| --- Log opened Sun May 13 00:00:17 2012 |
| 00:07 | | * ToxicFrog decides on a working name for his project: Warpcore |
| 00:25 | <&McMartin> | Ha ha ha ha ha |
| 00:25 | <&McMartin> | "You are expected to be able to read C and reasonable C++ code." |
| 00:25 | <&McMartin> | Emphasis mine |
| 00:35 | < Rhamphoryncus> | heh |
| 00:36 | <&McMartin> | What the Hell is this |
| 00:37 | | * McMartin takes the section on "Vector multiplication", has it taken out and shot |
| 00:37 | <~Vornicus> | Quite. |
| 00:39 | | * Rhamphoryncus gives you more ammo. You'll need it once you've started shooting bad explanations :P |
| 00:39 | <&McMartin> | This one isn't merely bad, it's false in every meaningful respect, and actively harmful. |
| 00:39 | <&McMartin> | There are two forms of vector multiplication, neither of which he presents, and both of which have geometrical interpretations. |
| 00:39 | < Rhamphoryncus> | heh |
| 00:40 | <&McMartin> | "I'm not going to assume you know any linear algebra [because if you did you would punch me in the face right now]" |
| 00:41 | < Rhamphoryncus> | hehe |
| 00:42 | < Rhamphoryncus> | I'm trying to understand depth buffer precision. It takes a float, normalizes it to 0..1, then stores that in an integer. However, if the float was 32 bit and the integer is also 32 bit what possible value is there in converting to an integer? You've already lost the precision by using a float |
| 00:42 | < Rhamphoryncus> | and you lose MORE by switching to an integer |
| 00:43 | <&McMartin> | If you're doing order comparisons, integer oprations will be simpler in hardware, I think. |
| 00:43 | < Rhamphoryncus> | hrm |
| 00:44 | <&McMartin> | This is for "what goes in the Z-buffer"? |
| 00:44 | < Rhamphoryncus> | yeah |
| 00:45 | <&McMartin> | Yeah, I bet it's so that you don't need a floating point unit to do occlusion tests on fragments |
| 00:45 | < Rhamphoryncus> | I'm planning to have a very large world (long view distances) that also lets you zoom in close. I know that's going to be problematic so I'm trying to educate myself on the options |
| 00:45 | <&McMartin> | Hrm |
| 00:46 | <&McMartin> | You shouldn't have to worry about this, since the clip space gets smaller (and thus depth precision increases) as you zoom in. |
| 00:46 | < Rhamphoryncus> | wait a sec |
| 00:46 | < Rhamphoryncus> | http://www.sjbaker.org/steve/omniv/love_your_z_buffer.html |
| 00:46 | < Rhamphoryncus> | The behaviour there isn't flat. It's still behaving like floating point |
| 00:47 | < Rhamphoryncus> | The intent is to let you look out the window of a train. That will leave your distance pretty high |
| 00:50 | < Rhamphoryncus> | So I think you're right about making the occlusion tests simpler |
| 00:50 | < Rhamphoryncus> | But I still don't know what the actual precision is |
| 00:51 | | * Vornicus thinks. a relatively simple bit transform should make floats directly comparable as integers... |
| 00:52 | < Rhamphoryncus> | It seems very biased towards near locations |
| 00:52 | <&McMartin> | That sounds like the right way to do it, tbh. |
| 00:52 | < Rhamphoryncus> | Perhaps it was designed to maximize what you get out of 16 bits and the extension to 24 and 32 was incidental |
| 00:53 | <&McMartin> | Also, stuff is likely to be "bunched up" more in the back so that the end result isn't visible to the user |
| 00:53 | <&McMartin> | Which speck of mud is in front back there isn't important; it's just mud |
| 00:53 | <&McMartin> | But which sheep is in front of which sheep next to the train is very important. |
| 00:53 | <~Vornicus> | (the multiplication presented there by the way is called the "cartesian product" and is used pretty much exclusively for non-square scaling) |
| 00:54 | < Rhamphoryncus> | Yeah, I'm not bothered by having a bias. I just get the impression that it's more than the float itself. I could be wrong though |
| 00:55 | <&McMartin> | The mapping may not be surjective. |
| 00:55 | <&McMartin> | (surjective?) |
| 00:55 | <&McMartin> | It might not use all the integers. |
| 00:56 | < Rhamphoryncus> | hrm. The sjbaker page does give the formula |
| 00:57 | < Rhamphoryncus> | which is.. a + b / z. a and b are constants derived from your znear/zfar. The result is then scaled by 2**depth |
| 01:01 | < Rhamphoryncus> | And poking at it in qalculate.. every doubling of distance gives you half as much precision |
| 01:02 | | Janus [NSwebIRC@Nightstar-afc5b963.res.rr.com] has joined #code |
| 01:03 | <&McMartin> | That sounds about right |
| 01:03 | < Rhamphoryncus> | Obviously that means the upper end is pretty close to insignificant |
| 01:04 | <&McMartin> | The closer you are to zfar, the closer you are to the pop point |
| 01:04 | <&McMartin> | If the user can distinguish errors back there then your zfar isn't ze far enough. |
| 01:05 | < Rhamphoryncus> | hrm |
| 01:06 | < Rhamphoryncus> | Add a couple orders of magnitude and you WILL gain more precision at long distances |
| 01:06 | <&McMartin> | Sure |
| 01:06 | <&McMartin> | I guess the danger is if you're shuffling cards in the foreground. |
| 01:07 | < Rhamphoryncus> | or only a negligible amount :/ |
| 01:07 | <&McMartin> | THis sounds like one of those things you shouldn't worry about until you can prove you have a proble, imo |
| 01:07 | <&McMartin> | *problem |
| 01:09 | < Rhamphoryncus> | I just want to understand it |
| 01:09 | < Rhamphoryncus> | And in particular I've seen mentions of having 32bit depth buffers, which could be a significant improvement |
| 01:12 | <~Vornicus> | (it still feels like you should be able to take an iee754 single and convert it with simple bit ops to things you can compare with integer comparison. it's probably a complement on the negatives so their values go largest to smallest. |
| 01:13 | | You're now known as TheWatcher[T-2] |
| 01:16 | < Rhamphoryncus> | 24 bit at 100km should only have a precision of around half a km, but 32 bit is down to only a few meters |
| 01:16 | < Rhamphoryncus> | So I'll have to experiment when I get to that point. It's a "maybe" |
| 01:17 | < Rhamphoryncus> | (8 bits -> 256 times the precision? Seems pretty obvious) |
| 01:19 | | You're now known as TheWatcher[zZzZ] |
| 01:20 | | * Vornicus pokes. Yes, okay, this can be done very simply. |
| 01:20 | <&McMartin> | (Tricky because 'precision' has a bunch of metrics, and they might sneak in some logarithmics somewhere) |
| 01:21 | <&McMartin> | (Not sure if 'resolution length' will be one of them) |
| 01:22 | <&McMartin> | Hmm, I wonder if I can get these gltut things running on my MBP. |
| 01:35 | < Rhamphoryncus> | main battle processor? |
| 01:35 | <&McMartin> | MacBook Pro. |
| 01:35 | <&McMartin> | Looks like the graphics card is just good enough |
| 01:37 | <&McMartin> | Oh C |
| 01:37 | <&McMartin> | I guess the other reason for "reasonable C++ code" is because C code in OpenGL is intrinsically unreasonable |
| 01:38 | <&McMartin> | Also, OK, looking through the first couple of chapters here |
| 01:39 | <&McMartin> | It's pretty clear that a good "modern OpenGL" project would simply be to rewrite the svaf renderer library to use buffer objects and vertex/fragment shaders instead of CVAs. |
| 01:39 | <&McMartin> | But I was using CVAs in a manner unnervingly similar to these VAOs. |
| 01:40 | <&McMartin> | There's only one call I'm not seeing yet |
| 01:41 | <&McMartin> | (CVAs had something like glDrawArrays but which let you submit the vertices repeatedly, out of order, for rendering purposes.) |
| 01:42 | <&McMartin> | Oh man, that is *dirty* |
| 01:42 | <&McMartin> | I'd better learn about geometry shaders before walking that path too far |
| 01:44 | | cpux|2 [cpux@Nightstar-c5874a39.dyn.optonline.net] has joined #code |
| 01:46 | | cpux [cpux@Nightstar-c5874a39.dyn.optonline.net] has quit [Ping timeout: 121 seconds] |
| 01:46 | | Noah [maoranma@Nightstar-cabc1a29.pools.spcsdns.net] has joined #code |
| 01:47 | <&McMartin> | Hmm. Looks like OpenGL 3.x deprecated explicitly defined outputs for color and normal from the vertex shader, at least if this tutorial isn't playing silly buggers again. |
| 01:48 | < Noah> | How do you color them? Draw with with dryerase on the screen? |
| 01:50 | <&McMartin> | Pass a 3-vector to the fragment shader by whatever means (possibly as a defined output of the vertex shader!) that the fragment shader may then interpret as a color, should it so choose. |
| 01:50 | < Noah> | I see *puts markers away( |
| 01:51 | <&McMartin> | Oh hey |
| 01:51 | <&McMartin> | My respect for this tutorial is marginally restored |
| 01:51 | < Noah> | Oh good |
| 01:52 | < Noah> | Now you're life can continue undaunted by such silly things such as bad tutorial |
| 01:52 | < Noah> | TheNewBoston gave me a lot of info about Python, but if I have to slice roastbeef[:] again... |
| 01:53 | <&McMartin> | Heh |
| 01:53 | <&McMartin> | Well, here, it's "oh, it talks about gimbal lock and how to not have it, hooray" |
| 01:54 | < Noah> | What the hell is a gimbal lock? |
| 01:54 | <&McMartin> | "Gimbal lock" is how homing-missile-style setups go Horribly Wrong (tm) if you store their position as a pair of rotations. |
| 01:55 | < Noah> | Ah |
| 01:55 | <&McMartin> | A gimbal is one of those steampunk gyroscope mount things. |
| 01:55 | <~Vornicus> | Noah: open an FPS |
| 01:55 | <~Vornicus> | Look straight up |
| 01:55 | <~Vornicus> | Spin around. |
| 01:55 | <~Vornicus> | Note how what you're aiming at doesn't change in the slightest. |
| 01:55 | < Noah> | Ah |
| 01:55 | <~Vornicus> | That's gimbal lock. |
| 01:57 | <&McMartin> | And it means that if you're using FPS-like controls for a game agent and it has an AI routine like "now go point at target X", if it has to change direction it will freak the fuck out |
| 01:57 | < Noah> | And if the coding in the game is so bad that your aimer DOES move? |
| 01:57 | <&McMartin> | It's actually deliberately blocking 'straight up' to prevent this, most likely. |
| 01:57 | < Noah> | I see |
| 01:57 | <&McMartin> | Gimbal lock is considered "bad" |
| 01:57 | < Noah> | Like a tank gun |
| 01:57 | <&McMartin> | Right. |
| 01:58 | < Janus> | I'm sure an alternative is to allow the AI to bend backwards and shoot upsidedown |
| 01:58 | < Noah> | that's what I do |
| 01:58 | < Noah> | then I get banned |
| 01:58 | <&McMartin> | The correct solution basically involves putting an invisible globe around your homing missile or whatever and plot a course on it from where it's pointed and where it's intended to point, then follow that path. |
| 01:58 | <&McMartin> | It turns out that you can model that with an extention to complex numbers called "quaternions", which this tutorial also covers |
| 01:58 | <&McMartin> | So good for it! |
| 01:58 | <&McMartin> | MCMARTIN BOOMS, "I AM WELL PLEASED" |
| 02:01 | < Noah> | quaternions, I've heard that before somewhere |
| 02:02 | < Noah> | Oh shit |
| 02:02 | < Noah> | It's got numbers like i, k, AND j? |
| 02:02 | <&McMartin> | And all of them are square roots of -1! |
| 02:03 | < Noah> | FFFFF |
| 02:03 | <&McMartin> | This is one of those things like exponentials of imaginary numbers that turn out to be really handy when used in specific ways. |
| 02:03 | < Noah> | Such as making your FPS ai NOT freak out |
| 02:04 | <&McMartin> | It's more important for flight sims, actually |
| 02:04 | <&McMartin> | You want enemy ships to turn to point at the target and not try to simultaneously turn backflips and end up flying 45 degrees off of horizontal while upside-down as a result |
| 02:05 | < Noah> | It seems like I've seen that issue in a FPS before, when an AI goes to turn around it just swings up then rotates |
| 02:05 | < Noah> | Maybe that was just more bad programming |
| 02:06 | <&McMartin> | That's a pretty classic case of gimbal lock, though |
| 02:06 | <&McMartin> | I mean, you're kind of restating it. |
| 02:06 | < Noah> | Oh |
| 02:06 | < Noah> | Oh right, I get it now |
| 02:06 | <&McMartin> | Setting up your aiming system so that gimbal lock can happen is bad programming, and wildly changing height while trying to yaw is a major part of the aforementioned the-fuck-outfreaking. |
| 02:07 | < Noah> | The gimbal lock is the alternative to using a quaternion complex number plane |
| 02:07 | < Noah> | And it's bad |
| 02:08 | <&McMartin> | Right |
| 02:08 | <&McMartin> | Though it *is* easier to implement in terms of glRotate. |
| 02:08 | < Noah> | Which is why everyone and your mom does it |
| 02:08 | <&McMartin> | Which isn't a function anymore in OpenGL, perhaps precisely because it encourages this. |
| 02:08 | <&McMartin> | Right. |
| 02:10 | <&McMartin> | This is also reminding me that I never really got the hang of impostors |
| 02:10 | <&McMartin> | And I'm wondering if fragment shaders mean now I don't have to. |
| 02:22 | <~Vornicus> | "impostors"? |
| 02:23 | <~Vornicus> | Note that in fps, gimbal lock is in a certain sense unavoidable - that's how humans work, approximately. |
| 02:34 | < Rhamphoryncus> | McMartin: I don't think most FPS's can have gimbal lock because they don't have roll. Pitch and yaw are controlled independently |
| 02:37 | < Rhamphoryncus> | imposters aka billboards aka sprites. There's some distinction but that's the basic idea |
| 02:39 | < Rhamphoryncus> | And AI might exhibit gimbal lock if you jump over them, but you could just as well call that modeling of human behaviour.. and I expect they'd cheat and turn fast enough to keep up just fine |
| 02:43 | < Rhamphoryncus> | eh, wikipedia disagrees with me and says 2d is still gimbal lock |
| 02:45 | <~Vornicus> | ah those. |
| 02:45 | <&McMartin> | Yeah, basically, the trick I never mastered was "put stuff in clip space that's oriented in camera space |
| 02:47 | < Rhamphoryncus> | with an imposter you mean? |
| 02:48 | <&McMartin> | Yeah |
| 02:48 | < Noah> | I never look up when someone jumps over me in a FPS, I spin around while moving opposite the direction they're moving, they gotta land after all, and bullet will be waiting |
| 02:49 | < Rhamphoryncus> | Off hand.. apply coordinates to camera matrix to find center point, then draw without the camera matrix |
| 02:49 | <&McMartin> | Ofc, in TF2 that won't work because they'll have by then rocket-jumped off your hat |
| 02:49 | <&McMartin> | Rhamphoryncus: Yeah, something like that. |
| 02:49 | <@rms> | http://www.telusplanet.net/public/stonedan/source.txt <-- SSDS (It's VB 5 code. Yes 5. Does not work on 6.) Good for a laugh, or a cry. |
| 02:50 | < Rhamphoryncus> | rms: it's on telus. I'm already trying to decide between laughing and crying |
| 02:50 | <@rms> | Heh |
| 02:51 | <&McMartin> | SSDS? |
| 02:51 | < Noah> | No, they would've jumped into my hats stacked ontop of each other like so much Hatris |
| 02:51 | <@rms> | SpectateSwamp Desktop Search |
| 02:51 | <@rms> | Either the guy is completely retarded (but able to "code"); or a really dedicated troll. |
| 02:52 | < Rhamphoryncus> | wait, wasn't that a guy that posted on thedailywtf? |
| 02:52 | <@rms> | Yes |
| 02:52 | < Noah> | Some trolls are highly dedicated, so that's not an easy call |
| 02:54 | < Rhamphoryncus> | I'm surprised thedailywtf is still up. In one piece that is |
| 02:54 | <@rms> | Heh |
| 02:55 | <@rms> | The unofficial channel merged into #codelove on Slashnet a while ago |
| 02:55 | < Rhamphoryncus> | Everybody must be saying "naw, too easy, too obvious" |
| 03:04 | | Noah [maoranma@Nightstar-cabc1a29.pools.spcsdns.net] has quit [Ping timeout: 121 seconds] |
| 03:05 | | Noah [maoranma@490720.C448F4.1BE4BF.6CE5B6] has joined #code |
| 03:14 | | Kindamoody[zZz] is now known as Kindamoody |
| 03:16 | | Attilla [Obsolete@Nightstar-604551a6.as43234.net] has quit [Ping timeout: 121 seconds] |
| 04:03 | | Janus [NSwebIRC@Nightstar-afc5b963.res.rr.com] has quit [Ping timeout: 121 seconds] |
| 05:04 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 05:04 | | celticminstrel [celticminst@Nightstar-5d22ab1d.cable.rogers.com] has quit [[NS] Quit: KABOOM! It seems that I have exploded. Please wait while I reinstall the universe.] |
| 05:17 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 05:17 | | mode/#code [+o himi] by ChanServ |
| 06:31 | | Kindamoody is now known as Kindamoody|afk |
| 06:32 | | Derakon is now known as Derakon[AFK] |
| 06:34 | | AnnoDomini [annodomini@A08927.B4421D.B81A91.464BAB] has joined #code |
| 06:35 | | mode/#code [+o AnnoDomini] by ChanServ |
| 06:42 | | Kindamoody|afk is now known as Kindamoody |
| 06:57 | | ErikMesoy|sleep is now known as ErikMesoy |
| 07:43 | | AnnoDomini [annodomini@A08927.B4421D.B81A91.464BAB] has quit [Ping timeout: 121 seconds] |
| 08:22 | | eckse [eckse@Nightstar-3b021371.dsl.sentex.ca] has quit [Connection reset by peer] |
| 09:40 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 09:53 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 09:53 | | mode/#code [+o himi] by ChanServ |
| 10:06 | | Rhamphoryncus [rhamph@Nightstar-5697f7e2.abhsia.telus.net] has quit [Client exited] |
| 12:31 | | jeroid [jerith@17C60D.A0B60C.F8FD5C.75B227] has joined #code |
| 12:31 | < jeroid> | o/ |
| 12:31 | < jeroid> | In line at customs in Washington DC. |
| 12:32 | < jeroid> | Arrived late, connection leaves in 30 minutes. Exciting times. |
| 12:36 | | jeroid [jerith@17C60D.A0B60C.F8FD5C.75B227] has quit [Ping timeout: 121 seconds] |
| 12:38 | | jeroid [jerith@17C60D.A0B60C.F8FD5C.75B227] has joined #code |
| 12:39 | | Kindamoody is now known as Kindamoody|afk |
| 12:43 | | jeroid_ [jerith@17C60D.A0B60C.F8FD5C.75B227] has joined #code |
| 12:43 | | jeroid [jerith@17C60D.A0B60C.F8FD5C.75B227] has quit [Client closed the connection] |
| 12:46 | | jeroid_ is now known as jeroid |
| 12:57 | | jeroid [jerith@17C60D.A0B60C.F8FD5C.75B227] has quit [Client closed the connection] |
| 13:00 | | Attilla [Obsolete@Nightstar-02604040.as43234.net] has joined #code |
| 13:17 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 13:29 | | You're now known as TheWatcher |
| 13:29 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 13:30 | | mode/#code [+o himi] by ChanServ |
| 13:49 | < sshine> | has anyone here used k-dimensional trees for effectively accessing many points in high dimensions (3+)? |
| 13:52 | < sshine> | I've only read the Wikipedia article ( http://en.wikipedia.org/wiki/K-d_tree ) and used the datastructure a while back. I wonder if it's always the case that dimensions are cycled through linearly, or if it makes sense to partition the space in some other way that better relates to the weight of the space. |
| 13:55 | | Kindamoody|afk is now known as Kindamoody |
| 13:58 | < sshine> | ah, I'm lazy. they're called adaptive k-d trees |
| 14:30 | | gnolam [lenin@Nightstar-202a5047.priv.bahnhof.se] has joined #code |
| 14:30 | < gnolam> | http://alan.dipert.org/post/153430634/the-march-of-progress |
| 14:58 | < ShellNinja> | Hahaha. |
| 14:58 | | * EvilDarkLord wtfs, finds that his Ubuntu installation appears to have vanished after rebooting. The relevant partition seems to be almost empty, and the installer doesn't detect it being installed at all. And GRUB is crashing to boot. |
| 14:59 | < EvilDarkLord> | Well, I had backups for all except my most recent work, no biggie. Just in case, though, anything obvious I should try before reinstalling? |
| 15:02 | < sshine> | EvilDarkLord, are you sure it's not just the partition table that is gone? |
| 15:02 | < EvilDarkLord> | sshine: I can't say. How would I check? Read random places from the partition and see if there's anything left? |
| 15:03 | < sshine> | EvilDarkLord, or recreate it if you know exactly where it was. |
| 15:05 | < EvilDarkLord> | sshine: The partition (ntfs) still appears to be present, it just has very little stuff on it. |
| --- Log closed Sun May 13 15:11:13 2012 |
| --- Log opened Sun May 13 15:11:31 2012 |
| 15:11 | | TheWatcher [chris@Nightstar-3762b576.co.uk] has joined #code |
| 15:11 | | Irssi: #code: Total of 24 nicks [7 ops, 0 halfops, 0 voices, 17 normal] |
| 15:11 | | mode/#code [+o TheWatcher] by ChanServ |
| 15:12 | | Irssi: Join to #code was synced in 36 secs |
| 15:17 | | Kindamoody is now known as Kindamoody|out |
| 15:17 | | * EvilDarkLord reads across it, finds insufficient nonzero stuff to account for an installation. |
| 15:42 | <@ToxicFrog> | EvilDarkLord: er, you installed Ubuntu to an NTFS partition? |
| 15:42 | <@ToxicFrog> | I didn't know that was supported. |
| 15:42 | <@ToxicFrog> | Also, if you can still see the partition at all, it's not that the partition table is gone (if it were you would probably want to use testdisk to recover it) |
| 15:48 | < EvilDarkLord> | ToxicFrog: Sort of. You can install it inside an NTFS partition, but you cannot see the contents trivially from inside Windows. |
| 16:08 | <@ToxicFrog> | Oh, it installs a filesystem image or something |
| 16:08 | <@ToxicFrog> | Did you install it under windows using Wubi? |
| 16:22 | < EvilDarkLord> | I think I figured out why it broke. Suffice to say it was embarrassing. |
| 16:24 | < EvilDarkLord> | I installed it once under Wubi, not this time though. |
| 16:39 | | ToxicFrog [ToxicFrog@ServerAdministrator.Nightstar.Net] has quit [Connection reset by peer] |
| 16:40 | | ToxicFrog [ToxicFrog@ServerAdministrator.Nightstar.Net] has joined #code |
| 16:40 | | mode/#code [+o ToxicFrog] by ChanServ |
| 16:43 | | eckse [eckse@Nightstar-3b021371.dsl.sentex.ca] has joined #code |
| 16:44 | | mode/#code [+o eckse] by ChanServ |
| 17:07 | | ToxicFrog [ToxicFrog@ServerAdministrator.Nightstar.Net] has quit [Operation timed out] |
| 17:46 | | Derakon[AFK] is now known as Derakon |
| 18:05 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 18:05 | | Rhamphoryncus [rhamph@Nightstar-5697f7e2.abhsia.telus.net] has joined #code |
| 18:15 | | ToxicFrog [ToxicFrog@ServerAdministrator.Nightstar.Net] has joined #code |
| 18:15 | | mode/#code [+o ToxicFrog] by ChanServ |
| 18:19 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 18:19 | | mode/#code [+o himi] by ChanServ |
| 18:45 | | Noah [maoranma@490720.C448F4.1BE4BF.6CE5B6] has quit [Client closed the connection] |
| 18:45 | | Noah [maoranma@490720.C448F4.1BE4BF.6CE5B6] has joined #code |
| 19:21 | | jeroid [jerith@83BF1D.4E7445.8DFF6A.5145AD] has joined #code |
| 19:23 | | Kindamoody|out is now known as Kindamoody[zZz] |
| 19:28 | | jeroid_ [jerith@83BF1D.4E7445.8DFF6A.5145AD] has joined #code |
| 19:28 | | jeroid [jerith@83BF1D.4E7445.8DFF6A.5145AD] has quit [Client closed the connection] |
| 19:35 | | jeroid [jerith@83BF1D.4E7445.8DFF6A.5145AD] has joined #code |
| 19:35 | | jeroid_ [jerith@83BF1D.4E7445.8DFF6A.5145AD] has quit [Connection reset by peer] |
| 19:38 | | jeroid_ [jerith@83BF1D.4E7445.8DFF6A.5145AD] has joined #code |
| 19:39 | | jeroid [jerith@83BF1D.4E7445.8DFF6A.5145AD] has quit [Client closed the connection] |
| 19:39 | | jeroid_ [jerith@83BF1D.4E7445.8DFF6A.5145AD] has quit [[NS] Quit: Bye] |
| 19:47 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 20:00 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 20:00 | | mode/#code [+o himi] by ChanServ |
| 21:49 | | Noah [maoranma@490720.C448F4.1BE4BF.6CE5B6] has quit [Ping timeout: 121 seconds] |
| 21:51 | | ErikMesoy is now known as ErikMesoy|sleep |
| 22:23 | | jeroid [jerith@Nightstar-7658afb4.tmodns.net] has joined #code |
| 23:09 | | * TheWatcher eyes github |
| 23:13 | <@TheWatcher> | Oh, right, that's why the icons have all gone to shit. Brilliant/ |
| 23:20 | <@rms> | What |
| 23:21 | <@TheWatcher> | They've replaced the old graphical icons with a custom font |
| 23:21 | <@TheWatcher> | "Octicons Regular" |
| 23:21 | <@TheWatcher> | contains glyphs that are the icons |
| 23:21 | <@TheWatcher> | which is fucking mental |
| 23:22 | <@rms> | :/ |
| 23:23 | <@TheWatcher> | Especially when you've turned off web fonts because you don't like the crap some web designers pull with them |
| 23:24 | <@rms> | Ha |
| 23:33 | <@TheWatcher> | Aaand, I can't find a download of an actual bloody .ttf/.pfb/.afm, either |
| 23:33 | < Rhamphoryncus> | Can't you use a svg as an image now? |
| 23:33 | <@rms> | svg is an image. |
| 23:53 | | jeroid [jerith@Nightstar-7658afb4.tmodns.net] has quit [[NS] Quit: Bye] |
| 23:55 | | * TheWatcher fixes it, by converting the woff to ttf, and using a user stylesheet |
| --- Log closed Mon May 14 00:00:37 2012 |