| --- Log opened Sun Nov 26 00:00:31 2017 |
| 00:12 | | McMartin [mcmartin@Nightstar-rpcdbf.sntcca.sbcglobal.net] has quit [[NS] Quit: leaving] |
| 00:56 | | Derakon[AFK] is now known as Derakon |
| 01:14 | | macdjord is now known as macdjord|slep |
| 01:27 | | McMartin [mcmartin@Nightstar-rpcdbf.sntcca.sbcglobal.net] has joined #code |
| 01:27 | | mode/#code [+ao McMartin McMartin] by ChanServ |
| 01:58 | | Alek [Alek@Nightstar-7or629.il.comcast.net] has quit [Ping timeout: 121 seconds] |
| 02:01 | | Alek [Alek@Nightstar-7or629.il.comcast.net] has joined #code |
| 02:01 | | mode/#code [+o Alek] by ChanServ |
| 02:31 | | Vornicus [Vorn@Nightstar-1l3nul.res.rr.com] has joined #code |
| 02:31 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 03:19 | | Alek [Alek@Nightstar-7or629.il.comcast.net] has quit [Ping timeout: 121 seconds] |
| 03:23 | | Alek [Alek@Nightstar-7or629.il.comcast.net] has joined #code |
| 03:23 | | mode/#code [+o Alek] by ChanServ |
| 03:28 | | Degi [Degi@Nightstar-dgfles.dyn.telefonica.de] has quit [Connection reset by peer] |
| 04:43 | | Vornlicious [Vorn@Nightstar-k8mmr0.sub-70-211-138.myvzw.com] has joined #code |
| 04:45 | | Vorntastic [Vorn@Nightstar-1l3nul.res.rr.com] has quit [Ping timeout: 121 seconds] |
| 05:12 | | Derakon is now known as Derakon[AFK] |
| 06:14 | | ppg [ouynx@Nightstar-b4p.ktq.236.134.IP] has joined #code |
| 06:14 | < ppg> | ARE YOU INTERESTED IN THE WHITE NATIONALIST MOVEMENT? |
| 06:14 | < ppg> | DO YOU WONDER WHY WHITE PEOPLE ARE SMARTER THAN NIGGERS? |
| 06:14 | < ppg> | PLEASE JOIN #/JOIN ON FREENODE IRC NETWORK AND MESSAGE |
| 06:14 | < ppg> | VAP0R ON HOW TO JOIN THIS GROWING MOVEMENT!! |
| 06:14 | | ppg [ouynx@Nightstar-b4p.ktq.236.134.IP] has quit [Z-Lined: Killbot has judged you to be a spammer. Begone. (ID: KL5VX41NP5)] |
| 06:37 | <@Alek> | well... it's debatable whether on average white people are smarter than black. and the only reason people like him don't drag down the white average is because there's plenty of black idiots too. |
| 07:25 | | Kindamoody[zZz] is now known as Kindamoody |
| 07:40 | | celticminstrel is now known as celmin|sleep |
| 07:44 | <@celmin|sleep> | s/GROWING MOVEMENT/CULT/ |
| 07:51 | <@Alek> | normally sure, but this is IIRC an Anon messing with the Vap0r guy. |
| 07:56 | < Vornlicious> | ^ |
| 08:21 | | Kindamoody is now known as Kindamoody|afk |
| 08:23 | < Vornlicious> | Thinking once more about programming in ways amenable to mobile input. I think a big issue is that standard text editing on mobile is utterly terrible; it would work vastly better if proper import |
| 08:27 | < Vornlicious> | Buttons are available, because right now typing a = b % 5 + 7 is like 13 keystrokes on a visual keyboard with several pages that you can't see until you toggle to them. |
| 08:28 | < Vornlicious> | Err, if proper input were available |
| 08:28 | | Kindamoody|afk [Kindamoody@Nightstar-eubaqc.tbcn.telia.com] has quit [Client exited] |
| 08:29 | < Vornlicious> | Never mind braces or any of the other madness. I think you'd want decent static analysis and parsing state tools to provide contextual data on what's available |
| 08:31 | | Kindamoody|autojoin [Kindamoody@Nightstar-eubaqc.tbcn.telia.com] has joined #code |
| 08:31 | | mode/#code [+o Kindamoody|autojoin] by ChanServ |
| 08:34 | < Vornlicious> | Oh God, some of the toys I've just thought of I want for regular programming. Show me as I'm coding what each variable in local scope looks like as I program |
| 08:36 | | Kindamoody|autojoin is now known as Kindamoody|afk |
| 08:36 | <&[R]> | That seems like it'd have the possibility of being very laggy |
| 08:39 | < Vornlicious> | Yeah you'd want to have it on its own thread with a thinking bar and a timeout |
| 08:39 | <@Alek> | I think there was one coding environment that promised that, some years back. |
| 08:40 | <@Alek> | It cost money though so I reluctantly abandoned all thought of it. |
| 08:42 | | Vornicus [Vorn@Nightstar-1l3nul.res.rr.com] has quit [Ping timeout: 121 seconds] |
| 09:06 | <&[R]> | Man, I'm really liking using the JS console to do some basic screen scraping. |
| 09:06 | <&[R]> | r=/([A-Za-z]*)-([0-9]*)\.([0-9]*)(\.([0-9]*))?\.tar\..z$/;JSON.stringify(Array.f rom(document.getElementsByTagName('a')).filter((a) => { return r.test(a.href) }).map((a) => { return r.exec(a.href) }).reduce((a, b) => { c = a[b[1]]; if (!c || c[2] < b[2] || c[3] < b[3]) { a[b[1]] = b; b.toJSON = function() { return this[0] } } return a }, {})) |
| 09:11 | < Vornlicious> | And then people do *that* and good gravy that looks terrible to even attempt in general but I know I've written the same level of madness |
| 09:13 | <&[R]> | Yeah, REPL and multiple line code don't mix well |
| 09:30 | | * McMartin finally gets around to writing something non-retro. |
| 09:30 | <&McMartin> | https://bumbershootsoft.wordpress.com/2017/11/26/gtk3-aspect-corrected-image-sca ling-with-opengl/ |
| 09:31 | <&McMartin> | Digging through some of my old papers, it turns out vertex shaders show up way earlier than I thought they did. |
| 09:32 | <&McMartin> | Also this probably would have looked better with MathML or something, but oh well |
| 09:36 | | * McMartin grumbles, fixes a bunch of typos and formatting infelicities. |
| 10:03 | | macdjord|slep is now known as macdjord |
| 10:24 | < Vornlicious> | Hooray aspect correction |
| 10:38 | <&McMartin> | Meanwhile, Cairo has an API that looks a lot like Legacy OpenGL, but in that one, (0,0) is in the upper left corner |
| 10:39 | <&McMartin> | That is A Bummer all on its own, but to add insult to injury, as far as we can tell it also then proceeds to not bother accelerating it |
| 10:40 | < Vornlicious> | I have determined that graphics setup also hates the living |
| 10:40 | <&McMartin> | Pretty much. |
| 10:40 | <&McMartin> | And the entire legacy API is like that |
| 10:41 | <&McMartin> | Manually disabling lighting/blending/materials, setting wind direction, etc. etc. |
| 10:41 | <&McMartin> | That stuff mostly carries over, but you get to skip a good chunk of it |
| 10:41 | <&McMartin> | A lot of the hungarian noise around it is basically manual mangling for a C API |
| 10:42 | <&McMartin> | so glUniform4f is really something more like gl::uniform(int, float, float, float, float); |
| 10:42 | <&McMartin> | and glUniform1i is gl::uniform(int, int); |
| 10:42 | <&McMartin> | I once likened programming legacy OpenGL to programming a robot to go toggle a program into a mainframe computer at the bottom of the ocean |
| 10:42 | < Vornlicious> | ...i read that as breeze not coil and it was silly |
| 10:43 | <&McMartin> | You did that for lights; particularly important when they were infinitely far away~ |
| 10:44 | <&McMartin> | (Which you could make happen by setting its perspective divisor to zero) |
| 10:44 | <&McMartin> | Anyway |
| 10:44 | < Vornlicious> | (i have written at some length on projective space) |
| 10:45 | <&McMartin> | Since that's all still there for setup, in the modern world it's more like standing in front of a wall-sized switchboard *yourself* and then toggling everything into the appropriate starting position. |
| 10:47 | <&McMartin> | ... so yeah, actually, you know what? |
| 10:47 | <&McMartin> | It doesn't hate the living. |
| 10:47 | <&McMartin> | It's just made of spiders. |
| 10:47 | <&McMartin> | *getting the functions to call* hates the living. |
| 10:47 | < Vornlicious> | Heh |
| 10:48 | <&McMartin> | OpenGL symbol loading is a horrific mess and includes such delights as "there is no guarantee that you get the same function back for the same symbol for two rendering contexts in the same program" |
| 10:48 | <&McMartin> | "Or that anything exists, like, at all, in the library, which will prevent your file from loading if you dare to treat it like a normal external symbol" |
| 10:49 | <&McMartin> | ("But if nothing exists, how can I get them out?" you ask. The functions for doing that are in a different, OS-specific, library.) |
| 10:50 | <&McMartin> | (The embedded space apparently has a standard for it, and Wayland apparently stole it for its own use for real OpenGL too. I don't know anything about either because to hell with touching that stuff myself; that's what other library writers are for) |
| 10:50 | < Vornlicious> | Ay |
| 10:51 | <&McMartin> | Offer void on macOS, where you totally do just link directly against OpenGL.framework, but you aren't allowed to use the same header files as literally everyone else. |
| 10:51 | <&McMartin> | And for which the GTK team has failed to get GL contexts to work, which baffles me because SDL has no trouble whatsoever, legacy -or- core. |
| 10:52 | <&McMartin> | And that's not jus tbecause you include <OpenGL/gl.h> instead of <GL/gl.h> |
| 10:53 | <&McMartin> | No, you need to include <OpenGL/gl3.h>, which is clearly marked as STOP USING THIS FILE ONCE OPENGL 3.0 IS STABLE |
| 10:53 | <&McMartin> | We are now at, uh |
| 10:53 | <&McMartin> | 4.6 |
| 10:53 | <&McMartin> | Though Zombie Steve has apparently forbidden compatibility beyond 4.1 |
| 10:54 | < Vornlicious> | I think zombie Steve hates the living at this point |
| 10:54 | <&McMartin> | (Not that this is a huge issue because it seems like the standard API target is 3.3) |
| 10:54 | <&McMartin> | (and if you want to do more than 3.3 provides, maybe bust out Vulkan) |
| 10:54 | | Reiver [quassel@ServerAdministrator.Nightstar.Net] has quit [Connection closed] |
| 10:54 | | Orthia [quassel@Nightstar-ksqup0.co.uk] has quit [Connection closed] |
| 10:55 | <&McMartin> | ... which in fact ends up making OpenGL 2.1, 3.3, and Vulkan map pretty cleanly to Direct3D 9.0c, 11, and 12, respectively~ |
| 10:55 | <&McMartin> | I glance briefly at Vulkan and was pleased to learn it did the OpenGL 2.1 thing there too |
| 10:56 | <&McMartin> | You can totally be Vulkan compatible by having all functions fail except for the one that fires up the whole system and the one that reports back as to how many GPU processors you have available that Vulkan can use |
| 10:56 | < Vornlicious> | There's a cab that stops here every morning. its number is 6502 |
| 10:56 | <&McMartin> | Because it is allowed to return 0 |
| 10:57 | <&McMartin> | (OpenGL 2.1 supports textures for vertex shaders, but "supports" here includes "doesn't support, but reports that 0 vertex texture units are present") |
| 10:58 | <&McMartin> | A fine number. There is likely a 6509 nearby! |
| 10:58 | | gnolam [lenin@Nightstar-ego6cb.cust.bahnhof.se] has joined #code |
| 10:58 | | mode/#code [+o gnolam] by ChanServ |
| 10:58 | <&McMartin> | Also gosh it is 3 AM, I should sleep |
| 10:58 | <&McMartin> | I spent more than half of Saturday in a Healing Coma, so this has clearly not helped my schedule any. |
| 11:01 | < Vornlicious> | Ni |
| 11:09 | | Orthia [quassel@Nightstar-ksqup0.co.uk] has joined #code |
| 11:09 | | mode/#code [+o Orthia] by ChanServ |
| 11:09 | | Reiver [quassel@Nightstar-ksqup0.co.uk] has joined #code |
| 11:09 | | mode/#code [+ao Reiver Reiver] by ChanServ |
| 12:12 | | Degi [Degi@Nightstar-dgfles.dyn.telefonica.de] has joined #code |
| 15:10 | | Derakon [Derakon@Nightstar-fr5qel.ca.comcast.net] has joined #code |
| 15:10 | | mode/#code [+ao Derakon Derakon] by ChanServ |
| 15:13 | | Derakon[AFK] [Derakon@Nightstar-fr5qel.ca.comcast.net] has quit [Ping timeout: 121 seconds] |
| 15:31 | | Vornicus [Vorn@Nightstar-1l3nul.res.rr.com] has joined #code |
| 15:31 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 15:42 | | celmin|sleep is now known as celticminstrel |
| 15:47 | | Vornlicious [Vorn@Nightstar-k8mmr0.sub-70-211-138.myvzw.com] has quit [Connection closed] |
| 15:47 | | Vorntastic [Vorn@Nightstar-1l3nul.res.rr.com] has joined #code |
| 15:50 | | RchrdB [RchrdB@Nightstar-qe9.aug.187.81.IP] has joined #code |
| 16:02 | <&[R]> | <McMartin> Meanwhile, Cairo has an API that looks a lot like Legacy OpenGL, but in that one, (0,0) is in the upper left corner <-- that always confused the hell out of me. Nothing else has you work in quadrant IV. Everything's quadrant I. |
| 16:03 | <~Vornicus> | technically it's still quadrant I |
| 16:04 | <~Vornicus> | since it's positive for both |
| 16:04 | <@Tamber> | it's just upside down? |
| 16:04 | <@ErikMesoy> | You read from the top left. Problem, comrade? :P |
| 16:04 | <~Vornicus> | but the handedness of the system is not what it is on paper |
| 16:04 | <&[R]> | Yeah, but it's where quadrant IV should be |
| 16:06 | < RchrdB> | I remember a document explaining PostScript's semantics describing working in quadrant I as "you wipe your feet before going up the stairs" |
| 16:06 | <&[R]> | wat |
| 16:06 | <&[R]> | What is that even supposed to mean? |
| 16:07 | <@ErikMesoy> | move along and then up? |
| 16:08 | < RchrdB> | wiping your feet is horizontal (unless you come from some kind of weird society in which people put floormats on their walls) and going up the stairs is… up. |
| 17:22 | <@celticminstrel> | I thought OpenGL was actually the exception in defaulting (0,0) at the bottom left of the window. |
| 17:23 | <&[R]> | Hmm |
| 17:23 | <@celticminstrel> | AFAIK all the older APIs place it at the top left as well, eg QuickDraw, SDL... |
| 17:23 | | * celticminstrel isn't sure about WinAPI. |
| 17:23 | <&[R]> | Yeah, I haven't done game coding in a long-ass time |
| 17:23 | <@celticminstrel> | This isn't necessarily just a game coding thing, mind you. |
| 17:23 | <&[R]> | I could very well have the facts mixed up |
| 17:23 | <&[R]> | I do know OpenGL is different from everything else though |
| 17:23 | <@celticminstrel> | IIRC Swing also has the origin at the top left. |
| 17:25 | <&[R]> | Speaking of weird X,Y axises. Minecraft. Any overhead map works via X,Z. |
| 17:26 | <@celticminstrel> | Yeah, that is indeed weird. |
| 18:26 | < RchrdB> | OpenGL is right-then-up too. The WebGL standard adds a `gl.UNPACK_FLIP_Y_WEBGL` flag for conveniently flipping the y axis when reading HTML images into textures because of this. |
| 18:27 | < RchrdB> | I think. |
| 18:27 | <~Vornicus> | minecraft follows one of the relatively standard bits of coordinate layout where y is up |
| 19:02 | | Vornicus [Vorn@Nightstar-1l3nul.res.rr.com] has quit [Ping timeout: 121 seconds] |
| 19:30 | <&jerith> | Most of the graphics APIs I've used put (0, 0) in the top left. |
| 19:30 | <&jerith> | Probably because scanlines. |
| 19:31 | | Kindamoody|afk is now known as Kindamoody |
| 20:07 | | himi [sjjf@Nightstar-v37cpe.internode.on.net] has quit [Connection closed] |
| 20:30 | | Reiv [NSkiwiirc@Nightstar-ih0uis.global-gateway.net.nz] has joined #code |
| 20:30 | | mode/#code [+o Reiv] by ChanServ |
| 20:32 | | Metalkeidran [MTKD@Nightstar-ko9rba.dyn.telefonica.de] has joined #code |
| 20:33 | | Metalkeidran [MTKD@Nightstar-ko9rba.dyn.telefonica.de] has left #code [] |
| 20:46 | <&McMartin> | Raster counts are the obvious reason for it, yes; smaller numbers appear earlier in the video signal |
| 20:47 | <&McMartin> | And, of course, on the earliest systems you're drawing a scanline at a time in software |
| 20:47 | <&McMartin> | Character cells are *always* 0,0 top left |
| 20:47 | <&McMartin> | The Spectrum, gloriously, has character cells 0,0 top left and pixels 0,0 bottom left |
| 20:47 | <&McMartin> | It's... honestly not a great look |
| 20:52 | <&McMartin> | Especially since bottom left can, IIRC change where it is depending on the size of the input buffer, so if you want to draw an accent on a character there isn't a single formula to get the location of the bitmap behind said character |
| 21:59 | | himi [sjjf@Nightstar-dm0.2ni.203.150.IP] has joined #code |
| 21:59 | | mode/#code [+o himi] by ChanServ |
| 22:14 | | Degi [Degi@Nightstar-dgfles.dyn.telefonica.de] has quit [Connection closed] |
| 22:17 | | Degi [Degi@Nightstar-dgfles.dyn.telefonica.de] has joined #code |
| 22:52 | <&jerith> | Yay, OM bonus missions complete. Except now I need to finish enough journal puzzles to get the new cheevo. |
| 22:52 | <&jerith> | But not tonight, being that it's 1am and I have work in the morning. |
| 22:53 | <&jerith> | McMartin: I also took the liberty of fixing a couple of situations where you had managed to get smaller numbers than me.~ |
| 23:02 | | Kindamoody is now known as Kindamoody[zZz] |
| 23:23 | | Degi_ [Degi@Nightstar-qqgrvp.dyn.telefonica.de] has joined #code |
| 23:26 | | Degi [Degi@Nightstar-dgfles.dyn.telefonica.de] has quit [Ping timeout: 121 seconds] |
| 23:40 | | Degi_ [Degi@Nightstar-qqgrvp.dyn.telefonica.de] has quit [[NS] Quit: Leaving] |
| --- Log closed Mon Nov 27 00:00:32 2017 |