IRC logs for #openttd on OFTC at 2020-06-08
⏴ go to previous day
00:00:01 <Speeder> but the other variable goes from 4, to 8
00:00:07 <Speeder> thus you actually quadrupled production
00:01:04 <michi_cc> No, if the production level changes, the production multiplier does not. The resulting production changes.
00:02:36 <Speeder> what production multiplier?
00:04:12 <michi_cc> And of course all this only applies if you use default production behaviour. If a production callback us used, none of that matters.
00:06:19 <Speeder> michi_cc, so, production_rate cargo label from the cargotable NML 0.5 For the given cargo label, amount of output cargo that is produced every 256 ticks. Initial value from the cargo_types property, but it will change accordingly when the production level changes. Returns 0..255. Example: production_rate("SUGR").
00:08:08 <Eddi|zuHause> which part of that should be wrong?
00:08:47 <michi_cc> I don't know offhand which VA2 variable that maps to, but I'm quite sure that returns the resulting production, i.e. multiplier * level.
00:09:43 <michi_cc> This is not the same as the value from the produce_cargo property.
00:09:44 <Speeder> Eddi|zuHause, it says two things: 1. it will return the initial value at first (so for my example earlier, 4), and then that it will return later, the value changed as production level changes.
00:10:26 <michi_cc> production = multiplier * level / 16. Default for level is 16.
00:10:48 <FLHerne> Speeder: production_rate tells you what the current production rate *is*
00:11:15 <michi_cc> The one thing wrong in this line is that a callback exists that allows to change the inital level from 16 to something else.
00:12:01 <glx> production_rate is given by openttd, you don't change it yourself
00:12:20 <glx> well you don't modify it directly
00:12:58 <Speeder> so step by step: 1: produce_cargo("GOOD", 4) 2: production_rate("GOOD") returns 4, production_level is 16. 3: a callback returns CB_RESULT_IND_PROD_DOUBLE, production_rate("GOOD") returns 8, production_level is 32
00:14:25 <Speeder> FLHerne, this is what I understood from the wiki. thing is, doing this way it means doubling production level, results in 4x output, not 2x
00:14:38 <FLHerne> And now you've lost me again
00:14:50 <FLHerne> production has gone from 16 to 32
00:14:56 <Speeder> I will try again my step by step
00:15:00 <michi_cc> The 8 IS the production.
00:15:05 <Speeder> 1: produce_cargo("GOOD", 4)
00:15:30 <Speeder> 2: production_rate("GOOD") returns 4, production_level variable is 16. Thus actual output is 4*16
00:16:05 <Speeder> 3: a callback retruned CB_RESULT_IND_PROD_DOUBLE
00:16:24 <Speeder> 4: production_rate("GOOD") returns 8, production_level variable is 16. Actual output is 8*32
00:16:39 <Speeder> 4: production_rate("GOOD") returns 8, production_level variable is 32. Actual output is 8*32
00:17:18 <Speeder> glx, what is 4*32/16?
00:17:31 <FLHerne> Speeder: > 4: production_rate("GOOD") returns 8, production_level variable is 16. Actual output is 8*32 is wrong
00:17:41 <Speeder> FLHerne, that line was a typo
00:17:46 <Speeder> I mean the one right after it
00:17:50 <glx> production_rate is the actual output
00:18:01 <FLHerne> Speeder: That one is also wrong
00:18:14 <Speeder> glx, so the initial value is NOT, what you set?
00:18:26 <FLHerne> Speeder: Again, > production_rate tells you what the current production rate *is*, it doesn't set it
00:18:39 <FLHerne> Speeder: It's a variable, you read it
00:18:43 <Speeder> FLHerne, I don't understand why you keep saying that
00:18:53 <FLHerne> It tells you how much cargo the industry is producing
00:19:02 <Speeder> I never used that to set anything
00:19:03 <FLHerne> The value isn't fed back into anything
00:19:10 <Speeder> I used produce_cargo to set
00:19:19 <FLHerne> Ten why are you saying " 8*32"?
00:19:23 <michi_cc> Speeder: production_rate("GOOD") is the value AFTER production_level is applied, not before.
00:19:26 <andythenorth> I had this *exact* conversation with dalestan when I started FIRS :P
00:19:52 <FLHerne> You haven't changed the produce_cargo
00:20:02 <Speeder> michi_cc, so... when it says the INITIAL VALUE, is whatever you set using produce_cargo
00:20:35 <andythenorth> are you just saying production_rate has a bug Speeder? :)
00:20:36 <Speeder> it means produce_cargo value is fudged before saving?
00:20:45 <Speeder> andythenorth, I am trying to understand the wiki
00:20:54 <Eddi|zuHause> you're not making any sense
00:21:00 <michi_cc> produce_cargo is a NewGRF property, not something that is stored in any savegame.
00:21:08 <Eddi|zuHause> "initial value" usually comes from some property, and some callback can change it
00:21:17 <andythenorth> nml or nfo wiki?
00:21:24 <Speeder> from wiki: production_rate: Initial value from the cargo_types property, but it will change accordingly when the production level changes.
00:21:46 <Speeder> so in cargo_types you used: produce_cargo("GOOD", 4)
00:22:03 <Speeder> thus production_rate("GOOD") will return 4?
00:22:04 <andythenorth> when game starts
00:22:22 <andythenorth> production rate is (prod_multiplier / 16) * prod_level
00:22:24 <FLHerne> Because production_rate is based on both the multiplier and the level
00:22:25 <michi_cc> For an industry directly after construction.
00:22:29 <andythenorth> and prod_level is 16 at game start
00:23:39 <michi_cc> produce_cargo("GOOD", 4) IS NOT the production_rate.
00:23:43 <glx> multiplier is produce_cargo
00:23:46 <andythenorth> you can't learn any actual coding from the nml docs
00:24:06 <Speeder> michi_cc, production rate "initial value" then refers to WHAT ?
00:24:14 <Speeder> it says it is intial from "cargo_types"
00:24:16 <michi_cc> (prod_multiplier / 16) * prod_level
00:24:29 <Speeder> then the wiki is wrong
00:24:40 <andythenorth> but don't use the nml wiki
00:24:42 <michi_cc> With inital value of prod_level being 16.
00:24:58 <andythenorth> nml wiki is just for looking up names of action 0 / varaction 2 / callbacks
00:25:12 <andythenorth> others may disagree
00:25:29 <andythenorth> use the actual spec
00:25:52 <FLHerne> Speeder: The value returned by production_rate is initially the same as you set in cargo_types
00:26:03 <Eddi|zuHause> it heavily depends whether you want a specification or an explanation. the two things are often not identical
00:26:08 <FLHerne> When you change the production_level, it isn't the same anymore
00:26:43 <Eddi|zuHause> an explanation tends to simplify things, while a specification must be as detailed as possible
00:26:47 <Speeder> FLHerne, so it IS, or NOT, initially, the same as "produce_cargo" ?
00:27:03 <FLHerne> But that doesn't affect the value you set in cargo_types
00:27:23 <FLHerne> Are you confused about the definition of "initially" ?
00:27:34 <Speeder> what is the actual formula then?
00:27:41 <andythenorth> as posted several times above
00:27:48 <glx> because prod_level start value is 16
00:27:49 <Eddi|zuHause> the formula was written like 4 times already
00:27:51 <andythenorth> (prod_multiplier / 16) * prod_level
00:28:26 <andythenorth> I was really confused by prod_multiplier and prod_level when I started FIRS
00:28:39 <andythenorth> and that was prior to nml
00:28:48 <andythenorth> nml makes it worse by renaming the concepts
00:29:05 <Speeder> so is this line in the wiki, that is ambiguous: produce_cargo("GOOD", 4) specifies the industry produces GOOD at a rate of 4 every production tick (256 game ticks), this production rate is scaled by the industry's production level.
00:29:06 <andythenorth> otoh, the original names aren't always great
00:29:18 <michi_cc> Technically, the NML wiki is indeed wrong, because it ignores the existence of callback 0x15F, which may be used the change the "initial value" of prod_level from 16 to something else.
00:29:32 <andythenorth> Speeder the line is 100% true but not easy to read
00:29:58 <FLHerne> Hm, I think it's "this production rate" that's confusing
00:30:07 <Speeder> that line implies final result is whatever you set * 16 (since elsewhere it says that is the initial production level)
00:30:11 <FLHerne> "this production rate" has nothing to do with the production_rate variable
00:30:47 <michi_cc> Speeder: No, because production_level is a fixed point value with 4 fractional bits.
00:31:33 <andythenorth> it can be 16, 1/16, or 16 * 16
00:31:33 <michi_cc> Which is why people have repeatedly wrote about "/ 16".
00:31:59 <andythenorth> Speeder it *is* confusing yes
00:32:05 <andythenorth> you'll get there
00:32:09 <Speeder> that is what I mean, the /16 is not mentioned in the wiki
00:32:21 <andythenorth> I looked it up in openttd src
00:32:29 <michi_cc> But it is, just indirectly by the word "scaled".
00:32:38 <andythenorth> I quite often ignore the newgrf docs and read the src in industry_cmd.cpp
00:32:39 <Speeder> michi_cc, thus why I said that thing is ambiguous
00:33:09 <glx> it's implicit because production can be halved too
00:33:18 <Eddi|zuHause> Speeder: imagine it like %. if you give a factor of 5, the result will be 500%
00:33:33 <andythenorth> speeder's mind is blow :)
00:33:50 <glx> or 8 depending on production level evolution
00:33:52 <andythenorth> irc is hard medium to get taught with
00:34:21 <Eddi|zuHause> i still have no real clue which part he's struggling with
00:34:36 <andythenorth> conceptually understanding what is going on
00:34:37 <michi_cc> "scaled" means "is proportional to", not "is multiplied with". Thus, the inital default of the production level is irrelevant.
00:34:56 <Speeder> Eddi|zuHause, it is the /16 formula is not mentioned on the wiki
00:35:05 <andythenorth> we could change that
00:35:09 <andythenorth> it may not help :)
00:35:16 <Speeder> but for a non-english speaker that might not be 100% obvious
00:35:19 <Eddi|zuHause> why does the wiki have to mention the /16?
00:35:24 <Speeder> changing "scaled" to "proportional" for example would help
00:36:06 <Speeder> Eddi|zuHause, because the way the wiki is written implies production_rate is the same as the multiplier
00:37:24 <michi_cc> I'd say it exactly does not imply that: " Initial value from the cargo_types property, but it will change accordingly when the production level changes."
00:38:04 <michi_cc> "will change accordingly" is to me quite clearly saying that if production level e.g. doubles, production rate doubles.
00:41:02 <FLHerne> michi_cc: But the cargo types array paragraph says "...this production rate is scaled by..." when describing the multiplier
00:43:02 <michi_cc> Which I would not read as equivalence either.
00:44:51 <FLHerne> michi_cc: It's very confusing to describe the multiplier as "this production rate" when production_rate has a completely different meaning and the multiplier isn't the production rate :p
00:44:55 <Speeder> FLHerne, well, you got what my gripe with the wiki is :D
00:46:15 <michi_cc> But is is the initial production rate unless you use callback 0x15F to change the initial production level.
00:46:25 <supermop_Home_> andythenorth these Scottish 37s are very pretty
00:46:43 <supermop_Home_> but like do they have no other locomotives up there?
00:46:53 <michi_cc> NML seems to call that CB build_prod_change
00:47:17 <andythenorth> supermop_Home_ it's a very specific requirement for high power, low axle load on that route
00:47:25 <andythenorth> most locomotives can't do it
00:47:28 <FLHerne> It's initially *equal to* the production rate, but it's semantically a different thing
00:47:34 <glx> The properties specify the amount produced at default production level, they are scaled equally on production changes.
00:48:04 <FLHerne> And it stops being equal whenever the production level changes, which is what makes it a semantically different thing
00:52:37 <andythenorth> oof sleeping time
01:24:29 <arikover> nielsm: Thank you for your quick answer, and sorry, I dozed off. I'll be posting the logs tomorrow...
08:33:42 *** andythenorth has joined #openttd
09:11:45 *** arikover has joined #openttd
09:19:10 <DorpsGek_III> [OpenTTD/OpenTTD] telk5093 updated pull request #8207: Add: 'cheat' console command to open the cheat window https://git.io/JfM57
09:21:01 <DorpsGek_III> [OpenTTD/OpenTTD] telk5093 commented on pull request #8207: Add: 'cheat' console command to open the cheat window https://git.io/JfDZ6
09:27:00 *** sla_ro|master has joined #openttd
09:32:31 *** Wormnest has joined #openttd
09:34:58 *** gelignite has joined #openttd
10:41:22 <DorpsGek_III> [OpenTTD/OpenTTD] fsimonis commented on issue #8201: Consider upgrading CMake to latest release https://git.io/JfMP8
11:00:58 <andythenorth> LordAro you tagging it? :)
11:34:00 *** Smedles has joined #openttd
11:40:11 *** Smedles has joined #openttd
12:29:13 *** Gustavo6046 has joined #openttd
12:38:14 *** WormnestAndroid has quit IRC
12:38:26 *** WormnestAndroid has joined #openttd
12:45:48 *** WormnestAndroid has quit IRC
12:46:01 *** WormnestAndroid has joined #openttd
13:04:24 <DorpsGek_III> [OpenTTD/OpenTTD] fsimonis commented on issue #8201: Consider upgrading CMake to latest release https://git.io/JfMP8
13:09:33 <Flygon> MFW - I try to come back to a game, and can't find a vehicle group I need to find because there's no Search function. :D]
13:44:16 *** gelignite has joined #openttd
14:19:20 <supermop_Home_> maybe i should buy a Z scale JR 103
14:19:48 <supermop_Home_> there's also a nice T scale HST out there
15:19:14 <supermop_Home_> just switch away from petrochem in 2030s, have all the oil wells close
15:32:08 <spnda> How does DoCommandP work? I need to change something about building and placing road stops.
15:33:00 <glx> I think you never need to modify DoCommandP
15:33:26 <spnda> but what functions does it execute
15:35:25 <spnda> there's a list of Commands and you can pass the commands to DoCommandP, but what handles what command was passed and where is defined what it does afterwards?
15:42:17 <supermop_Home_> andythenorth thoughts on z or T? all my BR stuff is N but its all from like 1997
15:45:33 <_dp_> spnda, you can see what function handles the command in commpand.cpp and then grep for that function
15:46:11 <supermop_Home_> has anyone ever considered a model railroad baseset?
15:46:33 <_dp_> spnda, well, or just go to definition if your editor supports it
15:48:39 <glx> usually I just search for the CMD_XXX
15:49:22 <_dp_> glx, that will only give you the calls, not the handler
15:49:42 <glx> because it's a comment on DEF_CMD lines
15:50:10 <glx> DEF_CMD(CmdBuildRoadStop, CMD_NO_WATER | CMD_AUTO, CMDT_LANDSCAPE_CONSTRUCTION), // CMD_BUILD_ROAD_STOP
15:50:30 <glx> the easier way to find the mapping
15:50:44 <_dp_> glx, well, that't just the same command.cpp I mentioned
15:51:25 <spnda> Yes I found CmdBuildRoadStop and the DEF_CMD(CmdBuildRoadStop... . What do I do now?
15:51:35 <spnda> I can't find any real definition or function call with it
15:51:56 <glx> you won't find direct call
15:52:29 <_dp_> spnda, you either search/grep the whole source base for CmdBuildRoadStop or use goto definition function if your editor has it
15:52:44 <spnda> Tried both, found nothing
15:53:19 <_dp_> spnda, then you're doing it wrong, mine jumps to station_cmd.cpp instantly
15:54:17 <spnda> Ok I found it manually by searching for RoadStop in station_cmd.cpp. But really, searching for definition or declaration gave me no results
15:54:21 <spnda> But thanks, I got it now.
15:54:44 <andythenorth> supermop_Home_ I only know about OO and HO :)
15:55:20 <glx> spnda: when "go to definition" fails just use global search
15:57:04 <glx> oh I see why it fails for CmdBuildRoadStop, there's a TILE_AREA_LOOP() in the function killing intellisense
15:57:26 <supermop_Home_> andythenorth no room for that!
15:57:43 <_dp_> lol, luckily sublime has none of that stupidsense xD
15:57:48 <andythenorth> supermop_Home_ I think I'm building a train set in OpenTTD :P
15:58:07 <andythenorth> the irony is that Iron Horse definitely includes trains I had OO models of
15:58:19 <andythenorth> but now I am buying OO models just because they're in Iron Horse
15:58:32 <supermop_Home_> commission a line of injected molded Iron Horse models
15:58:41 <andythenorth> trains are taking up a lot of space here
15:58:54 <andythenorth> I have got into a buy-look-at-sell-again
15:59:00 <andythenorth> that kind of collecting :P
15:59:08 <andythenorth> buy, sell, lose 60% of value,
15:59:36 <glx> spnda: global search is Ctrl+Shift+F for me
16:00:21 <spnda> havent been doin stuff for a while now
16:00:37 <glx> and it was worse with all FOR_ALL_XXX loops used before
16:00:50 <supermop_Home_> do the kids like the trains?
16:00:55 <glx> but there are still some loop macros in place
16:01:55 <andythenorth> supermop_Home_ some days they like the trains yes
16:02:02 <andythenorth> one of them likes them more than the other
17:23:17 *** virtualrandomnumber has joined #openttd
17:44:12 *** Wormnest has joined #openttd
17:53:14 <andythenorth> interesting has been adding more grass and stuff to industries
17:53:28 <andythenorth> even really heavy industry tends to have grass areas
17:54:08 <supermop_Home_> combination steel mill / power plant / golf course island as industry?
17:55:49 <supermop_Home_> interesting to see the coal stained vs ore stained areas
17:56:43 <supermop_Home_> i assume they are making very long products in those sheds
17:58:56 <supermop_Home_> few other towns along Seto sea have smaller versions of that
18:00:02 *** virtualrandomnumber has quit IRC
18:23:21 *** frosch123 has joined #openttd
18:53:43 *** Progman has joined #openttd
19:05:06 *** gelignite has joined #openttd
19:17:18 *** WormnestAndroid has quit IRC
19:17:23 *** WormnestAndroid has joined #openttd
20:02:36 <supermop_Home_> how's it over there?
20:08:23 <Wolf01> I liked more the lockdown
20:12:37 <supermop_Home_> more stores allowed to re-open today here
20:18:48 *** WormnestAndroid has quit IRC
20:19:47 *** WormnestAndroid has joined #openttd
20:21:32 *** ChanServ sets mode: +o Yexo
20:24:42 <frosch123> Wolf01: can you really play ottd with them?
21:02:26 * andythenorth should test that
21:03:06 <andythenorth> glx can we merge nml 155 and 156 yet?
21:03:20 <andythenorth> doubt anyone else is going to provide test cases until it's released
21:04:08 <spnda> I'm a bit confused right now. There's a bunch of different Station structs and I can't find the one I need. For a RailStation I saw "BaseStation" and what would be the RoadStop counterpart for that?
21:05:29 <glx> oh I broke nightlies again
21:05:55 <LordAro> eh, no one uses them anyway
21:07:19 <spnda> I see Station has references to bus and truck stops and airports, so I don't think that's what I am looking for
21:07:44 <spnda> I'm trying to get the struct for a RoadStop that I can then give it the RoadStopSpec for that stop
21:09:11 <spnda> oh it seems that this is all only defined for rail stations
21:09:31 <glx> BaseStation is the root of Stations and Waypoints
21:10:03 <glx> and Stations have a "facilities" flag
21:10:21 <glx> to determine what kind are present
21:10:58 <spnda> so I guess I should change up the StationSpecList inside BaseStation?
21:11:00 <glx> and there should be a link to the road/bus/truck stop in the station struct
21:12:41 <spnda> Should I have a list of specs inside the RoadStop definition or inside the BaseStation definition
21:12:51 <spnda> The rail stations currently use a list of specs in the BaseStation definition
21:13:00 <spnda> so i'd probably need to change that somehow
21:15:09 <glx> because there's only rail in station newgrf spec
21:15:52 <glx> and they are shared between stations and waypoints
21:16:46 <glx> I guess if your feature is only for RoadStops it should go in RoadStop
21:17:37 <glx> like airport grf spec is in Airport
21:26:25 <spnda> use of undefined type 'RoadStop'? what?
21:32:49 <Yexo> spnda: OpenTTD (used to) make heavy use of forward declarations for pointers. If there is a "struct Roadstop;" visible anywhere, you can define a "RoadStop *bus_stops" variable. You can't use member variables that unless you include the right header to get the full definition of the RoadStop struct
21:33:09 <spnda> i didn't include roadstop_base.h
21:33:17 <spnda> only included newgrf_roadstops.h
21:34:19 *** y2kboy23 has joined #openttd
21:48:47 <spnda> Different approach.... How should I add the RoadStopSpec to the RoadStop Station? I don't see a direct reference to it being added anywhere in the Airport build command, I only see the AllocateSpecToStation inside the Rail Station build command.
21:52:25 <glx> usually all this stuff uses indirections
21:54:43 <TrueBrain> in case nobody noticed it, the nightly failed :)
21:54:49 <TrueBrain> not sure who receives those emails ;)
21:55:00 <glx> yeah I know it failed, and I know why
21:55:26 <glx> but yesterday it built fine, but didn't make its way to the website
21:55:51 <glx> (and I don't receive emails)
21:57:14 <glx> dunno seems to happen sometimes
21:57:29 <TrueBrain> I will check it out
21:57:30 <spnda> so... hmm, should I just have a RoadStopClassID and then a int to get the spec from that class referenced inside the RoadStop struct?
21:57:34 <TrueBrain> nightly rarely fails :)
21:58:31 <andythenorth> worth reading the signs :)
22:04:41 *** D-HUND is now known as debdog
22:13:47 <TrueBrain> glx: will check it in a sec :)
22:22:20 *** jottyfan has joined #openttd
22:22:41 <TrueBrain> glx: the release job attached to it failed
22:22:45 <TrueBrain> with a very weird error :) I will check it out
22:24:49 <TrueBrain> seems the AWS API made a boo-boo
22:25:55 <EER> TrueBrain: when I use azure devops in my day-job, there's a button to just retry the deployment. Might have just been a fluke with aws :)
22:28:35 <TrueBrain> yeah, I know :) But first I needed to check out why it failed :P
22:28:39 <TrueBrain> but there should be no reason
22:31:24 <TrueBrain> well, a first time for everything I guess :) First time I have seen it fails because AWS is acting up :)
22:31:31 <TrueBrain> on DO this happened almost every other week
22:31:33 <TrueBrain> so I cannot complain
22:32:50 <TrueBrain> I always say: 1 error is no error, so, lets see if it ever happens again :)
22:33:18 <EER> That's usually my first approach to troubleshooting build failures, try again first, if it fails ... investigate :)
22:35:12 <TrueBrain> and tnx EER ; was not reading chat tbh, but you were spot on ;)
22:36:49 <Yexo> While the upload to AWS step is the one that failed, the "Download artifact - bundles - bundles" step apparently took 18h on the failed attempt
22:38:38 <Yexo> 2020-06-07T18:16:18.4920685Z ##[section]Starting: Upload to AWS
22:38:39 <Yexo> 2020-06-06T23:56:21.9629075Z ##[section]Finishing: Upload to AWS
22:38:48 <Yexo> Time in logs goes backwards
22:39:47 <Yexo> Oh well, not worth spending more time on, still interesting
22:44:43 *** Wormnest has joined #openttd
22:45:48 <DorpsGek_III> [OpenTTD/OpenTTD] MiguelHorta opened pull request #8209: Fix: Display banlist's indexes correctly https://git.io/JfDNJ
22:50:05 <DorpsGek_III> [OpenTTD/OpenTTD] nielsmh approved pull request #8209: Fix: Display banlist's indexes correctly https://git.io/JfDN3
23:18:44 *** supermop_Home_ has quit IRC
23:50:13 *** y2kboy23 has joined #openttd
continue to next day ⏵
