| 12:19.58 | Notify | 02GCI:shardulc * 6455158829481984 : Task Claimed - I would like to work on this task. |
| 12:29.51 | Notify | 02GCI:rossberg * 5020355882123264 : Task Closed - Congratulations, this task has been completed successfully. |
| 12:32.41 | *** join/#brlcad d_rossberg (~rossberg@66-118-151-70.static.sagonet.net) |
| 12:33.39 | *** part/#brlcad andrei_ (c35a6e7d@gateway/web/freenode/ip.195.90.110.125) |
| 12:33.44 | *** join/#brlcad andrei_ (c35a6e7d@gateway/web/freenode/ip.195.90.110.125) |
| 12:45.52 | *** join/#brlcad ries (~ries@D979C47E.cm-3-2d.dynamic.ziggo.nl) |
| 13:20.49 | *** join/#brlcad FreezingCold (~FreezingC@135.0.41.14) |
| 13:46.15 | Notify | 02GCI:etzernal * 5831779759947776 : Ready for review - The work on this task is ready to be reviewed. |
| 13:48.45 | raptor | in works https://www.google-melange.com/gci/task/view/google/gci2014/5264182400253952 |
| 13:48.46 | gcibot | [[ Categorize all of BRL-CAD's commands into a spreadsheet || BRL-CAD || Reopened || Popescu Andrei, Hardeep Singh Rai ]] |
| 13:49.47 | raptor | All utilities on each accounted values: |
| 13:49.55 | raptor | ['NAME', 'SYNOPSIS', 'DESCRIPTION', 'OPTIONS', 'FORMAT', 'TODO', 'EXAMPLE', 'DIAGNOSTICS', 'AUTHOR', 'COPYRIGHT', 'REPORTS', 'EXAMPLES', 'ALSO', 'BUGS', 'METHOD', 'EOF', 'PROCESSING', 'TABLES', 'MISC', 'PLY', 'WARNING', 'B', 'C', 'E', 'F', 'J', 'P', 'Q', 'R', 'Z', 'HINTS', 'UNIX', 'MDQS', 'NOTES', 'N', 'T', 'RETURN', 'DISCUSSION', 'REPORTING', 'AUTHORS', 'FONTS', 'PATH', 'DIR', 'DESCRIPTIONS', 'DEFINITIONS', 'CLASS', 'VARIABLES', 'FILES', 'PROMPTS', 'KEYWOR |
| 13:49.55 | raptor | DS', 'SKIP', 'VALUES', 'SIGNALS', 'SCRIPTS', 'TIPS', 'HINT', 'DEFICIENCIES', 'TRADEMARKS', 'A', 'RLE', 'MGED', 'VTK', 'S', 'PRISM', 'END', 'CAVEAT', 'AXIS', 'CREATED', 'VRML', 'REQUIREMENTS', 'HISTORY', 'DISCLAIMER', 'RESTRICTIONS', 'VALUE', 'SEEDVAL', 'GVALUE', 'WIDTH', 'HEIGHT', 'CAVEATS', 'CHANNELS', 'OUTPUT', 'NOTE', 'ASCII', 'RGB', 'THEORY', 'LIMITATIONS', 'L', 'CHECKING', 'SPLINE', 'INSPIRATIONS', 'FLOAT', 'INT', 'STR', 'TEMPERATURES', 'RESULTS'] |
| 13:50.39 | raptor | I think it's a lot to create the table |
| 13:51.01 | raptor | What I take from them |
| 13:51.03 | raptor | ? |
| 14:15.11 | Notify | 02GCI:izak * 5831779759947776 : Task Closed - Congratulations, this task has been completed successfully. |
| 14:43.02 | Notify | 02GCI:deepakkumarsharma * 5820470842621952 : Task Assigned - This task has been assigned to utkarshsingh. You have 100 hours to complete this task, good luck! |
| 15:02.39 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 15:12.13 | YashM | https://www.google-melange.com/gci/task/view/google/gci2014/5843395230564352 |
| 15:12.14 | YashM | Are these repeats? |
| 15:12.14 | gcibot | [[ Create a gallery of all 3D models #3 || BRL-CAD || Open || Sean ]] |
| 15:23.45 | Notify | 02GCI:xinjie * 5895161666076672 : Ready for review - The work on this task is ready to be reviewed. |
| 15:48.42 | Notify | 03BRL-CAD:starseeker * 63964 (brlcad/trunk/src/libbrep/shape_recognition.cpp brlcad/trunk/src/libbrep/shape_recognition.h): Still got some memory issue somewhere - can't turn on the cylindrical_loop_planar_vertices test without hanging. |
| 15:57.11 | Notify | 03BRL-CAD:starseeker * 63965 brlcad/trunk/src/libbrep/shape_recognition.cpp: forgot to free local filter copy. |
| 16:02.04 | *** join/#brlcad shardulc (~shardul@122.169.69.66) |
| 16:02.51 | Notify | 02GCI:shardulc * 6455158829481984 : Ready for review - The work on this task is ready to be reviewed. |
| 16:06.03 | shardulc | brlcad: are you online? |
| 16:16.21 | *** part/#brlcad shardulc (~shardul@122.169.69.66) |
| 16:21.14 | gcibot | 4 days left... What are your feelings :-)? |
| 16:26.16 | Notify | 03BRL-CAD:starseeker * 63966 (brlcad/trunk/src/libbrep/shape_recognition.cpp brlcad/trunk/src/libbrep/shape_recognition.h): Surface tests are always assumed to be destructive, so we need a new duplicate after each test. |
| 16:32.37 | unicodesnowman | andromeda-galaxy, andromeda_galaxy, andromedagalaxy, i didn't know there are 3 galaxies of you xD |
| 16:33.00 | andromeda_galaxy | unicodesnowman: sorry, I'm in the process of moving between irc clients... |
| 16:53.44 | *** join/#brlcad andrei_ (bc1b5712@gateway/web/freenode/ip.188.27.87.18) |
| 17:08.56 | *** join/#brlcad gaganjyot (~gaganjyot@118.102.186.30) |
| 17:18.33 | *** join/#brlcad balder_ (7d1409f9@gateway/web/freenode/ip.125.20.9.249) |
| 17:20.31 | balder_ | Hey! I have installed brlcad in ubuntu but when I type mged in terminal the graphics window does not open! Can any one say me the solution! |
| 17:20.49 | andromeda_galaxy | balder_: does it print an error message? |
| 17:21.36 | balder_ | Nope! In command line it shows this this "attach (nu|txt)[nu]? " |
| 17:22.38 | andromeda_galaxy | balder_: try 'mged -a X' or 'mged -a ogl' |
| 17:22.47 | balder_ | ok! |
| 17:23.46 | andromeda_galaxy | balder_: it's kind of odd, it seems like mged is being run in classic mode... how did you install brl-cad? |
| 17:25.33 | balder_ | Well from this link http://brlcad.org/wiki/Building_from_SVN |
| 17:26.24 | andromeda_galaxy | what kind of system are you on? |
| 17:26.45 | Notify | 02GCI:shivam691999 * 4596409726664704 : Ready for review - The work on this task is ready to be reviewed. |
| 17:26.55 | balder_ | This is ubuntu 14.04 running on intet i5 |
| 17:28.18 | andromeda_galaxy | you're just running 'mged' in a terminal? let me check something in the source quickly, it looks like its not detecting the x dm |
| 17:29.24 | andromeda_galaxy | balder_: can you put the output of the cmake configuration command you did at the beginning in a pastebin? |
| 17:30.01 | balder_ | andromeda_galaxy: yes 1sec |
| 17:33.48 | andromeda_galaxy | balder_: it seems like it might have been a misconfiguration error (build without X11/OpenGL support), so I needed to see the logs... |
| 17:34.19 | andromeda_galaxy | *output |
| 17:34.56 | Notify | 03BRL-CAD:starseeker * 63967 brlcad/trunk/src/libbrep/shape_recognition.cpp: Don't grab edges from planar faces when doing the secondary break-out |
| 17:36.12 | *** join/#brlcad raptor_ (~raptor@194.28.75.245) |
| 17:40.36 | brlcad | MarcTannous: I knew you had a task pending, it was just to let you know that there are more icon tasks (so you're aware they exist for *whenever* you're able/ready) :) |
| 17:41.01 | brlcad | your icons and yash's I believe were two of the more usable ones iirc |
| 17:41.11 | brlcad | and one other... escaping me |
| 17:42.21 | gaganjyot | brlcad, hi :) |
| 17:42.44 | gaganjyot | brlcad, had a question related to binary, thought I could get anwer here |
| 17:42.56 | gaganjyot | I want to read 2 bits from a byte |
| 17:42.56 | brlcad | andrei_: no worries on closing the forgeries ... it's impossible for one mentor to know about all the other activity -- next year we'll need to have a gallery in place right from the get-go for design tasks so it's easy to check and students have to put their name up on a board where forgeries would be obvious and embarrasing to them |
| 17:43.16 | gaganjyot | and the position of pointer in byte is x |
| 17:43.22 | gaganjyot | how can I get those 2 bytes |
| 17:43.27 | gaganjyot | 2 bits* |
| 17:43.49 | balder_ | andromeda_galaxy: I hope this is what you want to see http://pastebin.com/KeckYkK7 . And after creating a new database called and creating a sph1.s and rightclicking and opening it with mged(as there it was in options) it works! |
| 17:44.15 | brlcad | andromeda_galaxy: next step -- visualize a few examples... like a tgc at an angle, a box, and something beefier, maybe m35 |
| 17:44.33 | andromeda_galaxy | brlcad: that makes sense, though it would need some more changes to rtshot |
| 17:44.47 | andromeda_galaxy | (I actually did some of them as part of the last one to make sure that everything worked, but it's very rudimentary) |
| 17:45.21 | brlcad | and I did see your comment about having an option, but don't think it's necessary (we're just saving them a little math) |
| 17:45.43 | andromeda_galaxy | balder_: try building with 'cmake -DBRLCAD_ENABLE_X11=ON -DBRLCAD_ENABLE_OPENGL=ON |
| 17:45.46 | andromeda_galaxy | ' |
| 17:45.59 | andromeda_galaxy | brlcad: sounds good, I can remove that option |
| 17:46.30 | andromeda_galaxy | brlcad: did you also see my point about the way that rt_shootrays() callbacks have to re-implement significant code that could be the same for most of them? |
| 17:47.12 | andromeda_galaxy | also, I was thinking about changing the rtshot bundle parameter interface to, e.g., -B <bundle_type> <bundle_opts>, it would make it easier to not have to find letters that fit for some of the options (though some, like -p, could be preserved) |
| 17:48.12 | brlcad | andrei_: we're quickly approaching the deadline, that's why there are increasing claims -- always a big rush to the finish |
| 17:48.40 | andrei_ | brlcad: yeah, but isn't it like .. too late? they can finish at most one or two tasks |
| 17:50.34 | andromeda_galaxy | brlcad: do you want to make a task to do some of those visualizations for the stencil bundle? |
| 17:50.58 | andromeda_galaxy | (and maybe work on improving the rtshot option structure at the same time) |
| 18:00.10 | brlcad | balder_: that attach (nu|txt) prompt means you compiled brl-cad without support for X11 |
| 18:00.47 | brlcad | you need to make sure you have the X11 developer headers installed (see several of the dependencies on the Building_from_SVN page) |
| 18:00.54 | balder_ | andromeda_galaxy: ok! Now I did build it with cmake -DBRLCAD_ENABLE_X11=ON -DBRLCAD_ENABLE_OPENGL=ON. And it works thank You |
| 18:01.15 | andromeda_galaxy | balder_: no problem, glad it helped |
| 18:02.44 | brlcad | gaganjyot: you can shift the byte by x, and bitwise & the value(s) with 1 |
| 18:03.56 | brlcad | or just & the byte with a bitmask and you'll have four possible results (values) |
| 18:04.45 | brlcad | andromeda_galaxy: I did not see your point about callbacks |
| 18:05.02 | andromeda_galaxy | brlcad: for rt_shootrays(), it uses both b_hit/miss and a_hit/miss |
| 18:05.04 | andromeda_galaxy | functions |
| 18:05.19 | andromeda_galaxy | one of a_hit/a_miss is called for each ray, b_hit/b_miss for the whole bundle, right? |
| 18:05.39 | andromeda_galaxy | sometimes (like in my patch), it makes sense to get information for each individual ray, instead of just using the whole list of partitions |
| 18:05.40 | brlcad | I haven't looked, but sounds likely |
| 18:06.03 | brlcad | so that'd be an a_ case |
| 18:06.16 | andromeda_galaxy | rt_shootrays() requires you to define a whole bunch of boilerplate in a_hit that moves the partition/seg lists into the bundle struct for later use |
| 18:06.34 | andromeda_galaxy | the default handler function is declared static, so right now everyone who needs to use a_hit with rt_shootrays() |
| 18:06.49 | andromeda_galaxy | has to copy a whole bunch of boilerplate code that detaches the segs/partition lists and reattaches them to the bundle |
| 18:07.29 | andromeda_galaxy | brlcad: I was thinking that it would be nice if the default functions were exposed, because then people who had to redefine a_hit could just start it with a call to e.g, rt_shootrays_default_hit() |
| 18:07.29 | brlcad | so what are you wanting to do about that? |
| 18:07.40 | brlcad | hm |
| 18:08.03 | andromeda_galaxy | also, the a_uptr that the user originally passed in the struct application is somewhat hidden from a_hit, you have to do partition_bundle->ap->a_uptr, instead of ap->a_uptr |
| 18:08.23 | andromeda_galaxy | so having a utility function/macro to get the *original* application struct, might be nice |
| 18:09.18 | brlcad | that last bit won't pay for itself in terms of added API, so probably a no-go on having a nicity helper for that |
| 18:09.39 | andromeda_galaxy | brlcad: sure, it would be nice to document it somewher, though |
| 18:09.44 | brlcad | absolutely |
| 18:09.57 | andromeda_galaxy | I had to figure it out from the source for rt_shootrays() |
| 18:10.30 | brlcad | exposing a default hit function isn't out of the norm, but just have to make sure it is just some generic portion that can be well-documented |
| 18:11.04 | brlcad | e.g., if there's some portion that someone would always need to undo, then you'd probably still have a static default that calls this public default and then does a little bit more |
| 18:11.32 | andromeda_galaxy | brlcad: so currently, the default function does two things: |
| 18:11.32 | brlcad | but then new callers would have their "super" class function they could call |
| 18:11.41 | andromeda_galaxy | increments 'hit' |
| 18:12.08 | andromeda_galaxy | and moves the partitions/segs from the partition/seg lists passed in to those in the bundle |
| 18:12.09 | brlcad | anyone have a screenshot of the two copied book layouts? |
| 18:12.27 | brlcad | ( MarcTannous? ) |
| 18:12.32 | andromeda_galaxy | (well, right now it also checks that it was actually a hit, and increments misses if it wasn't, by checking the length of partitions) |
| 18:13.00 | andromeda_galaxy | I think that those things are all pretty sane default things to do, do you see any reason to not use all of them? |
| 18:13.07 | andrei_ | brlcad: wait a sec |
| 18:14.06 | andromeda_galaxy | brlcad: what should we call the helper? |
| 18:14.24 | brlcad | andromeda_galaxy: no, but then I haven't really put any serious design thought into the question either and that is usually warranted when introducing new API |
| 18:14.38 | brlcad | andromeda_galaxy: look at the default overlap handler, see what it is named |
| 18:14.44 | andromeda_galaxy | brlcad: true... well, something to think about, at least... |
| 18:14.51 | andrei_ | brlcad : http://www.google-melange.com/gci/task/view/google/gci2014/5020355882123264 and http://www.google-melange.com/gci/task/view/google/gci2014/5591216443883520 |
| 18:14.53 | gcibot | [[ Create a book layout for Doc Camp Book #6 || BRL-CAD || Closed || Kesha Shah, Sean ]] |
| 18:14.54 | gcibot | [[ Create a book layout for Doc Camp Book || BRL-CAD || Closed || Deepak, Harmanpreet ]] |
| 18:15.10 | andrei_ | they're not 1:1, but Marc mentioned that the majority of elements belonged to his design |
| 18:15.13 | brlcad | andromeda_galaxy: I'd say go ahead and do it if you like (are there still new API/refactoring tasks?) |
| 18:15.18 | andromeda_galaxy | rt_default_multioverlap()? |
| 18:15.18 | Notify | 03BRL-CAD:ejno * 63968 (brlcad/trunk/src/libged/simulate/collision.cpp brlcad/trunk/src/libged/simulate/collision.hpp and 4 others): fix several bugs with the collision algorithm; don't store velocity as attributes (will be replaced by storing Bullet's own serialization into a binary database object) |
| 18:15.27 | andromeda_galaxy | brlcad: yeah, there's a design an api to reduce code duplication task |
| 18:15.28 | brlcad | sounds about right |
| 18:15.40 | brlcad | so yeah, use one of those |
| 18:15.50 | brlcad | technically reducing potential code duplicaiton |
| 18:16.04 | andromeda_galaxy | I'll claim it and then look into the api a little bit... I'll probably include the implementation with the design, since it's quite simple |
| 18:16.20 | brlcad | the work will be on the /** */ comment in the raytrace.h header, to properly document what this default function is doing |
| 18:16.48 | Notify | 02GCI:o7p9bxbnyj * 5239535713648640 : Task Claimed - I would like to work on this task. |
| 18:16.55 | brlcad | so follow the same naming pattern, maybe rt_default_bundlehit() |
| 18:16.58 | andromeda_galaxy | brlcad: indeed |
| 18:17.07 | andromeda_galaxy | I was thinking of either rt_default_bundlehit or rt_bundlehit_helper() |
| 18:17.14 | andromeda_galaxy | the only advantage of the second one |
| 18:17.31 | andromeda_galaxy | is that it still is appropriate if the default handler does end up doing more than the helper does |
| 18:18.06 | andromeda_galaxy | brlcad: (i.e., if it seems useful to add something to the default handler that most overriding functions would have to undo) |
| 18:19.10 | brlcad | I don't think that's a great name, but I do think it needs to mirror the other handler (and both should later get renamed) just so it's clear they're the same "type" of function |
| 18:19.39 | andromeda_galaxy | so rt_default_bundlehit() is better, even if it eventually becomes not-actually-the-defualt? |
| 18:19.48 | Notify | 02GCI:deepakkumarsharma * 5239535713648640 : Task Assigned - This task has been assigned to Andromeda Galaxy. You have 100 hours to complete this task, good luck! |
| 18:19.49 | brlcad | could name it something else, but then a review would be warranted to deprecate the old handlers to new names |
| 18:20.00 | brlcad | and that is a process |
| 18:21.27 | brlcad | and no it's not better and it would almost preclude letting it do something that callers would always have to undo |
| 18:21.41 | brlcad | not better and better simultaneously ;) |
| 18:22.02 | andromeda_galaxy | brlcad: interesting... I'm also not sure which to use... if they're both non-optimal, what do you think is best? |
| 18:22.12 | brlcad | the real answer isn't possible without itemizing all librt functions and creating a mapping of old names to new names |
| 18:22.49 | andromeda_galaxy | brlcad: that sounds like quite a large project... which one do you think I should use in the interim? |
| 18:22.49 | brlcad | as that's not something that's going to happen today, it should match the existing convention |
| 18:23.01 | andromeda_galaxy | ah, okaay --- so rt_default_bundlehit()? |
| 18:23.07 | brlcad | yep |
| 18:23.25 | brlcad | not a large project, but it would take several days |
| 18:24.03 | andromeda_galaxy | brlcad: sounds good! also, the current default _miss routine increments the bundles miss count |
| 18:24.16 | andromeda_galaxy | should I make a 'default' versionf or that one as well, since the default hit routine does increment the hit count |
| 18:24.19 | andromeda_galaxy | ? |
| 18:29.36 | brlcad | is it really just one line? |
| 18:29.52 | andromeda_galaxy | brlcad: pretty much |
| 18:30.01 | brlcad | pretty much means what? |
| 18:30.04 | MarcTannous | brlcad: did you see my discussion with andrei regarding "forgery"? |
| 18:30.07 | andromeda_galaxy | just one line with a variable declaration to get it |
| 18:30.09 | andromeda_galaxy | *get the bundle |
| 18:30.16 | andromeda_galaxy | and one to increment bundle->miss() |
| 18:30.18 | brlcad | MarcTannous: did you see the responses? |
| 18:30.20 | andromeda_galaxy | *bundle->miss |
| 18:30.21 | andromeda_galaxy | anad one to return 0 |
| 18:30.48 | brlcad | andromeda_galaxy: so one line of actual logic (and that could be reduced to a one-line function if we wanted to |
| 18:31.09 | andromeda_galaxy | brlcad: yes, at least for now... |
| 18:31.33 | andromeda_galaxy | on the other hand adding api is a lot easier than removing it, so it might make sense to err on the side of not having it if it doesn't seem necessary |
| 18:31.34 | MarcTannous | brlcad: no, that's why I am asking.. |
| 18:31.38 | brlcad | andromeda_galaxy: so no, not for one line -- that can be part of the comment that they can/should increment bundle->miss if they need miss tracking. |
| 18:31.41 | MarcTannous | my computer auto-sleeps and it disconnects from IRC |
| 18:32.02 | brlcad | MarcTannous: yeah, book design, but I haven't seen it yet |
| 18:32.15 | andromeda_galaxy | brlcad: sounds good, I'll mention that in the comment for rt_default_bundlehit |
| 18:32.24 | brlcad | can you make a quick side-by-side comparison (or is andrei_ working on one?) |
| 18:32.39 | andrei_ | no, I'm not, I also asked Marc |
| 18:32.42 | andrei_ | and he detailed |
| 18:32.48 | andrei_ | what's "too" similar |
| 18:33.01 | andrei_ | I suppose he can write on IRC |
| 18:33.18 | brlcad | right, but that's why I need a side-by-side (not at a machine with photoshop at the moment) |
| 18:33.53 | andrei_ | aah |
| 18:34.01 | gaganjyot | thanks brlcad :) |
| 18:34.03 | MarcTannous | as far as I remember, exactly the same pages ( suspicious, as these were not mentioned on the brief ) , exactly the same layout, very simialr footer, and exactly the same image inserted, that I took as a mock-up image |
| 18:34.04 | gaganjyot | worked :) |
| 18:34.12 | MarcTannous | will make a side-by-side.. |
| 18:34.32 | andrei_ | MarcTannous: awesome, thanks ! :) |
| 18:34.36 | brlcad | MarcTannous: there are big swaths of gray here .. as many students participating don't know about citing sources and giving proper attribution (as obvious as that may seem) |
| 18:35.05 | Notify | 02GCI:popescuandrei * 5778240039813120 : Task Closed - Congratulations, this task has been completed successfully. |
| 18:35.09 | MarcTannous | not to even mention about UI icons... |
| 18:35.14 | MarcTannous | I mean that was plainly theft IMO |
| 18:35.22 | brlcad | so our first response is to educate (I will contact the student), and it's on repeat instances that they get in trouble ... or on blatant copying |
| 18:36.36 | brlcad | MarcTannous: have to be culturally sensitive too, not all countries have the same IP/ownership tenants as the US/EU |
| 18:37.19 | brlcad | so for you as a participant, I see it as reinforcing the value and quality of your work and that will reflect well on that task's grading |
| 18:37.36 | brlcad | and obviously not so well on the student that just tweaked a few things |
| 18:37.47 | MarcTannous | thank you then |
| 18:37.51 | MarcTannous | should I get one of the icon tasks? |
| 18:37.54 | MarcTannous | or some coding ones |
| 18:38.07 | brlcad | icon tasks! |
| 18:38.10 | brlcad | please :) |
| 18:38.27 | MarcTannous | sure thing |
| 18:38.30 | MarcTannous | coming right up |
| 18:38.48 | brlcad | the five new ones that are up there now form almost a complete set (or at least ARE a usable set) |
| 18:40.26 | MarcTannous | can I use some icon packs I own? |
| 18:40.37 | MarcTannous | or do you want them from scratch |
| 18:40.57 | MarcTannous | ( distribuitive license on these packs ) |
| 18:41.22 | brlcad | I'm not sure what that implies, but we need something that can be assigned full rights, so we can use it in our tools without worrying about any future legal issues |
| 18:41.30 | MarcTannous | yep |
| 18:42.26 | brlcad | if you can submit it like the other work, i'm not sure I see a difference |
| 18:42.45 | brlcad | we need to be able to relicense contributions under our license(s), distribute without pause |
| 18:43.35 | brlcad | which includes rights to create derivative works, rights to promote, rights to sell, rights to use, rights to ... unlimited rights :) |
| 18:43.37 | Notify | 02GCI:o7p9bxbnyj * 5239535713648640 : Ready for review - The work on this task is ready to be reviewed. |
| 18:44.02 | Notify | 02GCI:Melange * 6650599705673728 : Task Reopened - Melange has detected that the final deadline has passed and it has reopened the task. |
| 18:44.15 | andromeda_galaxy | brlcad: that's my stab at usingn consistent naming and documenting what it tries to do |
| 18:44.21 | brlcad | if we don't have that, it's probably not wanted or at least not usable |
| 18:44.31 | *** join/#brlcad teepee_ (~teepee@gateway/tor-sasl/teepee) |
| 18:44.40 | MarcTannous | let's go icon tasks! :D |
| 18:44.52 | Notify | 02GCI:tannousmarc * 5835108829364224 : Task Claimed - I would like to work on this task. |
| 18:44.55 | andromeda_galaxy | I also added a little thing to the description of rt_shootrays() saying how to get back a_uptr |
| 18:45.47 | brlcad | MarcTannous: there might be a problem is if you received/purchased icons from someone else, or derived yours off of someone else's work |
| 18:46.15 | MarcTannous | brlcad: will make sure that is not an issue then |
| 18:52.23 | andromeda_galaxy | brlcad: so on the rt-bundle-integration front, I know you made the task to look more for where the rounding error is coming from... do you have any tips on how to do that? I tried for a couple of hours, but I could never get it any better than this (~1-2e-7, I believe) |
| 19:17.00 | andrei_ | brlcad, MarcTannous: there was a task which andromeda did recently, involving commit iterrations |
| 19:17.07 | andrei_ | the functionality was covered by andromeda |
| 19:17.28 | andrei_ | but since Marc is oriented on design, and since he s finishing tasks rather fast, from what I see |
| 19:17.52 | andrei_ | perhaps we could make a task with designing the theme or "looks" for that? |
| 19:18.13 | MarcTannous | woudl love it |
| 19:24.05 | andromeda_galaxy | MarcTannous: it uses dojos dgrid/dojox charts for all the output, both are reasonably themeable... if you can do it without changing any of the js that would be great, since that js is all structured to make it easy to hook up to the server |
| 19:24.29 | MarcTannous | can I get a link? |
| 19:24.32 | MarcTannous | to the task |
| 19:24.41 | andromeda_galaxy | the one I've done so far? |
| 19:24.48 | MarcTannous | the one andrei is mentioning |
| 19:24.49 | andromeda_galaxy | https://www.google-melange.com/gci/task/view/google/gci2014/5275647798673408 |
| 19:24.50 | gcibot | [[ PROJECT: Implement web application for testing commit ranges || BRL-CAD || Closed || Mihai Neacsu, Dishank ]] |
| 19:25.34 | MarcTannous | will look into it |
| 19:25.48 | andromeda_galaxy | that's my current work on it... also, try not to change any of the links yet, I plan on making them all work with a nice RESTful api (HATEOAS) on the server |
| 19:26.08 | MarcTannous | what I'll probably do is export stuff that is being created as divs |
| 19:26.15 | MarcTannous | atm how are you adding it to the page? |
| 19:26.23 | MarcTannous | actually let me check the js lol |
| 19:29.37 | andromeda_galaxy | MarcTannous: there's a page on dgrid theming somewhere... |
| 19:30.01 | andromeda_galaxy | http://www.sitepen.com/blog/2012/05/03/css-styling-of-dgrid/ |
| 19:30.24 | andromeda_galaxy | MarcTannous: pretty much everything is being added through either dojo dgrid or dojo charts |
| 19:30.38 | andromeda_galaxy | really nice customizable, performant, etc. |
| 19:31.29 | MarcTannous | yeah this seems nice |
| 19:31.42 | MarcTannous | thought you were appending elements to the html body and that would've been more tricky |
| 19:33.25 | *** join/#brlcad gaganjyot (~gaganjyot@101.209.4.180) |
| 19:33.59 | andromeda_galaxy | MarcTannous: indeed :) I have a bunch of styling in there already to do things like the little indicator lights |
| 19:36.43 | MarcTannous | I'll get my icon task done and then check this out |
| 19:37.21 | andromeda_galaxy | MarcTannous: sounds good.. when I'm done with the rt stuff, I intent to write a server-side component for that to make it finish working.. |
| 19:37.44 | MarcTannous | that will get us a really nice working and clean looking project |
| 19:37.45 | MarcTannous | :D |
| 19:38.19 | brlcad | andromeda_galaxy: e^7 is approximately the limit of floating point precision, so you could first check your data types at EVERY single stage of processing, every pass to/from *every* function, to make sure they're all double-precision |
| 19:38.39 | andromeda_galaxy | brlcad: currently, all area fastf_t |
| 19:38.52 | MarcTannous | brlcad: where should I submit the design part of andromeda's current project? |
| 19:39.01 | brlcad | next would be to simply itemize the starting values to make sure they're the same for both .. if you have fuzz in the start, you're not going to get to the same end |
| 19:39.27 | andromeda_galaxy | brlcad: true... there are a couple of places that the problem could be, but both of them are quite difficult to eliminate: |
| 19:39.58 | andromeda_galaxy | one is calculating the angle to use, the other is just in the way that it builds up the numbers by adding <increment> (which is the same as the <increment> for the old method, but that involved multiplying...) |
| 19:41.02 | brlcad | andrei_: can you write up the task in question for MarcTannous ? |
| 19:41.29 | andrei_ | yes, I will, just not at this very moment |
| 19:41.35 | andromeda_galaxy | brlcad: I'm really not sure how either of those can be eliminated, especially the angle one |
| 19:41.38 | andrei_ | I'll put it directly on melange, ok? |
| 19:41.39 | brlcad | okay, thanks! |
| 19:41.41 | brlcad | yeah |
| 19:41.50 | brlcad | not doing any more bulk uploads |
| 19:41.51 | andromeda_galaxy | the problem is that the angle has to be just less than perspective angle, because of rts half-open behavior |
| 19:43.06 | andromeda_galaxy | so, it currently uses this to get it: atan(tan(rt_perspective / 2.0 * M_PI / 180) / width * (width - 1)); |
| 19:43.18 | brlcad | andromeda_galaxy: you're talking/thinking about how to eliminate a method without confirming that is the source |
| 19:43.36 | brlcad | you have to confirm where the fuzz is getting injected, very specifically |
| 19:43.43 | andromeda_galaxy | brlcad: true... I guess actually my question is more about how to find out whether any particular possible error is the source |
| 19:44.14 | brlcad | here's what you can do that is easily actionable |
| 19:44.14 | andromeda_galaxy | brlcad: the angle is especiaally hard to check |
| 19:44.33 | andromeda_galaxy | because the old method doesn't directly use an angle like that, it just goes in increments through a width/height |
| 19:44.47 | brlcad | itemize the lines of code that are involved one by one, and for just two rays show the values at every step |
| 19:44.50 | andromeda_galaxy | which is a pretty horrible thing to have for a generic api, since it requires you to know things like how long the direction vector is |
| 19:45.08 | brlcad | that's okay that they're using different methods |
| 19:45.34 | andromeda_galaxy | but doesn't using different methods make it hard to compare intermediate values? |
| 19:45.53 | brlcad | it depends on what all the itemized steps end up being |
| 19:46.22 | andromeda_galaxy | if the steps are different, comparing the results of each step becomes difficult... |
| 19:47.06 | brlcad | it depends entirely on the steps involved |
| 19:47.14 | andromeda_galaxy | hmm... |
| 19:47.20 | brlcad | perhaps you already know the steps well enough to say that there aren't any that can be compared? |
| 19:47.58 | andromeda_galaxy | brlcad: here's the quick sketch |
| 19:48.09 | andromeda_galaxy | (but that's what I had been thinking) |
| 19:48.47 | andromeda_galaxy | in the old method, it uses width/height and dy/dx in model space (computed far far away) and basically loops through each i, j and makes a ray with r_dir = VJOIN2(center_dir, i, dx, j, dy) |
| 19:49.12 | andromeda_galaxy | sorry, not center_dir but bottom_left_dir (I think) |
| 19:49.27 | andromeda_galaxy | in the new method, it calculates an angle of divergence in both x and y dimensions |
| 19:49.47 | andromeda_galaxy | and then calculates a new width/height appropriate for a vector with that length inside the gen_* function |
| 19:50.00 | andromeda_galaxy | and then uses a num steps to get its own dx/dy |
| 19:50.09 | andromeda_galaxy | and then uses those in an accumulator loop to get each pixel |
| 19:50.36 | andromeda_galaxy | comparing the dx and dy is hard |
| 19:50.49 | andromeda_galaxy | because they're scaled differently |
| 19:50.54 | andromeda_galaxy | but might be possible |
| 19:51.10 | andromeda_galaxy | brlcad: what do you thinK? |
| 19:51.44 | brlcad | I think you're not reading what I wrote ;) |
| 19:52.21 | andromeda_galaxy | brlcad: sorry? this was my 'do I know the steps well enough to know their aren't any that can be compared?' |
| 19:52.54 | andromeda_galaxy | brlcad: showing that I couldn't tell any way to compare the results (and I tried earlier), but wondering if you saw something I was missing |
| 19:53.04 | brlcad | that's not answering whether you know the steps well enough to know they can't be compared |
| 19:53.10 | brlcad | you itemized your understanding |
| 19:53.17 | brlcad | is that well enough? |
| 19:53.33 | brlcad | that's not a question you're going to answer by dumping your knowledge on the table :) |
| 19:53.40 | brlcad | it's something for you to think about |
| 19:53.56 | brlcad | it's also not something I can answer for you without me doing the work |
| 19:53.58 | andromeda_galaxy | well, I'd thought it was pretty well... the thing is that for all the steps that I could figure out during the last task, I tried and couldn't compare the results effectively |
| 19:54.37 | brlcad | I'd have to break out all the calculations going on in each and see if there are any syncing points |
| 19:54.58 | andromeda_galaxy | true... well, I'll try just doing that again, see if looking at it for more time leads to anything that I missed the first time throguh |
| 19:55.00 | andromeda_galaxy | *through |
| 19:55.26 | brlcad | so a word of caution, you mentioned atan(tan(rt_perspective / 2.0 * M_PI / 180) / width * (width - 1)); |
| 19:55.44 | brlcad | the trig functions have very minimal tolerance guarantees |
| 19:55.51 | andromeda_galaxy | in the comparing results stage, are the results ever "close enough" without being exactly the same? i.e. like how most comparisons of floating point numbers use a tol... |
| 19:57.05 | andromeda_galaxy | about the trig functions: interesting, I hadn't known that... I'll have to see if it's possible to do without as much trig/inverse trig |
| 19:57.13 | brlcad | I'd say that passing regression tests is probably close enough |
| 19:57.26 | andromeda_galaxy | the problem is that the regression tests compare pixel values |
| 19:57.32 | andromeda_galaxy | but what is being changed by this change is pixel coordinataes |
| 19:57.35 | brlcad | if regression tests still fail, that really begs for there being something more substantive going on |
| 19:57.49 | andromeda_galaxy | so if there's a pixel directly on an edge, |
| 19:58.07 | andromeda_galaxy | very very small changes in coordinate can result in (relatively) huge changes in value |
| 19:58.21 | brlcad | could, is that what's happening here? prove that |
| 19:58.44 | andromeda_galaxy | brlcad: I can try to.. like I said, I had some code that showed that pixel coordinates never differed by more thatn about 2e-7, but that might still be too much |
| 19:59.24 | brlcad | never differeing by more than 2e-7 doesn't really mean much |
| 19:59.48 | brlcad | need semantic context to know if that matters, not absolute values |
| 20:00.27 | andromeda_galaxy | true... the numbers are all mostly in the 1-10 range, coordinates in model space, at least for the 'solids' model |
| 20:00.42 | brlcad | ideally, my intuition says it's possible to get the rays to match exactly (perhaps by using a completely different calculation method -- with the same inputs -- or in a different order of operations) |
| 20:01.05 | andromeda_galaxy | might work.. one of the problems is just that the inputs that rt takes |
| 20:01.11 | andromeda_galaxy | just aren't good for a general function for this |
| 20:01.29 | andromeda_galaxy | rt just does a rectangular grid style approach, assuming that the viewbase_model &c is in the right place to make it all the right angle |
| 20:02.19 | andromeda_galaxy | brlcad: but semantically that is not a great approach to take for a library function, because for generating a frustum, an angle makes a lot more sense than 'this far above the point that is at the end of the ray in the input' |
| 20:02.44 | andromeda_galaxy | brlcad; but what is exactly for floating point numbers? even C compilers advise against comparing them with == |
| 20:03.13 | brlcad | don't confuse input parameters with method being used to calculate the rays |
| 20:03.36 | andromeda_galaxy | sorry, maybe I got confused... Ithought that by 'with the same inputs', you meant using the same inputs as the original method |
| 20:03.48 | brlcad | I could certainly have a function that takes a point+dir+angle and not use highly-sensitive angles to base my neighboring rays off of |
| 20:04.26 | brlcad | there's a variety of ways I could get that end-result with the same inputs |
| 20:04.42 | brlcad | input to a method != values used within that method |
| 20:04.54 | brlcad | rather, doesn't imply a particular method |
| 20:05.03 | andromeda_galaxy | sorry, I'm slightly confused... I was talking about the inaccuracies of converting values to pass them in and then converting them back in order to use the same metod |
| 20:05.09 | andromeda_galaxy | *method |
| 20:07.32 | brlcad | look, this is getting nowhere quickly .. |
| 20:07.37 | brlcad | you're brain dumping a lot |
| 20:07.50 | brlcad | and expecting some grand insight perhaps, which doesn't exist |
| 20:08.39 | brlcad | not to put words in your mouth, but I'm not getting the sense that my suggestions are getting absorbed very deeply amidst your other thoughts of the issues |
| 20:08.56 | brlcad | which is just to say that I think there's gotta be a step back taken |
| 20:09.30 | brlcad | I'm not going to have as much insight as you without me doing the task too |
| 20:09.36 | andromeda_galaxy | brlcad: sorry... that's a good idea, the problem is probably that I waas working on it for long enough earlier that I just keep thinking back to what didn't work... |
| 20:11.47 | brlcad | you have made many and statements that simply aren't strictly true (too many at this point to revisit), and I'm getting the sense that you're rushing too hard for a task like this that really requires in-depth meticulous attention to details |
| 20:12.37 | brlcad | what I can suggest, and you can fully ignore, are 1) to see if there are any sync points .. this is not an easy task |
| 20:12.50 | brlcad | 2) see if you can avoid the trig functions (find a different method) |
| 20:13.19 | brlcad | 3) consider changing the calculation method being used inside the function and/or the order of operations |
| 20:13.28 | brlcad | that's pretty much it |
| 20:13.52 | andromeda_galaxy | brlcad: thanks, I'll try those... sorry, as I said, I keep thinking about the things that didn't work earlier |
| 20:14.08 | *** join/#brlcad sofat (~sofat@202.164.45.198) |
| 20:14.18 | sofat | maths22, hello |
| 20:14.29 | brlcad | i'm sure, and you may have exhausted options (except I don't see you having mentioned anything about trying different methods) |
| 20:14.39 | sofat | please accept my pull request |
| 20:14.55 | andromeda_galaxy | brlcad: the trig one is the big one that I didn't try at all, in terms of methods/order of operations |
| 20:14.55 | sofat | brlcad, hell |
| 20:15.03 | sofat | brlcad, hello |
| 20:15.04 | andromeda_galaxy | I did try a couple of things, but there might be more that I missed |
| 20:15.32 | andromeda_galaxy | well, thanks for the discussion anyway, I'll try those techniques and see if any more accuracy can be gained |
| 20:15.37 | brlcad | but there many many ways to calculate a grid/set of rays that diverge, and the (*implementation*) method you chose may have tolerance issues (or maybe the existing method has issues, it would take a lot of work to prove/disprove that) |
| 20:16.14 | brlcad | any time you have a calculation based on a small number, it's going to accumulate error very quickly |
| 20:16.44 | brlcad | any time you have a calculation based on values that are not normalized (they are not near 0), they are going to accumulate error very quickly |
| 20:17.41 | andromeda_galaxy | that's true... another thing to consider, at least... |
| 20:19.14 | brlcad | for the first point, it would suggest calculating a dx/dy angle of divergence is not a good idea because that angle will likely be something very small like 0.000123 |
| 20:19.56 | brlcad | calculating the actual angle per ray would be one way (of several) to avoid that |
| 20:20.08 | brlcad | calculating without angles would be another |
| 20:20.20 | andromeda_galaxy | brlcad: that might work.. I did it as dx/dy in absolute coordinates (not angles) to match the way that the original code did it |
| 20:20.46 | andromeda_galaxy | it uses the angles and the length of the center ray to figure out what a dx/dy model coordinates vector should be, and uses that |
| 20:21.12 | Notify | 02GCI:vkolova * 5877505323958272 : Task Claimed - I would like to work on this task. |
| 20:21.15 | sofat | brlcad, you have any work for me ? |
| 20:21.20 | brlcad | for the second point, if the ray is way out in space and you're using it's point for all calculation, that is probably not a good idea because it's not normalized |
| 20:21.57 | brlcad | calculating a translation vector/matrix that works on the origin and then applies the inverse to the result would likely give a better result (as one of several possible solutions) |
| 20:23.29 | andromeda_galaxy | that makes sense, I'll try changing that too, if the other things don't help significantly (enough)... thanks for the suggestions |
| 20:24.13 | brlcad | this all gets at possibilities for changing the method of calculation .. there's dozens of possibilities that will all give different ray sets with different amounts of error |
| 20:24.52 | brlcad | here's one more idea for you ... you could use a fixed precision calculator to figure out *exactly* what the actual ray values should be for a given input |
| 20:25.20 | brlcad | if you can show that your outputs are more faithful to "actual" results (i.e., less error), then that becomes easier to justify |
| 20:25.47 | andromeda_galaxy | good point, I should try doing that... now I'm curious if using gmp for fastf_t makes a bunch of regression tests built around errors fail... |
| 20:26.55 | brlcad | you'd need a good calculator like mathematica or bc or something where you can set the precision to at least 50 decimal places |
| 20:26.55 | brlcad | gmp could work too, more setup though |
| 20:27.19 | brlcad | there was a set of tasks last year that looked into integrating a C++ class under fastf_t so it could do fixed precision math |
| 20:27.24 | andromeda_galaxy | right, mathematica is definitely doable... I'd probabaly need to write some mathematica code that (for a specific case) calculated all the ray values andn then find some way to automatedly compare them... |
| 20:27.30 | brlcad | for exactly this kind of inspection |
| 20:27.37 | andromeda_galaxy | brlcad: yeah, I saw that, there was a followup this year |
| 20:27.50 | brlcad | you'd need to make sure any trig functions you used in mathematica don't suffer similar tolerance issues |
| 20:28.07 | andromeda_galaxy | but currently brl-cad won't build with fastf_t != double for a multitude of reasons, so I didn' try to do it |
| 20:28.14 | brlcad | nods |
| 20:28.28 | andromeda_galaxy | indeed.. my first step is going to be trying to get rid of the trig functions overall |
| 20:29.45 | andromeda_galaxy | (probably) |
| 20:31.36 | sofat | brlcad, you have any new task for me ? please reply |
| 20:32.04 | brlcad | sofat: I have years of work for you |
| 20:32.19 | brlcad | but I can't tell you what to do, you need to figure out what you *want* to do :) |
| 20:33.09 | brlcad | sofat: and I'm not ignoring you, just fyi... it's just busy and open-ended questions aren't very constructive |
| 20:33.41 | brlcad | (ditto with saying hello on irc, but I admit I sometimes violate that myself) |
| 20:34.00 | sofat | sorry |
| 20:34.05 | brlcad | sofat: don't be sorry :) |
| 20:34.06 | sofat | :-( |
| 20:34.26 | brlcad | be productive :) |
| 20:34.30 | MarcTannous | that was deep |
| 20:34.34 | brlcad | whta interests you |
| 20:34.37 | sofat | okay |
| 20:35.23 | brlcad | figure out where you want to go, and take it to the next step |
| 20:35.49 | brlcad | you've been working a lot in one area, and are just waiting on maths22 to review? or what? |
| 20:37.17 | brlcad | sofat: do you want to keep working on docbook, keep working on website stuff, something else? |
| 20:38.13 | sofat | i am working on website |
| 20:38.36 | sofat | but i want to know the status of docbook project |
| 20:38.37 | brlcad | this I know :0 |
| 20:38.51 | brlcad | open-ended question |
| 20:38.58 | brlcad | status: exists |
| 20:39.03 | sofat | what the next step in docbook project |
| 20:39.03 | brlcad | ask a different question :) |
| 20:39.22 | sofat | okay |
| 20:39.23 | brlcad | there are many many many possible next steps, depending on who works on it next |
| 20:39.46 | brlcad | in terms of long-term goals, we need our docbook documents sync'd to the web |
| 20:40.12 | andromeda_galaxy | brlcad: oh, one other unrelated thing to think about for later ---- what do you think of making (some) of the parallel datastructure functions use userspace RCU on linux? |
| 20:40.13 | brlcad | we need stylesheets, updated pdfs, and web sync |
| 20:40.29 | brlcad | we need bidirectional web-editing |
| 20:40.37 | sofat | ok |
| 20:40.56 | brlcad | we need to convert a slew of non-docbook docs to docbook |
| 20:41.31 | brlcad | we need to eliminate/scavenge/archive docs that are out-of-date |
| 20:41.44 | sofat | ok |
| 20:42.09 | sofat | i ready for this |
| 20:42.14 | brlcad | we need someone to take ownership of the project (by consistently working on it) |
| 20:43.47 | sofat | i am ready to take the ownership of this project if you allow . |
| 20:45.05 | brlcad | sofat: I suggest writing up a wiki page with a todo list so the scope and vision of activity remains consistent |
| 20:45.44 | brlcad | taking ownership isn't something that gets granted, you own it by working on it (consistently, vigorously, more than anyone else) |
| 20:46.25 | brlcad | at least it's not needed for this, it just needs someone to work on oit |
| 20:46.36 | brlcad | I suggest a page like http://brlcad.org/wiki/STEP |
| 20:47.48 | brlcad | sofat: maybe create http://brlcad.org/wiki/Docbook and fill it out with the "we need" requirements you know of, then expand them with more specific sub-task bullets |
| 20:48.02 | sofat | ok |
| 20:48.38 | brlcad | until you've decomposed the problem at least into individual tasks that should take someone less than a couple days each |
| 20:49.02 | brlcad | do you feel directed? I hope not ;) |
| 20:50.04 | brlcad | once you have it all in front of you, then you might have a better idea deciding what you *want* to do |
| 20:50.17 | brlcad | and we'll all benefit |
| 20:50.31 | sofat | yes i understand . |
| 21:13.13 | *** join/#brlcad teepee_ (~teepee@gateway/tor-sasl/teepee) |
| 21:55.36 | Notify | 03BRL-CAD:n_reed * 63971 (brlcad/trunk/src/libbrep/intersect.cpp Property Changed: and 3 others): merge r63481 from brep-debug branch, fixes an intersection bug |
| 22:02.29 | brlcad | I need some brainstorming suggestions |
| 22:04.35 | brlcad | what's a good word that we can use to describe a bit of work that needs some attention, could be considered incomplete, is discovered but not yet investigated/worked, etc |
| 22:06.02 | brlcad | for example, I find a particular bug, I work on it for a few days, find out a LOT but don't get the bug fixed .. so I write down all my notes, stash away this beautifully self-contained example, and fully intend to come back to it some day but really anyone could pick up where I left off |
| 22:06.15 | brlcad | what's a good name for that? single word ideally |
| 22:07.15 | brlcad | not just bugs too .. these are sort of somewhere between a GCI task and a GSoC project .. mini projects, mega tasks |
| 22:08.08 | brlcad | pendings |
| 22:08.12 | brlcad | scraps |
| 22:08.27 | brlcad | piltrafa |
| 22:09.24 | brlcad | incompletes |
| 22:10.23 | brlcad | dilettantes |
| 22:11.09 | brlcad | rough |
| 22:11.14 | brlcad | roughage (heh) |
| 22:11.21 | brlcad | wantings |
| 22:17.17 | *** join/#brlcad merzo (~merzo@105-127-133-95.pool.ukrtel.net) |
| 22:27.26 | Notify | 02GCI:deepakkumarsharma * 5835108829364224 : Task Assigned - This task has been assigned to Marc Tannous. You have 90 hours to complete this task, good luck! |
| 22:27.41 | Notify | 02GCI:deepakkumarsharma * 5877505323958272 : Task Assigned - This task has been assigned to vkolova. You have 90 hours to complete this task, good luck! |
| 22:28.37 | Notify | 03BRL-CAD:carlmoore * 63973 brlcad/trunk/doc/docbook/system/man1/en/pixfade.xml: add note about default |
| 23:22.40 | *** join/#brlcad infobot (ibot@rikers.org) |
| 23:22.40 | *** topic/#brlcad is Topic for #brlcad: BRL-CAD || http://brlcad.org || logs: http://ibot.rikers.org/%23brlcad/ || Welcome GCI 2014 participants! If you're participating, be patient when asking questions... we're very busy and sometimes have packed schedules. We will respond if you are still on IRC: learn screen+irssi. |
| 23:24.47 | *** join/#brlcad teepee (~teepee@gateway/tor-sasl/teepee) |
| 23:31.13 | andromeda-galaxy | brlcad: thanks for the tip about trig functions being inaccurate (which I didn't know), it looks like that may have been it! |
| 23:31.15 | *** join/#brlcad Ko0lGuY_ (uid61616@gateway/web/irccloud.com/x-orglypdnskwbdkqm) |
| 23:39.47 | *** join/#brlcad nmz787_i (~nmccorkx@134.134.137.71) |
| 23:40.42 | nmz787_i | is there someone available to take all the mged examples (i.e. the PDFs), create files for each example, then commit those files to a github repo or similar? |
| 23:40.50 | nmz787_i | like one of the students |
| 23:44.22 | *** join/#brlcad wei1006 (~fircuser@183.90.37.120) |
| 23:44.37 | *** join/#brlcad wei1006 (~fircuser@183.90.37.120) |
| 23:44.57 | *** join/#brlcad wei1006 (~fircuser@183.90.37.120) |
| 23:46.05 | wei1006 | https://www.google-melange.com/gci/task/view/google/gci2014/5869143559503872 |
| 23:46.05 | gcibot | [[ Beginner: Install and run BRL-CAD! #59 || BRL-CAD || NeedsReview (3 days 11 hrs 36 min) || Harmanpreet, Ishwerdas ]] |
| 23:47.14 | tofu | andromeda-galaxy: it's not that they're inaccurate, it's that they're simply not guaranteed to be accurate ;) |
| 23:47.43 | brlcad | wei1006: tasks are announced to the mentors, you don't need to do anything in here |
| 23:47.54 | brlcad | other than talk if you're performing the communication piece ;) |
| 23:49.02 | wei1006 | I am directed to post the link of my work here and explain how I have done it |
| 23:49.28 | wei1006 | what should I do then? |
| 23:51.31 | brlcad | it does not direct you to post the link of your work here |
| 23:51.39 | brlcad | where do you think it tells you to do that? |
| 23:52.09 | andromeda-galaxy | brlcad: thanks for the tip about trig functions being inaccurate (which I didn't know), it looks like that may have been it! |
| 23:52.31 | brlcad | 18:47 < tofu> andromeda-galaxy: it's not that they're inaccurate, it's that they're simply not guaranteed to be accurate ;) |
| 23:52.37 | andromeda-galaxy | brlcad: oops, didn't mean to resend that |
| 23:52.42 | brlcad | ah, k |
| 23:52.56 | brlcad | they do that for performance |
| 23:52.59 | andromeda-galaxy | brlcad: indeed... well, anyway, that appears to have been the problem, now all the regress tests pass |
| 23:53.14 | andromeda-galaxy | I couldn't totally get around using the trig functions, but instead of atan(tan(...)...) |
| 23:53.17 | brlcad | it's expensive to do a chebyshev iteration a sufficient number of times to converge on a more accurate solution |
| 23:53.18 | andromeda-galaxy | it's now atan(...) |
| 23:53.38 | andromeda-galaxy | unfortunately, doing it without any trig seems impossible... |
| 23:53.51 | brlcad | so you probably still have error, but not enough to affect a result (which is good) |
| 23:54.04 | andromeda-galaxy | brlcad: right! it's now close enough that the regress tests all pass |
| 23:54.06 | wei1006 | I have posted my explanation of the task hother people posting their |
| 23:54.17 | brlcad | can you still print a comparison of the two for some simple example? |
| 23:54.22 | wei1006 | explanation |
| 23:54.31 | wei1006 | and link to the task |
| 23:54.36 | andromeda-galaxy | brlcad: you mean a comparison of the old and new ray values to ~12 digits? |
| 23:55.10 | brlcad | wei1006: my question is where did you read that it said to past a link to your task on IRC? |
| 23:55.54 | brlcad | andromeda-galaxy: more than 12 digits would be ideal |
| 23:56.08 | andromeda-galaxy | brlcad: sure... why? |
| 23:56.09 | brlcad | %.20lf should be good |
| 23:56.19 | brlcad | just to know how much of a change we're talking about |
| 23:56.25 | n_reed | brlcad: setups, weekenders, dust-offs, orphan/adoptable tasks |
| 23:56.27 | andromeda-galaxy | ahh, sure |
| 23:56.35 | andromeda-galaxy | I can upload the debug file from the 'solids' regression test if you want |
| 23:56.41 | brlcad | n_reed: mm, those are good |
| 23:56.42 | andromeda-galaxy | that shows the high-precision values |
| 23:56.47 | brlcad | i like orphans :) |
| 23:56.55 | brlcad | adopt an orphan... |
| 23:57.18 | andromeda-galaxy | brlcad: would that be enough? |
| 23:57.47 | brlcad | if it shows before and afters, sure |
| 23:57.51 | andromeda-galaxy | (the debug file containing my printout of old vs new) |
| 23:58.05 | brlcad | more than 12 digits? |
| 23:58.14 | andromeda-galaxy | brlcad: I haven't produced it yet, but it'll be easy |
| 23:58.20 | brlcad | okay, cool thanks |
| 23:58.47 | brlcad | wei1006: you did all you need to do (except for answering my question...) |
| 23:59.09 | andromeda-galaxy | brlcad: sounds good! then we'll have a regression-test-passing rt and a good idea of how different it is... |
| 23:59.18 | andromeda-galaxy | though rt is still mildly bloated from data structure conversion problems... |