| --- | Log | opened Tue Jun 12 00:00:35 2007 |
| 00:02 | |-| | setrodox [~setrodox@83-65-235-58.dynamic.xdsl-line.inode.at] has joined #openttd |
| 00:09 | |-| | Osai [~Osai@pD9EB68C5.dip.t-dialin.net] has quit [Quit: Osai] |
| 00:41 | <Phazorx> | . |
| 00:47 | |-| | Frostregen_ [~sucks@dslb-084-058-146-131.pools.arcor-ip.net] has joined #openttd |
| 00:47 | |-| | Taikaponi [~Zavior@d195-237-7-209.elisa-laajakaista.fi] has quit [Read error: Connection reset by peer] |
| 00:47 | |-| | Zavior [~Zavior@d195-237-7-209.elisa-laajakaista.fi] has joined #openttd |
| 00:48 | |-| | Taikaponi [~Zavior@d195-237-7-209.elisa-laajakaista.fi] has joined #openttd |
| 00:53 | |-| | Frostregen [~sucks@dslb-084-058-153-145.pools.arcor-ip.net] has quit [Read error: Operation timed out] |
| 00:53 | |-| | Frostregen_ changed nick to Frostregen |
| 00:59 | <stillunknown> | Truebrain: How much faster is the rendering now? |
| 01:02 | <Phazorx> | stillunknown: http://openttdcoop.ppcis.org/blog/2007/06/07/patch-train-collision-cache/ |
| 01:03 | <stillunknown> | Phazorx: Is that the cpu usage? |
| 01:04 | <stillunknown> | Or reduction? |
| 01:04 | <Phazorx> | spu usage |
| 01:04 | <stillunknown> | 5% cpu usage, that seems very low/ |
| 01:05 | <Phazorx> | it's actualy flips between 3 and 5 |
| 01:05 | <stillunknown> | Too low imo. |
| 01:05 | <Phazorx> | i am surprised too |
| 01:05 | <@peter1138> | :o |
| 01:06 | <Phazorx> | but that game is weird in a sense |
| 01:06 | <Phazorx> | lots of tracks and trains are longer |
| 01:06 | <Phazorx> | longest of all games |
| 01:06 | <@peter1138> | me too. i didn't get that amount of improvement |
| 01:06 | <Phazorx> | peter1138: testing teqnique makes a lot of difference too |
| 01:06 | <stillunknown> | peter1138: btw, i pretty much done with my own hashmap thingie, just to let you know |
| 01:06 | <Phazorx> | that is done with no animation and sounds |
| 01:06 | <Phazorx> | as well as viewpoint not having any action on screen |
| 01:07 | <Phazorx> | since sprite managing takes a lot of power |
| 01:07 | <Phazorx> | so this is pretty much only pathfining and collision cycles only |
| 01:07 | <stillunknown> | There is much more at work, than just that. |
| 01:08 | <Phazorx> | well... i have a feeling these two are largest consumer in terms of CPU power |
| 01:08 | <Phazorx> | since it is per vehivle per tick |
| 01:08 | <stillunknown> | TrainLocoHandler is a a big load as well |
| 01:08 | <Phazorx> | rest is event based or pert game timeunit |
| 01:09 | <Phazorx> | well my numbers are not absolute, however they are taken in same situation for all tested cases |
| 01:09 | <Phazorx> | whcih emans they can be compared against each other and represent state of things on my platform |
| 01:10 | <stillunknown> | Either you have a blazingly fast cpu, or something is wrong. |
| 01:11 | <stillunknown> | Also keep in mind that rendering stuff has been improved in the last few revs. |
| 01:11 | <@peter1138> | not a great improvement |
| 01:12 | <@peter1138> | most of the reason for those changes are code separation |
| 01:12 | <@peter1138> | *is |
| 01:12 | <@peter1138> | or something. my english went wrong there :o |
| 01:13 | <Phazorx> | rendfering as in graphics? |
| 01:14 | <@peter1138> | yes |
| 01:14 | <@peter1138> | which you ruled out :) |
| 01:14 | <Phazorx> | stillunknown: CPU is clocked t-bred 2100+ |
| 01:14 | <stillunknown> | peter1138: first version of my patch, http://paste.openttd.org/97 |
| 01:14 | <Phazorx> | yeah for the test cases screen is pointed at water in corner of the map |
| 01:14 | <Phazorx> | there is nothing moving on screen and animation and full details are disabled |
| 01:15 | <Phazorx> | the aim was trully tests performance of logic part rather than anything else |
| 01:16 | <@peter1138> | http://fuzzle.org/o/newhash3.diff |
| 01:17 | <stillunknown> | 6 bits is a bit low |
| 01:17 | <stillunknown> | But i must go now. |
| 01:18 | <@peter1138> | i tried increasing it. didn't affect much to be honest |
| 01:18 | <@peter1138> | didn't actually benchmark it though |
| 01:26 | |-| | SpBot [terom@zapotekII.paivola.fi] has quit [Ping timeout: 600 seconds] |
| 01:26 | |-| | Taikaponi [~Zavior@d195-237-7-209.elisa-laajakaista.fi] has quit [Read error: Connection reset by peer] |
| 01:26 | |-| | Zavior [~Zavior@d195-237-7-209.elisa-laajakaista.fi] has quit [Write error: connection closed] |
| 01:26 | |-| | SpComb [terom@zapotek.paivola.fi] has quit [Ping timeout: 600 seconds] |
| 01:35 | |-| | TinoM [~Tino@i5387CAC1.versanet.de] has joined #openttd |
| 01:47 | <@peter1138> | bananas |
| 01:48 | <@peter1138> | eek EvL uses ancient grfs |
| 01:48 | <@peter1138> | and dbsetxl with dbset?? |
| 01:50 | |-| | |Gekkko| [~Gekko@CPE-60-229-217-120.nsw.bigpond.net.au] has joined #openttd |
| 01:50 | |-| | |Gekkko| changed nick to Gekko |
| 01:51 | |-| | Gekko [~Gekko@CPE-60-229-217-120.nsw.bigpond.net.au] has quit [] |
| 01:51 | |-| | |Gekkko| [~Gekko@CPE-60-229-217-120.nsw.bigpond.net.au] has joined #openttd |
| 01:51 | |-| | |Gekkko| changed nick to Gekko |
| 01:54 | <@peter1138> | hmm, yes, EvL is a lot faster |
| 01:54 | <@peter1138> | becomes smooth on my athlon 1250 |
| 01:55 | |-| | Zr40 [~Zr40@zr40.xs4all.nl] has joined #openttd |
| 01:57 | |-| | Gekko [~Gekko@CPE-60-229-217-120.nsw.bigpond.net.au] has quit [Quit: KVIrc 3.2.6 Anomalies http://www.kvirc.net/] |
| 01:58 | |-| | G [~nigel@202.154.148.243] has quit [Ping timeout: 600 seconds] |
| 02:01 | |-| | |Gekkko| [~Gekko@CPE-60-229-217-120.nsw.bigpond.net.au] has joined #openttd |
| 02:10 | |-| | boekabart [~boekabart@81.58.27.138] has joined #openttd |
| 02:13 | <Phazorx> | peter1138: grfs are funky there sorry about that |
| 02:14 | <Phazorx> | you see improvement w/ ur patch tho? |
| 02:14 | <@peter1138> | well |
| 02:14 | <@peter1138> | the game speeds up |
| 02:14 | <@peter1138> | as my cpu is too slow |
| 02:14 | <@peter1138> | i'm doing profiling now |
| 02:14 | <@peter1138> | although maybe 10000 frames was too much |
| 02:14 | <Phazorx> | reduce animation/details |
| 02:14 | <Phazorx> | or run dedicated server |
| 02:14 | <@peter1138> | heh, i saved the game in the top corner |
| 02:15 | <Phazorx> | i also disabled sound and tweaked grfx |
| 02:15 | <@peter1138> | ./openttd-6-1 -g save/speedtest.sav -v null -m null -s null |
| 02:15 | <@peter1138> | hash bits 6, hash res 1 |
| 02:15 | <Phazorx> | that's a good one |
| 02:21 | |-| | |Gekkko| changed nick to Gekko |
| 02:23 | |-| | Tobin [~Tobin@c58-107-50-36.eburwd7.vic.optusnet.com.au] has left #openttd [] |
| 02:23 | |-| | Tobin [~Tobin@c58-107-50-36.eburwd7.vic.optusnet.com.au] has joined #openttd |
| 02:23 | [~] | Tobin waves |
| 02:25 | <@peter1138> | hello |
| 02:26 | <@peter1138> | ohhh |
| 02:26 | <@peter1138> | 35.13 11.36 11.36 95542065 0.00 0.00 FindTrainCollideEnum(Vehicle*, void*) |
| 02:27 | <@peter1138> | 15.28 16.30 4.94 956793 0.00 0.00 VehicleFromPos(unsigned int, void*, void* (*)(Vehicle*, void*)) |
| 02:27 | <@peter1138> | ^^ trunk |
| 02:27 | <@peter1138> | 5.63 3.78 1.09 8817163 0.00 0.00 FindTrainCollideEnum(Vehicle*, void*) |
| 02:27 | <@peter1138> | 3.25 8.21 0.63 803832 0.00 0.00 VehicleFromPosXY(int, int, void*, void* (*)(Vehicle*, void*)) |
| 02:27 | <@peter1138> | 0.88 15.16 0.17 152961 0.00 0.00 VehicleFromPos(unsigned int, void*, void* (*)(Vehicle*, void*)) |
| 02:27 | <Phazorx> | on EvsL? |
| 02:27 | <@peter1138> | yes |
| 02:28 | <Phazorx> | what does VFP do? |
| 02:28 | <@peter1138> | no idea |
| 02:28 | <Phazorx> | it is used alarmingly often |
| 02:29 | <@peter1138> | oh, VehcileFromPos |
| 02:29 | <@peter1138> | hehe |
| 02:29 | <@peter1138> | it's the hash lookup |
| 02:29 | <@peter1138> | in trun, VFP is used for collisions, signal updates, etc... |
| 02:29 | <@peter1138> | +k |
| 02:29 | <@peter1138> | with the patch, VFPXY is used for collisions |
| 02:30 | <@peter1138> | VFP is used for the rest |
| 02:30 | <Phazorx> | i see |
| 02:30 | <@peter1138> | VFP only checks one hash area |
| 02:30 | <@peter1138> | VFPXY will check 1 to 4 as necessary |
| 02:30 | <Phazorx> | can you use stillunknown caching on top of that to reduce load even further tho? |
| 02:30 | <@peter1138> | i don't think it's worth it |
| 02:31 | <@peter1138> | not least because it has invalid assumptions |
| 02:31 | <@peter1138> | hmm, irrc |
| 02:31 | <@peter1138> | 6.93 18.54 2.24 1104000 0.00 0.00 TrainLocoHandler(Vehicle*, bool) |
| 02:31 | <@peter1138> | trunk |
| 02:31 | <@peter1138> | 4.85 4.72 0.94 803964 0.00 0.00 TrainController(Vehicle*, bool) |
| 02:31 | <@peter1138> | err |
| 02:31 | <@peter1138> | 13.89 2.69 2.69 1104000 0.00 0.00 TrainLocoHandler(Vehicle*, bool) |
| 02:31 | <@peter1138> | patch |
| 02:31 | <Phazorx> | well is any part of that data is static for tick and calculated for more than one vehicle? |
| 02:31 | <@peter1138> | 2.24 -> 2.69? :o |
| 02:32 | <Phazorx> | does it do that much extra ? |
| 02:32 | <@peter1138> | a tiny bit |
| 02:33 | <@peter1138> | it has to set a flag to update signals, as the signal update routine relied on the vehicle not changing hash positions |
| 02:35 | <Phazorx> | sort of got that one :) |
| 02:35 | <Phazorx> | back to my question - cahing state of tracks is a good idea imho |
| 02:36 | <@peter1138> | it may not be necessary though, as now the lookup is much faster |
| 02:36 | <@peter1138> | and the caching processing was quite involved |
| 02:36 | <@peter1138> | so instead of caching something that's slow, it's caching something that's faster |
| 02:37 | <@peter1138> | i will try it though :p |
| 02:39 | <@peter1138> | http://paste.openttd.org/98 are results |
| 02:41 | <@peter1138> | hmm, the larger the hash resolution the less time spent in UpdateNewVehiclePosHash() |
| 02:42 | <Noldo> | what is it that you are tweaking? |
| 02:43 | <@peter1138> | oh, lol |
| 02:43 | <@peter1138> | trunk: |
| 02:43 | <@peter1138> | 3.09 19.54 1.00 16985534 0.00 0.00 SignalVehicleCheckProc(Vehicle*, void*) |
| 02:43 | <@peter1138> | patch: |
| 02:44 | <@peter1138> | 0.00 19.37 0.00 33876 0.00 0.00 SignalVehicleCheckProc(Vehicle*, void*) |
| 02:44 | |-| | Zr40 [~Zr40@zr40.xs4all.nl] has quit [Quit: Leaving] |
| 02:48 | <Phazorx> | heh nice |
| 02:52 | |-| | Desolator [Desolator@82.79.212.95] has joined #openttd |
| 02:52 | |-| | Maedhros [~jc@host86-137-64-30.range86-137.btcentralplus.com] has joined #openttd |
| 02:53 | <Maedhros> | morning |
| 02:53 | <Desolator> | mornin' |
| 02:54 | |-| | Desolator [Desolator@82.79.212.95] has quit [] |
| 02:57 | |-| | Nickman [~nick_defr@d54C1C327.access.telenet.be] has joined #openttd |
| 03:08 | <@peter1138> | bah, i wish gaim supported msn's "free text" field |
| 03:12 | |-| | Zavior [~Zavior@d195-237-7-209.elisa-laajakaista.fi] has joined #openttd |
| 03:12 | |-| | XeryusTC [~irc@cc480157-b.sneek1.fr.home.nl] has joined #openttd |
| 03:13 | <Kjetil> | Why use msn at all ? |
| 03:14 | <@peter1138> | so i can keep in contact with people? |
| 03:14 | <@peter1138> | msn messenger, of course |
| 03:36 | |-| | maddy [~maddy@88-136-244-96.adslgp.cegetel.net] has joined #openttd |
| 03:37 | |-| | Vikthor [~Vikthor@snat1.spoje.net] has joined #openttd |
| 03:38 | <Kjetil> | peter1138: get them to use irc :P |
| 03:39 | <Biff> | peter1138: what is the free text field? |
| 03:40 | <mikegrb> | Biff: it is wherey you type to not get charged |
| 03:40 | [~] | mikegrb runs |
| 03:45 | |-| | maddy [~maddy@88-136-244-96.adslgp.cegetel.net] has quit [Remote host closed the connection] |
| 03:48 | |-| | maddy [~maddy@88-136-244-96.adslgp.cegetel.net] has joined #openttd |
| 04:09 | |-| | pro [~http@80-240-220-232.dnat.migtel.ru] has joined #openttd |
| 04:39 | |-| | TheJosh [~josh@d220-238-156-217.dsl.vic.optusnet.com.au] has joined #openttd |
| 04:40 | <TheJosh> | hey |
| 04:40 | <TheJosh> | i am having a problem doing a 'svn update': "svn: File 'src/texteff.cpp' has inconsistent newlines" |
| 04:41 | <CIA-1> | OpenTTD: rubidium * r10109 /trunk/src/ (spritecache.cpp spritecache.h): |
| 04:41 | <CIA-1> | OpenTTD: -Fix [FS#838]: some NewGRFs use the same (unused in the "current" climate) |
| 04:41 | <CIA-1> | OpenTTD: sprite IDs. Normally this gives some artefacts, but when one NewGRF expects it |
| 04:41 | <CIA-1> | OpenTTD: to be a sprite and another NewGRF overwrites it with a non-sprite nasty things |
| 04:41 | <CIA-1> | OpenTTD: happen (drawing a non-sprite crashes OTTD). |
| 04:41 | <CIA-1> | OpenTTD: maedhros * r10110 /trunk/src/newgrf.cpp: -Fix: Reset NewGRF errors along with all the other NewGRF data so that errors get loaded again when pressing "Apply". |
| 04:56 | |-| | Zr40 [~Zr40@zr40.xs4all.nl] has joined #openttd |
| 05:02 | |-| | geoffk [~geoffk@host86-130-159-207.range86-130.btcentralplus.com] has joined #openttd |
| 05:19 | |-| | pro [~http@80-240-220-232.dnat.migtel.ru] has quit [Ping timeout: 600 seconds] |
| 05:31 | |-| | lolman [~lolman@cpc3-leds2-0-0-cust55.leed.cable.ntl.com] has joined #openttd |
| 05:54 | <TrueBrain> | [07:59][07:59] <stillunknown> Truebrain: How much faster is the rendering now? <- it in fact is a bit slower, but in return we get much more freedom :) |
| 05:54 | |-| | kaan [~Klaus@82.192.152.195] has joined #openttd |
| 05:55 | <kaan> | hello |
| 05:55 | <Noldo> | TrueBrain: :P |
| 05:57 | |-| | lolman [~lolman@cpc3-leds2-0-0-cust55.leed.cable.ntl.com] has quit [Quit: Leaving] |
| 05:57 | |-| | lolman [~john@cpc3-leds2-0-0-cust55.leed.cable.ntl.com] has joined #openttd |
| 06:04 | |-| | re06011988 [~RE0601198@vol21-2-82-226-46-162.fbx.proxad.net] has joined #openttd |
| 06:06 | <TheJosh> | i have found rendering a tad better. when i ran Pile it went a tad faster |
| 06:07 | <TrueBrain> | zoom-outs run faster |
| 06:07 | <TrueBrain> | much faster in fact |
| 06:07 | <TheJosh> | althouth turning off trees is still the best way to make Pile run well in my opinion |
| 06:07 | <TheJosh> | yeah i noticed that as well |
| 06:19 | |-| | Gekko [~Gekko@CPE-60-229-217-120.nsw.bigpond.net.au] has quit [Quit: KVIrc 3.2.6 Anomalies http://www.kvirc.net/] |
| 06:22 | <CIA-1> | OpenTTD: peter1138 * r10111 /trunk/src/ (roadveh_cmd.cpp train_cmd.cpp vehicle.cpp vehicle.h): |
| 06:22 | <CIA-1> | OpenTTD: -Codechange: Add new vehicle hash table for collision detection and finding |
| 06:22 | <CIA-1> | OpenTTD: vehicles on a tile. The hash area scanned is far smaller than the old hash |
| 06:22 | <CIA-1> | OpenTTD: table, which is now used for viewport updates only. This should give a |
| 06:22 | <CIA-1> | OpenTTD: significant performance improvement for games with many vehicles. (Based on work |
| 06:22 | <CIA-1> | OpenTTD: by 'B. N. SmatZ!' and 'madman2003') |
| 06:24 | <TheJosh> | what is "many vehicles"? Pile Transport? |
| 06:24 | <TheJosh> | or would even 200 trains be considered 'many'? |
| 06:24 | <hylje> | :o |
| 06:24 | <boekabart> | TheJosh: many vehicles is so many vehicles that the game starts to slow down because of it |
| 06:24 | <hylje> | make that around 500 |
| 06:24 | <boekabart> | depends on map and hardware :) |
| 06:25 | <@peter1138> | well |
| 06:25 | <@peter1138> | it'll be faster even with only a few vehicles |
| 06:25 | <@peter1138> | but you won't notice it |
| 06:25 | <boekabart> | hey, peter1138, how does the collision handling know not to collide with other cars in same train |
| 06:25 | <TheJosh> | im gonna give pile a try |
| 06:25 | <@peter1138> | magic? |
| 06:26 | <boekabart> | seriously |
| 06:26 | <boekabart> | go to Engine and compare? |
| 06:26 | <boekabart> | (isn't that also kind of slow?) |
| 06:26 | <hylje> | self-collision |
| 06:26 | <hylje> | heh |
| 06:26 | <TheJosh> | wouldint it go to engine, looking at everything along the way, and then look at everything after itself? |
| 06:27 | <@peter1138> | boekabart: not really |
| 06:27 | <TheJosh> | isnt a train a linked list? |
| 06:27 | <@peter1138> | boekabart: the first engine is cached, so you just test if a wagon's first vehicle is yourself |
| 06:27 | <boekabart> | peter1138: maybe is (me = other->next || other = me->next) would be enough to prevent self colliding? or is 'looping' explicitly allowed... |
| 06:27 | <boekabart> | boekabart: only engine is collision detected anyway? |
| 06:28 | <boekabart> | in that case np. |
| 06:28 | <@peter1138> | boekabart: looping is explicitly allowed |
| 06:28 | <@peter1138> | it's really not that slow though |
| 06:28 | <TheJosh> | Pile is running very smootly |
| 06:28 | <@peter1138> | the slow bit was the hash lookup, heh |
| 06:29 | <boekabart> | no, if only engine checks vs others and others have a direct engine*, ok. nothing to gain at all |
| 06:29 | <boekabart> | question answered. |
| 06:29 | <@peter1138> | ah, that's why it's still slow |
| 06:29 | <@peter1138> | was still a profile build |
| 06:29 | <hylje> | :o |
| 06:30 | <TheJosh> | whats the default blitter? |
| 06:30 | <@peter1138> | 8bpp-optimized |
| 06:37 | |-| | Sug [~graeme@88-104-120-205.dynamic.dsl.as9105.com] has joined #openttd |
| 06:52 | <TheJosh> | is the blitter code changes to allow for a 32bpp blitter in the future? |
| 06:52 | <@peter1138> | *cough* present *cough* |
| 06:53 | <hylje> | :o |
| 06:53 | <hylje> | unpossible |
| 06:53 | <@peter1138> | although it only draws 8bpp sprites at the moment |
| 06:53 | <TheJosh> | unpossible? |
| 06:53 | <@peter1138> | 12:26 < TheJosh> Pile is running very smootly |
| 06:53 | <@peter1138> | did it not before? |
| 06:54 | <TheJosh> | i only see 8bpp optimised, debug and simple |
| 06:54 | <@peter1138> | yes, 32bpp is not in trunk |
| 06:54 | <TheJosh> | peter1138: more smoothly, and my other large games are too |
| 06:54 | <TheJosh> | is it much of an improvement? i guess you need 32bpp grfs to take advantage of it |
| 06:55 | <hylje> | duh |
| 06:55 | <@peter1138> | none at all |
| 06:55 | <boekabart> | hylje: not so duh, i can think of lots of improvements as soon as the buffer is 32bpp |
| 06:56 | <CIA-1> | OpenTTD: celestar * r10112 /branches/gamebalance/ (193 files in 10 dirs): [gamebalance] -Sync: r9520:9620 from trunk |
| 06:57 | <hylje> | oo |
| 06:57 | <hylje> | sync |
| 06:57 | <TheJosh> | will the 32bpp support opacity? thus allowing for semi-transperent gui objects, or chat messages/text effects to fade out? |
| 06:58 | <@peter1138> | who knows |
| 06:58 | <hylje> | think we want hw accel for silly effects |
| 06:59 | <TheJosh> | doesnt the display system (sdl) use hw acceleration? |
| 06:59 | <Eddi|zuHause2> | hw accel is supposed to be on a different level of graphics driver |
| 06:59 | <@peter1138> | no |
| 06:59 | <TheJosh> | just make the silly effects a patch option for those of us with good cpus |
| 07:00 | <TheJosh> | i can see it now: "[ x ] enable silly effects: on" |
| 07:02 | <@peter1138> | what would be nice is a (2D) opengl renderer |
| 07:03 | <TheJosh> | another video device (instead of sdl?) |
| 07:05 | |-| | lolman [~john@cpc3-leds2-0-0-cust55.leed.cable.ntl.com] has quit [Remote host closed the connection] |
| 07:07 | |-| | NukeBuster [~opera@195-241-212-152.dsl.ip.tiscali.nl] has joined #openttd |
| 07:09 | <Nickman> | peter1138: you want a 2D openGL renderer? or are you gonna make one? :) |
| 07:10 | <@peter1138> | either? |
| 07:11 | <Nickman> | :D |
| 07:11 | <Nickman> | Well, some friends and I are planning on making something like that ;) |
| 07:11 | <@peter1138> | it is "possible" now |
| 07:11 | <Nickman> | Were gonna start this summer with a project of our own |
| 07:11 | <@peter1138> | boo |
| 07:11 | <Nickman> | you just have to make another blitter? |
| 07:11 | <Nickman> | to use OpenGL |
| 07:11 | <Nickman> | hehe :D |
| 07:12 | <@peter1138> | :o |
| 07:12 | <@peter1138> | hmm |
| 07:12 | <Nickman> | To make the OpenGL renderer, do you just need to make another blitteR? or is it somewhat more complicated? :p |
| 07:13 | <hylje> | video driver |
| 07:14 | <@peter1138> | needs a bit more than a blitter, yeah |
| 07:14 | <@peter1138> | crap |
| 07:14 | <@peter1138> | GL_EXT_paletted_texture |
| 07:14 | <@peter1138> | Future NVIDIA GPU designs will no longer support paletted textures. |
| 07:14 | <@peter1138> | silly nvidia |
| 07:14 | <@peter1138> | thinking everything needs to go 32bpp |
| 07:15 | <hylje> | good excuse! |
| 07:16 | <Eddi|zuHause2> | probably they think everything else is old enough to not need to be accelerated |
| 07:16 | <@peter1138> | it might've been handy for things like palette animation... |
| 07:16 | <@peter1138> | might still be, i suppose |
| 07:17 | <hylje> | well with gpu acceleration you can do full-blown animation |
| 07:18 | <Eddi|zuHause2> | the problem is converting the old animation system to the new one |
| 07:18 | <Eddi|zuHause2> | which might not be trivial |
| 07:18 | <hylje> | possible |
| 07:19 | <Eddi|zuHause2> | i loved the palette animation in civ 1 |
| 07:19 | <Eddi|zuHause2> | shame civ 2 lost that feature |
| 07:28 | <CIA-1> | OpenTTD: peter1138 * r10113 /trunk/src/ (10 files in 2 dirs): -Fix (r10092): Missing svn properties and some Id/@file comments |
| 07:28 | <Nickman> | peter1138: are there any plans to start an openGL renderer? |
| 07:28 | <@peter1138> | not currently |
| 07:29 | <Nickman> | Well, if a branch starts, i'd be happy to try and help ;) |
| 07:39 | <hylje> | heh my gpu starts heating up when running ottd |
| 07:40 | <TrueBrain> | poor gpu |
| 07:42 | <TheJosh> | hey whats the proper process for a bug being fixed? i have just fixed a small bug, and added the patch to flyspray. is that all I have to do? |
| 07:42 | <Nickman> | :D |
| 07:42 | <boekabart> | add it as a BUG in flyspray though, not a patch. |
| 07:42 | <boekabart> | if it's a bug |
| 07:42 | <boekabart> | (so make report type Bug, and attach your Patch as suggested solution) |
| 07:42 | <TheJosh> | no I added the patch as a comment to an existing bug |
| 07:43 | <boekabart> | that is proper. |
| 07:43 | <TrueBrain> | TheJosh: we noticed that you added the patch, and someone will look at it now :) |
| 07:43 | <TheJosh> | cool |
| 07:43 | <TheJosh> | do people get emails when stuff happens? |
| 07:43 | <TrueBrain> | [14:41][14:41] <_42_> Flyspray: [FS#850] Attachment 'filename_fixes.patch' (1207 bytes) added to Bug Report 'Invalid save game name offered' by josh - http://bugs.openttd.org/task/850 |
| 07:44 | <TrueBrain> | #openttd.notice |
| 07:46 | <TheJosh> | im going to bed now |
| 07:46 | <TheJosh> | everyone have a good night/day |
| 07:46 | <TrueBrain> | have a good night sleep |
| 07:46 | <TheJosh> | i wil; |
| 07:46 | <TheJosh> | i hope that patch is of some use |
| 07:46 | |-| | TheJosh [~josh@d220-238-156-217.dsl.vic.optusnet.com.au] has left #openttd [] |
| 07:47 | |-| | glx [~glx@bny93-6-82-245-156-124.fbx.proxad.net] has joined #openttd |
| 07:47 | |-| | mode/#openttd [+v glx] by ChanServ |
| 07:47 | <boekabart> | hm, the code duplication could have been less in that patch |
| 07:47 | <boekabart> | plus, on *nix, all of those except / are allowed, right? |
| 07:48 | <Eddi|zuHause2> | i'd not be surprised if even / was allowed |
| 07:48 | <@peter1138> | heh |
| 07:48 | <boekabart> | Eddi|zuHause2: no, i just tried, not allowed |
| 07:56 | <@Belugas> | hello |
| 07:56 | <boekabart> | 'sup |
| 08:22 | <CIA-1> | OpenTTD: maedhros * r10114 /trunk/src/ (6 files): |
| 08:22 | <CIA-1> | OpenTTD: -Fix: Only load newgrf error messages if the language matches the current |
| 08:22 | <CIA-1> | OpenTTD: language. Since only one error can be loaded anyway, if the language didn't |
| 08:22 | <CIA-1> | OpenTTD: match you'd get "Undefined string". Also since we're only loading one language |
| 08:22 | <CIA-1> | OpenTTD: there's no need to use AddGRFString any more. |
| 08:25 | <Biff> | / is not allowed |
| 08:25 | <Biff> | unless you hack the filesystem :p |
| 08:30 | <TrueBrain> | I like this blitter-layer: null dirver finally really doesn't draw anything to the screen |
| 08:30 | <TrueBrain> | neither does the dedicated server (in my local working copy that is) |
| 08:30 | <TrueBrain> | poor Briannetta, no more dedicated server screenshots :p |
| 08:40 | |-| | NukeBuster [~opera@195-241-212-152.dsl.ip.tiscali.nl] has quit [Ping timeout: 600 seconds] |
| 08:41 | <@peter1138> | still wastes time doing sprite sorting |
| 08:42 | <TrueBrain> | very true |
| 08:46 | |-| | Osai [~Osai@pD9EB5E01.dip.t-dialin.net] has joined #openttd |
| 08:52 | |-| | Zr40 [~Zr40@zr40.xs4all.nl] has quit [Ping timeout: 600 seconds] |
| 08:52 | <TrueBrain> | so there, dedicated servers can still assign a blitter if they like :p |
| 08:55 | |-| | Zr40 [~Zr40@zr40.xs4all.nl] has joined #openttd |
| 08:55 | <Nickman> | :D, Briannetta will be pleased ;) |
| 08:56 | <TrueBrain> | I tihnk so too :p |
| 08:58 | |-| | lolman [~john@cpc3-leds2-0-0-cust55.leed.cable.ntl.com] has joined #openttd |
| 08:58 | |-| | lolman [~john@cpc3-leds2-0-0-cust55.leed.cable.ntl.com] has quit [Remote host closed the connection] |
| 09:05 | <kaan> | Is there any reason that improved loading algorithm is for trains only? |
| 09:06 | <@peter1138> | is it? |
| 09:07 | |-| | Osai^2 [~Osai@pD9EB5E01.dip.t-dialin.net] has joined #openttd |
| 09:07 | |-| | Osai [~Osai@pD9EB5E01.dip.t-dialin.net] has quit [Read error: Connection reset by peer] |
| 09:07 | <kaan> | hmmm,. now i cant find the patch setting :P |
| 09:07 | <kaan> | disregard me until i get control of myself please :D |
| 09:08 | <kaan> | there, well what do you know, it *was* turned off ;) |
| 09:09 | <TrueBrain> | then it mostly doesn't work, no |
| 09:09 | <TrueBrain> | dunno why, but that always happens |
| 09:09 | <TrueBrain> | when you disable something, it doesn't work! |
| 09:09 | <Nickman> | hehe :D, it's not a bug, it's a real feature! |
| 09:10 | <kaan> | maybe someone that isnt as stupid as i am today will make a bugreport :P |
| 09:12 | <kaan> | "Dear dev team, the game is working as intended! Please take action immidiately." |
| 09:17 | <Nickman> | yeah :D |
| 09:17 | |-| | re06011988 [~RE0601198@vol21-2-82-226-46-162.fbx.proxad.net] has quit [Quit: KVIrc 3.2.6 Anomalies http://www.kvirc.net/] |
| 09:18 | <Eddi|zuHause2> | what was that report? "Error: Download is much faster than upload. PS: only happens on ADSL connections" |
| 09:18 | <@peter1138> | hehe |
| 09:22 | |-| | Osai^2 [~Osai@pD9EB5E01.dip.t-dialin.net] has quit [Read error: Connection reset by peer] |
| 09:22 | |-| | Osai [~Osai@pD9EB5E01.dip.t-dialin.net] has joined #openttd |
| 09:22 | <CIA-1> | OpenTTD: belugas * r10115 /trunk/src/industry_map.h: |
| 09:22 | <CIA-1> | OpenTTD: -Codechange: Create accessors to triggers and random bits for industries. |
| 09:22 | <CIA-1> | OpenTTD: Implementation will follow soon. |
| 09:23 | |-| | Sug [~graeme@88-104-120-205.dynamic.dsl.as9105.com] has quit [Quit: Leaving] |
| 09:28 | |-| | Osai^2 [~Osai@pD9EB5E01.dip.t-dialin.net] has joined #openttd |
| 09:28 | |-| | Osai [~Osai@pD9EB5E01.dip.t-dialin.net] has quit [Read error: Connection reset by peer] |
| 09:35 | |-| | NW|Aerandir [magic.powe@90-230-201-111-no37.tbcn.telia.com] has quit [Quit: - nbs-irc 2.3 - www.nbs-irc.net -] |
| 09:38 | |-| | Aerandir [magic.powe@90-230-201-111-no37.tbcn.telia.com] has joined #openttd |
| 09:46 | |-| | lolman [~john@cpc3-leds2-0-0-cust55.leed.cable.ntl.com] has joined #openttd |
| 09:48 | |-| | boekabart [~boekabart@81.58.27.138] has quit [Quit: Leaving.] |
| 09:48 | |-| | boekabart [~boekabart@81.58.27.138] has joined #openttd |
| 09:55 | |-| | UndernotBuilder [~chatzilla@168.226.104.12] has joined #openttd |
| 09:58 | |-| | Osai^2 changed nick to Osai |
| 10:05 | |-| | helb [~helb@84.244.90.159] has quit [Quit: Logout] |
| 10:07 | <stillunknown> | peter1138: there was no performance increase in seperating the viewport and normal hashmap? |
| 10:08 | <@peter1138> | hmm? |
| 10:08 | <@peter1138> | wasn't therE? |
| 10:09 | <stillunknown> | peter1138: Please make more sense? |
| 10:09 | <@peter1138> | you said there was no performance increase |
| 10:09 | |-| | Progman [~progman@p57A1D026.dip.t-dialin.net] has joined #openttd |
| 10:09 | <stillunknown> | When did i say that? |
| 10:09 | <@peter1138> | 16:04 < stillunknown> peter1138: there was no performance increase in seperating the viewport and normal hashmap? |
| 10:09 | <stillunknown> | ? |
| 10:10 | <stillunknown> | it's a question |
| 10:10 | <stillunknown> | since i looked at the diffstat |
| 10:10 | <@peter1138> | no "there was" is a statement |
| 10:10 | <@peter1138> | "was there" is a question |
| 10:10 | <stillunknown> | Sorry for confusion. |
| 10:10 | <@peter1138> | besides, it doesn't make sense as the hashes as separated |
| 10:11 | <stillunknown> | They are seperated. how? |
| 10:11 | <stillunknown> | correction: How are they seperated? |
| 10:12 | <@peter1138> | there are... two of them |
| 10:13 | <@peter1138> | hmm, variable names suck though. never mind :p |
| 10:14 | <stillunknown> | You let the old one exist for the viewport? |
| 10:14 | <@peter1138> | yes |
| 10:15 | <@peter1138> | it's not a bottle neck for rendering, so i didn't touch it |
| 10:15 | <stillunknown> | Maybe the names should reflect the viewport. |
| 10:15 | <@peter1138> | probably it could be optimized, but some other time |
| 10:15 | <@peter1138> | yeah |
| 10:16 | |-| | lolman [~john@cpc3-leds2-0-0-cust55.leed.cable.ntl.com] has quit [Quit: Konversation terminated!] |
| 10:18 | |-| | helb [~helb@84.244.90.159] has joined #openttd |
| 10:19 | <stillunknown> | peter1138: You should also consider looking at a replacement for "realistic" acceleration, since it's quite hungry for what it does. |
| 10:21 | <@peter1138> | hehe |
| 10:21 | <hylje> | caching! |
| 10:21 | <@peter1138> | stillunknown |
| 10:21 | <@peter1138> | http://fuzzle.org/o/trunk.txt |
| 10:21 | <@peter1138> | http://fuzzle.org/o/patch-7-0b.txt |
| 10:21 | |-| | RamboRonny [magic.powe@90-230-201-111-no37.tbcn.telia.com] has joined #openttd |
| 10:22 | <@peter1138> | trunk is profiling from before the patch |
| 10:22 | <@peter1138> | 7-0b is profiling with the chosen hash values (i.e. trunk now) |
| 10:22 | <stillunknown> | You used some huge scenario for that ;-) |
| 10:22 | <@peter1138> | i really need some kind of averaging profiler, though |
| 10:23 | <@peter1138> | this "EvL" game |
| 10:23 | <@peter1138> | only 550 trains |
| 10:23 | <TrueBrain> | exactly the same amount of ticks, nice ;) |
| 10:23 | |-| | Peakki [antti@cs181000195.pp.htv.fi] has joined #openttd |
| 10:23 | <@peter1138> | TrueBrain: strange that :p |
| 10:23 | <TrueBrain> | -vnull rules :) |
| 10:23 | |-| | RamboRonny [magic.powe@90-230-201-111-no37.tbcn.telia.com] has quit [] |
| 10:24 | <@peter1138> | problem is |
| 10:24 | <stillunknown> | peter1138: results like nice |
| 10:24 | <@peter1138> | now TrainLocoHandler sucks ;( |
| 10:24 | <hylje> | :o |
| 10:24 | <TrueBrain> | lol, the blitter is rather fast :p Haha :) |
| 10:25 | <@peter1138> | hmm? |
| 10:25 | <TrueBrain> | in the profile :) |
| 10:25 | <@peter1138> | oh |
| 10:25 | <@peter1138> | yes |
| 10:25 | <@peter1138> | it would be |
| 10:25 | <@peter1138> | i'd scrolled to the top of the map so there was nothing going on |
| 10:25 | <TrueBrain> | it wasn't with -vnull? |
| 10:25 | <hylje> | heh |
| 10:25 | <@peter1138> | with -vnull yes |
| 10:25 | <TrueBrain> | ah, yes :p Then blittering already is very low :p |
| 10:25 | <TrueBrain> | with my patch even less :p |
| 10:26 | <TrueBrain> | lol |
| 10:26 | <stillunknown> | peter1138: I just hope that optimizations continue. |
| 10:26 | <stillunknown> | Fortunately there will always be the next biggest cpu consumer. |
| 10:27 | <hylje> | (un) |
| 10:27 | <@peter1138> | well TrueBrain is doing a tremendous job with blitting and rendering |
| 10:27 | <hylje> | tremendous! |
| 10:27 | <TrueBrain> | but not really optimizations :p |
| 10:35 | <stillunknown> | I know just the thing to make TrainLocoHandler look good. |
| 10:35 | <stillunknown> | Using 50-100 ships ;-) |
| 10:36 | <hylje> | wut |
| 10:36 | <stillunknown> | Ships are extremely expensive, at least for pathfinding. |
| 10:37 | <stillunknown> | expensive == use a lot of cpu power |
| 10:37 | <hylje> | oh, really? |
| 10:42 | <@peter1138> | hehe |
| 10:42 | <stillunknown> | Have you looked at this simplified acceleration patch on the forum? |
| 10:43 | |-| | lolman [~lolman@cpc3-leds2-0-0-cust55.leed.cable.ntl.com] has joined #openttd |
| 10:44 | <@peter1138> | yeah |
| 10:45 | <Noldo> | ships just have a bit more options |
| 10:46 | <CIA-1> | OpenTTD: rubidium * r10116 /trunk/src/ (fileio.cpp fileio.h misc_gui.cpp screenshot.cpp): -Fix [FS#850]: remove invalid characters (for the file system) from savegame names. Based on a patch by TheJosh. |
| 10:48 | <stillunknown> | peter1138: is it any good? |
| 10:48 | <Noldo> | We visioned with someone about making kind of rails to ships from those b-things |
| 10:49 | |-| | Hendikins [~wolfox@CPE-121-209-208-13.nsw.bigpond.net.au] has quit [Ping timeout: 600 seconds] |
| 10:49 | |-| | Progman [~progman@p57A1D026.dip.t-dialin.net] has quit [Remote host closed the connection] |
| 10:51 | |-| | Progman [~progman@p57A1D026.dip.t-dialin.net] has joined #openttd |
| 10:51 | <Rubidium> | Noldo: problems arise when people don't use buoys, so making "rails" between buoys doesn't solve the issue |
| 10:51 | <UndernotBuilder> | what about using the trams topic as an general announcements one? |
| 10:52 | <Noldo> | Rubidium: too bad for them, ships won't find their way |
| 10:52 | <Biff> | do the ship algo use A*? |
| 10:54 | <Rubidium> | Biff: yes they do |
| 10:54 | <Biff> | ok |
| 10:54 | |-| | Progman [~progman@p57A1D026.dip.t-dialin.net] has quit [] |
| 10:58 | |-| | Progman [~progman@p57A1D026.dip.t-dialin.net] has joined #openttd |
| 11:03 | |-| | Prof_Frink [~proffrink@82-43-58-81.cable.ubr04.croy.blueyonder.co.uk] has joined #openttd |
| 11:07 | |-| | HMage [Queneex@89-178-51-189.broadband.corbina.ru] has joined #openttd |
| 11:15 | |-| | HMage` [HMage@89-178-132-170.broadband.corbina.ru] has quit [Ping timeout: 600 seconds] |
| 11:17 | <Eddi|zuHause2> | hm, the bad weather arrived here... |
| 11:23 | |-| | HMage` [HMage@89-178-38-144.broadband.corbina.ru] has joined #openttd |
| 11:32 | |-| | HMage [Queneex@89-178-51-189.broadband.corbina.ru] has quit [Ping timeout: 600 seconds] |
| 11:41 | |-| | KritiK [Maxim@ppp83-237-102-185.pppoe.mtu-net.ru] has joined #openttd |
| 11:42 | <Phazorx> | morning |
| 11:44 | <TrueBrain> | he who can guess what is special about this image, gets a cookie: http://devs.openttd.org/~truelight/art10.png |
| 11:45 | <stillunknown> | It's at a low resolution. |
| 11:45 | <+glx> | no try again |
| 11:46 | <stillunknown> | One of the fields is bounded by stones. |
| 11:46 | <+glx> | no try again |
| 11:46 | <Eddi|zuHause2> | do i win if i say "nothing"? :p |
| 11:46 | <+glx> | no |
| 11:46 | <Eddi|zuHause2> | i'd expect something about a new (32bpp?) blitter |
| 11:47 | <stillunknown> | It's made with the new blitting code? |
| 11:47 | <+glx> | yes |
| 11:47 | <Phazorx> | blitting code? |
| 11:47 | <stillunknown> | But that's not special ;-) |
| 11:48 | |-| | Hendikins [~wolfox@CPE-121-209-208-13.nsw.bigpond.net.au] has joined #openttd |
| 11:48 | <+glx> | Eddi|zuHause2 said something :) |
| 11:49 | |-| | Bjarni [~Bjarni@0x50c79adc.virnxx14.adsl-dhcp.tele.dk] has joined #openttd |
| 11:49 | |-| | mode/#openttd [+o Bjarni] by ChanServ |
| 11:49 | <Eddi|zuHause2> | you kinda get that idea if you follow the "art#" series :) |
| 11:49 | <TrueBrain> | I like my art series :) |
| 11:49 | [~] | TrueBrain slips Eddi|zuHause2 a cookie: www.amazon.com |
| 11:50 | <Eddi|zuHause2> | it's definitely great :) |
| 11:50 | <Eddi|zuHause2> | i'd sell it :) |
| 11:50 | <stillunknown> | Is 32bpp still moving forward? |
| 11:50 | <UndernotBuilder> | there is a new tree |
| 11:51 | <Eddi|zuHause2> | stillunknown does not see the forest because those trees are in the way... |
| 11:51 | <UndernotBuilder> | oh, and what do you mean with 'blitting code'? |
| 11:52 | <Eddi|zuHause2> | the code that does blitting? |
| 11:53 | <Eddi|zuHause2> | i'd try the files in the "blitter" directory :) |
| 11:53 | <UndernotBuilder> | there is no "blitter" directory in my r10067 |
| 11:54 | <+glx> | UndernotBuilder: latest is r 10116 |
| 11:54 | <@Belugas> | blitter is a technical term that describe the code/object responsible of painting the pixels |
| 11:55 | <UndernotBuilder> | nope latest is r10104 |
| 11:55 | <@peter1138> | $ svn up |
| 11:55 | <@peter1138> | At revision 10116. |
| 11:55 | <@peter1138> | nope latest is r10116 |
| 11:56 | <Eddi|zuHause2> | UndernotBuilder: the blitter directory was introduced in r10092 |
| 11:56 | <Eddi|zuHause2> | "Aktualisiert zu Revision 10116." |
| 11:58 | <UndernotBuilder> | well, in one hour there will be r10116 in the nightlies page |
| 11:59 | <Eddi|zuHause2> | not if there are more commits till then :) |
| 12:00 | |-| | Wolf01 [~wolf01@host202-235-dynamic.5-87-r.retail.telecomitalia.it] has joined #openttd |
| 12:00 | <Wolf01> | hello |
| 12:01 | |-| | HMage [HMage@89-178-95-196.broadband.corbina.ru] has joined #openttd |
| 12:01 | <Wolf01> | i tried articulated vehicles, they have some glitches but they are really cool :D |
| 12:02 | <Eddi|zuHause2> | what kind of glitches? |
| 12:02 | <Phazorx> | stillunknown: is thre anything you know of that would prevent a save amde in 10048M (with your patch) to load in any other version? |
| 12:02 | <XeryusTC> | Wolf01: i got a game where an artic part ran into a house! :P |
| 12:03 | <Wolf01> | Eddi|zuHause2 sometimes the central wagon "disappear" because it is under the tail car |
| 12:03 | <Wolf01> | and the tail car is always not aligned with the first and the second, which seem to be perfectly aligned |
| 12:04 | <@peter1138> | sounds more like dodgy sprite offsets to me |
| 12:05 | |-| | HMage` [HMage@89-178-38-144.broadband.corbina.ru] has quit [Ping timeout: 600 seconds] |
| 12:05 | |-| | lolman [~lolman@cpc3-leds2-0-0-cust55.leed.cable.ntl.com] has quit [Quit: Leaving] |
| 12:08 | <XeryusTC> | Wolf01: that happens when you use the hiroshima tram set |
| 12:08 | <XeryusTC> | i know of that set that it has some odd offsets |
| 12:08 | <Wolf01> | yeah |
| 12:09 | <Wolf01> | i use that |
| 12:09 | <Wolf01> | is the only set which has articulated vehicles, isn't it? |
| 12:09 | <XeryusTC> | that set has some sprite offsets that are not entirely correct |
| 12:09 | <XeryusTC> | afaik it does indeed |
| 12:10 | |-| | e1ko [~L@205.117.broadband9.iol.cz] has joined #openttd |
| 12:10 | <Wolf01> | i hope for a serbian tram set update with arv |
| 12:11 | <CIA-1> | OpenTTD: maedhros * r10117 /trunk/src/economy.cpp: |
| 12:11 | <CIA-1> | OpenTTD: -Fix [FS#863]: When deleting the vehicles of bankrupt players, delete trains as |
| 12:11 | <CIA-1> | OpenTTD: a whole rather than each part individually, as that leads to invalid tests on |
| 12:11 | <CIA-1> | OpenTTD: parts that have already been deleted. |
| 12:12 | |-| | re06011988 [~RE0601198@vol21-2-82-226-46-162.fbx.proxad.net] has joined #openttd |
| 12:14 | <stillunknown> | Phazorx: yes, probably |
| 12:15 | <stillunknown> | It's larger and has a newer revision than openttd expects. |
| 12:17 | <Phazorx> | stillunknown: i'm very interested in somehow fixing that |
| 12:17 | <Phazorx> | as a one time deal |
| 12:17 | <Phazorx> | since i have developd on that build |
| 12:17 | <Phazorx> | something rather large |
| 12:18 | <Phazorx> | but i want to be able to load it on other versions as well for testing purposes |
| 12:18 | <@Bjarni> | hmm |
| 12:18 | <XeryusTC> | stillunknown: you save the cache to the savegame? |
| 12:19 | <@Bjarni> | I say you should use a clean build and modify it to load the savegame in question and clear the cache and then save it as a normal savegame |
| 12:19 | <stillunknown> | XeryusTC: no, an extra variable was added, if the train was crossing itself or not |
| 12:19 | <Phazorx> | Bjarni: no builds i have can load any of saves made with 10048M |
| 12:19 | <stillunknown> | the cache was stored in the map |
| 12:20 | <@Bjarni> | Phazorx: it should be made. Nobody coded anything to solve this issue |
| 12:20 | <stillunknown> | You would need to modify to undo the changes, then somehow fake it into lowering the version number. |
| 12:20 | <UndernotBuilder> | for when a new feature for openttd? :( |
| 12:22 | <Phazorx> | stillunknown: modify to undo the changes? |
| 12:23 | <stillunknown> | Phazorx: Your savegame is larger than the usual savegame, it needs to be loaded by a version which shrinks it back to the usual size. |
| 12:24 | <Phazorx> | stillunknown: would you happen to have that version ? |
| 12:24 | <Phazorx> | cuz i have about dozen differen bulds and most of them can not load that |
| 12:24 | <stillunknown> | At the moment i can't help you with this, when i have time i will see if i write some code to undo this. |
| 12:24 | <stillunknown> | (i will ask for the savegame then) |
| 12:25 | <Phazorx> | is it possible to patch the save to make other builds to belive it is okay to load? |
| 12:26 | <stillunknown> | Phazorx: i hope it's possible to fake the savegame into an older version, i have no experience with this however |
| 12:27 | <Eddi|zuHause2> | the easiest solution would be to cut out all changes other the savegame changes from the patch |
| 12:27 | <Eddi|zuHause2> | and then compare the two changed versions |
| 12:27 | <stillunknown> | The version needs to be reduced as well. |
| 12:28 | <Phazorx> | my idea was - pausing sometihng |
| 12:28 | <Phazorx> | loading with vanilla 48 and 48m |
| 12:28 | <Phazorx> | saving in both |
| 12:28 | <Phazorx> | then comparing |
| 12:28 | <Phazorx> | then patching 48 save with 48m differences |
| 12:28 | <Eddi|zuHause2> | no, i mean if he wants to just compare the behaviour, he can just modify the normal build to load the newer version, not change the version back |
| 12:28 | <Phazorx> | and if that works - apply same idea to the save i want to use |
| 12:29 | <Phazorx> | Eddi|zuHause2: i'd like to be able to use thats ave game in general |
| 12:29 | <Eddi|zuHause2> | that's gonna be much more problematic, Phazorx |
| 12:30 | <Eddi|zuHause2> | as the savegame versioning is not supposed to handle that case |
| 12:30 | |-| | Osai [~Osai@pD9EB5E01.dip.t-dialin.net] has quit [Read error: Connection reset by peer] |
| 12:30 | |-| | Purno [~Purno@5357D37C.cable.casema.nl] has joined #openttd |
| 12:30 | |-| | Osai [~Osai@pD9EB5E01.dip.t-dialin.net] has joined #openttd |
| 12:31 | <Phazorx> | hmm |
| 12:31 | |-| | thgergo [~th_gergo@dsl51B65DAF.pool.t-online.hu] has joined #openttd |
| 12:32 | <Phazorx> | so i take it code modification is necessary to be able to resave the game |
| 12:32 | <Rubidium> | yes |
| 12:32 | <Phazorx> | is it just the bump of save version which eneds to be dealt with of there is more? |
| 12:32 | <Rubidium> | OpenTTD's savegames are backward compatible, not forward compatible |
| 12:33 | <Phazorx> | i'm trying to load the save in newever version |
| 12:33 | <Phazorx> | not older |
| 12:33 | <stillunknown> | In a way, that version was "newer". |
| 12:33 | <Eddi|zuHause2> | no, the modificated build is "newer" than the umodificated build |
| 12:33 | <Phazorx> | than any unmodified? |
| 12:34 | <Phazorx> | 1M is newer than 9999 ? |
| 12:34 | <Rubidium> | Phazorx: not necessarily |
| 12:34 | <Eddi|zuHause2> | Phazorx: not if the change made in 1M was merged back to trunk later |
| 12:34 | <Phazorx> | change is not in truck for sure |
| 12:34 | <Rubidium> | for example a savegame made with trunk r10001 won't load in 0.5-r10083, even though the svn revision number is higher for 0.5 |
| 12:35 | <Phazorx> | however i dont get how tracking of that is done in code asuide of numbers |
| 12:35 | <Phazorx> | Rubidium: different trees |
| 12:35 | <Phazorx> | my case is same tree different branches i guess |
| 12:35 | <XeryusTC> | Phazorx: there's a save version number in saveload.[hcpp] IIRC |
| 12:36 | [~] | Phazorx goes to fetch buildttd |
| 12:36 | <Rubidium> | Phazorx: so, applying stillunknowns patch made you "go" on a different tree too |
| 12:36 | |-| | Osai [~Osai@pD9EB5E01.dip.t-dialin.net] has quit [Read error: Connection reset by peer] |
| 12:36 | <Phazorx> | Rubidium: at this stage i want to be able just to fake it so i can load the game somewhere |
| 12:36 | |-| | Osai [~Osai@pD9EB5E01.dip.t-dialin.net] has joined #openttd |
| 12:37 | <Phazorx> | and btw EvsL widely used for trsing now is smade with modifed code as well |
| 12:37 | <Phazorx> | it has copy/paste patch on top of that |
| 12:37 | <Phazorx> | but loads just fine everywhere :/ |
| 12:37 | <Rubidium> | uhm, and in English? |
| 12:37 | <Rubidium> | EvsL? |
| 12:38 | <Phazorx> | ughm... this was badly spelled english but i think it's comprehendable :) |
| 12:38 | <Rubidium> | trs? |
| 12:38 | <Phazorx> | http://openttdcoop.ppcis.org/blog/2007/04/22/express-and-local-ml-separation |
| 12:38 | <Eddi|zuHause2> | Phazorx: try removing all files from stillunknown's patch except saveload.*, and apply this modified patch to a clean build |
| 12:38 | <Rubidium> | Eddi|zuHause2: won't work |
| 12:38 | <+glx> | Phazorx: I think copy/paste doesn't touch saveload code |
| 12:38 | <Phazorx> | Eddi|zuHause2: i tihnk i need reverse of that |
| 12:39 | <Rubidium> | the problem is that Phazorx's vehicle savegame chunck isn't the right size for trunk |
| 12:39 | <Phazorx> | glx but it is modded copraed to regular |
| 12:39 | <Phazorx> | different branch just like stillunknowns |
| 12:39 | <Phazorx> | glx, i'm trying to figure out what needs to get changed to convert save into somethingf eatable for other versions |
| 12:40 | <+glx> | remove the savegame changes |
| 12:40 | <Phazorx> | Rubidium: is that confirmed? |
| 12:40 | <Rubidium> | where are EvsL and trs defined? |
| 12:40 | <Phazorx> | Rubidium: "Express vs Local" |
| 12:41 | <Phazorx> | trs = tes :) |
| 12:41 | <Phazorx> | i meant testing |
|
