| --- Log opened Fri Jul 22 00:00:01 2016 |
| 00:03 | | catadroid [catadroid@Nightstar-rjs.159.132.82.IP] has joined #code |
| 02:45 | | gizmore|2 [kvirc@Nightstar-sh49ni.dip0.t-ipconnect.de] has quit [[NS] Quit: KVIrc 4.9.2 Aria http://www.kvirc.net/] |
| 02:56 | <~Vornicus> | "the best optimizer is between your ears" |
| 02:56 | <~Vornicus> | having seen the kinds of shenanigans a modern compiler can get up to |
| 02:56 | <@Alek> | wow, +5/*3 suggests potential for bruteforcing pathfinding. |
| 03:22 | <~Vornicus> | "+5/*3"? |
| 03:25 | <@Alek> | function findSolution(target) { |
| 03:25 | <@Alek> | function find(start, history) { |
| 03:25 | <@Alek> | if (start == target) |
| 03:25 | <@Alek> | return history; |
| 03:25 | <@Alek> | else if (start > target) |
| 03:25 | <@Alek> | return null; |
| 03:25 | <@Alek> | else |
| 03:25 | <@Alek> | return find(start + 5, "(" + history + " + 5)") || |
| 03:25 | <@Alek> | find(start * 3, "(" + history + " * 3)"); |
| 03:25 | <@Alek> | } |
| 03:25 | <@Alek> | return find(1, "1"); |
| 03:25 | <@Alek> | } |
| 03:28 | <@Alek> | Nothing doing for roads, but game pathfinding could work. If you store length and multiple working paths outside the basic set, you can find the more efficient ones as well. |
| 03:28 | <~Vornicus> | ...what |
| 03:29 | <@Alek> | I'm learning JS. and thinking of other ways to use the example code. |
| 03:30 | <@Alek> | this is example code from the book. |
| 03:30 | <~Vornicus> | oh |
| 03:30 | <@Alek> | it bruteforces a path to a target number until the first is found or until it goes through all and none are found. |
| 03:32 | <@Alek> | paths end once they pass the target number, but if you have paths that backtrack you can follow them too. but drop any that are longer than the ones you've already found, and trim off longer found ones when you find shorter ones. etc. |
| 03:37 | | * ASCII sees the word pathfinding, nopes right out |
| 03:44 | | * [R] has no clue how that findSolution function is supposed to work. |
| 03:45 | <&[R]> | Alek: Also when doing JS and you get into classes, it's strongly recommended that you make your functions start with a capital letter when its a constructor and lower-case every other time. |
| 03:48 | <~Vornicus> | <3 pathfinding |
| 03:50 | | * Vornicus sees ascii, is reminded of the sphere noise thing, thinks he might have thought of something. but it's kinda nasty and he has no idea the power of the dark si-- shaders. |
| 03:51 | < ASCII> | Can't be worse than when you taught me dijkstra's... ten years ago or so? |
| 03:51 | < ASCII> | what'd you think of? |
| 03:53 | <~Vornicus> | Something like: you can do perlin-type noise with 3d value functions and then proj that onto the actual normal (from a mathematical description of the object?) at the target point to get an altitude. I suspect this *might* give something differentiable so you can pull normals off, but I haven't tried it. |
| 03:58 | <&[R]> | Oh, findSolution isn't actually doing something useful, derp |
| 04:07 | < ASCII> | Currently I am cheating somewhat |
| 04:08 | < ASCII> | for my normals |
| 04:08 | < ASCII> | I;m using the heightmap to blend between flat normals and a mountain texture I made. |
| 04:09 | < ASCII> | also working on using sobel to convert the heightmap into normals directly, though that is tricky since the noisemap isn't in UV space |
| 04:10 | < ASCII> | so I'm repurposing the texture lookup I use to color the heightmaps to feed into the sobel function in uv space |
| 04:10 | < ASCII> | unity handles normals in a very annoying way though |
| 04:10 | < ASCII> | which I'm running into |
| 04:12 | <~Vornicus> | I saw that. |
| 04:13 | <~Vornicus> | I'm thinking that using the 3d value functions and a proj using the true object normal might give an altitude function that is itself differentiable. |
| 04:14 | < ASCII> | It is possible. Probably not from within hlsl though, probably would need to do it in c# |
| 04:16 | < ASCII> | (also I am a really terrible coder) |
| 04:19 | < ASCII> | Though given how complicated it has gotten, redoing the planet generator as a script might not be the worst idea |
| 04:20 | <~Vornicus> | you could then do the differentiation "by hand" and do *that* as a shader |
| 04:20 | < ASCII> | my head asplode |
| 04:21 | | * Vornicus is the mathematician, fear not |
| 04:27 | < ASCII> | Now that I have all the individual pieces, I need to refactor the shader significantly, and redo the lighting pretty much from scratch |
| 04:27 | < ASCII> | Since the model I used initially gives a nice terminator but it doesn't work well with a perturbed surface at all |
| 04:27 | | Turaiel is now known as Turaiel[Offline] |
| 04:28 | < ASCII> | probably will attempt some nicer atmospheric scattering. Probably completely fake but nicer than the simple dot of the view angle |
| 04:31 | < ASCII> | The nice thing about blending premade normal maps is that, well, they just look a lot nicer. Can include much more interesting high frequency detail that I really can't make with perlin noise without running all sorts of absurd things like erosion on them |
| 04:31 | <~Vornicus> | (idunno shit 'bout shaders though) |
| 04:31 | < ASCII> | http://i.imgur.com/pufS4UW.jpg |
| 04:32 | < ASCII> | this is with just a single normal map, blended by altitude |
| 04:32 | <~Vornicus> | oh that is good. |
| 04:33 | <&[R]> | That valley near the sea almost dead center of the pic... |
| 04:33 | < ASCII> | You can see how it breaks my simplified lighting model |
| 04:33 | <&[R]> | Should be flooded so bad |
| 04:34 | < ASCII> | No, it is all above water. |
| 04:35 | < ASCII> | Terrain features are blended into the coloration. |
| 04:35 | < ASCII> | But the shading is broken, as I mentioned=P |
| 04:36 | < ASCII> | It won't be perfect, though once I add a second normal map for non mountains to use for lower altitudes that should held with that problem. |
| 04:40 | < ASCII> | But yeah, it looks like I'm gonna need to pretty much start from scratch on the lighting, so hopefully I can take the chance to make everything else more organized and less insane. |
| 04:42 | < ASCII> | honestly, for a stellaris scale game(for instance) what I have is probably good enough. |
| 04:42 | | * Vornicus will poke at it over the weekend 'cause this one's got him intrigued |
| 04:42 | < ASCII> | One of the problems ofnot having an actual game in mind is not having any set constraints of goals |
| 04:44 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has quit [Connection reset by peer] |
| 05:02 | | Thalass [thalass@Nightstar-rfp.o7s.158.104.IP] has quit [Server shutdown] |
| 05:02 | | VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has quit [Server shutdown] |
| 05:02 | | jerith [jerith@ServerAdministrator.Nightstar.Net] has quit [Server shutdown] |
| 05:02 | | celticminstrel [celticminst@Nightstar-nhhr58.dsl.bell.ca] has quit [Server shutdown] |
| 05:02 | | himi [sjjf@Nightstar-v37cpe.internode.on.net] has quit [Server shutdown] |
| 05:02 | | Reiver [quassel@Nightstar-ksqup0.co.uk] has quit [Server shutdown] |
| --- Log closed Fri Jul 22 05:02:23 2016 |
| --- Log opened Fri Jul 22 05:16:46 2016 |
| 05:16 | | TheWatcher [chris@GlobalOperator.Nightstar.Net] has joined #code |
| 05:16 | | Irssi: #code: Total of 38 nicks [33 ops, 0 halfops, 0 voices, 5 normal] |
| 05:16 | | mode/#code [+o TheWatcher] by ChanServ |
| 05:17 | | Irssi: Join to #code was synced in 52 secs |
| 05:51 | | catadroid` [catadroid@Nightstar-bop2u3.dab.02.net] has joined #code |
| 05:54 | | catadroid [catadroid@Nightstar-rjs.159.132.82.IP] has quit [Ping timeout: 121 seconds] |
| 06:53 | <@Alek> | R, you give it a positive integer and it uses a combination of +5 and *3 to find a path to it from 1. |
| 06:53 | | * Alek goes to bed. -_- |
| 06:54 | <@Alek> | if a path exists, of course. |
| 06:54 | <@Alek> | some numbers have multiple paths, some have none. |
| 06:54 | < catadroid`> | Without constraint creativity is often unhelpful |
| 06:55 | <@Alek> | and as is, the function only returns the first found path, which isn't necessarily the shortest. XD |
| 07:04 | | celticminstrel is now known as celmin|sleep |
| 07:12 | | catadroid` [catadroid@Nightstar-bop2u3.dab.02.net] has quit [[NS] Quit: Bye] |
| 08:34 | | Emmy [NSkiwiirc@Nightstar-41pbej.static.chello.nl] has joined #code |
| 08:34 | | mode/#code [+o Emmy] by ChanServ |
| 09:14 | | Azash [Azash@wizard.engineering] has joined #code |
| 09:14 | | mode/#code [+o Azash] by ChanServ |
| 09:30 | <@ErikMesoy> | What a glorious future: the blinking twelve problem has come to my window. |
| 09:31 | <@ErikMesoy> | I'm alone in the office and the sunshades on the window closed earlier this morning, making it pretty fucking dark in here since most of the building lights are off, and now they reopened. Not sure if script or light-detection or what. |
| 09:32 | <@ErikMesoy> | I have no manual control over them and no idea where to set the machine control. There is no visible sunshade switch/toggle/knob/thing in the office or the neighboring walls or anywhere in the vicinity that I can see. |
| 09:51 | <@abudhabi> | So you're... Alone in the Dark? |
| 09:52 | | catadroid [catadroid@Nightstar-bop2u3.dab.02.net] has joined #code |
| 10:03 | < catadroid> | Yarr |
| 10:04 | <@abudhabi> | Is it Talk Like A Pirate Day? |
| 10:07 | <@ErikMesoy> | Well, I was for an hour. |
| 10:08 | <@ErikMesoy> | I dunno how to turn the building lights on, either. I searched the entry area and the reception extensively and found *four* separate switches all to open the entry door to let visitors in. No light switches. |
| 10:09 | < Pink> | catadroid, that is quite true |
| 10:09 | < Pink> | Most of the works I've been happy with were done under stupid/absurd constraints:D |
| 10:32 | < catadroid> | Huh? |
| 11:15 | < catadroid> | Ah |
| 11:15 | < catadroid> | The creativity thing |
| 11:15 | < catadroid> | The point of human creativity is to develop solutions to problems given constraints |
| 11:15 | < catadroid> | If you don't constrain yourself, you're searching with no goal and you're not going to be able to evaluate anything interesting you come up with as interesting anyway |
| 11:41 | | Emmy [NSkiwiirc@Nightstar-41pbej.static.chello.nl] has quit [[NS] Quit: http://www.kiwiirc.com/ - A hand crafted IRC client] |
| 11:49 | < catadroid> | ...why can't a version control system just use structural sharing between documents |
| 11:50 | < catadroid> | Hm |
| 11:51 | <@TheWatcher> | Because Reasons |
| 11:52 | < catadroid> | I wonder if that's what git dies |
| 11:52 | < catadroid> | Does* |
| 11:53 | < catadroid> | I mean, a diff is essentially the same operation as an immutable assoc operation |
| 11:53 | < catadroid> | If you apply it |
| 12:21 | | catadroid` [catadroid@Nightstar-mk9nt0.dab.02.net] has joined #code |
| 12:24 | | catadroid [catadroid@Nightstar-bop2u3.dab.02.net] has quit [Ping timeout: 121 seconds] |
| 13:09 | <@ErikMesoy> | The cleaning service has come by and turned on lights. The light switch turns out to be cleverly disguised in what I thought was a read-only display of red diodes and black panels. |
| 13:28 | | Emmy [Emmy@Nightstar-9p7hb1.direct-adsl.nl] has joined #code |
| 13:28 | | mode/#code [+o Emmy] by ChanServ |
| 13:54 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 13:54 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 13:55 | < VirusJTG> | Do any of you have any experience adjust SSL timeouts in IIS. Specifically I am trying to increase the time the handshake can run before it times out. |
| 13:56 | < VirusJTG> | Reason, I have a server in the US and end nodes in Australia, ping times are over 400ms |
| 14:53 | | catadroid` is now known as catadroid |
| 15:36 | | Emmy [Emmy@Nightstar-9p7hb1.direct-adsl.nl] has quit [Ping timeout: 121 seconds] |
| 15:54 | < catadroid> | Hrngh |
| 15:54 | < catadroid> | C++ templates are just awful macros |
| 15:54 | < catadroid> | I am annoyed |
| 16:00 | <@abudhabi> | Then use C++ macros. |
| 16:00 | < catadroid> | Those are neither C++ nor worthwhile macros |
| 16:01 | <@abudhabi> | Write in a language that compiles to C++! |
| 16:08 | | celmin|sleep is now known as celticminstrel |
| 16:20 | < catadroid> | Or I could just avoid using C++ |
| 16:26 | <&ToxicFrog> | Clearly the end result of this is going to be you writing a clojure to C++ compiler so that you can use clojure at work~ |
| 16:27 | <&ToxicFrog> | catadroid: as for structural sharing, git can do that, but it's more computationally expensive than just storing each blob as is |
| 16:27 | <&ToxicFrog> | (it does do structural sharing between trees) |
| 16:28 | <&ToxicFrog> | So it activates or not on a per-blob basis based on the space/time tradeoff |
| 16:31 | <&ToxicFrog> | Git objects are immutable, so if you create a new commit it actually behaves similarly to updating the contents of a clojure map or vector -- the parts that haven't changed are re-used and new objects are constructed to hold the parts that have. |
| 16:53 | | starkruzr [quassel@Nightstar-m85rjf.fios.verizon.net] has joined #code |
| 16:54 | | mode/#code [+ao starkruzr starkruzr] by ChanServ |
| 17:09 | <@celticminstrel> | I can't find any clues to why my text isn't drawing... |
| 17:11 | <@celticminstrel> | ...wait a second... |
| 17:13 | | * celticminstrel apparently told SDL_TTF to render using a colour with alpha 0... though fixing that hasn't helped... |
| 17:15 | <@celticminstrel> | I've managed to verify that SDL_TTF is rendering properly by temporarily disabling OpenGL and using SDL_BlitSurface. |
| 17:17 | <@celticminstrel> | So I guess it must be something about the texture code. |
| 17:18 | <@celticminstrel> | But I see nothing wrong when comparing it with various examples I've found online... |
| 17:19 | <@celticminstrel> | Hmm, if I pass 4 instead of GL_BGRA as the third argument to glTexImage2D, the quad is filled with the text's colour instead of its assigned colour... |
| 17:27 | <@celticminstrel> | Which makes me suspect the texture coordinates are wrong. |
| 17:31 | <@celticminstrel> | Fiddling with them doesn't seem to make any difference though. |
| 17:54 | | catadroid` [catadroid@Nightstar-qu4hm2.dab.02.net] has joined #code |
| 17:55 | | catadroid [catadroid@Nightstar-mk9nt0.dab.02.net] has quit [Ping timeout: 121 seconds] |
| 18:03 | <@celticminstrel> | :( |
| 18:06 | | catadroid` [catadroid@Nightstar-qu4hm2.dab.02.net] has quit [Ping timeout: 121 seconds] |
| 18:16 | <@celticminstrel> | The quad colour seems to be blending (sort of?) with the text colour even though the quad colour has 0 alpha... |
| 18:17 | <@celticminstrel> | (And the text colour still fills the quad for some reason.) |
| 18:38 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has quit [Ping timeout: 121 seconds] |
| 18:47 | | gnolam [lenin@Nightstar-t1tbf0.cust.bahnhof.se] has quit [Connection reset by peer] |
| 18:48 | | gnolam [lenin@Nightstar-t1tbf0.cust.bahnhof.se] has joined #code |
| 18:48 | | mode/#code [+o gnolam] by ChanServ |
| 19:51 | | Alek [Alek@Nightstar-9qtiqv.il.comcast.net] has quit [Ping timeout: 121 seconds] |
| 20:00 | | Alek [Alek@Nightstar-9qtiqv.il.comcast.net] has joined #code |
| 20:00 | | mode/#code [+o Alek] by ChanServ |
| 20:29 | | catadroid [catadroid@Nightstar-e0sppa.dab.02.net] has joined #code |
| 21:19 | | Shady [ShadyGuru@Nightstar-8v9css.leh.ptd.net] has joined #code |
| 21:41 | | Kindamoody|afk is now known as Kindamoody |
| 21:41 | | Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 21:41 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 22:04 | | Alek [Alek@Nightstar-9qtiqv.il.comcast.net] has quit [Ping timeout: 121 seconds] |
| 22:13 | | Alek [Alek@Nightstar-9qtiqv.il.comcast.net] has joined #code |
| 22:13 | | mode/#code [+o Alek] by ChanServ |
| 22:38 | <&McMartin> | std::lock_gourd |
| 22:38 | <&McMartin> | This doesn't work as well as one might like |
| 22:46 | <@gnolam> | Nah. That's only for CalaBash scripting. |
| 22:46 | | Turaiel[Offline] is now known as Turaiel |
| 23:01 | | * catadroid snrk |
| 23:05 | | Kindamoody is now known as Kindamoody[zZz] |
| 23:50 | <@celticminstrel> | What's wrong with it? |
| 23:51 | < catadroid> | It's more of a watermelon than you might expect |
| 23:56 | <&ToxicFrog> | m. |
| 23:56 | <&ToxicFrog> | Hmm., |
| 23:56 | <&ToxicFrog> | I find myself needing to implement undefok in my command line flag parsing library. |
| 23:56 | <&ToxicFrog> | This will require more work than I had hoped. |
| --- Log closed Sat Jul 23 00:00:18 2016 |