Single Player bugs and crashes v37 plus (SVN) - After the 24th of December 2016

[Toffer90]

I wouldn't go that far. I still think that tiles should be distinct from each other, otherwise...

My interpretation does not make one terrain type less distinct than a different terrain type.
Rather the opposite: Two flatland lush terrain plots are not two perfectly uniform rectangles of identical form, topography, size, and content. (I chose two identical plots to really emphasize how my interpretation makes all plots, regardless of terrain and plot type, more distinct)
It also makes the plot types (coast - ocean - flatland - hills - peaks) more distinct by giving them a general size rule (peak < hill < flatland < coast < ocean < space > other planet).

One of the reasons for why I prefer to interpret the map this way, is that I don't like the idea of having modern ships with 20-40 movement points. It should be closer to 5-10 because I feel it would provide a more enjoyable game-experience.

You couldn't have a tile the size of Russia on a giant map of earth, since this tile would certainly need more than 8 neighbours.

I don't see how that example is particularly relevant to the discussion, but it could of course have exactly 8 neighbors. Why do you say that that is impossible?

 

[tmv]

My interpretation does not make one terrain type less distinct than a different terrain type.

No, not terrain type. I mean two tiles shouldn't interpenetrate (we already discussed this).

One of the reasons for why I prefer to interpret the map this way, is that I don't like the idea of having modern ships with 20-40 movement points. It should be closer to 5-10 because I feel it would provide a more enjoyable game-experience.

There are purely game-reasons for this - there are many ship-generations in the game, and every new generation should offer something new. Increased movement is the most useful advantage. Another option would be to forbid older ship types from entering certain terrain completely, and only the newer ship types (and after a certain tech) would be allowed to. I agree that this high movement makes amphibious invasions too easy, with naval warfare hardly happening, if at all.

Why do you say that that is impossible?

A tile has 8 neighbours (3² - 1²). Then there are 16 second-order-neighbours (5² - 3²), 24 third-order-neighbours (7² - 5²), and so on (let's ignore the map border for the moment - this is a giant map and the part of the map that close to the border is negligible). After a while it gets difficult to "place" the tiles on a real map while keeping the tile connections you have (Russia has about 17 million sq. km, which is about one ninth of the total landmass of earth - roughly 150 million sq. km according to https://hypertextbook.com/facts/2001/DanielChen.shtml).

 

[Toffer90]

I would say it's easier to avoid voids and overlapping when wrapping a rectangular 2D grid around a 3D shape (e.g. a sphere) if it is allowed to warp/morph and stretch the 2D grid.
Sounds to me that you think the opposite, that warping and stretching makes it more difficult to wrap the 2D grid around a 3D object without getting overlaps and voids.
A cylinder is an exception though, minimal warping is necessary to wrap a rectangular grid of uniform squares around it.

The way I see it, the grid map in civ is by mathematical necessity a morphed representation of the 3D object it is derived from, and much of your arguments supports my interpretation of the map rather than challenge it.

A tile has 8 neighbours (3² - 1²). Then there are 16 second-order-neighbours (5² - 3²), 24 third-order-neighbours (7² - 5²), and so on (let's ignore the map border for the moment - this is a giant map and the part of the map that close to the border is negligible). After a while it gets difficult to "place" the tiles on a real map while keeping the tile connections you have (Russia has about 17 million sq. km, which is about one ninth of the total landmass of earth - roughly 150 million sq. km according to https://hypertextbook.com/facts/2001/DanielChen.shtml).

If I were to make an earth specific map where I wanted Russia to be represented by e.g. one flatland plot, I could go real low resolution and make a giant map consisting of 16 (X and Y axis must be dividable by 4 in Civ IV maps) plots, there would be no smaller editions of the map but I might make a gigantic version that has 64 plots (lol).

e.g. 12 could be flatland plots and 4 could be coastal plots
F - Flatland
C - Coast
▬▬▬▬
---------↓ Russia could perhaps be this one on the top right corner.
F C F F
C F F F
F C F C
▬▬▬▬
It is then obvious that the plots does not represent the actual planet that it is depicting without you, in your mind, warping those plot to fit the contours of an actual earth map.
One has to imagine that the flatlands contains hills, mountains, forest, bamboo, all the terrains and features that is on Earth than did not fit into the low resolution map, that Australia is represented by one coastal plot, that Africa is represented by 2 and a half flatland plots and that Europe is represented by one half flatland plot, that the atlantic and pacific ocean are present in several flatland plots.

Since your example was extreme my response to it must also be extreme.

 

[tmv]

I would say it's easier to avoid voids and overlapping when wrapping a rectangular 2D grid around a 3D shape (e.g. a sphere) if it is allowed to warp/morph and stretch the 2D grid.

Do you really have a globe in mind when building a map in C2C? Then why do most of these maps (and the animal spawn areas) resemble the Mercator projection so much? Because if you choose a certain projection, you get a flat map which you can tessellate (https://en.wikipedia.org/wiki/Tessellation). In case of Mercator, every tile spans a certain angle latitude-longitude-wise, which is why Mercator breaks down near the poles. Some other projections cannot be used here because they are not cylindrical (like https://en.wikipedia.org/wiki/HEALPix). Here is a (more or less) full list: https://en.wikipedia.org/wiki/List_of_map_projections

I could go real low resolution and make a giant map consisting of 16 (X and Y axis must be dividable by 4 in Civ IV maps) plots

Giant map has a certain meaning here (even if every mapscript can deviate from the norm). You said there was no right or wrong with these interpretations and I wanted to point out that this is only true within reason. Sorry.

 

[Toffer90]

Do you really have a globe in mind when building a map in C2C?

Not really, but I like to imagine that the world seen in civ represent a three dimensional world.
That It's either a doughnut shaped planet (Toroidal Map option), or a spherical planet with no access to the poles (Cylinder map option).
I've never really made a civ map manually, but I did make a map script that use a 4 dimensional noise field (Two coordinates [X,Y,Z,W] that are close together have only a small difference in value, a value that have an upper and lower limit, like 0.0 to 1.0.) that provides a height map that wrap both at the y and x axis without any distortion of the noise field values.
[Commercial alert ^^] The only map script that can effortlessly have a continuous landmass that crosses both the x and y edges of the map without any mirroring or sudden jumps in height values.
[Edit: Bad commercial, mental note: Commercials should mention the product name ▬ World ▬ lol]

Then why do most of these maps (and the animal spawn areas) resemble the Mercator projection so much?

The earth maps I've played have had obvious warping of land areas, far north and south areas usually have way more plots than they should have compared to (e.g.) Africa, which suggest to me that the land area cannot be the same for all the plots.
Mercator projection is well in line with my way of interpreting the map. it distorts the size of map features more and more the higher the latitude. Although I go farther, and say that there is no reason why there may not be local warping of the grid all over the map (At each and every plot) instead of only one uniform warping effect globally across the map due to the mathematical projection that might have been used as a base. That each plot square seen in the game represent a uniquely shaped puzzle piece of the spatial area that the map represents.

Giant map has a certain meaning here (even if every mapscript can deviate from the norm). You said there was no right or wrong with these interpretations and I wanted to point out that this is only true within reason. Sorry.

Yeah sure, if I made an earth map with 20 000 plots where only one of the plots were to represent entire Russia then I would have to seriously warp and stretch the world map as we know it by enlarging (adding more plots to) the ocean and land neighbors around Russia; it would be completely possible but the end result would make the player think, where is Russia, and why is China/Mongolia and Scandinavia so insanely big compared to the rest?

I think you are interpreting what I say outside of what is reasonable to expect.
It would be unnatural for me to think that one plot may represent something that much bigger than what the other plots represents, but if someone wants to do that, it is as I said, there's no right or wrong when it comes to something as subjective as interpreting things that are abstract in its nature.
Detour:
When I play a game like Fallout New Vegas and the game report (graphically) that a magnum 44 round was planted smack in the middle of the forehead of a completely normal human without a helmet who just keeps on fighting afterward, I choose to interpret that I either missed or hit my opponent differently than what the game showed me. I might even interpret that one enemy represents several enemies and that I just killed one of them.
When I read a fictional book, I fill in the blanks, imagine how it looks like and read between the line, my experience of the book is different from all others but there is no right or wrong there.

 

[tmv]

Although I go farther, and say that there is no reason why there may not be local warping of the grid all over the map (At each and every plot) instead of only one warping effect that has a uniform warping effect globally across the map due to the mathematical projection that earth map makers might use. That each plot square seen in the game represent a uniquely shaped puzzle piece of the spatial area that the map represents.

I don't think it should be possible is to pick any C2C map and call it an earth map. IMO there should be some way to recognize what the map is supposed to represent - that would be better served by a uniform warping effect.

When I play a game like Fallout New Vegas

Shooters are perhaps the worst games in that sense - not just fatal wounds that are shrugged off, but even non-fatal wounds would often require weeks (if not month) of surgery, rehabilitation, etc.

I might even interpret that one enemy represents several enemies and that I just killed one of them.

In that case that enemy's firepower should be reduced - if that is the case, fine.

When I read a fictional book, I fill in the blanks, imagine how it looks like and read between the line, my experience of the book is different from all others but there is no right or wrong there.

Certain movies are so bad in that regard that I wonder how they can call that a film adaptation of the book that (usually) just happens to have the same name (e.g. (in no special order) Dune (1984), Prince Caspian, Troy, The Time Machine (2002), The Hobbit).

 

[Toffer90]

I don't think it should be possible is to pick any C2C map and call it an earth map. IMO there should be some way to recognize what the map is supposed to represent - that would be better served by a uniform warping effect.

I agree that earth maps should resemble actual earth maps, but I've never seen an earth map that resembles earth very well, without me using my imagination to warp all the plot sizes and shapes to better represent the actual real life map features better. One would need an insane amount of plots to even get close, e.g. one real life mountain is represented by 40 000 peak plots, and even then there would be anti aliasing effects on the plots at the intersection between that mountain and the next map object/feature/terrain/biome, before you actually have confined that mountain to only be inside the plots that represent it. So even then there are some plots that have a morphed shape (something that is not exactly squared is morphed to be square on the game map). Game developers have long ago realized that there is little point in graphically and game-mechanically describing reality accurately, it is better to abstract reality to save an immeasurably amount of computing power that can be used better in other areas of the game than at the reality factor, a game with a warped version of reality is usually a much more fun game as well.
Since the game map typically is a very low resolution of the earth topology one has to imagine that a peak plot does not represent a perfectly square piece of land that in its whole has a shared common ruleset, but that the peak plot might represent a piece of land with a complex shape that share a common ruleset.

Here's an illustration of how I may interpret a Civ map:

Spoiler Crude Illustration :

When I see this:

I interpret it to represent something like this:

If you feel that this is totally unreasonable, then we would have to agree to disagree.

Certain movies are so bad in that regard that I wonder how they can call that a film adaptation of the book that (usually) just happens to have the same name (e.g. (in no special order) Dune (1984), Prince Caspian, Troy, The Time Machine (2002), The Hobbit).

I agree, book to movie adaptations never really do the book justice, like "The War of the Worlds" where the alien invasion from Mars was changed to the monster invasion from the subterranean depths of earth. That was not even close to a subjective interpretation of something abstract in the book, it was objectively speaking a deviation from the very concrete parts of the book.
A terrible movie from 2005, but a great sci-fi book written in 1898, also a great real time strategy game adaptation from 1998 (other adaptations exist).

 

[Raledon]

After a couple of versions ago I had a repeated crush, I updated to the latest svn and took over this city.
the borders seems to have gone a little wonky, as a 3rd party got the tiles around the city for no reason. My units in the stack got kicked out a tile as I captured the city, as I'm in peace with that guy. In addition, some units in the city (a great general and two workers) were left inside, instead of being converted to my side. This was used with stack attack and without. If I raze the city, orange gets the city's tile as well

Spoiler :

map after i took over sao paulo

 

[tmv]

I agree that earth maps should resemble actual earth maps, but I've never seen an earth map that resembles earth very well, without me using my imagination to warp all the plot sizes and shapes to better represent the actual real life map features better.

I think (most of) the common projections are rather good, I think Mercator is still one of the best (the poles are a problem, but they are close to unusable anyway).

Game developers have long ago realized that there is little point in graphically and game-mechanically describing reality accurately

There was one attempt in 1993 to do away with tiles completely in a strategy game: http://www.mobygames.com/game/dos/patriot - unfortunately that game was so bug-ridden that it vanished very quickly (in theory it was excellent for a very hard strategy game).

one real life mountain is represented by 40 000 peak plots

With the current movement points you would simulate snails, not humans.

Crude Illustration

That's actually a rather good transformation IMO, although the desert plots might be a bit large (turning the Sahara almost into a non-problem even with terrain damage).

I agree, book to movie adaptations never really do the book justice, like "The War of the Worlds" where the alien invasion from Mars was changed to the monster invasion from the subterranean depths of earth. That was not even close to a subjective interpretation of something abstract in the book, it was objectively speaking a deviation from the very concrete parts of the book.

I haven't actually seen the movie myself (I have read a warning about the little girl, played by an excellent actress, being incredibly annoying), but - are you serious? So these earthbound monsters are defeated by the common cold, coming from earth? They take the one weak point of the novel (spacefaring aliens without NBC defense - which was completely unknown in 1898 and not at all the author's fault) and make it worse?

 

[Thunderbrd]

After a couple of versions ago I had a repeated crush, I updated to the latest svn and took over this city.
the borders seems to have gone a little wonky, as a 3rd party got the tiles around the city for no reason. My units in the stack got kicked out a tile as I captured the city, as I'm in peace with that guy. In addition, some units in the city (a great general and two workers) were left inside, instead of being converted to my side. This was used with stack attack and without. If I raze the city, orange gets the city's tile as well

Spoiler :

View attachment 483922
map after i took over sao paulo

Looks like the orange player once owned sao paulo. Keep going and see how it plays out once you get the city out of anarchy. Those units left inside, were they even units owned by the enemy or were they there visiting from another player? That might be an interesting issue. Also, sometimes, with older saves, you can end up getting some index references out of whack and the whole game goes a bit wonky.

 

[Raledon]

Looks like the orange player once owned sao paulo. Keep going and see how it plays out once you get the city out of anarchy. Those units left inside, were they even units owned by the enemy or were they there visiting from another player? That might be an interesting issue. Also, sometimes, with older saves, you can end up getting some index references out of whack and the whole game goes a bit wonky.

Same player. Having units enter the city caused the units to "initiate combat" and lose, once per unit.
It's possible that this is due to the update, as it's the first I encountered this issue since I disabled BUG's stack attack.

 

[Thunderbrd]

Same player. Having units enter the city caused the units to "initiate combat" and lose, once per unit.
It's possible that this is due to the update, as it's the first I encountered this issue since I disabled BUG's stack attack.

I'd be interested to see a save that captures the moment before you instruct the attack by the last unit that will capture the city.

 

[Raledon]

I'd be interested to see a save that captures the moment before you instruct the attack by the last unit that will capture the city.

This is the save, just attack with the stack near sao paulo. I've uploaded the version before the update as it might grant some insight, though the bug seem to be the same as after applying the update.

Another bug I noticed is that the gold income from priest and engineer seems to be off the scales. It claims to grant me +130/-125 actual gold for adding/removing a priest, while the number seem to be impossible. A priest shouldn't give me much more than 3-4 base gold (+0.75, +1 from civics +bonuses from religious), and even with all the multipliers, should be capped at somewhere around 10-15.
It's possible it claims to give me as much as all the priests currently in the city, though it seem strange as the number appears low due to modifiers.

 

[Toffer90]

I think (most of) the common projections are rather good, I think Mercator is still one of the best (the poles are a problem, but they are close to unusable anyway).

Map projection theory is not really the core of the subject we are discussing, but rather the nature of the grid plot system, which I feel makes everything too "jagged" (jagged mountain ranges, costs line, and biome borders), which is why I don't interpret one plot as representing a strict size and shape of space taken directly from the idea that the game map represent.

With the current movement points you would simulate snails, not humans.

True, if the game operated at such a map resolution (trillions of plots) the entire game would have to be changed, unless we wanted units to have hundreds of movement points one would have to change tech inventing to take millions of turns and each turn would have to represent something closer to one hour.
I'm glad the game map is operating at a low resolution (high level of abstraction) like every good strategy game should.
The example I used was perhaps a bad attempt at proving a point, as some implications of the example could act distracting.

That's actually a rather good transformation IMO, although the desert plots might be a bit large (turning the Sahara almost into a non-problem even with terrain damage).

That unambitious drawing doesn't really do the grid warping I imagine justice, but it is close enough to give others an idea about what I mean.
About the size of Sahara, that's a question about map size. I'm not really sure where in Libya that the Sahara begins, and where it ends in the south, but it should be at least 6 plots vertically if you extrapolate the illustration I made further south.

I haven't actually seen the movie myself (I have read a warning about the little girl, played by an excellent actress, being incredibly annoying), but - are you serious? So these earthbound monsters are defeated by the common cold, coming from earth? They take the one weak point of the novel (spacefaring aliens without NBC defense - which was completely unknown in 1898 and not at all the author's fault) and make it worse?

It's a long time since I saw the movie, so they may have been tied to Mars in one way or another. I mostly remember that the movie was real inconsistent with the original novel in ways that I found poorly justified, I left the cinema theater quite dissatisfied.

I don't agree that the weakness you mentioned in the novel is completely reasonable to call a weakness.
The novel describes the aliens as beings similar to humans, that long ago (not specified but a couple of thousands of years might be reasonable) once lived on the surface of Mars, but that they were forced to move deep underground around the cataclysm when their atmosphere deteriorated. During their time underground they evolved rapidly into something else (read the book for more details). The aliens probably never achieved space travel until they actually invaded earth by shooting a couple dozen pods with no navigational systems that probably slingshot around mars moons toward earth.

They had probably observed earth for a couple hundred years while living underground without invading, which suggest that they neither had the technology nor the resource capacity to do so until they suddenly was forced to execute their plan which had been in the works for decades. It is described that the invasion was a last ditch attempt for their species to survive ( they had very short time left on Mars before going extinct, I don't really remember the reason, 16 years since I read the book, but it might have been connected to resource shortage, breakdown of civilization and maybe even a cooling Mars core was mentioned ) and that they, even though they had a huge brain, and was supposedly real geniuses at math, no good premise to understand and properly prepare for all the obstacles they may meet during the invasion (not enough empiric and experimental data).

The pods contained something like 5-10 aliens, some war- and work- machines, and what they needed to set up a makeshift factory/lab/outpost in the crater they landed in. They did have equipment that protected them, all things considered, very well against the environment. They were described to have issues with breathing the air on earth and needed to stay inside their protective gear and machines until they had made some serious terraforming of the plant life on earth, they had special plant life with them that produced a certain gas that was essential to them. It was not before they in late stages of the invasion that they started to get careless.

There are many weak point in the novel if we look at it with todays common knowledge though, e.g. The pods traveled from Mars to Earth in just a couple of days, Wells (the author) probably had no idea how far away Mars really was.
To me that book will always represent a truly impressive work of sci-fi storytelling.

Edit: Perhaps we should end this friendly conversation now, considering this is the bug reporting thread and all, it's a "bit" off-topic.

 

[JosEPh_II]

This is the save, just attack with the stack near sao paulo. I've uploaded the version before the update as it might grant some insight, though the bug seem to be the same as after applying the update.

Another bug I noticed is that the gold income from priest and engineer seems to be off the scales. It claims to grant me +130/-125 actual gold for adding/removing a priest, while the number seem to be impossible. A priest shouldn't give me much more than 3-4 base gold (+0.75, +1 from civics +bonuses from religious), and even with all the multipliers, should be capped at somewhere around 10-15.
It's possible it claims to give me as much as all the priests currently in the city, though it seem strange as the number appears low due to modifiers.

You do do a Re-calc after each update right?

And it can take several turns for the Base modifier values to stabilize again.

 

[strategyonly]

i just looked at the other nations in my game, and i am in the Information era and the cities have ONLY 1 unit guarding their cities on each city?? WTH, and they are n't even sending out workers to do anything at all, in the pic i have a stone resource next to the city, and nothing and in the other city all that is there is a mule?? no roads being built, nothing?? and infact on the whole map only 1 maybe 2 units are in the city guarding them, this is not good . . . .

game in zip if u want to look??

 

[Raledon]

You do do a Re-calc after each update right?

And it can take several turns for the Base modifier values to stabilize again.

I do, I have updated at the same turn, though. When I updated, the same battle caused a constant CTD, and the update itself "fixed" it.
In regards to stabilizing, roughly 15 turns later I took over another city, and had a similar issue units of the opponent being kept alive in the conquered city.

 

[Thunderbrd]

This is the save, just attack with the stack near sao paulo. I've uploaded the version before the update as it might grant some insight, though the bug seem to be the same as after applying the update.

Another bug I noticed is that the gold income from priest and engineer seems to be off the scales. It claims to grant me +130/-125 actual gold for adding/removing a priest, while the number seem to be impossible. A priest shouldn't give me much more than 3-4 base gold (+0.75, +1 from civics +bonuses from religious), and even with all the multipliers, should be capped at somewhere around 10-15.
It's possible it claims to give me as much as all the priests currently in the city, though it seem strange as the number appears low due to modifiers.

OK, I'll take a look.

i just looked at the other nations in my game, and i am in the Information era and the cities have ONLY 1 unit guarding their cities on each city?? WTH, and they are n't even sending out workers to do anything at all, in the pic i have a stone resource next to the city, and nothing and in the other city all that is there is a mule?? no roads being built, nothing?? and infact on the whole map only 1 maybe 2 units are in the city guarding them, this is not good . . . .

game in zip if u want to look??

Looks like we're still getting crime implosions. I've noticed a critical problem with the AI in my game that I need to evaluate that may be partially responsible. Your game may give me some insights as well. I'll look into it. It's amazing how many folks are getting into the info age and beyond now though.

 

[Raledon]

Is the AI still susceptible to bad crime handling?
The AI seem to be plagued by revolution. Looking at a few nearby cities, it seem to build/station its armies poorly.
For example, in the pic below there are 15 units which causes crime while stationed in the city, and only one LE. This will surely cause issues for the AI, sooner or later.
The other cities don't seem much better. There are a few guards walking around (you can see a group of 8 LE on the peaks), but they are not stationed in the cities, indicating the AI might be moving them around all the time instead of creating additional units.

Spoiler :

 

[Raledon]

Another couple of issues:
It's unclear if guilds really give gold as intended. From my understanding, each guild should give +5 gold for each building of that type, in the city the building is built. However, I don't seem to get the gold in the city, so I'm wondering how it works.
Carrack Merchant returns to the capital instead of the city it was built, resulting in the merchant ship popping up in a land city. I'd expect it to be destroyed or pop at some random coastal city, at the very least.

Edit: mercy rule (from BUG) seems to work based on Mastery Victory's settings instead of the score as indicated. This leads to a sudden lose if an opponent decides to go for "cultural victory" (even if it's set to off).