--- Log opened Sun Jul 27 00:00:17 2008 |
00:00 | <@McMartin> | Didn't we have this discussion yesterday~ |
00:01 | <@McMartin> | (And the answer is: "it would be nice if the language knew about function signatures") |
00:01 | <@McMartin> | (Or at least had a separate type for "callable") |
00:08 | <@MyCatVerbs> | McMartin: pretty much. |
00:09 | <@MyCatVerbs> | McMartin: I think the final verdict is, "it'd be nice if there was an alternative to dlopen() which actually paid attention to types". |
00:09 | <@MyCatVerbs> | (As a compromise, I mean. Since that takes less work (in terms of implementation and breaking old code) than either shifting away from C to soem |
00:09 | <@MyCatVerbs> | gah. |
00:10 | <@MyCatVerbs> | to something saner, like Java, or than attempting to formally prove that all your programs never trash the heap or stack. :) |
00:10 | <@McMartin> | Or use sane hardware that does that enforcement, but then you're breaking a lot of code that uses trampolines. |
00:11 | < Shoukanjuu> | SLI or Crossfire? |
00:12 | < Shoukanjuu> | I mean, it's the same thing with different cards |
00:13 | < Shoukanjuu> | But if I were to spend the same amount to get a pair of cards, ATI or Nvidia, which would have better performance? [/idiocy] |
00:14 | <@McMartin> | That varies by era. |
00:14 | < Shoukanjuu> | Current era. |
00:14 | <@McMartin> | "Since the last time I checked was like five years ago, don't ask me" |
00:14 | <@McMartin> | Also target OS, as which drivers explode in your face all the time varies by era. |
00:14 | < Shoukanjuu> | Since the forgone conclusion of Macsoft releasing GoW and UT3 is taking entirely too long |
00:14 | <@McMartin> | Unless you are TF, nVidia drivers seem to be more stable. |
00:15 | < Shoukanjuu> | I was going to attempt to swallow my pride and make a windowsbox |
00:15 | <@McMartin> | Go XP if you do |
00:15 | < Shoukanjuu> | Well, of course |
00:16 | < Shoukanjuu> | It's still not the right time to switch to vista yet |
00:16 | < Shoukanjuu> | from XP |
00:16 | < Shoukanjuu> | Maybe about the time when Microsoft releases another OS, Vista will be as supported as XP still is |
00:17 | < Shoukanjuu> | Another person I was consulting, however, remarked about how 64 bit XP sucks, and so I should just go for Vista |
00:17 | < Shoukanjuu> | But I wasn't paying much attention, because I rarely do when I've been up for upwards of 40 hours |
00:17 | < Shoukanjuu> | And I didn't know exactly why |
00:18 | < Shoukanjuu> | And now, since Colloquy doesn't know how to arrange logs correctly, all the conversation has been lost in walls of irrelevant text \o/ |
00:20 | <@McMartin> | Hum |
00:20 | <@McMartin> | I believe it is true that you're better off with 32-bit XP than any other MS OS right now, at least if you mainly want to play games. |
00:20 | <@McMartin> | Windows 7 may mean Vista becomes as well-supported as XP, or it may mean Vista becomes stillborn like Me. |
00:21 | < Shoukanjuu> | What exactly are the (de)merits of 64 bit OS? |
00:21 | <@McMartin> | merits: Faster vector operations = often faster graphics performance |
00:21 | <@McMartin> | demerits: code stability, driver availability |
00:21 | <@McMartin> | Many applications are not 64-bit clean and either won't run or will need to run in emulation. |
00:21 | < Shoukanjuu> | So not much that I don't know |
00:22 | <@McMartin> | If you really need the integer capability on a 64-bit machine for scientific computation or w/e go with Linux generally and Fedora specifically |
00:22 | <@McMartin> | Fedora has, IME, the best 64&32-bit-simultaneous install I've dealt with. |
00:22 | <@McMartin> | As of about 7 or so. |
00:22 | < Shoukanjuu> | Cool |
00:22 | < Shoukanjuu> | I'll go with 32bit XP for now |
00:22 | <@McMartin> | (There are a few wobbles in the initial install but after that yum keeps everything straight.) |
00:23 | < Shoukanjuu> | Since I happen to have a license key and the disc with me right...here |
00:23 | <@McMartin> | Yeah. My 64-bit machine uses 32-bit windows when it's in it. |
00:23 | < Shoukanjuu> | somewhere :/ |
00:23 | <@McMartin> | Ah yes. Give Mr. B. Torrent my regards. |
00:24 | <@Vornicus> | hee |
00:24 | < Shoukanjuu> | How dare you! I did nothing of the sort >_> |
00:24 | < Shoukanjuu> | newgroups :B |
00:25 | <@McMartin> | I have no reason to doubt your word |
00:25 | < Shoukanjuu> | I'm pretty sure that's where my dad got it from, but I confirmed it working and able to accept updates |
00:25 | < Shoukanjuu> | So *shrug* |
00:25 | < Shoukanjuu> | Doesnt' matter where it came from as long as it works |
01:05 | | androsch [~androsch@Nightstar-5896.pools.arcor-ip.net] has joined #code |
01:34 | | androsch [~androsch@Nightstar-5896.pools.arcor-ip.net] has quit [Ping Timeout] |
02:05 | | Thaqui [~Thaqui@Nightstar-13764.jetstream.xtra.co.nz] has joined #code |
02:05 | | mode/#code [+o Thaqui] by ChanServ |
03:03 | | Attilla [~The.Attil@92.8.206.ns-26222] has quit [Quit: <Insert Humorous and/or serious exit message here>] |
07:43 | < Doctor_Nick> | Shoukanjuu: install linux porblem sorlved |
07:45 | | Vornicus is now known as Vornicus-Latens |
08:34 | | AnnoDomini [AnnoDomini@Nightstar-29006.neoplus.adsl.tpnet.pl] has joined #Code |
08:34 | | mode/#code [+o AnnoDomini] by ChanServ |
09:24 | | You're now known as TheWatcher |
10:11 | | Attilla [~The.Attil@92.8.206.ns-26222] has joined #code |
10:11 | | mode/#code [+o Attilla] by ChanServ |
10:54 | <@McMartin> | Doctor_Nick: "Problem" was "Play Unreal Tournament 3", so fuck no. |
11:01 | < Doctor_Nick> | ha |
11:01 | < Doctor_Nick> | it's been over 8 months since they promised the linux client |
11:02 | <@McMartin> | 16:16 < Shoukanjuu> Since the forgone conclusion of Macsoft releasing GoW and UT3 is taking entirely too long |
11:03 | | Thaqui [~Thaqui@Nightstar-13764.jetstream.xtra.co.nz] has left #code [MORE constitution LESS destitution MORE pros...perity.] |
11:03 | < Doctor_Nick> | the guy who is porting them says its some poltical hangup |
13:22 | | You're now known as SnugglesZombie |
13:23 | | You're now known as TheWatcher |
13:53 | < Shoukanjuu> | >_> |
13:53 | < Shoukanjuu> | Is there a link for that? |
13:56 | < Shoukanjuu> | Ah, it seems like they need opengl specialists |
13:58 | < Shoukanjuu> | And there's the 'political hangup' |
16:11 | | Attilla [~The.Attil@92.8.206.ns-26222] has quit [Ping Timeout] |
16:13 | | Attilla [~The.Attil@92.8.206.ns-26222] has joined #code |
16:13 | | mode/#code [+o Attilla] by ChanServ |
16:28 | | Vornicus-Latens [~vorn@Admin.Nightstar.Net] has quit [Ping Timeout] |
16:37 | | Vornotron [~vorn@Admin.Nightstar.Net] has joined #code |
16:37 | | Vornotron is now known as Vornicus |
16:47 | | Attilla_ [~The.Attil@92.8.206.ns-26222] has joined #code |
16:47 | | Attilla [~The.Attil@92.8.206.ns-26222] has quit [Ping Timeout] |
16:49 | | Attilla_ is now known as Attill |
16:49 | | Attill is now known as Attilla |
17:08 | | * TheWatcher ponders |
17:09 | <@TheWatcher> | Opinions on a c++ style question - empty destructor: explicitly show it as "~Foo() { } // do nothing" or just leave it out completely |
17:09 | <@TheWatcher> | I generally do the latter, but.. |
17:13 | < Vornicus> | The latter. |
17:13 | | * TheWatcher nods |
17:54 | <@MyCatVerbs> | TheWatcher: if there's some specific reason why the destructor is empty, then put the empty destructor in and a comment about why. |
17:55 | <@MyCatVerbs> | TheWatcher: personally, if I were working on C++, I'd always put the empty destructors in to save peoples' sanity when they start browsing my source. But that's just me, and do be aware that I've been reading more code than writing it, lately. :) |
17:56 | <@MyCatVerbs> | (Save people ever wondering "where the Heck is the destructor for this thing?", naturally.) |
18:01 | | AnnoDomini is now known as Elizabeth |
18:37 | | EvilDarkLord is now known as Rico |
19:09 | | Vornotron [~vorn@Admin.Nightstar.Net] has joined #code |
19:10 | | Vornicus [~vorn@Admin.Nightstar.Net] has quit [Ping Timeout] |
19:10 | | Vornotron is now known as Vornicus |
19:25 | | Vornicus [~vorn@Admin.Nightstar.Net] has quit [Ping Timeout] |
19:30 | | Vornotron [~vorn@Admin.Nightstar.Net] has joined #code |
19:33 | | Vornotron is now known as Vornicus |
20:10 | <@McMartin> | MCV: Aren't C++ destructors nonvirtual if you don't declare them? |
20:11 | <@McMartin> | Which in turn makes you class fundamentally incapable of being subclassed safely, producing a wide variety of hilarious and wacky errors? |
20:11 | <@jerith> | That sounds very C++y... |
20:43 | | AnnoDomini [AnnoDomini@Nightstar-28988.neoplus.adsl.tpnet.pl] has joined #Code |
20:43 | | mode/#code [+o AnnoDomini] by ChanServ |
20:43 | | AnnoDomini is now known as Liz |
20:44 | | Elizabeth [AnnoDomini@Nightstar-29006.neoplus.adsl.tpnet.pl] has quit [Killed (NickServ (GHOST command used by Liz))] |
20:44 | | Liz is now known as Elizabeth |
20:49 | <@TheWatcher> | McMartin: c++ destructures are nonvirtual unless you explicitly declare them to be virtual |
20:49 | <@TheWatcher> | *destructors |
20:50 | <@TheWatcher> | And since this is for a subtractive method random number generator, I strongly suspect there is little to no chance of it being subclassed |
20:50 | <@McMartin> | The question was whether implicit destructors |
20:50 | <@McMartin> | Er, whether implicit destructors had this. |
20:50 | <@McMartin> | Since implicit copy constructors, IIRC, do. |
20:51 | <@TheWatcher> | I;m honestly not sure, when writing classes I expect to be subclassed, I always bung in a virtual destructor |
20:58 | | Vornicus is now known as Finerty |
21:18 | <@MyCatVerbs> | McMartin: ohhh yes, now I remember. |
21:19 | <@MyCatVerbs> | McMartin: Bruce Eckel reccommends always giving an empty virtual destructor for exactly that reason, I think. |
21:20 | <@MyCatVerbs> | McMartin: sorry for forgetting that, but I don't hack in C++... pretty much at all. Resultingly most of my knowledge of its brain damage has been paged out into longer-term memory (or just dropped on the floor). |
21:23 | | Attilla [~The.Attil@92.8.206.ns-26222] has quit [Ping Timeout] |
21:27 | | Attilla [~The.Attil@92.10.170.ns-10975] has joined #code |
21:27 | | mode/#code [+o Attilla] by ChanServ |
21:33 | | * McMartin wishes C++ could be laid to an honorable death. |
21:33 | <@Elizabeth> | Or even a dishonorable one. |
21:33 | <@McMartin> | I should learn more about D to see if it handles the things C++ needs remotely acceptably. |
21:33 | | * jerith would prefer a crossroad at midnight. |
21:33 | <@jerith> | I'll even bring the stake. |
21:34 | <@McMartin> | No, it had its place, and even now that place isn't merely "an unmarked grave" |
21:34 | < Finerty> | Speaking of C, I recently had the unique displeasure of reading some LPC code. |
21:34 | <@McMartin> | Let's Play C? |
21:34 | <@Elizabeth> | LOL. |
21:34 | < Finerty> | Lars Pensjo C |
21:34 | < Finerty> | Which looks like C. But is decidedly not C. |
21:34 | <@Elizabeth> | I'd prefer Let's Play Battletoads. :P |
21:34 | < Shoukanjuu> | XD |
21:35 | < Shoukanjuu> | "Let's Play C++" |
21:35 | | * McMartin recommends Let's Play GTA:SA again |
21:35 | | * Shoukanjuu attempts to make a video of him messing with C++ |
21:35 | < Finerty> | It is enough like C, superficially, that I thought I was looking at logic errors. |
21:36 | < Finerty> | Apparently (among other things) sizeof works like python's len |
21:36 | <@McMartin> | >_< |
21:36 | <@McMartin> | "Let's learn a lot of bad habits!" |
21:36 | < Finerty> | Yes. |
21:39 | <@McMartin> | Hrm. I was going to link Snack, but its SF page has no project page. |
21:40 | <@McMartin> | Which is aggravating. I swear I've seen demonstrations somewhere. |
21:42 | < Finerty> | What's Snack? |
21:43 | <@McMartin> | Somebody else's pet project for going from JS-level sorts of languages into more standard bytecode languages. |
21:43 | < Finerty> | ah. |
21:44 | | * Finerty wants better psyco support. |
21:47 | <@ToxicFrog> | psyco being the JIT compiler for Python? |
21:47 | < Finerty> | Yes. |
21:48 | <@ToxicFrog> | LP GTA:SA is indeed excellent. |
21:48 | <@ToxicFrog> | And has gotten me playing SA again ?? |
21:49 | <@ToxicFrog> | There's a luaJIT, which is cool, but it only supports 32-bit x86. |
21:52 | <@McMartin> | Really, your first JIT target should be JVM or CIL. |
21:53 | <@McMartin> | Then you have Sun/MS/IBM/the Mono team do your native code generation. |
21:53 | <@McMartin> | (Snack is by a guy who did a lot of under-the-hood work for I7, and also wrote a Z-Code->CIL jit) |
22:00 | <@ToxicFrog> | "Some places are more difficult to get to than others, but necessity is the mother of invention... plus jumping off of high shit looks awesome and makes chicks dig you - young impressionable children please take note!" |
22:00 | <@ToxicFrog> | Someone else is working on a Lua->JVM compiler, I think |
22:00 | <@McMartin> | Wouldn't be surprised. |
22:00 | <@McMartin> | I need to learn more about CIL |
22:01 | <@ToxicFrog> | Or possibly it's a VM written in Java. Or both. |
22:01 | <@McMartin> | Everything I hear about it in passing implies that it's an easier target to hit than JVM, though it's not as well-run. |
22:01 | <@MyCatVerbs> | ToxicFrog: where'd you see/hear that? Sounds like an advertisement for a Parkour club. |
22:01 | <@McMartin> | LP GTA:SA, I pet |
22:02 | <@McMartin> | bet |
22:02 | <@MyCatVerbs> | Oh of course. |
22:05 | <@McMartin> | Snack started targeting Glulx, apparently because Glulx has a more consistent runtime than CIL atm |
22:12 | <@McMartin> | C# also needs more widespread compilers. =P |
22:13 | <@McMartin> | Though some of that high opinion is due to knowing that it autosolves a number of extremely common Java bugs. |
22:22 | | Elizabeth is now known as AnnoDomini |
22:22 | | Rico is now known as EvilDarkLord |
22:58 | | AnnoDomini [AnnoDomini@Nightstar-28988.neoplus.adsl.tpnet.pl] has quit [Quit: Some people have evil spirits. You, you have stupid spirits. Go see shaman. Get hole in head. Big hole. Very big. Huge!] |
23:01 | | Attilla [~The.Attil@92.10.170.ns-10975] has quit [Ping Timeout] |
23:05 | <@McMartin> | AhAHAHAHAHAAH |
23:06 | <@McMartin> | "Fortran is at the door, it wants a word with about how god is real" |
23:06 | < Finerty> | huh? |
23:10 | <@McMartin> | Fortran types variables based on the first letter of their name. |
23:11 | < Finerty> | oh, heh |
23:11 | < Finerty> | and so something like i through... s or something like that are integers and the rest are floats or something |
23:11 | <@McMartin> | I think some are guaranteed strings too |
23:11 | <@McMartin> | But yeah, float/real is the default, IIRC |
23:28 | | You're now known as TheWatcher[T-2] |
23:34 | | You're now known as TheWatcher[zZzZ] |
23:40 | | Attilla [~The.Attil@92.11.6.ns-26403] has joined #code |
23:40 | | mode/#code [+o Attilla] by ChanServ |
--- Log closed Mon Jul 28 00:00:28 2008 |