• Our friends from AlphaCentauri2.info are in need of technical assistance. If you have experience with the LAMP stack and some hours to spare, please help them out and post here.
Taurus

Taurus 1.03

I'm going to go with on big displays it's hard to tell apart.
The FoR options tab problem seems to be another Play Now problem. Don't have it fully confirmed, but at least every time I've tried Play Now in that exe session that's happened; whereas the only times it's worked have been when not having Played Now.
I compiled the dll after only changing the player cap; didn't remove anything. The size was 4.96 vs the original's 4.98. Should I be concerned about that at all?
 
Last edited:
does Bug show the information from "ho ho ho, I have an archer" anywhere as "best military unit" or similar? If not I'd make that suggestion to save time reopening diplo until it displays.
 
I'm going to go with on big displays it's hard to tell apart.
A slightly bigger bite doesn't seem to help much (mock-up attached). Maybe food lost through negative health should just be shown as a food icon with a negative number. That's the only thing the eaten-food icon is used for: when previewing the effects of worker builds and city buildings, and when breaking down a city's net food yield.
Edit (December): For future reference, there are seven game texts that reference ICON_EATENFOOD in BUG_CIV4GameText.xml and BULL_CIV4GameText.xml, and EATEN_FOOD_CHAR gets used in two places in dynamically generated text in the DLL (CvGameTextMgr.cpp, CvDLLWidgetData.cpp). Revising all that text is a bit too much work by my standards.
The FoR options tab problem seems to be another Play Now problem. Don't have it fully confirmed, but at least every time I've tried Play Now in that exe session that's happened; whereas the only times it's worked have been when not having Played Now.
Was going to take a first look at this, and I don't even see FoR listed when I go via Play Now (screenshot attached). Nor Maze, Ring_World or Balanced – which the mod doesn't even touch. Is that what the isAdvancedMap functions is for ... nope, adding that function doesn't cause Maze to appear. :undecide: I don't suppose you've changed anything to be able to select FoR via Play Now? Oh, I guess you're saying that accessing the Play Now screen - selecting whichever map - may later cause problems with FoR on the Custom Game screen. That sounds like an issue that could've escaped or eluded the script's authors.
I compiled the dll after only changing the player cap; didn't remove anything. The size was 4.96 vs the original's 4.98. Should I be concerned about that at all?
Seems a little counterintuitive and, for AdvCiv, my 48-civ DLL is a tiny bit larger than the 18-civ DLL. That being said, the max-players/ teams preprocessor defines are used in a lot of for loops (most of which I've replaced at some point in AdvCiv edit: i.e. they probably don't affect the file size in the case of that mod), and it's not so clear how compiler optimizations play out for different values. 18 iterations might get optimized by triplicating the body and running only 6 iterations or something like that, which may result in slightly more verbose binary code than with a higher limit. But the size difference could also be due to the exact version of the libraries we each use. I've been using these specific versions: https://github.com/f1rpo/BtSGameCoreDLLDepends
But I don't think this matters (even if it should slightly affect the DLL size) so long as compilation succeeds without warnings. And my DLLs on GitHub use the FinalRelease build configuration, but I'm guessing that you've used that one as well because, otherwise, I'd expect the difference in file size to be greater.
does Bug show the information from "ho ho ho, I have an archer" anywhere as "best military unit" or similar? If not I'd make that suggestion to save time reopening diplo until it displays.
I would've expected that info on the Strategic Advantages tab, but it's apparently not there. In a test (screenshot attached), Boudica brags about her Gallic Warrior, but the screen doesn't show it. Ideally, perhaps, an alert could (additionally) inform the player when the AI starts announcing a new unit, but I don't think any modder will want to go to such lengths. In AdvCiv, I'm already letting the AI brag with 100% odds (instead of 25%) when it has gotten a new unit since the last meeting. Not really in the spirit of BUG. I think the units should just be placed in the "Known" column of "Strategic Advantages".
 

Attachments

  • food-eaten_mockup.jpg
    food-eaten_mockup.jpg
    41.9 KB · Views: 122
  • map-selection.jpg
    map-selection.jpg
    349.1 KB · Views: 104
  • strategic-advantages.jpg
    strategic-advantages.jpg
    234.6 KB · Views: 120
Last edited:
Oh, I guess you're saying that accessing the Play Now screen - selecting whichever map - may later cause problems with FoR on the Custom Game screen. That sounds like an issue that could've escaped or eluded the script's authors.
Yeah, that's the problem. I can't get maze and this one to show up in the play now screen either, but stuff like mapscriptools maps do. :hmm:
I've also noticed that even though Big and Small has the sealevel dropdown enabled, it doesn't do anything; and Arboria disables both sealevel and climate, but both actually do affect it. Thinking changed versions could go in the mod.
Also started fixing the Tectonics map to make the land count somewhat consistent relative to map size by shrinking the grid for some options like Mediterranean and lakes...by default at huge they have 2000 more land tiles than the earth setting...
Here's a 50 civ dll for version 1.03 for anyone to use. Haven't made any other changes to it.

Also: information for the Advanced Combat Odds feature seems to have been left out of the help file.
 
Last edited:
Cool that you're getting stuff done while I'm being all talk. :thumbsup:
I've given the K-Mod version of Big & Small ("Not-too-big-or-small") an operational sea level setting, so I should be able to port that easily to Big & Small. As always, with map script inclusions, this will inadvertently bump it up in the map selection dropdown, but, via PrivateMaps, Fractal and Pangaea will be kept at the very top, and Big & Small is a pretty popular script. Good to know about Arboria.

Tectonics – the options do say (in their names) that e.g. Earthlike and Lake have different land-sea ratios; but it may well be that Mediterranean is too extreme or arbitrary. I've noticed that the Earthlike water percentage aren't nearly accurate either and seem to vary a lot. My notes about changes I made to Tectnoics in AdvCiv also say that the Tectonics generates a lot more rivers and elevation compared with the original scripts. Though this may be part of its charm.

The BUG help file was discontinued in Oct 2008, so none of the BULL features are documented. Might accomplish more to improve the help text on the BUG Options screen. I've already revised many of the texts on that screen in AdvCiv, but I've also rearranged it a lot, which BULL users won't appreciate (ACO tab attached for example).
 

Attachments

  • aco-menu.jpg
    aco-menu.jpg
    126.8 KB · Views: 109
Tectonics – the options do say (in their names) that e.g. Earthlike and Lake have different land-sea ratios; but it may well be that Mediterranean is too extreme or arbitrary. I've noticed that the Earthlike water percentage aren't nearly accurate either and seem to vary a lot. My notes about changes I made to Tectnoics in AdvCiv also say that the Tectonics generates a lot more rivers and elevation compared with the original scripts. Though this may be part of its charm.
The ratios are labelled, but the overall grid size should change based on the option. ie, at Huge size every option should generate at least roughly 2000 land tiles and not more than about 3800. So the grid size should be bigger for 70% water and reduced a lot for mediterranean etc. Otherwise, to get a size for mediterranean which won't be abnormal, the map size option has to be used, which makes all the modifiers skewed from the worldinfo xml.
 
With the earth 18 civs scenario, the culture growth alert doesn't work. Also had a bug in which julius destroyed louis, whom I assume was his worst enemy, but that turn it displays blank space as his worst enemy. attached a save; current 1.03
The bundled version of the low random map script also appears out of date: https://forums.civfanatics.com/resources/random-map-script-v5-final.3574/
 

Attachments

The greeting text also (at random) shows an empty string for the worst enemy, like in the screenshot here. The game is still storing player number 9 (Louis, I'm guessing) as Julius' worst enemy; it apparently fails to update the worst enemy upon player defeat. Fixing this could clearly affect the course of the game a little bit because Julius would have a proper worst enemy sooner. Giving him no worst enemy for a little while would avoid the visual artifacts and would make sure that things don't somehow get even weirder. But it might be even more confusing to an observant player – "why no worst enemy? seems that Catherine should qualify ..." Well, a player who can figure out that an empty string stands for Louis probably also can guess that the absence of a worst enemy is a temporary result of Louis' defeat. Still, kind of covers up the underlying issue. :undecide:

Low's map: Turns out that BULL already comes with v5 (it says so on line 2 and I've also compared the file with the one you linked to through WinMerge), so the file only needs to be renamed. I'm still unsure whether the scripts should stay with Taurus at all; it'll probably become clearer when I finally get around to going through your unaddressed bug reports and suggestions from earlier this year. Maybe the download page should only offer an optional link to an archive.

Culture growth alert: Gets the ID of the human player wrong if it's not 0 – and, in scenarios, the human player can have any ID. EmperorFool had already noticed the issue according to his comment here, but I guess it wasn't obvious to him how to fix it. It does seem simple though, I've pushed my bugfix to GitHub: link
I've hardly ever enabled that alert; seems quite useless to me (probably has some niche use that I can't think of). I think K-Mod also disables it by default and perhaps for that reason it was never noticed and fixed in that context either. (In AdvCiv, I've even removed it from the BUG menu.)
 
It's useful for fogbusting; since the expansion happens the end of the same turn; if the culture growth would reveal tiles that had previously needed fogbusting (particularly expanding over a hill), then the unit can move out faster. I wouldn't want it disabled.
 
Is there a way to change the worker queue functionality to match AdvCiv? In AdvCiv, I can queue worker actions across multiple tiles (e.g. build a farm --> move to a hill tile --> build a mine). In Taurus, this doesn't seem to work.
 
I believe this mostly works; screenshot attached. When I just tried it, I had one situation where the Shift key was, effectively, ignored, i.e. the worker started the Farm straight away, but I didn't keep a savegame and haven't been able to reproduce this issue in a new game. When I try it with just BtS or BUG, it seems that the move-to plot (the hill) is usually ignored, i.e. there is no button for the Mine. Though, if I fiddle around with canceling missions and queueing them again for a bit, it does work eventually. Maybe this problem also can occur in Taurus. I don't remember if I changed anything about that in Taurus; I guess not, but BULL may well have (largely) fixed that issue. I do think it works reliably in AdvCiv, probably already does in K-Mod. Do you have a Taurus savegame where worker missions can't be properly queued?
 

Attachments

  • taurus-queue-mine.jpg
    taurus-queue-mine.jpg
    326.2 KB · Views: 72
Adding to the imperialistic governor still overvaluing food when building settlers, the generic governor for worker also overvalues food above +2. If there's a 3:food: food tile and one with 3:hammers:1:commerce: , it'll work the food; but if the former is replaced with a forested grassland it will switch to the latter. This also happens in vanilla civ 4. I don't know if it's fixed in advciv.
 

Attachments

I have attached a save, if I try to queue actions with the worker near Prague or Aachen I don't receive the worker action options I would expect, hopefully the 2 attached images illustrate this.

Worth noting that I am on Linux playing through Wine so it could be some unusual quirk of my setup, I know that Linux isn't supported.

Edit: updated screenshots
 

Attachments

Last edited:
@Deaf Metal: Seems that this term ...
C++:
	// tiny food factor, to ensure that even when we don't want to grow, 
// we still prefer more food if everything else is equal
iValue += (aiYields[YIELD_FOOD] * 1);
... in the AI tile yield evaluation is responsible. It cancels out the low valuation of the 1 commerce. K-Mod has overhauled the AI code for citizen assignments, to the point where it's difficult to compare the AI logic. But this food-based tiebreaker still exists in K-Mod, and a food production check (whether food is getting converted into production) has been added. This indeed fixes the problem in the savegame. It might be nice to still err on the side of more food when food is treated as production. Or at least not to rely on the modifier from the Imperialistic trait when working a high-food tile is just as effective on the bottom line. Because the player tends to forget about that modifier and may then be confused. I could move the tiebreaker to the end of the function and scale the value up by a factor of 10 before applying it. But it's difficult to anticipate all consequences of putting that value on a higher scale. So I guess I'll just go with the food production check (no food-based tiebreaker then) – and only for human cities, don't want to change AI behavior.

Hopefully, this will also fix the issue with the Imperialistic trait you noticed earlier. If the modifier isn't taken into account at all, that would be another problem to address; but I tend to think/ recall that the modifier is already accounted for by the BtS AI.
 
Last edited:
@selnik: I can reproduce this issue from this savegame, but, oddly, not reliably. If I just load it, select the worker and use right-click hover or hold shift, the Mine button will sometimes be missing; usually, it appears. Perhaps worth noting that the desert hill can't be mined because it is unowned – though a grayed-out Mine button should still appear. If it's somehow caused by undefined program behavior, then I could imagine that there's a hardware-/ platform-dependent aspect to it. I'll try to take a closer look tomorrow.
 
My best bet is that it's a problem with the Shift key not being detected reliably by BtS. I've tentatively adopted this K-Mod change ...
K-Mod 1.20 changelog said:
Replaced all calls to gDLL->altKey, shiftKey, and ctrlKey, with equivalent native Windows functions. This should fix the problem of keys not always registering correctly.
... and it seems that I can no longer reproduce the issue now. But, since I had trouble reproducing it in the first place, perhaps you want to confirm that it's fixed. Hopefully that native Windows function corresponds to something equally reliable under Wine. I've pushed my change to the v1.03 branch on GitHub. Though, if installing from GitHub (archive to be unzipped and then renamed to just "Taurus") is inconvenient, I can just (eventually) release v1.03 with the change, or maybe I'll have an opportunity to test it further myself.
 
@f1rpo I've given the updated version a quick test and it seems to have solved the problem, thank you so much for your quick response! I'll do more testing and let you know if the issue crops up again.
 
The mouseover for creative leaders in the sevopedia says it doubles production of madrassa instead of library:
 

Attachments

  • Civ4ScreenShot0000.JPG
    Civ4ScreenShot0000.JPG
    124.7 KB · Views: 79
Last edited:
Still broken in K-Mod/ AdvCiv, curiously never reported anywhere (not for BUG/BULL either, I suppose). Sevopedia fails to pass along the civ ID for the tooltip, and, curiously, uses 1 as a dummy value for unused tooltip data, which corresponds to the civ ID of Arabia. Will fix. Edit: Pushed the code to Github.
 
Last edited:
@Deaf Metal: Finally gotten around to sorting out the map scripts and going through the minor issues you pointed out in spring. Hoping to publish v1.03 soon. Regarding the maps, I intend to include the PublicMaps folder only as a ZIP archive in the release version, meaning that players will have to manually unpack it to get the additional/ enhanced maps. Install instruction still to be updated accordingly. Players who don't do so will no longer be burdened with the wasteful memory use of SmartMap and the visual clutter of scripts they never use. I'll keep everything unpacked in Git. I'd like to keep the PrivateMaps copies of Fractal and Pangaea in any case, i.e. not to be unpacked manually, because pushing these two to the top of the list on the game setup screens seems quite helpful to me; however, having two map folders is perhaps too confusing when the player is supposed to fiddle around with one of them.

I've added Ring and Wheel to my workaround for the minStartingDistanceModifier issue in the DLL. Including copies of those just to additionally work around (a little more thoroughly) the same problem inside the scripts doesn't seem worth it (even when the map scripts are optional). For Tectonics, I've only added some more water percentages to the names of the custom option choices and made the existing percentages more accurate. Players can then at least choose a suitable player count. From Big&Small and Medium&Small, I've removed the sea level option. I also committed a change that instead gives the existing sea level option its proper impact, but then decided that it's more appropriate for Taurus to only make a UI change, i.e. to hide the options which, in BtS, have no impact.

Regarding sea level and climate options, are you sure that they affect the generated maps for some of the scripts that disable those options, e.g. Arboria? Looking at Arboria maps and (very briefly) at the code, this isn't apparent to me. For reference, the full list of BtS scripts that disable both options is: Arboria, Boreal, Donut, Highlands, Rainforest, Tectonics and, from WL, Great_Plains, Oasis. Additionally, Team_Battleground and (WL) Mirror, Maze only disable sea level and Fantasy_Realm (WL) and Ice_Age (base game) only disable climate. Could well be that some of those still refer to the map's sea level or climate at some point – and will then be affected by the settings of a previously played or selected map script. I think this situation can be, let's say, sanitized for all map scripts by the DLL by setting the sea level to Normal and climate to Temperate when the map script disables those. I've made such a change in AdvCiv now (that mod also slightly adjusts tech costs to the sea level), but, for Taurus, I don't want to add random bugfixes that break BULL network-compatibility. Fixes in a couple of individual scripts would be better – as scripts get sent across the network at game launch.

The "Minor" civilization being listed in Sevopedia – had never noticed this myself, but it's true; was easy to remove. Hotseat disabling the Resource Display at turn end is a puzzling choice by Firaxis. It's a line in the DLL that explicitly does this. Had disabled this long ago in AdvCiv. In Taurus, it's now tied to the option that shows the Resource Display at game start, i.e. the DLL will not turn it on or off when that option is enabled. Lastly ...
In single player team games (maybe real multiplayer too); if a player later in the turn sequence is the one who finishes a tech that an earlier player had selected, when the earlier player's next turn starts, there's no tech selection popup.
Should I expect this to happen frequently if I just put myself in slot 0 and an AI teammate in slot 1? Or does it hinge on special settings/ circumstances?
 
Back
Top Bottom