IRC logs for #openttd on OFTC at 2016-04-29
⏴ go to previous day
00:04:13 <V453000> just photoshop improvements :)
00:07:04 *** orudge` has joined #openttd
00:07:09 <Wolf01> I really like the animal farm
00:12:59 <andythenorth> V453000: what’s new in those? o_O
00:13:09 *** ToneKnee has joined #openttd
00:13:35 <andythenorth> they all look quite badass, but what’s changed?
00:14:11 <V453000> sec making comparison page already
00:14:16 *** minimoo has joined #openttd
00:14:39 <andythenorth> the mines are substantially awesome, I need to draw a bunch of FIRS mines, I should make them even less realisms
00:16:51 <andythenorth> I was just sick in my mouth
00:16:59 <andythenorth> there is not realisms, and then there is Bad Ideas
00:19:52 <andythenorth> Wolf01: how many Porsches will you buy then?
00:20:46 <andythenorth> V453000: you have added much weathering? o_O
00:20:48 <andythenorth> and some contrast?
00:21:00 <V453000> been painting on it for 2 days
00:21:06 <V453000> yeah basically everything is more fucked up
00:21:14 <V453000> somewhere some road stuff etc
00:21:56 <andythenorth> you are like a model train fan now :)
00:22:48 <Wolf01> why did you remov slugs?
00:22:57 *** ConductorCat has joined #openttd
00:24:06 <andythenorth> V453000: lighting more pleasing, you learnt more skills from BRIX? Or just more time?
00:24:25 <V453000> lighting is generally unchanged, I literally only touched photoshop
00:24:41 <andythenorth> what did you do, dump an S shape into curves? o_O
00:24:43 <V453000> utilizing photoshop for 3D postproduction is what I do for factorio a lot
00:25:21 <V453000> I see what you mean Wolf01 , but they never made it into official release :D
00:26:02 <andythenorth> V453000: boost highlights, lower shadows, clamp the mid-range
00:26:11 <andythenorth> is how I would have done it :P
00:26:27 <V453000> yeah, but that never gives a super nice result
00:26:49 <V453000> basically, have the render, duplicate it, set mode to multiply, put a mask to it, paint the mask
00:26:51 <andythenorth> have you literally painted all that?
00:27:01 <V453000> my hand is dying of wacom
00:27:26 <V453000> honestly, I have been using the tablet for like 8 months now and I can't imagine working without it anymore
00:27:36 <andythenorth> doesn’t any longer
00:27:42 <andythenorth> you get claw hand
00:27:57 <andythenorth> just wait until you get car seat wrist :P
00:28:00 <andythenorth> then you’ll be crippled
00:28:02 <V453000> luckily I don't paint all day long usually
00:28:37 <andythenorth> every parent I know has experienced this thing where they strain the tendons in one wrist by lifting car seat into car (or out)
00:28:59 *** andythenorth has left #openttd
00:44:18 *** Myhorta has joined #openttd
01:39:34 <Samu> when entering a custom number of towns, i can't input a value higher than 5000, but when generating with a high number of towns on a 4096x4096, it says it's generating +12k towns :o
01:44:28 <_dp_> minimap is a mess though)
01:45:13 <Samu> original land generator should lock snow height to 7
01:45:32 <Samu> terragenesis sets it to 15
01:45:41 <Samu> but that's good for terragenesis only
01:48:06 <Samu> ugh, height levels break some original land generator stuff
01:48:23 <Samu> minimap is too green even for mountains
01:48:37 <Samu> where are the brighter greens
02:30:27 *** minimoo has joined #openttd
02:57:17 *** supermop has joined #openttd
03:42:55 *** Cybertinus has joined #openttd
03:54:04 *** AdmiralKew has joined #openttd
04:47:54 *** Eddi|zuHause2 has joined #openttd
06:46:22 *** Klanticus has joined #openttd
06:52:06 *** sla_ro|master has joined #openttd
07:01:18 *** ConductCat has joined #openttd
07:46:08 *** Alberth has joined #openttd
07:46:08 *** ChanServ sets mode: +o Alberth
07:54:43 *** supermop_ has joined #openttd
07:57:32 *** Hiddenfunstuff has joined #openttd
08:00:27 *** Biolunar has joined #openttd
08:42:34 *** MonkeyDrone has joined #openttd
08:47:49 *** empezar has joined #openttd
09:12:46 *** Myhorta has joined #openttd
09:23:24 *** ConductingCat has joined #openttd
09:27:48 *** ConductorCat has joined #openttd
10:04:14 <Leanden> firstly ive uploaded my ssh public key to the dev page
10:04:26 <Leanden> but im trying to manually compile my test GRF using nmlc
10:04:31 <Leanden> but it keeps returning an error
10:04:38 <Leanden> Parsing ...Illegal character '#' (character code 0x23) at "src\BRTrains.pnml", line 9, column 1
10:05:10 <Alberth> paste the first say 20 lines into paste.openttdcoop.org
10:05:34 <Alberth> and yes, # is not in the NML language :)
10:06:18 <Alberth> ah, right, that's not an nml file
10:06:40 <Alberth> do you have cpp or gcc or cc ?
10:08:13 <Leanden> so what do i type into my nmlc to make it compile?
10:08:33 <Alberth> you don't, make does that for you
10:08:50 <Alberth> type make -n to see what it would do if you type make
10:09:16 <Alberth> many improvements in the yetis industries V
10:09:31 <V453000> yeah, all just photoshop painting :)
10:09:35 <Alberth> you made the yetis clean the mine platform
10:10:04 <Alberth> I was wondering about the oil wells thing, shouldn't there be a fence around the moving things?
10:10:14 <Leanden> "No repository found*
10:11:10 <Alberth> it tries to get revision information from the current directory, which fails of course, if your directory in not under source control
10:11:29 <Leanden> ok so ive updated my directory to be a .hg repository
10:11:34 <Leanden> make: *** No rule to make target `BRTrains.grf', needed by `all'. Stop.
10:11:39 <Leanden> I now get this error only
10:11:52 <V453000> one of the economies only has worker yards and food plants
10:11:53 <Alberth> V: your ore mine does have a fence around the hole :p
10:12:10 <V453000> you produce food from yetis, might as well not care about safety
10:12:22 <V453000> yeah point :P that is to prevent stealing of the precious brick-ore
10:12:56 <Alberth> doesn't look very dangerous, that fence, though :)
10:13:55 <Leanden> [ -f .version ] && [ "`cat .version`" = "r-1" ] || echo r-1 > .version echo "[MD5] BRTrains.md5" md5sum BRTrains.grf | sed "s/ / /;s/ / /" > BRTrains.md5 cat: BRTrains.md5: No such file or directory echo "[TXT] docs/readme.txt" cat docs/readme.ptxt \ | sed -e "s/{{GRF_TITLE}}/BRTrains r-1/" \ | sed -e "s/{{GRF_ID}}//" \ | sed -e "s/{{REPO_REVISION}}/-1/" \ | sed -e "s/{{FILENAM
10:14:18 <Alberth> yeah, don't paste in here, use a pastebin
10:14:39 <Alberth> np, we all make mistakes :)
10:15:13 <Leanden> so i created a blank file called BRTrains.grf for it to write into
10:15:22 <Leanden> but now im getting this missing .md5 error
10:16:14 <Alberth> creating the blank .grf file just makes that it skips building the grf
10:16:38 <Alberth> you want automatic building on the compile farm, I guess?
10:16:52 <Leanden> yes but i understand thats broken atm due to a http read fault
10:17:14 <Alberth> apparently, but ssh is on the way
10:17:37 <Alberth> nothing I can do with that, sorry
10:17:53 <Alberth> planetmaker: ^ can you help Leanden ?
10:18:35 <Alberth> if you have linux, md5 is easy to add
10:18:45 <Alberth> otherwise, euhm, no idea
10:20:00 <Leanden> nope windows unfortunately
10:20:01 <Alberth> you might want to compare with another project that uses makefiles (which is pretty much all of them)
10:20:17 <Leanden> im using SBBset as my comparison
10:20:34 <Leanden> (in fact ive literally copied their makefile and changed the filepaths)
10:21:10 <Alberth> that's a useful strategy in many cases :)
10:21:40 <Alberth> no md5sum thingie from where-ever you got gcc?
10:22:00 <Leanden> when i downloaded nmlc
10:22:05 <Leanden> there was a md5 with that
10:22:08 <Leanden> is that what im missing?
10:22:35 <Alberth> don't know, I have no idea what md5 it comes with
10:22:39 <Leanden> no actually, thats a stupid comment, that was the md5 for the installer ;)
10:26:27 *** kais58_ is now known as kais58|AFK
10:26:50 <Leanden> ok so i got an md5 generator and created the md5 hashfil
10:27:27 *** kais58|AFK is now known as kais58_
10:27:29 <Leanden> so was this successful or not? :P
10:28:12 <Leanden> i think not because i still dont have an NML file to run nmlc on
10:29:21 <Alberth> gcc -C -E -nostdinc -x c-header -o BTtrains.nml BTtrains.pnml
10:30:32 <Alberth> first command runs the C pre-processor, eliminating all the #define and #include, second command builds the grf
10:31:10 <Alberth> remarks like "it don't work" give no clues at all what to fix
10:31:17 <Leanden> fatal error: src/compatibility.pnml: No such file or directory #include "src/compatibility.pnml"
10:31:21 <Leanden> sorry i was getting there ;)
10:31:59 <Alberth> #include is relative to the source file it is in, I think
10:32:27 <Leanden> the base pnml shouldnt be in the src folder then
10:32:32 <Alberth> alternatively, you can add a -Isrc and kill all src/ prefixes
10:32:57 <Leanden> nmlc ERROR: "src/header.pnml", line 9: Expected a string literal of length 4 Included from: "BRTrains.pnml", line 10
10:33:34 <Alberth> now fix the nml errors :)
10:34:24 <Leanden> hmmm so theres an error in my header pnml?
10:34:44 <Leanden> oh line 9 is my GRF ID
10:34:48 <Alberth> nml detects it there, but that does not imply the fix should also be there
10:34:54 <Leanden> grfid: "JK\08\04\16";
10:35:53 <Leanden> nmlc ERROR: "src/header.pnml", line 13: Unrecognized identifier 'REPO_REVISION' encountered
10:37:30 <Alberth> but these are generated from the makefile
10:38:21 <Leanden> i think my make may have failed actually
10:38:22 <Alberth> you can fiddle with them manually, but it's messy
10:40:04 <Alberth> unix2dos doesn't do what the makefile expects, I think, I have such a unix2dos too
10:40:46 <Alberth> I fixed that somewhere, but where?
10:42:12 <Alberth> ah, the make-nml project, I think do you have a line like
10:42:12 <Alberth> UNIX2DOS_FLAGS ?= $(shell [ -n $(UNIX2DOS) ] && $(UNIX2DOS) -q --version 2>/dev/null && echo "-q" || echo "")
10:43:02 <Alberth> make-nml is sort of the base of the makefile stuff, which you are supposed to copy when making a new project
10:43:23 <Alberth> most people however copy from a random other project, and never update their files
10:43:31 <Alberth> which is fine, until it breaks :p
10:44:10 <Leanden> where do i find that file?
10:45:41 <Leanden> ye i saw that page earlier but couldnt find a download
10:46:04 <Alberth> not sure there is one, tbh
10:46:12 <Alberth> but it makes sense to have that
10:47:46 <Alberth> hmm, doesn't seem to work either
10:47:54 <Leanden> ive manually downloaded the makefile
10:49:01 <Leanden> lets see how this works
10:49:32 <Leanden> yay yhe make file works
10:50:23 <Leanden> im now back up to the REPO_VERSION error
10:53:56 <Alberth> you're running this in a repository directory?
10:54:41 <Leanden> its a local only repository in Tortoisehg
10:56:06 <Alberth> oh, that's fine, tortoisehg is just a frontend for hg
10:56:16 <Leanden> but its an unrecognized identifier
10:57:45 <Alberth> there is a point where they get replaced
10:58:05 <Alberth> maybe a custom_tags file or so that gets generated?
10:59:20 <Alberth> nvm, that's for your readme file
11:03:29 <Alberth> scripts/Makefile.def has the commands to derive revisions, it seems
11:03:47 <Alberth> there are tooo many makefiles in SBB set :(
11:04:21 <Alberth> hg id -n | cut -d+ -f1 gets the revision
11:05:02 <Alberth> but hg will only work in a directory in the repository
11:05:06 <Leanden> oh i scrapped the sbb ones
11:05:09 <Leanden> and im using the base ones now
11:06:22 <Leanden> ive temporarily overrident he identifier with the number 1
11:06:37 <Alberth> findversion.sh does that now
11:15:50 <Leanden> well i have findversion.sh in there
11:15:55 <Leanden> but it doesnt appear to be doing anything
11:16:13 <Leanden> its even generated a .version file
11:20:07 <Alberth> 0749fe4d44f9 5438 M v5438M (0749fe4d44f9) 2014-11-21 <-- afaik it produces this kind of output
11:20:58 <Alberth> does hg id do anything?
11:21:15 <Alberth> if not your working directory is wrong
11:24:00 <Leanden> it returns 0000000000000 tip
11:26:59 <planetmaker> revision 0000 tip certainly means there's nothing checked-out
11:30:40 <Leanden> well ive now managed to compile it all the way through to the end
11:33:39 <Leanden> obviously with the REVISION_VERSION identifier substituted
11:35:30 <Leanden> nothing in game though :D
11:35:59 <Alberth> ah, not added files and committed to the repo yet ?
11:40:39 <Leanden> ok hg id now returns: d62ab68b6469+ tip
11:40:48 <Leanden> but the REPO_VERSION identifier is still failing
11:45:18 * Leanden slaps planetmaker around a bit with a large fishbot
11:48:03 <Leanden> im stuck waiting for my hg key to be activated on the dev repository :D
12:02:37 *** Islacrusez has joined #openttd
12:13:15 *** Deactivated has joined #openttd
12:14:09 *** Deactivated has left #openttd
12:32:09 <V453000> Wolf01: I guess I should put slugs in the farm :)
12:33:32 *** Wolf01 is now known as Guest2236
12:33:32 *** Wolf03 is now known as Wolf01
12:42:37 *** JacobD88 has joined #openttd
13:11:03 *** Wolf01 is now known as Guest2239
13:11:04 *** Wolf03 is now known as Wolf01
13:11:08 <Samu> bool IsDead(CompanyID slot) const
13:14:59 <Samu> im not sure what void and const do though
13:15:45 <Alberth> const means the function does not change the object
13:16:15 <Alberth> virtual means derived classes may redefine the function
13:21:07 <Samu> static bool IsDead(CompanyID slot)
13:24:16 <Wolf01> static mean the function is callable without an instance (and has some other nice properties)
13:25:02 <Wolf01> also, it might have a cached value
13:27:45 <Alberth> and it's not available outside the .cpp file it is in
13:28:42 <Alberth> oh, it's a function member eh? the latter only holds for normal functions, sorry
13:33:28 <Samu> i got to copy paste the whole function?
13:36:21 <Samu> then i used the IsDead in lines 111 and 130 so far
13:36:28 <Samu> havent edited the rest yet
13:38:26 <Alberth> it's much better if you make the already available function available everywhere you need it
13:38:57 <Alberth> duplicating code creates major headaches if you want to change thigs
13:39:43 <Samu> it is also used in another big function, the AI Debug window function
13:39:50 <V453000> I wonder if the best way to build factorio railways is one huge 1-way loop
13:40:33 <Samu> hmm, so where do i put this IsDead at
13:41:03 <Alberth> V: if all trains must vist A -> B -> C -> ... -> A, then looks like it :)
13:41:16 <V453000> no, even when they don't
13:41:28 <V453000> building junctions is generally quite an issue without bridges
13:41:36 <V453000> chain signals are great but trains still need to cross
13:41:42 <Samu> sec, let me get the list of all functions it would help
13:41:46 <Alberth> no worm holes, just no bridges :p
13:42:37 <Samu> struct AIConfigWindow : public Window {
13:42:40 <Alberth> double track around the stations they don't need to transfer cargo?
13:42:44 <Samu> struct AISettingsWindow : public Window {
13:44:30 <Alberth> a number of connected circles like an 8 but more circles could work perhaps
13:44:31 <Samu> struct AIDebugWindow : public Window {
13:44:31 <Wolf01> V, do you bother about delivering speed in factorio?
13:44:59 <Wolf01> I bother about stuff getting delivered and not chewed by biters around the map
13:45:38 <Wolf01> also... all my bases are under attack...
13:46:14 <V453000> just blueprint more laser wall :D
13:47:04 <Wolf01> yes, you can defend every single outpost, but I really need a chinese great wall now
13:47:50 <Wolf01> aliens seem to have learned to pass between the outposts instead of attacking them
13:48:30 <Samu> there's an already available function inside struct AIDebugWindow : public Window {, it's where I copied the code from
13:48:46 <Samu> how do i make it available for the other 2
13:49:22 <Alberth> move it out of the window class/struct
13:50:14 <V453000> yeah, great wall is usually a good idea
13:50:15 <Alkel_U3> Yesterday I tried the oxygen mod with train outposts mod. I died so fast I didn't even get to see a biter on radar
13:57:11 <Wolf01> I think I'll install the shuttle train mod and the fat controller mod, at least I could call a train to go to different outposts
13:58:20 <Wolf01> :( the attachment does not exist anymore
14:00:31 <V453000> well the cursed rail thing is cute but it's just an unobstructed separater railway
14:00:40 <V453000> I don't have a problem with making my factory fit elsewhere
14:00:47 <V453000> I have the problem of trains colliding with each other
14:00:55 <V453000> which I bet this thing keeps
14:01:49 <Wolf01> how do you get colliding trains=
14:02:17 <V453000> at the junction of course
14:02:26 <V453000> if you can't have multilevel junction
14:02:57 <V453000> I guess tunnels would work
14:03:14 <Wolf01> I set up chain signals and trains wait for free block without blocking the junction
14:03:42 <Wolf01> I don't use it, no need to
14:14:00 <V453000> well that's bad as well, the extra tunnel should require quite a bit of space so that it costs you stuff
14:14:04 <V453000> not just "we can cross now"
14:29:37 <Samu> how do you pause a script? i see code about IsPaused
14:29:45 <Samu> how do i do this in a game?
14:30:50 <Wolf01> iirc IsPaused() is when the script is doing nothing because it doesn't need to do anything
14:31:40 *** Supercheese has joined #openttd
14:31:56 <Flygon> Where was that thing that put townnames onto scenario maps again?...
14:32:08 <Wolf01> mmmh, I don't want to know what will happen when the fire nation will attack the southern outpost...
14:32:12 <Flygon> I'm bored, and a friend of mine is daring me to make a 4096*4096 Australia scenario
14:32:19 <Flygon> So now I'm going "Fuggit, fine"
14:32:28 <Flygon> Even though I think a 16k*16k scenario is a better use of time :P
14:35:03 <V453000> one thing is realism fucking which is sensible, but just common sense is important
14:35:06 <V453000> having trains cross is weird
14:35:57 <Wolf01> I remember ottd having 90° turns and train crossing themselves
14:37:16 <Samu> i don't know how to trigger the execution of these lines in a game
14:40:16 <Alberth> looks like hitting a debugging breakpoint
14:42:05 <Samu> I don't see a Continue button in AI Debug window... what am i missing
14:46:18 <V453000> Wolf01: doesn't mean it is a good thing :P
14:46:32 <Wolf01> Samu: what you do need to continue? the AI is dead
14:47:11 <Alberth> Samu: no explanation at the wiki about the break thing?
14:48:29 *** Eddi|zuHause2 is now known as Eddi|zuHause
14:52:24 <Samu> ah i did not know about this
14:53:38 <Samu> interesting, the continue button is there
15:01:49 *** Prof_Frink has joined #openttd
15:08:39 <Samu> nice, my code replacement works
15:08:41 <Samu> !IsDead(this->ai_debug_company)
15:12:44 <Samu> i must have an IsDead compatible for all the 3 different functions
15:12:55 <Samu> or whatever they're called
15:17:30 <Samu> hmm i'm unsure if my IsDead adaptation works for all the 3 structs
15:27:36 *** sla_ro|master has joined #openttd
15:34:16 <Samu> i'm gonna ruin the entire code, see if it works
15:34:53 <Alberth> no "this" in the moved function
15:35:19 <Alberth> so it shouldn't access anything from the debug window
15:35:43 <Samu> !IsDead(ai_debug_company)
15:35:55 <Samu> it's originally !this-IsDead()
15:36:58 <Alberth> yes, the function is now not a member any more, so you can't access it through this
15:37:58 <Samu> I must have screwed somewhere, because if (!IsDead(this->ai_debug_company)) { was working
15:38:57 *** tokai|noir has joined #openttd
15:38:57 *** ChanServ sets mode: +v tokai|noir
15:56:54 <peter1138> er, you can make a heightmap however you like
15:57:01 <peter1138> open image editor, draw...
15:57:52 <Flygon> Yes, but from real world locations
15:59:14 <Samu> now I can replace this huge line "(this->slot == OWNER_DEITY && !Game::GetInstance()->IsDead()) || (Company::IsValidAiID(this->slot) && !Company::Get(this->slot)->ai_instance->IsDead())" to just this line "!IsDead(this->slot)"? or is it "!IsDead(slot)"?
15:59:37 <Samu> this->slot vs slot is what's still confusing me
16:00:13 *** supermop has joined #openttd
16:01:05 <Alberth> sometimes they are the same thing and sometimes not
16:01:27 <Alberth> this->slot is the "slot" variable in the class/struct
16:01:58 <Alberth> slot is a local variable in the code, or, if it does not exist, this->slot in the class/struct
16:02:44 <Alberth> openttd code however forbids the latter, hence all code uses "this->" prefix to refer to variables (and functions) in the class/struct
16:03:25 <Alberth> it needed a cliff hanger :p
16:08:25 <Wolf01> this.x vs this->x is that confuses me :)
16:10:23 <argoneus> so either (*this).x or this->x
16:11:19 <Wolf01> ok, then I'm confused about this and (*this)
16:11:26 <Wolf01> I just can't understand why C++ is not able to handle the difference by itself
16:11:32 <argoneus> *this resolves a value where it points
16:11:49 <Alberth> Wolf01: it can, that's why you get an error :p
16:12:17 <argoneus> "this" is a pointer to the current instance of the class/struct
16:12:33 <Alberth> it's largely historical, I think
16:12:38 <argoneus> it has to be like that
16:12:55 <argoneus> you want to access one specific instance of the class in the memory
16:13:10 <argoneus> so you need a pointer to the one specific instance
16:13:21 <Alberth> argoneus: what Wolf01 is arguing about is that there is no point in making the difference. You cannot do . on a pointer anyway
16:13:41 <Alberth> so why not automagically . to ->
16:14:01 <Alberth> like eg Java, and I guess C#
16:14:27 <argoneus> I guess that's historical then
16:14:34 <argoneus> you can't do (*this).bla in java
16:14:47 <argoneus> in java everything is a magic reference or whatever they are
16:15:02 <argoneus> and you pass around reference
16:15:25 <Alberth> I think the idea is that * and -> are different on objects, and you want consistency in operations
16:15:59 <Alberth> argoneus: not entirely, the simple types (int, float etc) are by value
16:16:05 <Alberth> to make it simpler :p
16:16:14 <argoneus> BUT WAIT, THERE'S MORE
16:16:17 <argoneus> you can do Integer x;
16:16:40 <Alberth> List<int> doesn't work
16:16:50 <Alberth> although they are working on it
16:16:53 <Wolf01> in C# everything is an object, and everything uses "."
16:17:04 <Alberth> nope, must be List<Integer>
16:17:12 <Wolf01> (and you can do List<int> in C#)
16:17:14 <argoneus> "the java specification forbids the use of primitives in generics"
16:17:23 <argoneus> for some reason I never needed this I guess
16:17:49 <argoneus> thanks, I learned something new
16:17:50 <Alberth> java spec is borked at lost of places :)
16:18:11 <peter1138> Does java have unsigned integers yet?
16:18:58 <Alberth> although you can just use the 32 bits, if you ignore the weird values
16:19:24 <argoneus> this is one of the reasons why packing structs is annoying with Java
16:19:26 <argoneus> you need to use byte arrays
16:19:36 <argoneus> I tried doing something like that for openttd and it was a pain
16:19:47 <argoneus> openttd uses packed structs for networking, right
16:20:16 <Alberth> quite normal to use packed structs :)
16:20:26 <argoneus> I can't say I've seen it somewhere else
16:20:32 <argoneus> then again I've never really done networking on a low level
16:20:37 <Alberth> although I always expand it to reading bytes and glueing them together again
16:21:21 <argoneus> I feel slightly bad for picking up software engineering at uni
16:21:32 <argoneus> I thought it would teach me how to code
16:21:38 <argoneus> and how to work with large codebases
16:21:48 <Alberth> I could have told you that :p
16:22:12 <Alberth> many people believe CS is about programming :)
16:22:14 <argoneus> well, in the end, these degrees are similar, right
16:22:22 <argoneus> in terms of practical usability
16:22:28 <argoneus> you have to learn most things in the field anyway
16:22:43 <argoneus> and I know the basics of CS and automata and stuff
16:23:06 <argoneus> a friend of mine picked up CS instead of softeng and the only difference is when I was drawing UML diagrams he was writing a front-end for gcc
16:23:30 <argoneus> (which I suppose is much more fun and useful)
16:23:56 <Alberth> euhm, not sure about that :p
16:24:34 <Alberth> in general, uni teaches you to find structure in apparent chaotic problems
16:25:03 <Alberth> and then use that structure to solve the problem
16:27:19 <Alberth> how to work with tools, documents, and programming is learned by experience, and the uni doesn't have enough hours to teach you that
16:27:32 <Alberth> since it takes several years at least
16:27:46 <argoneus> also each company has different styles and policies and technologies
16:27:50 <argoneus> and it's physically impossible to cover those
16:28:21 <Alberth> but your "find structure in apparent chaos" skills do work there
16:28:33 <Wolf01> <Alberth> in general, uni teaches you to find structure in apparent chaotic problems <- and I reaaly need to do that but I never made it to that point of the course :(
16:28:35 <Alberth> which is why you are able to find your way around it
16:29:49 <Alberth> Wolf01: ha :) Well, I found it doesn't teach you how to recognize apparent chaos from real chaos either :)
16:29:58 <argoneus> by the way, real talk, you have a lot more experience than me
16:30:04 <Alberth> I found out after 10 years :p
16:30:11 <argoneus> is it a bad idea to work at a start-up?
16:30:25 <argoneus> I used to think it was a hip thing where you have to work 80 hours a week and may or may not get results
16:30:39 <argoneus> but the flip side is you get to work on non-legacy code with enthusiastic people who don't suffer from depression
16:30:46 <argoneus> does it generally work or not work?
16:31:17 <Alberth> it really depends on what you like and need to have around you
16:31:22 <argoneus> I'm not sure what kind of dev job I want to do :(
16:31:43 <argoneus> I considered trying to apply for Factorio since that seems really fun and interesting
16:31:49 <argoneus> but they are only looking for senior developers
16:32:08 <argoneus> senior game developers, even
16:32:13 <Alberth> game industry is quite difficult, every kid wants to make games
16:32:13 <argoneus> I'm neither of those
16:32:14 <supermop> argoneus: start up = free beer
16:32:31 <supermop> my fiance's start up has two kegs on tap at all times
16:32:34 <argoneus> V453000: that's right, right
16:32:47 <supermop> also the fire people all the time, so trade off
16:33:12 <argoneus> I just want a job where I don't work on 30 year old legacy code
16:33:13 <supermop> but that's a NYC start up which are more ruthless than in california or boston or europe
16:33:19 <Samu> the biggest line i had on ai_gui.cpp has been reduced to just this:
16:33:20 <argoneus> and my job is actually important to the company
16:33:22 <Samu> bool editable = _game_mode != GM_NORMAL || IsDead(this->slot) || (config_item.flags & SCRIPTCONFIG_INGAME) != 0;
16:33:40 <argoneus> I did a part-time job at Oracle, doing QA
16:33:43 <argoneus> and it drained me mentally
16:33:55 <argoneus> not only was the work boring, no one even appreciated it or cared about it
16:34:10 <argoneus> and the devs sometimes gave me chores
16:34:14 <argoneus> that they didn't want to do
16:34:28 <argoneus> such is life being a part-time student at a super-company, right
16:34:46 <supermop> argoneus: also legacy code may be written by people who know what they are doing, at some start ups, most of the entry level devs might be straight out of a ruby bootcamp and have no idea yet
16:35:20 <Alberth> argoneus: legacy code is an order of magnitude more complicated than new code
16:35:59 <supermop> Alberth: my fiance's frustration is not with code but with people she works with
16:36:01 <argoneus> Alberth: I saw some Solaris code
16:36:17 <argoneus> it's basically super-optimized for the specific compiler version they're using
16:36:21 <supermop> but she is a data scientist so she is somewhat removed from their day to day work
16:36:26 <argoneus> some lines that seemed to make no sense actually helped it run faster somehow
16:36:40 <supermop> ie she writes and works on her own
16:36:41 *** TheMask96 has joined #openttd
16:36:47 <argoneus> and the code was hard to read
16:36:55 <Alberth> ha, that even happens with new code :p
16:37:21 <Alberth> I was trying to make a kinect decoder runs faster, and removed an always-false if condition. It slowed down :p
16:37:54 <Alberth> argoneus: yep, genius programmers can write complicated code :p
16:38:18 <supermop> argoneus: depends also on the start up scene and culture of the city you live in/want to live in
16:38:22 <argoneus> but yeah, basically, I have absolutely no clue what kind of job I want to do
16:38:30 <Alberth> I am guessing it has to do with branch prediction
16:38:34 <argoneus> I'm not one to wear thick glasses and drink coffee from a fancy mug
16:38:47 <argoneus> I can't even grow a beard
16:38:52 <supermop> what about fancy coffee from a thick mug?
16:39:24 <argoneus> I have no idea, I don't like coffee
16:40:09 <supermop> im off to buy fancy coffee from the above in a paper cup
16:40:33 <argoneus> why is their logo half my screen
16:40:37 <argoneus> I have to scroll down to even read text
16:40:49 <supermop> but only because there is no intelligentsia, counter culture, or stumptown nearer to the office
16:40:57 <supermop> yeah fine on my screen
16:41:44 <argoneus> V453000: I was just being sad you are only recruiting senior game developers
16:42:27 <supermop> V453000: argoneus instinctively highlighted you when i mentioned BEER
16:42:28 <argoneus> that looks nice supermop
16:42:31 <V453000> is the only difference between junior and senior that one knows his shit better?
16:42:47 <argoneus> V453000: I thought senior was "sorry you don't have 15 years of experience"
16:43:23 <argoneus> technically, it doesn't say you need experience
16:43:24 <V453000> idk what are the criteria and I doubt they are super strict
16:43:31 <argoneus> I might try applying for Factorio after finals
16:43:35 <argoneus> literally living in prague
16:44:10 <supermop> argoneus: if you dont like bitter, start with roasters like these, 3rd wave light roasts, especially the Kenyans and Ethiopians are fruiting and sweet
16:44:19 <supermop> slightly sour, not at all bitter
16:44:21 *** sim-al2 has joined #openttd
16:44:37 <argoneus> supermop: m8 I don't even know what this is
16:44:44 <argoneus> I just bought some cheap coffee at my local grocery store
16:44:55 <argoneus> and decided I don't like coffee
16:45:03 <supermop> i didnt drink coffee until abt 4 years ago when i go into this stuff
16:45:24 <supermop> im sure prague has a good 3rd wave roaster
16:48:55 <argoneus> Alberth: if I may ask, what kind of job do you personally think is the most "fun"?
16:48:58 <argoneus> in terms of development
16:49:10 <argoneus> or at least some jobs I should really avoid (other than QA)
16:49:25 <argoneus> you have a lot of experience, right
16:49:30 <supermop> idk if you can trust this argoneus but here is something:
16:49:35 <Alberth> that has no meaning to you
16:49:53 <V453000> I would avoid bathroom hygiene levels manager
16:50:13 <argoneus> I bet that's more fun than QA
16:50:14 <Alberth> you should ask yourself what environment do you need, what kind of work do I enjoy or hate, what languages
16:50:46 <Alberth> do you want to go deep, same kind of stuff for 10 years
16:50:56 <V453000> yeah, I don't think there is an universal scale of more fun > less fun
16:50:57 <argoneus> I suppose a small-ish stable company working on a cool project with C#/C++/Python/whatever could be nice
16:50:57 <Alberth> or do something else every 3 months
16:51:11 <argoneus> something like Factorio seems really comfy
16:51:29 <Alberth> is stable job important, or place where the work is, or duration of journey
16:51:44 <argoneus> so instead of looking for one specific thing
16:51:50 <argoneus> I should just filter existing things and pick the best one?
16:52:11 <Alberth> one solution is no solution
16:52:20 <argoneus> it really is a privilege being able to choose the job
16:52:25 <argoneus> I sure am glad I'm not a lawyer
16:52:26 <Alberth> you cannot judge it, as there is no alternative
16:52:41 <argoneus> or a mentally ill webdev
16:53:13 <Alberth> so yeah, looks what's there, and picture yourself in that job
16:53:44 <argoneus> to be honest, at one point I was even wondering if I even like programming, as I absolutely hated QA
16:53:55 <argoneus> but from the reactions of other people it's normal to......not love that job
16:53:58 <supermop> if you are still fairly young and have not financial obligations or debt, you could just do whatever jobs to try them out
16:54:12 <Alberth> QA is not even near programming
16:54:25 <argoneus> I did code some stuff in perl
16:54:30 <argoneus> but that was mostly maintaining legacy tools
16:54:34 <Alberth> although you get in contact with it
16:55:07 <argoneus> I was actually assigned to a new internal project and I had fun with it
16:55:21 <supermop> i thin it is good to try things from tiny garage start up with no funding up to huge government bureaucracy, just to find where you best fit
16:55:26 <argoneus> but then they hired a graduate full-time, kicked me off it, moved me to another desk and that was that
16:55:31 <supermop> that goes for any industry
16:55:53 <argoneus> supermop: I thought it was really bad for your CV to jump jobs a lot
16:56:02 <argoneus> is that not the case then?
16:56:19 <supermop> depends on where you are in your career
16:56:20 <Alberth> as long as you can explain things, I think it would be alright
16:56:40 <argoneus> so a fresh graduate having 3 jobs in the span of 2 years is fine?
16:56:41 <supermop> and obviously spend longer than a month each place
16:56:55 <Alberth> "I moved to a new employer because they paid $5 more / hour" :p
16:56:58 <argoneus> I thought I had to commit
16:57:05 <argoneus> well I feel a bit better now
16:57:45 <supermop> there has been a change
16:58:16 <supermop> and it depends on the place, but the culture of one job cradle to grave is gone
16:58:30 <supermop> you need to spin it as a learning experience
16:58:42 <argoneus> "I joined a company to work on this new project"
16:58:44 <argoneus> "and left when it was done"
16:59:06 <Alberth> there are also companies that hire you, and then send you to different companies for a period to do a job
16:59:22 <supermop> that way the person hiring can say, they built up this broad experience, seeking to learn what they needed to
16:59:22 <Alberth> not sure what's called in English, perhaps posting?
17:00:44 <supermop> also in 5-10 years when you are looking for a more senior position doing a specific thing, not only will you know what you want, but the company will know, this guy has seen enough to know that he wants to do this
17:01:04 <argoneus> I didn't see it that way
17:01:06 <supermop> he's not going to discover that he hates it and walk out in 2 months
17:01:11 <argoneus> like "okay he's been in like 6 different jobs"
17:01:18 <argoneus> "and he's signing up for something he's done before"
17:01:21 <argoneus> "that's a good sign"
17:01:30 <supermop> nor has he been doing the same thing all his live and now it has driven him crazy
17:01:50 <Alberth> there is no inherent good or bad in anything, the reasoning is at least as important
17:01:55 <supermop> but there is a balance and it depends on how you pitch it
17:01:57 <argoneus> I'm glad to hear that
17:02:01 <argoneus> I thought I had to pick and stick with something
17:02:32 <supermop> also it almost doesnt matter at all what yo do right after graduating
17:02:59 <Alberth> do a world tour for 50 years :p
17:08:59 <supermop> again, what Alberth said ^
17:21:45 <Wolf01> mmmh, I knew I would have started again this trend... mod it until it does crash, like Skyrim
17:23:38 <Doge> 1 Say fridge 5 times 2 Say hotdog 2 times 3 hold ur breath till u post this one another game check your voice
17:24:54 <Alberth> nah, you're not around often enough to notice I am not :)
17:25:40 <Doge> Im not always on the OpenTTD
17:25:47 <Doge> Because i have other stuff to do
17:25:59 <Doge> Make new operating system...
17:26:47 <Doge> I have linux operating system.
17:27:07 *** Progman has joined #openttd
17:27:57 <Doge> I mean Ubuntu,Windows 95,Xp,Windows 7,8.1 And 10
17:35:35 <Flygon> I'm still amused OTTD totally compiles for 9x still :3
17:36:11 <Flygon> Not unhappy, actually rather quite happy
17:36:37 <blathijs> Flygon: IIRC vim (the text editor) dropped support for MS-DOS only recently :-)
17:36:53 <Flygon> Kind of a sad moment... but
17:36:58 <Flygon> Er, I don't think anyone really used it O_o
17:37:48 <Flygon> I'm the guy that got upset when Ragnarok Online client didn't work on Windows 98SE
17:38:00 <Flygon> (I wanted to see how the 90s era internal graphics would handle it :P)
17:38:14 <Flygon> Turns out post-2010ish clients require XP and up
17:38:55 *** Wormnest has joined #openttd
17:43:40 <Alberth> with video display moving to the video card, the days of cpu blitting are pretty much over
17:46:44 *** Ketsuban has joined #openttd
17:46:59 <supermop> V453000: how many manhours do you spend painting up sprites post-render?
18:02:46 <Alberth> can't answer you now, he's too busy painting up sprites :p
18:02:55 <Samu> bah i found a bug with the move up / move down slot stuff, grrr
18:04:35 <supermop> i wonder what typical, non vocal players want in a new grf
18:04:46 <supermop> or if they do not want new grfs at all
18:06:16 <Alberth> ie firs is popular, since it's designed to play
18:06:48 <supermop> yet only like 5% of servers seem to run firs
18:07:05 <Alberth> yes, only reddit does afaik
18:07:34 <Alberth> a far bigger percentage of MP at least, just play vanilla
18:07:58 <supermop> needs to be an official Andy server playing FIRS CHIPS RH IH Squid 24/7
18:07:59 <Alberth> downloading grfs before play is a big hurdle
18:08:37 <Alberth> don't know about train sets
18:08:52 *** Quatroking has joined #openttd
18:08:54 <Alberth> you see a lot of nationalism there, I think
18:09:11 <Alberth> german players like DBset, dutch players the Dutch set, and so on
18:10:01 <Alberth> although most are not designed for play, but more as replicating the trains of a country
18:10:20 <Alberth> even if it makes no sense, game-play wise
18:10:20 <Samu> darn slot 15 is now bugging me
18:10:35 <Samu> i can move slot 15 down.... that's not supposed to happen
18:11:00 <Samu> it's my code that's bugged, must find out where
18:12:03 <Alberth> A problem in slow adaption is perhaps unclarity what a newgrf offers
18:12:44 <Alberth> just a name has no meaning for all but a few nerds that read graphics tt-forums and/or hangout at #openttd
18:13:38 <Alberth> as a new player, you got this insane list of newgrfs, and no pointers at all about what they do, or the quality
18:18:21 <Alberth> we need an hour or more to set up a MP game with newgrfs, a newbie is not going to manage it at all, imho
18:22:42 *** Islacrusez has joined #openttd
18:26:00 <Samu> okay, the bug is in IsEditable
18:26:20 <Samu> that for cycle is ruining something
18:26:58 <Samu> it is saying that a supposedly slot 16 is editable
18:27:03 <Samu> there are no 16 slots grrr
18:30:55 <Alberth> likely, it has undefined behavior for out-of-range slots
18:33:42 <Samu> i solved it, i got rid of that for
18:34:01 <Samu> also got rid of max_slot
18:34:20 <Samu> return slot < MAX_COMPANIES;
18:35:01 <Samu> I was always wondering what was the for cycle doing
18:35:15 <Samu> i didn't know for sure, so i let it stay there
18:43:49 <Samu> IsDead might need a better name
18:44:53 *** Islacrusez has joined #openttd
18:45:30 <Samu> the issue i have with the name is about human companies
18:46:04 <Samu> IsDead is also taking care of what to return when the slot is a human company
18:46:43 <Samu> but also what to return when the slot doesn't have any company started
18:47:11 *** Islacrusez has joined #openttd
18:48:10 <Samu> move up and down code is making use of IsDead, but sometimes there's not even a Human company in the slot above or below
18:48:21 <Samu> it assumes those slots are dead
18:49:01 <Samu> the assumption is correct, but the name "IsDead" makes little sense now
18:54:45 *** HerzogDeXtEr has joined #openttd
18:56:04 *** sim-al2 has joined #openttd
18:57:38 <Samu> IsNotStartedOrStartedButDead ?
18:59:54 *** _johannes has joined #openttd
19:05:20 <_johannes> I'm still writing that savegame->route-network-pdf exporter... Someone said it should not be included in the openttd main tree - is that correct?
19:06:23 <Samu> did someone say "savegame"?
19:07:46 <Samu> no idea if it's related to what you're doing
19:37:17 *** gelignite has joined #openttd
19:46:21 <supermop> Alberth: so just simple but pretty sets are not the answer?
19:46:33 <supermop> i guess that was the approach pineapple took
19:46:49 <supermop> and ive never seen a server running pineapple
19:47:02 <Alberth> simple but pretty may be the answer
19:47:08 *** ToneKnee_ has joined #openttd
19:47:35 <Alberth> but basing it on real-life, including speeds, costs, and other details, without regard to game play, is not, imho
19:47:53 <Alberth> at least not for all non-rail-enthousiasts
19:48:29 <Alberth> "you can buy the engine for 5 years"
19:48:52 <Alberth> wtf, before I finished a piece of track I am 10 years further in the game
19:48:54 <Alkel_U3> I've run a server with Pineapple for some time and people I had there liked it
19:49:17 <supermop> but it seems pineapple meets the simple but pretty goal, but you still run into the issue of new users don't know about it, don't want to have to download before playing mp
19:49:29 <supermop> Alkel_U3: never seen your servers online...
19:50:27 <Alkel_U3> supermop: just saying for good measure, I don't run the server too regularly
19:50:38 <supermop> V453000 was talking a bit ago about modularizing train sprites, i like that idea
19:51:24 <supermop> i think a lot of the wonder and fun of 'progression' in the game is lost if the train set is too prescriptive
19:51:43 <Alkel_U3> hm, I've been getting ready to get the server back up for at least a month now... :/
19:51:46 <supermop> there being only one thing that makes sense to use, you know when it comes about, wtc
19:52:13 <supermop> back in circa 1994, playing as a child
19:52:53 <supermop> the first time playing was so fun because i would be legitimately surprised when a new train came out to use
19:53:12 <supermop> and also i was not familiar with the BR prototypes
19:55:19 <supermop> no way to replicate that in this game though
19:56:46 <Alkel_U3> I had an idea - you would task the manufactuers for a certain category of vehicles, they would be too busy to develop other things for some time and then they would give you a new engine, for example. I'm sure it doesn't sound impossible, not too sure how difficuilt it would be to actualy get it into openttd (probly hella) and not at all sure how viable that would be from gameplay perspective
19:58:30 <Alkel_U3> I usually think about this in the shower to pass time :-)
19:59:47 <supermop> well it could be simpler than that, but
20:01:03 <supermop> if a train sprite can be randomly composited of 2-3 images, and if vehicle names could be generated like towns
20:01:28 <supermop> you could have vehicles that you do not yet recognize in each game
20:01:51 <supermop> assuming hp and intro date etc changes as well
20:02:32 <Islacrusez> one approach would be to build an entirely fictional tech tree and have their stats randomized (or perhaps procedurally generated might be a better term?); the graphics are technically secondary
20:03:23 <Islacrusez> you could also have a system where you pay a company to develop a design, unlocking retrofits that improve certain aspects of the design
20:03:32 <supermop> but i do not think there would be a way to have say, one game you have 4 electric locomotives over the course of 50 years, and the next game only 2
20:03:50 <Alkel_U3> that would be good. The graphics are important, imho, for the right feel
20:04:06 <supermop> Islacrusez: i dont think the idea of tech trees or research really helps openttd gameplay
20:04:26 <Islacrusez> tech tree is the wrong word, more like a timeline
20:04:44 <supermop> some railroads did in fact develop technology themselves, but it was never a case of "unlocking" anything
20:05:59 <supermop> for example, if got a 10B$ contract to build a railway tomorrow, i could lay miles and miles of track, or i could call up hitachi and go ahead buy a shinkansen
20:06:14 <supermop> i don't need to first figure out what a steam engine is
20:06:18 <Islacrusez> supermop: I'm using the term to represent a new technology being developed and thus being revealed to the player; no point in trying to surprise them if they can see all the retrofits ahead of time
20:07:02 <Alkel_U3> well, a tech tree from the manufacturing companis' points of view, right? They would just kinda at random decide, what area to advance next, so you may have a game with early advanced expensive el. trains and late diesels or vice versa, maybe some advancements to keep steam in the business longer. Right?
20:07:24 <Alberth> Islacrusez: in a newgrf, that is too complicated
20:07:31 <Alkel_U3> on an unrelated note, I need a new keabord :I
20:07:45 <supermop> well my main issue is less ambitious
20:08:04 <Alberth> supermop: add more engines that you give to the player, and skip a few at random each time?
20:08:27 <Alberth> eg have 8 or 10, and give 4or5
20:08:57 <Alberth> openttd also does a bit stat randomizing, not sure how much
20:09:27 <supermop> now, if i play default, by the late 70s/early 80s i know not to buy anymore manley morels because i can just feel that Dash around the corner
20:09:58 <Alberth> oh, you actually bother about such things? :)
20:10:14 <Alberth> I just run autoreplace on everything every now and then
20:10:31 <supermop> in 94 etc, i often went bankrupt early in game, the first time i even saw a DMU i had my mind blown
20:10:49 <supermop> then blown again the first time i got far enough to see the monorail
20:10:57 <Alberth> ah yes, the early games :)
20:11:03 <Islacrusez> Alkel_U3, yeah pretty much like that
20:12:30 <supermop> if the game could randomly introduce "DMU 1" with a 50s style locomotive cab one game, that would be exciting
20:12:55 <Islacrusez> Alberth, considering how much we managed to do before OTTD became a necessity, I doubt it's impossible; economy of effort is another matter though
20:13:24 <Islacrusez> I'm also speaking more generally, implementation is rarely the same as the concept that it follows
20:15:46 *** frosch123 has joined #openttd
20:18:51 <Islacrusez> having more engines than needed is certainly one way of doing it; if that can be made to work then it could perhaps be refined so that certain combinations become more or less likely (for flavour/balance reasons); and if one really wanted to go nuts you could design a set of alternate "solutions" to the problems that every engine made in reality was designed to solve, essentially creating a new alternate history of locomotion
20:18:51 <Islacrusez> every time the game was played
20:19:10 <Islacrusez> and isn't that what we play the game for?
20:22:57 <supermop> well there is also the issue that any railroad witht e insane profits that you make in the game could develop and commission any type of bespoke train it desired
20:23:53 <Alberth> I always desire fast maglevs early in the game for my back bone, never happened thus far :p
20:23:59 <supermop> or it could just build new luxury apartments for everyone right next to their place of work so that they do not even need a train
20:24:22 <Alberth> and let others get my money, nah
20:25:09 <supermop> it seems though that this idea only really has legs if it is somewhat likely to randomly screw over players some games
20:26:32 <supermop> like release an early prototype channel maglev, player spends billions building maglev network, then only release transrapid style maglevs from then on
20:27:13 <supermop> or, say have only a kirby paul until 2010
20:27:49 <Islacrusez> abandoning the idea of directing the research in any way (for now at least; I have)
20:28:06 <Islacrusez> it shouldn't be too crazy, or at least it should have a setting for crazy
20:28:54 <Islacrusez> because I have to say, a ______ challenge sounds hilarious
20:29:43 <supermop> Islacrusez: much of what makes trains and railways so fascinating and idiosyncratic is a result of nearly 200 years of frequent bad decisions and mistakes
20:30:35 <Islacrusez> supermop, all we're doing is looking at those bad decisions and mistakes (as well as good decisions) and looking to see what other choices there were
20:30:46 <Islacrusez> and then randomising the results
20:31:00 <Islacrusez> so you get about as many good and bad decisions, they're just in different places
20:31:04 <supermop> people out there love steam trains, but a time traveller from today would have gone all in on electrics from the 1890s on
20:32:02 <supermop> basically if you could direct the research with modern knowledge, you would always make the "best" choice
20:33:21 <supermop> and it gets even more boring than knowing that your uu37 soon is made obsolete by a floss 47
20:33:49 <Islacrusez> that's kinda the point I'm addressing, isn't it?
20:38:12 <Islacrusez> unless I'm missing something?
20:41:24 <supermop> Alberth: dozens of unused vehicles seems like best way, but then grf gets huge unless they have some way of reusing some standard graphic elements
20:42:26 <Alberth> "huge" is bigger then 200MB, with yetis around :)
20:43:27 <Alberth> but yeah, re-use of parts would be great, as it's a simple way to make loads of different engines
20:44:17 <supermop> esp if everything is same length
20:44:25 <Islacrusez> if you go with the alternative options theory, you could get away with simply reusing a lot of sprites from mutually exclusive engines; though ultimately modular engines would be very useful for a number of reasons
20:45:20 <supermop> cut every vehicle into 3 parts or so - 25% cab, 50% car body, 25% other cab
20:47:12 <supermop> where vehicle is "first generation diesel", chose 1 cab from list "mk1 cabs" chose mk1 carbody, etc
20:47:19 <Islacrusez> supermop, you could use the same technique to make vehicles of different length; though it may make some trickery to make it not throw a fit
20:50:31 *** Myhorta has joined #openttd
20:57:14 <supermop> hmm my tiny 3d printed monorail jewelry has too much slack in the coupler
20:58:37 <supermop> the space between the car and the 'bellows' can vary from 0-2mm depending on if you are pushing or pulling it along the track
20:58:41 <Samu> that's the full patch with everything already edited to make use of IsDead
20:58:56 <supermop> that is a lot of slack when the car is only 1cm long
20:59:04 <Samu> I'm not 100% sure about where IsDead should be placed, any advice?
21:01:42 <Alberth> it misses a description of the slot parameter
21:03:26 <Alberth> everybody can find it, without forward declaration, you can't get it much better than that
21:07:14 *** andythenorth has joined #openttd
21:07:51 <supermop> maybe i need alittle cotter pin or something to hold the coupler tight to the car
21:11:25 <Islacrusez> supermop, apparently I'm an idiot and can't read
21:11:38 <Islacrusez> I now understand what you meant
21:12:25 <Islacrusez> and I agree, too much agency in that sort of tech progression would be bad
21:12:39 <Islacrusez> kinda like rushing go-karts in RCT
21:15:20 <Islacrusez> and thinking about it, balancing that would be more trouble than it's worth; you'd need a civ3-esque tech tree and that'd not fit the game at all
21:21:50 <Islacrusez> I'm imagining creating a tech "tree", where you map the tech progression through history and with every new engine, design, or technology, you go through all the choices made against the requirements to produce that design; then you look at the other options, and make your tech "tree" choose randomly (weighted by previous decisions perhaps) as to which decision it makes... then you build the tree to completion, and make smaller
21:21:50 <Islacrusez> iterative jumps in designs more likely but larger jumps possible; a number of designs would be mutually exclusive (or a high likelihood of being so) to prevent spamming up the world with similar designs for the same uses...
21:21:59 <Islacrusez> In theory you could go monorail before you go diesel xD
21:23:52 <Islacrusez> in fact you could complete the game and never see a diesel engine, which would be amusing
21:27:12 * andythenorth wonders how you could short-cut that
21:27:20 <andythenorth> there’s probably a way to get the result without much complexity
21:28:14 <Islacrusez> the easy way is to fake it; come up with those scenarios and decisions and requirements based on the tech tree you want
21:28:24 <Islacrusez> cuts out all of that pesky research ;)
21:29:04 <Alberth> randomly add a bit, certain bit combinations enable adding an engine
21:29:46 <Alberth> if "sufficient" bits are on, make the engine available
21:30:21 <Alberth> could be problematic on eg model lifetime
21:30:51 <Alberth> but that also holds for the complicated idea
21:31:14 <Alberth> could add bits until next engine released
21:31:14 <Islacrusez> the complicated idea just needs a lot of protections like any procedurally generated anything
21:31:51 <Islacrusez> though the idea of a procedurally generated tech tree is a weird one
21:32:39 <Islacrusez> it sounds fun though, I'd love to see it done
21:33:35 <Islacrusez> you'd probably want to change the probability of a new tech/design being developed, increasing it as existing designs get old
21:34:10 <Islacrusez> the in-world analogue being designs being seen as unsuitable over time and new tenders being placed for better ones
21:35:43 <Alberth> nah, just add more bits to represent each piece of progress
21:38:10 <Islacrusez> I suspect that this is what the relationship between a designer and a developer feels like
21:58:22 <m4rek> well, gotta run a download and apparently even IRC won't cope with what's left of the bandwidth
22:08:26 <supermop> i wonder if this is safe for MP?
22:18:08 <Samu> oh, right, the description for IsDead, i nearly forgot about it
22:19:28 <Samu> description kinda depends on the structure
22:23:34 <Samu> it is compatible with the 3 windows that make use of it. debug window checks if the selected script is dead. config window checks if the selected slot is dead. parameters window checks if the parameters of the currently selected script is dead.
22:23:56 <Samu> but they're all compatible
22:24:10 <Samu> maybe i shouldn't call it "slot"
22:25:30 *** drac_boy has joined #openttd
22:26:11 <drac_boy> just curious but anyone here from england/europe area and able to name a particular large rail books publisher? (I mean I already know of Platform 5 from england but still, just curious)
22:26:14 <Samu> meh, slot it is, it's the least confusing name
22:30:26 <Samu> it can return 4 types of true's
22:42:26 <Samu> Description: * Check whether a Game Script is not running, or if it is running, if it is dead. Check whether an AI slot is not started, or if it is started, it is not an AI, or if it is an AI, if it is dead.
22:45:00 <Samu> every single outcome is useful
22:45:23 <Samu> that's why I'm having a hard time describing it
22:51:34 <Wolf01> I think that you should split it in more functions which do just one thing
22:51:45 <Wolf01> like IsAiDead() IsAiSlot()
22:52:30 <Wolf01> so if it's not an AI slot, it's not necessary to check if it's dead
22:53:17 <Wolf01> and you should ad asserts here and there to be sure to pass only the right thing
22:53:56 <Wolf01> for example you shouldn't pass a OWNER_DEITY slot to IsAiDead()
22:56:10 *** sim-al2 has joined #openttd
23:18:28 *** Hiddenfunstuff has quit IRC
23:45:05 <Samu> i orchestrated a function too much that it's now too hard to describe it
23:46:04 * drac_boy unfunctions samu? :)
23:46:50 <Samu> it works for what I intended it to do, but.... it's like a 5-in-1 function
23:48:06 <Samu> splitting it into 2 functions could mean.... larger lines on already large lines
23:49:04 <drac_boy> (I was just being a bit silly ofc)
23:50:08 <Samu> the function is used on 3 different struct's
23:54:11 <sim-al2> Lol I didn't know they have opening engine room windows
23:54:28 <drac_boy> well sim..some of the locomotives actually had center doors
23:54:41 <drac_boy> or were you talking about the smoke instead? :)
23:55:14 <sim-al2> The guy hanging out of the middle, doesn't look like the door
23:56:50 <drac_boy> the window location corresponds to the internals tho .. center window is inbetween the two engines .. and the outward windows are between engine and cab
23:56:55 <sim-al2> Yeah lots of locomotives of the 50s-70s have them, I wonder why they disapppeared anyway
23:57:22 <drac_boy> regulations may be to point at me think (especially only 1 or 2 person on a locomotive compared to before)
23:57:54 <drac_boy> heck even on emd geeps it wasn't too unusual to find at least one or more men just standing around on the walkway while train was underway
23:58:49 <Wolf01> oh, it's history channel time :)
23:59:13 * drac_boy whumps wolf01 with a pillow
23:59:40 <sim-al2> I don't know of any regulations on that. It seems cost would be a more pressing reason, with automatic light timers and stuff
continue to next day ⏵
