#gemrb@irc.freenode.net logs for 12 Dec 2015 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage

[00:00:44] * mcphail wouldn't have the space for CRT these days :)
[00:01:55] <mcphail> I use the proprietary nvidia driver. Open source drivers just don't work well with GOG's wine wrapper
[00:02:22] <kandros> I'm using AMD's open source drivers
[00:02:38] <kandros> which may be the cause of the missing .so files
[00:02:41] <kandros> It probably is
[00:02:45] <mcphail> kandros: They don't work for me. Tried them with my APU
[00:02:53] <kandros> Ah well
[00:03:09] <kandros> Good thing having that confirmed, so I don't waste any more time on it
[00:03:15] <kandros> GemRB seems to work well, enough
[00:03:19] <mcphail> kandros: never tried proprietary AMD drivers with Torment, as I hate the proprietary AMD drivers
[00:03:25] <kandros> except from a few obious issues
[00:03:32] <kandros> Yeah, same
[00:03:36] <kandros> Haven't even tried them
[00:03:58] <kandros> But the fact that they don't stay up to par with xorg is quite repulsive, just that
[00:04:09] <mcphail> They are a mess
[00:04:43] <mcphail> And the Open Source drivers work well with steam. Just GOG games seem to be the problem
[00:04:52] <kandros> yeah
[00:04:56] <kandros> Dota 2 works quite nice
[00:05:07] <kandros> not as good performance as the windows drivers, but hey
[00:05:09] <kandros> progress
[00:05:16] <mcphail> indeed
[00:05:39] <kandros> I do wish I could get GPU passthrough working
[00:06:04] <mcphail> That's a bit bleeding edge for me
[00:06:11] <kandros> Then I'd probably play most games on Windows anyway
[00:06:32] <kandros> Yeah, well, it's not really just a button press
[00:06:59] <kandros> and it seems I'll have to flash my card with a non-official UEFI-BIOS
[00:07:17] <kandros> I'll save that when I don't have finals :I
[00:07:24] <mcphail> Ha!
[00:07:48] <mcphail> Anyway, I'm off to bed. Goodnight
[00:07:57] <kandros> Goodnight!
[00:36:44] --> phao has joined #gemrb
[00:55:58] <-- phao has left IRC (Ping timeout: 272 seconds)
[02:07:39] --> Eli2_ has joined #gemrb
[02:08:32] <-- Eli2 has left IRC (Ping timeout: 246 seconds)
[05:55:49] --- u is now known as ilbelkyr
[10:18:13] --> phao has joined #gemrb
[13:15:55] --> edheldil_ has joined #gemrb
[13:22:29] <-- edheldil_ has left IRC (Ping timeout: 264 seconds)
[13:40:52] --> edheldil_ has joined #gemrb
[14:17:38] --> fizzle has joined #gemrb
[14:27:10] <fizzle> lynxlynxlynx: do you remember what exactly caused you to introduce GF_SKIPUPDATE_HACK?
[14:40:23] <lynxlynxlynx> isn't it in the log?
[14:40:53] <lynxlynxlynx> i try to put it there if not the comments
[14:51:30] <fizzle> well, it's in the comment, sort of
[14:51:53] <fizzle> but in my tests, that case works just fine without the GF
[14:51:57] <fizzle> unless I misunderstood the comment
[14:52:21] <fizzle> hence, the question
[15:09:50] <lynxlynxlynx> ok, let me look it up, i don't remember off-hand
[15:11:09] <lynxlynxlynx> oh, i see
[15:11:34] <lynxlynxlynx> i don't think it was added by me; initially we had it to fix totl
[15:13:29] <lynxlynxlynx> you have no problems with that perpetually invisible djinn anymore?
[15:13:47] <fizzle> no (unless there's more than one djinn...)
[15:14:07] <lynxlynxlynx> only one major one, in one of the corner towers
[15:14:16] <lynxlynxlynx> a baddy summons him with an item
[15:14:33] <lynxlynxlynx> could be all your unscheduled fixes helped
[15:14:37] <fizzle> yeah, that seems to work fine
[15:17:21] <lynxlynxlynx> hmpf, i get an odd crash earlier, unrelated
[15:18:07] <lynxlynxlynx> did you see any bad consequences of the current state, besides code pollution?
[15:18:35] <fizzle> no, I just stumbled across the GF when investigating something else
[15:19:00] <fizzle> something I have forgotten since...
[15:19:25] <fizzle> long time away creates problems like that
[15:19:36] <lynxlynxlynx> ok, i suggest you remove it later, i'm starting to wrap up for the next release and this could bite
[15:19:46] <fizzle> sure
[15:19:55] <lynxlynxlynx> thanks
[15:22:55] <lynxlynxlynx> ehh, more fallout from that "c++ guarantees this!=null"
[15:23:07] <lynxlynxlynx> should've just reverted it
[15:44:29] --> edheldil__ has joined #gemrb
[15:44:57] <lynxlynxlynx> ok, the djinn appeared and was normally killable
[15:47:23] <Pepelka> [commit] lynxlynxlynx: fixed more crash regressions from c3a52d2c29 https://github.com/gemrb/gemrb/commit/792167c3564a2bbac35b4e8f77b7bc1a580ccfde
[16:08:06] <-- fizzle has left #gemrb
[16:49:36] <-- edheldil__ has left IRC (Ping timeout: 256 seconds)
[17:01:28] --> brada has joined #gemrb
[17:02:07] <brada> lynx: i sure hope you didnt mean that! :)
[17:02:57] <lynxlynxlynx> why not? it was a far from complete job
[17:04:00] --> edheldil__ has joined #gemrb
[17:04:34] <brada> well sure, but how are we supposed to complete the job without letting it thresh those cases out?
[17:06:49] <lynxlynxlynx> doing it locally?
[17:09:43] <brada> sure, but obviously i havent had time to do much testing. i did loadup a game and open some windows and things like that. leaving it in would have cause problems with people with compilers set to high optimization according to the warning clang was producing.
[17:15:36] <lynxlynxlynx> just for those with -Werror
[17:15:57] <lynxlynxlynx> things like changing areas and just opening the records window were crashers
[17:16:28] <lynxlynxlynx> so it would be cleaner to revert and do properly
[17:16:36] <brada> supressing the warningwont prevent the compiler from optimizing it out tho :)
[17:17:06] <lynxlynxlynx> i doubt it did or people would notice
[17:20:42] <brada> ok, if thats what you mean by “revert” i cant argue there, but there is no reason for gemrb to not follow the standards :)
[17:21:15] <lynxlynxlynx> i don't think we were in violation
[17:21:57] <lynxlynxlynx> just pointers, no object instances
[17:32:33] --> brada_ has joined #gemrb
[17:39:26] <-- brada has left IRC (*.net *.split)
[17:39:30] <-- xrogaan has left IRC (*.net *.split)
[17:39:33] --- brada_ is now known as brada
[18:10:20] <-- phao has left IRC (Ping timeout: 240 seconds)
[18:21:55] <brada> lynx: according to the C++ standard (it follows from the paragraph 5.2.5/3 of the standard ISO/IEC 14882:2003(E)), calling any nonstatic method of any class through a null-pointer leads to undefined behavior
[18:28:26] <-- gembot has left IRC (Remote host closed the connection)
[18:31:06] <-- |Cable| has left IRC (Ping timeout: 272 seconds)
[18:31:16] --> phao has joined #gemrb
[18:32:42] --> xrogaan has joined #gemrb
[18:32:52] <-- brada has left IRC (Quit: brada)
[18:44:01] --> |Cable| has joined #gemrb
[19:21:53] <lynxlynxlynx> can't find it online, but '89 doesn't seem to mention it
[23:33:50] <-- edheldil_ has left IRC (Ping timeout: 272 seconds)