Log for #openttdcoop.devzone on 21st July 2010:
Times are UTC Toggle Colours
00:00:37  <Ammler> a bit smaller, but not much
00:00:45  <Rubidium> another bit now :)
00:01:23  <Ammler> hmm
00:02:48  <Brot6> NFORenum - Revision 393:5d5fc50ddc03: Fix: action0 feature 10 property 13 had a wrong size and pr... (Rubidium) @
00:03:06  <Ammler> well, that is for tomorrow
00:03:13  <Ammler> enough manual triggering :-)
00:03:16  <Ammler> good night
00:45:10  <Brot6> OpenMSX - Revision 91:3720730df0fa: Change: set non default build type: music (Ammler) @
00:45:10  <Brot6> OpenSFX - Revision 97:e88bd5bc8864: Change: set non default build type: sound (Ammler) @
01:06:27  *** Hirundo has quit IRC
01:07:01  *** Hirundo has joined #openttdcoop.devzone
02:31:02  *** Frankr has quit IRC
04:25:56  *** OwenS has quit IRC
04:26:11  *** OwenS has joined #openttdcoop.devzone
06:30:36  *** Alberth has joined #openttdcoop.devzone
07:34:04  <Yexo> Ammler / Rubidium: I didn't add the railtypes patch, I thought dalestan already added it
07:34:12  <Yexo> I'm sure he added at least parts of it
07:40:26  <Rubidium> yes... exactly, he added parts
07:42:26  <Yexo> ok, I'll complete it then
07:42:50  <Rubidium> I think I've already did that (but I might've forgotten some stuff)
07:47:31  <Rubidium> action 3 of railtypes gets //!!Warning (142): <num-cid> must be 0 for this feature.
07:47:49  *** Yexo has quit IRC
07:48:48  *** Yexo has joined #openttdcoop.devzone
07:49:08  <Rubidium> action 3 of railtypes gets //!!Warning (142): <num-cid> must be 0 for this feature.
07:49:10  <Yexo> windows was so nice to show me a blue screen again :(
07:49:11  <Rubidium> seems to fix that
07:49:36  <Yexo> @logs
07:49:36  <Webster> Logs:
07:53:06  <Yexo> that patch is ok
07:54:51  <Brot6> NFORenum - Revision 394:37118c19630b: Fix: action3 "cargos" for railtypes are valid as well (Rubidium) @
08:04:19  <Rubidium> DJNekkid, regarding nutracks: in sprites/nfo/header.pnfo you are, for the last 4 action0s, not counting correctly; there are 18 properties, but you say there are 16 in the action0. If you fix those you'll have 4 errors less
08:13:46  <Rubidium> DJNekkid: should fix those errors
08:27:24  <Brot6> NFORenum - Revision 395:4416c03f4659: Add: support action7/9 condition types 0D and 0E (railtype ... (yexo) @
08:48:02  <Rubidium> oh... swedishrails (compiled, decompiled, nforenumed) causes quite a bit of errors/warnings
08:50:36  <Brot6> NFORenum - Revision 396:ec5846b6e9e1: Add: varaction2 vehicles variable 4A (Rubidium) @
08:54:11  <Yexo> Rubidium: could you not touch data.cpp until I'm finished with adding airporttiles support?
08:54:36  <Yexo> should only be half an hour or so
08:54:44  <Brot6> NFORenum - Revision 397:95ebee3eaf3a: Add: support for the two railtype (pseudo) random bits (Rubidium) @
08:56:09  <Rubidium> Yexo: okay :)
09:00:54  * Ammler wonders about offering ChillCore commit rights to OpenGFX
09:05:06  <Yexo> anyone wants to review this?
09:06:00  <Rubidium> where can I find the specs?
09:06:15  <Yexo> hidden in the ttdpatch wiki / nowhere
09:06:25  <Yexo> ie openttd source code
09:06:49  <Yexo> contains most of it
09:06:51  <Webster> Title: Transport Tycoon Forums • View topic - NewGRF Airports spec (at
09:12:15  <Rubidium> first chunk looks okay :)
09:12:50  <Rubidium> while you're possibly doing not so much, could you look at var43 of railtypes? It's still missing. Just added it to the specs though
09:14:08  <Yexo> ok
09:15:47  <Rubidium> oh, and do you know what he difference between 83 and 84 is in varaction2 stuff (in data.cpp)
09:16:15  <Yexo> yes
09:16:21  <Yexo> 84 means dword sized
09:16:32  <Yexo> 83 is dword sized but only the first 3 bytes are used
09:18:25  <Rubidium> ah
09:20:20  <Yexo> you want a seperate commit for railtype var 43 or can I just throw it in the airporttiles commit?
09:20:56  <Rubidium> it means bumpding that number twice, doesn't it?
09:21:01  <Yexo> yep
09:22:02  <Rubidium> what piece of the code/specs says 20 num 83 bits?
09:22:20  <Rubidium> (or rather 32)
09:22:29  <Yexo> nothing (but stations have 32 (0x20) random bits)
09:23:40  <Rubidium> though for airport tiles you seem to be or-ing station random bits with tile bits
09:23:51  <Yexo> yes, that should be changed
09:24:06  <Yexo> just 4 random bits for the airporttile and 32 for the airport would be better
09:24:49  <Rubidium> industry tiles does something with scope; maybe that's what's needed for airport tiles as well
09:25:05  <Rubidium> anyhow, in that case it looks okay
09:26:13  <Yexo> current code (or-ing the station random bits in) is consistent with the newgrf station code, but the airport/airporttiles spec looks a lot more like industries/industrytiles anyway
09:26:23  <Yexo> so it's better to make this also consitent with industry tiles
09:27:04  <Rubidium> so station might not have 32 bits
09:27:29  <Yexo> oh, right :(
09:27:33  <Yexo> station has only 16 bits
09:27:50  <Yexo> but railstation tiles have 4 more bits
09:27:57  <Yexo> so nforenum is already wrong for newgrf station tiles
09:28:07  <Rubidium> so the x20 should be 20 and/or x15
09:28:17  <Rubidium> s/5/4/
09:28:17  <Yexo> value should be 0x14
09:29:33  <Yexo> diff updated, added in railtype var 43 and fixed the random bits
09:30:39  <Rubidium> yay :)
09:30:42  <Rubidium> looks okay
09:31:20  <Rubidium> bah.... my internet is *slow*
09:31:50  <Rubidium> as if I were on 14k4 with a download
09:32:21  <Brot6> NFORenum - Revision 398:1782d9805fdf: Add: support for airporttiles (yexo) @
09:32:40  <Rubidium> ~10 seconds for google to start rendering its page :(
09:32:57  <Yexo> ouch
09:34:39  <Ammler> 	M( 0xf54, PAL_NONE,  0x5b5, PAL_NONE,  1,  3, 14, 11,   7, 0), <-- means x-offset -1 and y-offset -3
09:35:09  <Ammler> I guess, the size does have influence too?
09:35:47  <Yexo> yes (but I don't know how exactly)
09:36:12  <Ammler> might be easiest to adjust all non-0 houses manually
09:37:16  <Ammler> which key does again enable bounding boxes?
09:37:38  <Ammler> b
09:40:33  <Yexo> easiest solution for the now invalid "First building sprite cannot share its bounding box with the previous sprite." warning seems to be to just disable it completely
09:42:19  <Rubidium> Yexo: nforenum thinks swedishrails contains some erros and such; probably bugs in nml
09:42:34  <Yexo> I'll check soon
09:43:51  <Yexo> remove the warning completely as it's no longer valid
09:45:28  <Rubidium> that's the "stacked ground tiles" stuff, right?
09:46:20  <Yexo> yes
09:47:13  <Rubidium> then that's probably the best thing to do
09:53:14  <Yexo> first warning nforenum gives seems to be invalid
09:53:30  <Yexo> according to you can use all special parameters from the action7 page
09:53:38  <Yexo> A1 is listed on the action7 page as "openttd version"
09:55:26  <Rubidium> ahha, so that's more nforenum brokeness
09:55:49  <Yexo> unless you read the spec in another way
09:55:52  <Brot6> NFORenum - Revision 399:867abea13af5: Fix: remove warning 135 (first sprite cannot share bounding... (yexo) @
09:56:01  <Yexo> but it works in openttd and the code seems to be consistent with the spec
09:56:52  <Yexo> all "//!!Error (49): Missing real sprite. Sprite" errors are because nforenum first removes the real sprites lines because it are png files
09:57:05  <Yexo> so it's true that it isn't strictly valid nfo
09:58:47  <Rubidium> I've took the compiled version, and decompiled that, then nforenumed it
09:58:58  <Rubidium> so I wouldn't have the png problems
09:59:18  <Yexo> ah, ok
09:59:27  <Yexo> I've let nml create an nfo file
10:02:21  <Yexo> /!!Warning (108): is worrying
10:02:39  <Yexo> but sprite 69 and 70 jump to label 12 at sprite 77, so I think that warning is invalid too
10:03:03  <Yexo> oh, nvm
10:03:13  <Yexo> it's just me reading wrong, that's an error in nml
10:08:27  <Brot6> NewGRF Meta Language - Revision 563:2acf0e70580f: Fix: disable the reuse of action10 labels for n... (yexo) @
10:10:52  <Yexo> warning 76 is valid and should be fixed (it can break things in ttdpatch, not in openttd)
10:11:28  <Yexo> warning 98 is valid if you just look at that single action but invalid because the relevant bytes are overwritten by an action6
10:12:24  <Yexo> the few warnings 100 and 170 are also valid but no problem at all
10:12:30  <Yexo> they can be fixed by changing the input nml
10:25:06  *** Seberoth has joined #openttdcoop.devzone
10:42:53  <Brot6> NFORenum - Revision 400:0924dbadeb3e: Fix: for some reason the special parameters A1 to A4 where ... (yexo) @
10:42:53  <Brot6> NFORenum - Revision 401:44a2ea885b0e: Revert (part of 400:0924dbadeb3e): accidentally committed s... (yexo) @
10:53:17  <Rubidium> that last commit doesn't quite look like reverting debug code :)
10:55:45  <Yexo> reverting to old code style then
11:07:38  <Ammler> what is "@param dz the virtual height of the sprite"
11:10:47  <Ammler> standard seems to be (..., 0, 0, 16, 16, dz, 0)
11:16:13  <Yexo> are there any open issues in nforenum that still need to be resolved?
11:19:12  <Ammler> Yexo: this evening, it will rebuild a lot newgrfs, maybe we could skip all @warnings?
11:20:15  <Yexo> you can try, but there might still be a lot of valid warnings that don't apply to a specific case so are disabled for good reason
11:20:30  <Yexo> like 98, 100, 170
11:20:44  <Ammler> well, bug not e.g. openttd gui count
11:22:05  <Ammler>
11:39:19  *** Seberoth is now known as Seb|afk
11:41:51  <Brot6> NFORenum - Revision 402:0a1a2f8df8a1: Update: action 5 variable 15 maximum sprite count (Rubidium) @
11:44:28  <Ammler> well, there might be more, that is why I would think, it might be useful to run it without @@warnings
12:59:55  *** Seb|afk is now known as Seberoth
13:19:20  <Brot6> GRFCodec - Revision 177:fc47d75bda2f: Update: the changelog (Rubidium) @
13:26:05  *** frosch123 has joined #openttdcoop.devzone
13:35:04  *** V4530000 has joined #openttdcoop.devzone
13:35:42  *** V453000 has quit IRC
13:35:42  *** V4530000 is now known as V453000
14:31:59  <Brot6> NFORenum - Revision 403:ff3ea3d4605f: Update the changelog (Rubidium) @
14:58:28  <Brot6> OpenGFX - Revision 469:39afaa99ef0f: Fix: adjust some houses, which use different bounding boxes ... (Ammler) @
15:26:21  <Ammler> he, also funny, this different bounding boxes did also fail my adjustment tests with the sprite2station newgrf
16:19:11  <Ammler> now let the channel spam with rebuilds :-)
16:19:27  * Alberth gives a magic wand to Ammler
16:20:31  <Brot6> grfcodec: update from r175 to r177 done -
16:21:57  <Rubidium> so far so good
16:22:43  <Brot6> nforenum: update from r392 to r403 done -
16:24:19  <Brot6> nml: update from r561 to r563 done -
16:25:21  <Rubidium> bros will still fails
16:25:26  <Rubidium> s/s$//
16:25:42  <Rubidium> nutracks probably has less warnings/errors
16:26:33  <Brot6> opengfx: update from r468 to r469 done -
16:27:01  <Yexo> Ammler: is this with all @@WARNING disabled?
16:27:08  <Ammler> no
16:27:48  <Brot6> openmsx: update from r90 to r91 done -
16:30:26  <Brot6> opensfx: update from r96 to r97 done -
16:30:32  <Brot6> Following repos didn't need a nightlies update: 2cctrainset (r573), 32bpp-extra (r36), airportsplus (r52), bros (r12), comic-houses (r70), firs (r1074), fish (r386), heqs (r362), newgrf_makefile (r124), nmts (r16), nutracks (r86), ogfxplus (r40), snowlinemod (r15), swedishrails (r140), worldairlinersset (r659)
16:32:37  <Brot6> 2cctrainset: rebuild of r573 done (5093 errors) (Diffsize: 30422) -
16:34:14  <Brot6> 32bpp-extra: rebuild of r36 done (3584 errors) -
16:34:17  <Rubidium> hmm... something nforenum-ish is broken
16:34:27  <Rubidium> um/4.dat contains information for fewer features than does feat.dat.
16:34:31  <Rubidium> Update it, or delete feat.dat.
16:35:42  <Brot6> airportsplus: rebuild of r52 done (Diffsize: 2) -
16:35:44  <Yexo> _dat4[] in data.cpp (nforenum) is broken
16:35:57  <Ammler> how can that happen, if I build the chroot everytime again?
16:36:01  <Ammler> oh
16:36:52  <Ammler> airport+ is change in the compiler
16:37:02  <Ammler> since I move the nfo to log
16:37:04  <Brot6> bros: compile of r12 failed -
16:37:41  <Rubidium> bros is just broken on itself; does anyone care to fix it?
16:37:55  <Ammler> I should disable compiling
16:38:09  <Ammler> I guess, the project is (again) dead
16:38:28  <Brot6> comic-houses: compile of r70 failed -
16:39:02  <Brot6> NFORenum - Revision 404:8472b7a48849: Fix (398:1782d9805fdf): dat4 wasn't updated with the new nu... (yexo) @
16:39:05  <Rubidium> Ammler: it's really easy to fix
16:39:25  <Ammler> change from renum to nforenum?
16:40:11  <Rubidium> something includes foo.pnfo whereas it should include foo.nfo
16:40:20  <Brot6> firs: rebuild of r1074 done (89 errors) (Diffsize: 23677) -
16:40:32  <Ammler> :-o
16:40:35  <Ammler> lol
16:42:02  <Brot6> fish: rebuild of r386 done (169 errors) (Diffsize: 3868) -
16:42:38  <Rubidium> the rebuilds are pointless though :(
16:42:48  <Rubidium> due to the nforenum bug
16:43:00  <Alberth> Ammler: a nice target to test your error re-compiling
16:43:00  <Rubidium> but from that point of view... the rebuilds were very useful!
16:43:02  <Ammler> hmm, yes, but thanks to the rebuilds you found the bug
16:43:18  <Ammler> :-)
16:43:23  <Brot6> NFORenum - Revision 405:ee8348369acd: Fix (r391): printed data file names were incorrect (Rubidium) @
16:43:43  <Brot6> heqs: rebuild of r362 done (3141 errors) (Diffsize: 10627) -
16:43:57  <Alberth> wow :)
16:44:08  <Ammler> well, now we should first build nforenum again
16:44:58  <Ammler> Alberth: adding a threshold, if that is reached, to exit rebuild
16:45:16  <Brot6> newgrf_makefile: rebuild of r124 done -
16:45:24  <Alberth> sounds like a plan :)
16:45:36  <Ammler> if diff > 100 lines
16:45:59  <Rubidium> a diff of more than 100 lines could still be okay
16:46:18  <Ammler> and errors someone
16:46:33  <Ammler> or only errors
16:46:40  <Brot6> nmts: rebuild of r16 done (247 errors) -
16:46:42  <Rubidium> especially if 100 lines of errors were removed by a bugfix in nforenum :)
16:47:19  <Alberth> an increase in errors, perhaps
16:47:43  <Ammler> the link could go directly to log
16:48:13  <Brot6> nutracks: rebuild of r86 done (768 errors) (Diffsize: 6271) -
16:49:31  <Brot6> ogfxplus: rebuild of r40 done -
16:50:54  <Brot6> snowlinemod: rebuild of r15 done (Diffsize: 2) -
16:52:19  <Rubidium> <- that looks really fishy
16:52:29  <Brot6> swedishrails: rebuild of r140 done (Diffsize: 6) -
16:53:05  <Rubidium> md5 checksum changed :)
16:53:39  <Ammler> now what should happen with REBUILD?
16:53:51  <Ammler> shall it replace the release?
16:53:58  <Brot6> worldairlinersset: compile of r659 failed -
16:54:02  <Ammler> shall it fail?
16:54:30  <Rubidium> Ammler: leave it, unless it is really known to fix a bug
16:55:22  <Ammler> any idea, what caused this in ser?
16:55:25  <Ammler> not a bug?
16:55:38  <Rubidium> nml change
16:56:25  <Rubidium> it might be a bug fix, but nobody noticed the issue besides nforenum
16:58:11  <Ammler> hehe, also a kind of regression test, decompile and nforenum :-)
16:58:41  <Brot6> nforenum: update from r403 to r405 done -
17:00:55  <Brot6> 2cctrainset: rebuild of r573 done (1 errors) -
17:02:33  <Rubidium> should the rebuild directory be removed upon a successful rebuild?
17:04:17  <Ammler> hmm, yes
17:04:25  <Ammler> I didn't copy it, but it should be removed
17:07:40  <Brot6> nutracks: rebuild of r86 done (164 errors) (Diffsize: 243) -
17:08:52  <Rubidium> there... that one is better than "before"
17:09:32  <Rubidium> 15360 pixel sprites?!?
17:09:35  <Rubidium> so many?!?
17:10:58  <Rubidium> @calc sqrt(15360)
17:10:58  <Webster> Rubidium: 123.935467079
17:12:12  <Ammler> I guess, it isn't possible to automatically check, if the rebuild is better or worse
17:12:34  <Rubidium> nope
17:15:22  <Rubidium> did you start there rebuilds manually? There's so much time between them
17:18:34  <Ammler> only the last 3
17:18:45  <Ammler> one build needs around 2 mins
17:19:00  <Ammler> (setup the whole chroot)
17:24:07  <Brot6> #openttdcoop - Revision 102:e9f4e8a44910: [Compiler] Fix: remove possible REBUILD on empty diff (Ammler) @
18:24:11  <Brot6> HEQS "Heavy Equipment" Set - Revision 363:a9215445bba7: Change: hide Speedytrac rail version (u... (andythenorth) @
18:44:27  <Brot6> HEQS "Heavy Equipment" Set - Revision 364:bbb1429b75e5: Change: hide Gmund Mog rail version (andythenorth) @
19:20:37  *** Seberoth has quit IRC
19:24:09  *** Seberoth has joined #openttdcoop.devzone
19:25:16  <Ammler> oh, Mercurial 1.6 available, might need to update :-)
19:28:11  <Rubidium> if there isn't something you really need, you should wait for your OS' packager to pick it up
19:35:33  <Ammler> yes, there was something
19:35:56  <Ammler> it is out longer, now it is available from our distro repo
19:36:19  <Ammler> I guess, something with rebase
19:36:41  <Ammler> or mq qpush --move
19:37:24  <Ammler> well, not on the server, but locally for working
19:48:05  <Ammler> hmm, I can't hg convert our svn repo anymore
19:48:49  <Ammler> marcel@inspiron:~/hg> hg convert --source-type svn
19:48:51  <Ammler> assuming destination trunk-hg
19:48:52  <Ammler> initializing destination trunk-hg repository
19:48:54  <Ammler> does not look like a Subversion repository
19:53:35  <Ammler> another important extension is eol, specially for the windows guys
19:59:06  <Alberth> we already had a eol thingie, there is now just a new extension afaik
20:01:21  <Alberth> good night
20:01:56  *** Alberth has left #openttdcoop.devzone
20:03:39  <Ammler> win32text
20:03:51  <Ammler> which is quite a pain
20:46:09  <Ammler> no idea, why it fails, I might use my shell script
21:42:32  *** frosch123 has quit IRC
23:05:43  *** Seberoth has quit IRC

Powered by YARRSTE version: svn-trunk