#eulora Logs for 02 Dec 2019



December 2nd, 2019 by Diana Coman
diana_coman: in CS questions-from-the-code: "ctxt->iview = other.ctxt->iview; // @@@ Is this right?" [13:38]
diana_coman: there are a few more of those, too; a very self-doubting code. [13:40]
asciilifeform: diana_coman: cpp programmers tend to end up as depressives (illustrated above) or maniacs ('you are not expected to understand this'(tm) & similar brag) [13:41]
diana_coman: the wonder is that nobody actually forced them to use CPP or something; nor XML for that matter; nor that mess of a SCF and VFS and all the plugins nightmare and all of it really. [13:43]
diana_coman: anyway, I don't suppose there's a lot to it. [13:44]
asciilifeform: diana_coman: are you baking a 3d engine in ada ? [13:46]
diana_coman: asciilifeform: atm the bill of work is theoretically less than that (though not sure in practice): http://logs.minigame.biz/2019-11-24.log.html#t12:02:30 [13:48]
lobbesbot: Logged on 2019-11-24 12:02:30: <diana_coman> hmm, not sure things are clear there somehow; let me see from the other end: do you suggest I dig in CS and figure out how to directly get from it what I want it to do, ditching all those shaders xmls and whatnots? [13:48]
asciilifeform: unrelatedly, diana_coman et al : this chan's subjline still contains a 'Please use http://p.bvulpes.com/ for ALL pastes.' that www has been dead for at least 6mo. [13:50]
diana_coman: hm, and the logs link should point to .biz huh [13:52]
asciilifeform: diana_coman: re 3d engines -- are you familiar with one dave baggett ? fella wrote a well-known game for 'playstation', where no 3d accelerator, and in fact wrote (and quickly) a quite efficient 3d system in something resembling 'tinyscheme'. unfortunately was never published openly. [13:57]
diana_coman: the crash bandicoot guy? [14:27]
diana_coman: iirc it was simply very, very overfitted basically - as those things tend to be anyway (this might apply as far as I can see to possibly ALL FUCKING COMPUTER GRAPHICS may it burn in hell already). [14:28]
asciilifeform: overfitted yes. ( recall that it had to sit on a 1994 cpu however. ) [14:44]
mircea_popescu: diana_coman, evidently it's been tested extensively :D [14:45]
mircea_popescu: the loglink DOES point to biz! [14:47]
mircea_popescu: diana_coman, : in other news, the results of me trying to follow wiki recipe to get game client up : http://paste.deedbot.org/?id=iAO0 [16:47]
diana_coman: mircea_popescu: we don't have a webpage yet, do we? [16:49]
mircea_popescu: i dun think we do. [16:49]
diana_coman: I mean:up; as to the versions there, it will most probably blow up later on :D [16:50]
mircea_popescu: nor much of a genesis really [16:50]
mircea_popescu: right [16:50]
diana_coman: well yes, but genesis aka what, from OS on, right? [16:50]
mircea_popescu: nfi [16:50]
mircea_popescu: the thing is, there's no practical drawback to arbitrary genesis. if you want to graft it later as a node on some tree, can always regrind. [16:51]
mircea_popescu: this is also incidentally how "too many small patches" gets resolved, all trees are intended to regrind occasionally anyways [16:51]
diana_coman: there is the practical drawback of the pain of genesising a huge mess that possibly includes all sorts though; with unclear usefulness since every system has different requirements anyway. [16:52]
mircea_popescu: the pain, for sure. [16:52]
mircea_popescu: the other drawbacks are actually advantages, though. once the mess is made definite it's thereby also made resolvable [16:52]
mircea_popescu: though not also resolved, tis true [16:53]
mircea_popescu: imo even the section from the wiki i was following, turned into a plain .sh script and genesised, would be an improvement over current state. [16:54]
diana_coman: supposedly anyone from the community could have made a .sh script out of that, no? [16:55]
mircea_popescu: indeed. [16:55]
diana_coman: as I re-installed recently the client on all sorts of system, I know rather clearly what a mess it is but a. CS even quite pushes actively the ftjam dependency because of the idiotic "plugins" system that even relies on ftjam packing stuff in a certain way; theoretically it's possible to make it work without that specific packing but I'm also sure it will be yet-another-pain; possibly unavoidable though; b. the only solution I can see ... [16:58]
diana_coman: ... to having a fixed environment and hence a fixed-working install would be still along the lines "here's the eulora machine"; as a soft approx one can say "here's this GNAT including gcc" but note that this still means OS-dependencies that might or might not be around as expected. [16:58]
mircea_popescu: but "anyone from the community" is too busy registering random nicknames because i mentioned them to do something useful. [16:59]
mircea_popescu: kinda lulzy, this dedication to failure. rando anal child ain't gonna do what works, because he's gonna SHOW US!!!! or whatever. [17:00]
mircea_popescu: diana_coman, this is the gaming whore, it has no problem installing everything in triplicate. [17:00]
diana_coman: because tarabostes, because "too aggressive" (with his stupidity) and so on, yes. [17:01]
mircea_popescu: myeah. [17:02]
diana_coman: it's not about triplicate, it's about whether they are still available/work with the rest/don't break anything because whatever-changed-meanwhile. [17:02]
mircea_popescu: well honestly i was kinda going to check that out. [17:02]
diana_coman: dunno, do you want me to put up the website somewhere? not like a huge issue really, can plonk it probably on the logger machine just a different vhost or something, if it's at that [17:03]
mircea_popescu: meh. [17:04]
mircea_popescu: im satisfied for now. [17:04]
mircea_popescu: don't have the time to spend with it, leaving in a few days anyways. [17:04]
diana_coman: with the state of brokedness, lolz. [17:04]
mircea_popescu: well, i tried, i reported, what can i say. [17:04]
diana_coman: report duly noted! [17:05]

Comments feed: RSS 2.0

Leave a Reply