Log for #openttdcoop.devzone on 7th September 2012:
Times are UTC Toggle Colours
00:01:02  <Brot6> World Airliners Set - Revision 965:dade8476c368: Coded 3 New Liveries into 777-200 XBeardie27X @
00:01:02  <Brot6> World Airliners Set - Revision 966:54f7bd2c0d7e: Added 3 New 777-300 PCX Files XBeardie27X @
00:01:02  <Brot6> World Airliners Set - Revision 967:fa8f02510f37: Added and Coded 8 New 777-300 Planes XBeardie27X @
00:01:03  <Brot6> World Airliners Set - Revision 968:93e6f61dce48: Fixed Number of Repaints Defined in NFO 777-300 Err... XBeardie27X @
00:03:43  <Brot6> worldairlinersset: update from r964 to r968 done (148 warnings) -
00:06:19  <Brot6> World Airliners Set - Revision 969:517e7c1b1649: 733F change XFrankX @
00:06:19  <Brot6> World Airliners Set - Revision 970:1d740d2950bb: changes XFrankX @
00:06:19  <Brot6> World Airliners Set - Revision 971:d56f888a8e17: From Gateway Error XFrankX @
00:07:44  <Brot6> worldairlinersset: update from r968 to r971 done (148 warnings) -
00:38:27  <Brot6> World Airliners Set - Revision 972:fbc7f3a98514: Fixed Missing Tails on most of the New 777-300 PCXs XBeardie27X @
00:38:27  <Brot6> World Airliners Set - Revision 973:2afa306f96f6: Merged With Frank XBeardie27X @
00:39:03  <Brot6> worldairlinersset: update from r971 to r973 done (122 warnings) -
00:42:39  *** Beardie has quit IRC
01:32:33  *** Frankr has quit IRC
12:07:17  <Brot6> DevZone Help Center - Membership #4202: Applying for project: Rythorn XSevestraRythornX @
12:08:18  <Brot6> DevZone Help Center - Membership #4202 (Closed): Applying for project: Rythorn XSevestraRythornX @
12:08:18  <Brot6> DevZone Help Center - Membership #4202 (Closed): Applying for project: Rythorn XAmmlerX @
12:23:37  <Brot6> Rythorn Airline AI - Revision 0:dfdce43927fe: Initial Commit. Start from version 3. XJames Knight@james-laptopX @
12:30:23  *** Frankr has joined #openttdcoop.devzone
14:52:04  *** Alberth has joined #openttdcoop.devzone
15:55:58  *** ODM has joined #openttdcoop.devzone
15:56:40  *** frosch123 has joined #openttdcoop.devzone
17:14:06  *** Zuu has joined #openttdcoop.devzone
17:23:28  <Brot6> worldairlinersset: update from r962 to r973 done (122 warnings) -
17:27:22  <Brot6> dutchtracks: update from r95 to r98 done -
17:27:47  <Brot6> ai-aroai: update from r62 to r63 done -
17:29:09  *** Frankr has quit IRC
18:40:25  *** andythenorth has joined #openttdcoop.devzone
19:06:45  <andythenorth> planetmaker: (I moved chat here, less noisy) so my hope is this alternative supplies mechanic is better for things like NoCarGoal
19:06:57  <andythenorth> delivery will instantly boost production for 3 months
19:07:13  <andythenorth> and some other problems might be solved
19:07:21  <andythenorth> - players want a reward for delivering more
19:07:35  <andythenorth> - delivering only 1t is counter-intuitive
19:08:17  <andythenorth> - the amount of FMSP on the map goes from 'none' to 'way too much' really fast
19:08:44  <andythenorth> and also the range of industry production will be narrower
19:08:55  <andythenorth> I don't so much like insane production levels
19:09:02  <Yexo> current code in produce_primary.pnml is all there is?
19:09:25  <andythenorth> I have been looking in templates/check_primary_production_level.pnml
19:09:44  <andythenorth> the current code should adapt quite well
19:10:19  <andythenorth> I didn't write it though :)
19:10:39  <Yexo> should the current increase/decrease code be kept?
19:11:23  <andythenorth> no
19:11:40  <andythenorth> well....this is a test, so maybe we have a branch or test or such
19:11:54  <andythenorth> but this new code would replace the current code entirely if it works
19:12:03  <andythenorth> people who want old behaviour can use old FIRS :P
19:15:16  <Yexo> how does the "default" production change?
19:25:33  <andythenorth> Yexo: by "default" do you mean the standard ottd behaviour?  Or the default production multiplier value?
19:25:49  <Yexo> the default production multiplier value
19:26:20  <andythenorth> FIRS sets it to a random value at construction (via cb 15F)
19:26:36  <Yexo> yes, but should it ever change?
19:26:40  <andythenorth> yes
19:26:46  <Yexo> how?
19:27:02  <andythenorth> production multiplier will be easiest way to implement this new
19:27:15  <Yexo> <- existing documentation about it
19:27:22  <andythenorth> I guess if I want 2x or 4x, we need to know the 'default' value when constructed
19:27:46  <andythenorth> maybe cb15f should store that
19:28:18  <andythenorth> and we use return value 0F of cb35 to set the multiplier as 1x stored value, 2x stored value, or 4x stored value
19:28:19  <andythenorth> brb food
19:29:52  <Yexo> can you forget about the actual implementation for a bit and tell me how the production should vary?
19:30:00  <Yexo> currently it has a random chance of increase/decrease
19:30:13  <Yexo> the chance varies depending on whether or not you deliver supplies
19:30:38  <Yexo> what should the new way be to change production values? (apart from the 200%/400% bonus for supplies)
19:37:57  <andythenorth> Yexo: no other method
19:38:04  <andythenorth> production changes only due to supplies
19:38:12  <andythenorth> ;)
19:38:46  <Yexo> so if you don't use supplies the production will stay exactly the same during the course of the game?
19:38:53  <andythenorth> yup
19:39:00  <Yexo> kk
19:39:02  <andythenorth> I haven't thought about whether decrease is desirable
19:39:11  <andythenorth> I always turn decrease off in my games
19:39:13  <andythenorth> so no data :P
19:39:47  <Alberth>    pseudo-random ramblings of andy about firs :p
19:40:17  <andythenorth> hmm I put it quite nicely there
19:40:24  <andythenorth> I was having a coherent moment
19:42:59  *** frosch123 has quit IRC
19:45:36  * andythenorth reads ECS manual
19:45:41  <andythenorth> so much detail :o
19:45:54  <andythenorth> FIRS is much less sophisticated
19:47:34  <andythenorth> ho ho
19:47:48  <Yexo> I think I found a nice and easy way to implement this :)
19:47:59  <Yexo> does it matter that production will revert at the first of the month always?
19:48:01  <andythenorth> I think it's easy to do in quite a dumb way :)
19:48:09  <andythenorth> first of month is what I expected
19:48:16  <Yexo> increase as soon as you deliver, but decrease at first of month
19:48:20  <andythenorth> yes
19:48:27  <andythenorth> and run a counter for 3 months?
19:48:33  <Yexo> something like that
19:48:37  <andythenorth> set counter to 3 when delivering
19:48:48  <andythenorth> decrease by one at start of each month
19:48:49  <Yexo> no, keep a running count for current month and previous 2 months
19:48:53  <andythenorth> ah ok
19:49:03  <andythenorth> I kind of had it half-figured, but couldn't see all of it :P
19:49:15  <andythenorth> I can see enough to know it's not that hard
19:49:16  <planetmaker> I'd actually store the future output in 3 variables
19:49:33  <planetmaker> that way you can also store like 4x (next), 4x (2nd next), 2x (3rd next)
19:49:38  <planetmaker> depending on delivered supplies
19:49:48  <andythenorth> I considered decreasing in a gradient
19:49:55  <andythenorth> thought it would be more complex to explain
19:49:57  <planetmaker> you can still do that
19:50:16  <andythenorth> when I say ECS is sophisticated, I'm not 100% being complimentary ;)
19:50:19  <planetmaker> but you just need to store how much was delivered n months ago. You could also store that
19:50:34  <Alberth> "production gradually decreaes back to the original level"
19:50:35  <planetmaker> err-no-parse complimentary in this context
19:50:59  <planetmaker> you mean it's too complicated?
19:51:23  <planetmaker> and complimentary != complementary, I guess
19:52:13  <andythenorth> typo sorry :)
19:52:28  <andythenorth> oh, no I was right :P
19:52:37  <andythenorth> now I see typos where there are none :P
19:52:44  * andythenorth needs a holiday
19:54:50  <andythenorth> I kind of like the idea that industries are reckless
19:55:01  <andythenorth> and will just use up all supplies as fast as possible to max production :)
19:55:24  <andythenorth> I considered a parameter: "Industries: Cautious | Gung Ho"
19:55:25  <andythenorth> :P
19:56:17  <andythenorth> also I think it would be quite easy to parameter-ise the number of months, and the multiples
19:56:21  <andythenorth> if desired
19:56:30  <andythenorth> I dislike parameters personally
19:56:40  <andythenorth> but I could be convinced
19:57:50  <planetmaker> they would make for easier testing for good default values
19:57:59  <planetmaker> important are good defaults. Not few parameters
19:58:07  <andythenorth> and for tuning for GS challenges
19:58:27  <andythenorth> hmm
19:58:30  <planetmaker> I dislike setting many parameters. But I like the option to set them, if I disagree with the author's defaults ;-)
19:59:02  <planetmaker> and OpenTTD has many ways to play which might warrant different parameters being useful
19:59:34  <andythenorth> on a tangent - I had the following idea, can't be implemented yet but...
19:59:49  <andythenorth> if FIRS only varies primary production by supplies
20:00:17  <andythenorth> by 1x, 2x, 4x, or amounts set by parameter
20:00:27  <andythenorth> then it might be interesting to then let GS vary the base factor
20:00:53  <andythenorth> e.g. "Farm cargo prices increase massively, farmers increase production in response"
20:01:00  <andythenorth> or "Drought causes farm production to fall"
20:01:45  <andythenorth> that could be handled if we're careful about how we code supply boost
20:01:50  <andythenorth> but later :P
20:02:26  <andythenorth> how to describe this cleanly....
20:02:48  <andythenorth> the word 'multiplier' is used too many ways wrt industries :P
20:03:46  <andythenorth> - newgrf delegates control of production multipler to OpenTTD
20:03:51  <Yexo> <- completely untested, but might accidentally work
20:03:56  <Yexo> for arable farm only for now
20:04:00  <andythenorth> - OpenTTD provides a way for GS to vary production multiplier
20:04:32  <Alberth> good night
20:04:33  <andythenorth> - actual production by newgrf is calculated based on (production multiplier x some factors)
20:04:36  <andythenorth> bye Alberth
20:05:07  * andythenorth tests
20:05:23  *** Alberth has left #openttdcoop.devzone
20:05:55  <andythenorth> ah
20:06:02  <andythenorth> I can't build newgrfs :P
20:06:09  <andythenorth> I have to fix some python things
20:06:30  <Yexo> ok, I'll test it first and give you a grf later
20:07:03  <andythenorth> ta
20:07:08  <andythenorth> I'll fix my python stuff :)
20:10:28  <andythenorth> ho
20:10:32  <andythenorth> the state of python packaging :P
20:26:43  <Yexo> for some reason an industry still produces cargo following the "default" production_rate even when you use the 256tick production callback
20:27:29  <andythenorth> yes
20:27:40  <andythenorth> it will produce the multipler x the prod level per cargo
20:29:20  <andythenorth> I'd probably just adjust the production multiplier once per month (on cb35)
20:29:23  <andythenorth> and when cargo is delivered
20:29:29  <andythenorth> or even just once per month only
20:37:45  <Yexo> updated and
20:37:55  <Yexo> still only for arable farm, but now tested so it actually works
20:38:20  <Yexo> +		STORE_TEMP(LOAD_TEMP(9) * production_level / 2, 10) <- I do not understand why I need that /2 there, but otherwise the production is too high
20:38:49  <andythenorth> I'd have to read industry_cmd.cpp to figure that :P
20:39:05  <andythenorth> maybe openttd adjusts for number of output cargos or such, not sure
20:39:06  <Yexo> that's what I've already tried :p
20:39:42  <andythenorth> if we binned all the smooth economy stuff (as per frosch would like), that code would be easier to read :P
20:39:56  <andythenorth> oh
20:40:01  <andythenorth> my nmlc has got into a loop
20:40:04  <andythenorth> that's intriguing
20:42:06  <andythenorth> I think it's the makefile that's looping
20:42:08  <andythenorth> not nmlc
20:42:19  <andythenorth> I have FIRS tip, unpatched, and nml tip
20:43:32  <andythenorth> built it with grfcodec ok
20:46:06  <Yexo> I seem to be mixing up prod_level and production_rate
20:46:12  <Yexo> that's why the /2 is needed I think
20:46:20  <Yexo> will need to look at that again another time
20:49:23  <andythenorth> that confuses everybody ;)
20:49:38  <andythenorth> it doesn't help that they are inconstently named :P
20:49:45  <andythenorth> newgrf specs use different terms iirc :P
20:50:25  <andythenorth> seems to be working
20:50:26  <Yexo> I need NFO vars 0x8E and 0x8F (=production_rate), not var 0x93 (=prod_level)
20:50:37  <Yexo> you'll notice the bonus isn't exactly 100%
20:51:10  <andythenorth> yes, if you do it on the 256 tick cb you need those vars
20:51:21  <andythenorth> I can't see any reason not to do it the 256 tick way
20:51:27  <andythenorth> my habit is to just cb 35 :P
20:51:32  <andythenorth> but I don't know why
20:51:45  <Yexo> this was the easiest way I could think of
20:51:50  <Yexo> no habits to get stuck in :p
20:52:26  <andythenorth> does it respect the initial offsets of prod_level?
20:52:32  <Yexo> but vars 8E and 8F are not documented in the NFO spec, nor are they available in nml
20:56:20  <andythenorth> hmm
20:56:38  <Brot6> NewGRF Meta Language - Revision 1992:0e531faa1414: Add: production_rate_1/2 variables for industries XyexoX @
20:56:41  <andythenorth> if we use prod_level, but leave its value unchanged, and then use our own multiplier
20:56:52  <andythenorth> then production level cheat will work quite nicely
20:56:58  <Yexo> I just added the necesary variable to nml and used that
20:57:00  <andythenorth> he
20:57:03  <Yexo> updated diff and grf are uploaded (same urls)
20:58:05  <Yexo> production level cheat works fine :)
20:58:17  * andythenorth tests
20:59:06  <Brot6> AroAI - Revision 64:84962403ce33: Fix: Missed some debug statements XLordAroX @
20:59:06  <Brot6> AroAI - Revision 65:ba3c57b194e4: Change/Fix: Limit the number of times the pathfinder is called per... XLordAroX @
21:01:22  <andythenorth> Yexo: seems to work well
21:01:29  <andythenorth> need a GS MP game to test it :)
21:01:35  <andythenorth> not tonight I guess
21:01:41  <Yexo> and implement it for other industries :)
21:01:45  <andythenorth> oh that :P
21:01:56  <andythenorth> can it not just replace existing template? :P
21:02:01  <Yexo> almost, yes
21:02:02  <andythenorth> oh it needs 256 tick cb adding :P
21:02:44  <andythenorth> copy paste :(
21:03:48  <andythenorth> k bedtime :)
21:03:50  <andythenorth> thanks, bye
21:04:03  *** andythenorth has quit IRC
21:18:43  <Yexo> <- new (optional by parameter) supplies algorithm for firs
21:23:00  <Brot6> FIRS Industry Replacement Set - Feature #4203 (New): New supply algorithm XyexoX @
21:29:34  *** ODM has quit IRC
21:36:47  *** Frankr has joined #openttdcoop.devzone
22:42:46  *** player23 has joined #openttdcoop.devzone
22:43:44  *** player23 has quit IRC
23:48:57  *** Zuu has quit IRC

Powered by YARRSTE version: svn-trunk