Quick Answers / 'Newbie' Questions

[Samson]

To the best of my (limited) understanding this is true. However, you cannot safely tell where the first and last entry is, when using the arrows there is no such distiction.

Usually your capital is the first city founded and thus the first in the database. This gives you an important point of reference. But you donnot always have that reference.

But you know which you get a popup about first when it finishes a build.

 

[vorlon_mi]

This is really interesting, thanks. To me it brings up an unrelated point:

I thought the order that you went through cities with the left/right arrows will in the city screen was the order in the database, so you can always determine the order. It is relevant in a few cases.

If not, are there two such orders that matter at different times?

I've noticed something interesting about that particular city order. In the early game, the first few cities are my self-founded ones, in the order they were founded. Later, when I conquer another civ's original capital -- and keep it -- that city often comes up in the second or third slot. It was founded by the AI, usually in 4000 BC, so it moves up in the order. If I conquer more than one original capital, they will often appear before my second self-founded city.

Another complicating factor seems to be loading a save file. During a play session, if I conquer (say) Persopolis, that city might appear last in my arrow-thru-cities order. I save the game, exit, and then start a new session the next day. That AI capital might now be 2nd or 3rd, as I noted above. My point being: if I tried a palace jump during one session after a conquer, or in a later session after resuming play, I might get different results.

 

[Samson]

I've noticed something interesting about that particular city order. In the early game, the first few cities are my self-founded ones, in the order they were founded. Later, when I conquer another civ's original capital -- and keep it -- that city often comes up in the second or third slot. It was founded by the AI, usually in 4000 BC, so it moves up in the order. If I conquer more than one original capital, they will often appear before my second self-founded city.

In my experience it seems like this is always city founded order unless cities have been raised. If I found my capital in the first turn it will be first whatever other capitals I capture. If I move my selter any capitals I capture will get their builds before my capital. As I tend to build a settler later than the AI if I take the whole core it is likely one or two will get thier turn before my second founded city.

Another complicating factor seems to be loading a save file. During a play session, if I conquer (say) Persopolis, that city might appear last in my arrow-thru-cities order. I save the game, exit, and then start a new session the next day. That AI capital might now be 2nd or 3rd, as I noted above. My point being: if I tried a palace jump during one session after a conquer, or in a later session after resuming play, I might get different results.

This is weird and I have not noticed anything like this.

 

[Fergei]

1) With the AI aggression slider in C3C, what actually changes? Is it only the AI's starting attitude to all other Civs? (i.e. poorer attitude increases potential for war against human and AI?). From reading old threads AI starting attitude against humans gets worse as difficulty increases so I am guessing the difficulty slider simply extends this to the attitude to AIs. But it would be good to know.

2) my very old machine grinds to a halt in a late game on any large or huge map. End of turn takes minutes and it is not viable for me. Will map size be a variable in slowing down the end of turn OR is it really only based on number of units / number of land tiles? For example, would a large map with 80% water have shorter end of turns in the late game than a large map with 60% water? I am assuming so. Could the number of Civs also be a variable?

Hypothetically I assume a huge map with 10 Civs and 800 land tiles with say 90% water would have quicker end turns than a standard map with 12 Civs and 900 land tiles and 40% water (yes, I know those settings may not be possible in the editor).

If nobody knows it is fine, I will try a modified Large map with the same number of land tiles as a standard 60% water map. I am looking for an experience that feels different from the norm (ideally 4 to 6 very large islands but massively spaced out. Standard map, 60% water archi essentially just creates continents rather than true islands). I am also looking for ways to further subtly adjust difficulty via map settings, specfically to keep the game pretty much too hard for me due to enemy army size but not quite so hard on the ancient & medieval tech front (where i typically fall too far behind to really be able to contemplate war until later on). I think more proper islands will help limit AI contact with other AI and slow down their runaway tech sharing but I don't know if I can create that sort of experience on my old computer.

 

[Takhisis]

2) my very old machine grinds to a halt in a late game on any large or huge map. End of turn takes minutes and it is not viable for me. Will map size be a variable in slowing down the end of turn OR is it really only based on number of units / number of land tiles? For example, would a large map with 80% water have shorter end of turns in the late game than a large map with 60% water? I am assuming so. Could the number of Civs also be a variable?

The game calculates the trade networks again every turn, so that's always a factor. Also, you don't get to see most of the map, so who knows how many unit movements and actions (including terraforming and bombardment, both of which affect trade) you're not seeing.

Basically, the more stuff on the map, the more interactions between all the tiles and their occupants the game has to calculate anew every turn.

 

[Spoonwood]

1) With the AI aggression slider in C3C, what actually changes? Is it only the AI's starting attitude to all other Civs? (i.e. poorer attitude increases potential for war against human and AI?). From reading old threads AI starting attitude against humans gets worse as difficulty increases so I am guessing the difficulty slider simply extends this to the attitude to AIs. But it would be good to know.

No, there's more. The aggression level affects the trade price between you and the AIs. Workers cost the least at minimum aggression, and more at higher levels. Also, the exchange rate of gold for gpt is 18 gold for 1 gpt at minimum aggression, but it's not that one higher levels. Also, it's a fixed rate on minimum aggression, while it varies on higher levels depending on the attitude of the AI you trade with. There could be more. I actually tried to contact Soren Johnson on this, but he said it had been so long that he did civ III that I probably had a better idea than he did.

 

[Samson]

Also, the exchange rate of gold for gpt is 18 gold for 1 gpt at minimum aggression

A negative interest rate on government debt? Those radical economists

 

[tjs282]

A loan of 18g requiring a total repayment of 20g (assuming no default) is like an interest rate of 11%...

 

[Civinator]

2) my very old machine grinds to a halt in a late game on any large or huge map. End of turn takes minutes and it is not viable for me. Will map size be a variable in slowing down the end of turn OR is it really only based on number of units / number of land tiles? For example, would a large map with 80% water have shorter end of turns in the late game than a large map with 60% water? I am assuming so. Could the number of Civs also be a variable?

Hypothetically I assume a huge map with 10 Civs and 800 land tiles with say 90% water would have quicker end turns than a standard map with 12 Civs and 900 land tiles and 40% water (yes, I know those settings may not be possible in the editor).

If nobody knows it is fine, I will try a modified Large map with the same number of land tiles as a standard 60% water map. I am looking for an experience that feels different from the norm (ideally 4 to 6 very large islands but massively spaced out. Standard map, 60% water archi essentially just creates continents rather than true islands). I am also looking for ways to further subtly adjust difficulty via map settings, specfically to keep the game pretty much too hard for me due to enemy army size but not quite so hard on the ancient & medieval tech front (where i typically fall too far behind to really be able to contemplate war until later on). I think more proper islands will help limit AI contact with other AI and slow down their runaway tech sharing but I don't know if I can create that sort of experience on my old computer.

Fergei, the R17_preview version of the Flintlock mod allows you to answer these questions if you use it, as it holds a tool to measure the interturn time:

https://forums.civfanatics.com/thre...rd-and-much-more.666881/page-85#post-16539233
https://forums.civfanatics.com/thre...rd-and-much-more.666881/page-85#post-16539023

 

[Lanzelot]

my very old machine grinds to a halt in a late game on any large or huge map. End of turn takes minutes and it is not viable for me.

Maybe the following helps: I also have an old Win XP desktop from 2006, but it handled large maps ok, after I updated the RAM to the maximum (I think it was 4 GB for that motherboard). I didn't try huge maps, though. But in general I think, that for good performance on big maps, sufficient RAM is more important than a faster CPU. Because if the entire game data does not fit into the available RAM, Windows starts "swapping" (meaning: memory (= game data) that is currently not needed, is written to the Windows pagefile, and when it is needed again for calculations, it is loaded back into the RAM from the pagefile (and other parts of the memory are paged out instead)). Once this is happening for every single calculation, the constant file read&write access of course slows down the process by several magnitudes...

So if you are a bit "hardware-savvy" and are not yet at the max RAM configuration of your mainboard, you could

• Check the type of RAM your mainboard needs (unfortunately, there are many different incompatible types of RAM these days, from DDR1 to DDR5 etc.)
• Go to eBay and buy the maximum supported by your mainboard, e.g. a typical configuration of early 2000s hardware was 4 modules a 1GB. On eBay you can still find even old DDR1 modules in brand-new state for a handful of bucks.
• Replace them in your PC. Of course there is always the risk to break something, especially if the old RAM modules are sitting in their socks very tightly, but the performance boost you get from e.g. an upgrade of 4 x 256MB to 4 x 1GB is quite big.

 

[justanick]

If RAM is any serious issue, then the maschine is indeed very old.

 

[Samson]

I also have an old Win XP desktop from 2006

Wow, that is pretty impressive. I assume you keep it offline?

I have been wondering about trying virtualisation as a way to play civ on linux. Some minimal version of XP, perhaps based on embedded, would be my first choice of "lightweight" windows.

 

[Fergei]

Maybe the following helps: I also have an old Win XP desktop from 2006, but it handled large maps ok, after I updated the RAM to the maximum (I think it was 4 GB for that motherboard). I didn't try huge maps, though. But in general I think, that for good performance on big maps, sufficient RAM is more important than a faster CPU. Because if the entire game data does not fit into the available RAM, Windows starts "swapping" (meaning: memory (= game data) that is currently not needed, is written to the Windows pagefile, and when it is needed again for calculations, it is loaded back into the RAM from the pagefile (and other parts of the memory are paged out instead)). Once this is happening for every single calculation, the constant file read&write access of course slows down the process by several magnitudes...

So if you are a bit "hardware-savvy" and are not yet at the max RAM configuration of your mainboard, you could

• Check the type of RAM your mainboard needs (unfortunately, there are many different incompatible types of RAM these days, from DDR1 to DDR5 etc.)
• Go to eBay and buy the maximum supported by your mainboard, e.g. a typical configuration of early 2000s hardware was 4 modules a 1GB. On eBay you can still find even old DDR1 modules in brand-new state for a handful of bucks.
• Replace them in your PC. Of course there is always the risk to break something, especially if the old RAM modules are sitting in their socks very tightly, but the performance boost you get from e.g. an upgrade of 4 x 256MB to 4 x 1GB is quite big.

Thanks very much. The thought of needing to pimp my computer to run a 20 year old game makes me chuckle. Unfortunately, my abilities are limited to replacing the battery every 10 years or so and I wouldn't do anything to jeopardise my wonderful 2001(?) XP computer, especially now I have bought dozens of cheap XP compatible games on GoG to last me until the grave! I've only got 2GB RAM and will just have to stick with it.

I am online on the computer and I partly work in Data Security - I am just not convinced a modern OS is required outside of computers used for business. The computer still does everything I need and plays the games I want to play (except for a huge map on Civ3). Most importantly I don't get ruddy Windows updates forced upon me. Its fine here living in the Industrial Era.

*waits for computer to get infected or blow up tomorrow*

EDIT: a young computer, just 2004. Also, I don't do anything to do with money or my identity on it. I use other devices for that sort of thing.

 

[Fergei]

Ooh, so a huge map with 80% water has only 20% more land tiles than a standard map with 60% land (5000 compared to 4000). Same for a large map with 70% water. I will try cramming 18 Civs onto these maps and see how the old computer holds up. I suspect naval activity will grind it to a halt.

I think Large at 80% water will be its limit.

 

[Takhisis]

I also have an old Win XP desktop from 2006

Wow, that is pretty impressive.

Mine from 2008 lasted until this year and, presumably, since all that happened was some hard drive damage, I could try and clone the hard drive if I wanted to. Specifically the system32 folder went down and had to be replaced with the backup copy, but who knows?

 

[Samson]

Are/were you were going on the internet and running javascript with an XP machine? I would not have believed it was possible, but then I have not used windows much recently.

 

[Takhisis]

Yeah, most sites worked fine. I used that partition mostly for programmes that wouldn't've worked any other way except perhaps a virtual machine, i.e. mostly games from the 1990s-2000s and some design software. Flash still ran because Adobe never managed to remotely disable it for older systems (Metanet's N is one game everyone should play).

btw I keep finding places that still run Windows XP for accounting, tills and so on. I wonder whether they're paying Microsoft for servicing them or not.

The thing is, Windows XP did start running slowly over the years compared to the Linux partition, so it was relegated more and more to specific gaming (or modding) sessions and for almost everything I used more modern, non-Microsoft systems.

 

[robbus]

Thanks very much. The thought of needing to pimp my computer to run a 20 year old game makes me chuckle. Unfortunately, my abilities are limited to replacing the battery every 10 years or so and I wouldn't do anything to jeopardise my wonderful 2001(?) XP computer, especially now I have bought dozens of cheap XP compatible games on GoG to last me until the grave! I've only got 2GB RAM and will just have to stick with it.

I am online on the computer and I partly work in Data Security - I am just not convinced a modern OS is required outside of computers used for business. The computer still does everything I need and plays the games I want to play (except for a huge map on Civ3). Most importantly I don't get ruddy Windows updates forced upon me. Its fine here living in the Industrial Era.

*waits for computer to get infected or blow up tomorrow*

EDIT: a young computer, just 2004. Also, I don't do anything to do with money or my identity on it. I use other devices for that sort of thing.

Nice to know there are others doing what I have done! Saving their XP computer
I have a 2001 XP rig that has been offline since 2014.
I use it Mainly to play SCARFACE and Call of Duty 2 and earlier.

But it also has a fire wire port which I have used to archive my vhs home videos.
I have newer rigs including a 3000 nVidia 1 tb ssd that has been idle for a year since I bought it

cheers

 

[Fergei]

The game calculates the trade networks again every turn, so that's always a factor. Also, you don't get to see most of the map, so who knows how many unit movements and actions (including terraforming and bombardment, both of which affect trade) you're not seeing.

Basically, the more stuff on the map, the more interactions between all the tiles and their occupants the game has to calculate anew every turn.

Ah, I had not noticed that. Now that you mention it I see if EoT turns takes 2mins, the first minute appears to be unit moves and alliances. Then I can have a city production update, then I can wait for a second minute, which always baffles me. This giant delay when it has been identified it is my turn. Maybe this is when the trade deals are calculated and as you suggest, further slows down turns during wartime. I can't see anything I can do about that except reduce the number of Civs or increase aggression so there is more AI extinction before naval activity.

Basically turn length seems a lot less dependent on the number of available land tiles than I was expecting. I got through a recent Huge map game (19 Civs) without it being that much worse than some Standard map game.

 

[justanick]

I can't see anything I can do about that except reduce the number of Civs or increase aggression so there is more AI extinction before naval activity.

As i understand it the numbers of civs has only a low to mediocre impact. It is cities that matter and therefore tiles. For every city it is checked whether it has a trade route to any other city. But how say a given 100 cities are distributed among 1 to 31 tribes has little impact.

Basically turn length seems a lot less dependent on the number of available land tiles than I was expecting.

If my understand is true, than your new understand is wrong, while your old understand was right.