Formula for growth

[J Bytheway]

Originally posted by Maquiladora
Im not sure I get you, what do you mean by "total output per worker"? I thought every single tile on each ring was worked by a percetage of current workers available to total workers available on a ring, regardless of number of tiles.

Yes, but the important part there is the fact that it's "to total workers available". If you decrease the "total workers available" for the ring, then you increase the output of each worker.

[wheathin]

You're right - but the number of tiles on the ring determines total output.

In ring 1, there are 8 tiles, worked by 8 workers: one tile per worker. In ring 2, there are 12 tiles, worked by 12 workers: one tile per worker. In ring three, there are 24 tiles, worked by 12 workers: it is as if each worker were working two tiles as opposed to one. Each worker in ring 3 is twice as productive as the workers in rings 1 and 2.

[Maquiladora]

Originally posted by J Bytheway
Yes, but the important part there is the fact that it's "to total workers available". If you decrease the "total workers available" for the ring, then you increase the output of each worker.

Im aware of that, it was the words "tiles per worker" that confused me to his point.

You're right - but the number of tiles on the ring determines total output.

In ring 1, there are 8 tiles, worked by 8 workers: one tile per worker. In ring 2, there are 12 tiles, worked by 12 workers: one tile per worker. In ring three, there are 24 tiles, worked by 12 workers: it is as if each worker were working two tiles as opposed to one. Each worker in ring 3 is twice as productive as the workers in rings 1 and 2.

Well now I see your point.

[wheathin]

Well now I see your point.

Thanks! Now the question is: why do it that way - is it a game balance issue? If it is, why not increase the productivity of workers in rings 4 and 5 too? Are there other ways to structure ring-expansions that would be better for city spacing, for game balance, or for the AI?

If anyone is interested, I can post a chart for alternate ring spacings...

It seems to me that this is something that was done in the original release, for reasons known only to the designers (if it was intentional at all), but which, despite 5 years of modding, has never been addressed. That's probably because the effect - while quite large - occurs in a way that makes it very hard to detect. I only found it because I was trying to determine optimal city spacings based on tile utilisation. Makes wonder what other kinds of strangeness are buried in the code, still hiding.

[Maquiladora]

If anyone is interested, I can post a chart for alternate ring spacings...

Of course.

What you suggested earlier, because cities begin to overlap mostly on the 3rd ring, is weak but perhaps the reason, so whether they overlap on the 4th and after is not an issue, so they only addressed the 3rd ring. Plus all other civ games have 2 rings, it wouldnt surprise me if their ideas for expanding rings werent completed or not playtested enough.

Ill always play for only 2 rings on the unmodded gigantic map, 3 or more rings and you waste early space and time moving, only small maps or a small island ill play for 1 ring space.

Bigger default maps and earlier ring expansions would help balance this, so larger city spacing is a must, but then you likely cant have equal workers to tiles in all rings, unless you significantly increase growth rates instead, but I think growth is already too fast.

[wheathin]

Here's a city radius chart. The numbers are the SquareRadius values in the CitySizeX.txt files. It's just x^2 + y^2 = r^2 (circle! - thanks Euclid!) The colors correspond to the Cradle city-size rings. The outer yellow ring is at R=30, which could be a new fifth ring if the existing fourth and fifth rings were combined.

[wheathin]

Here's that larger layout. The chart with this one includes intervals of 10 pop for each level - longer initial expansion, but very productive outer rings at the end. You'd want a very high limit in the final ring, like 100+ to get all your specialists.

Vision is the square of the VisionRadius value in the .txt file, to make it easier to see where the vision boundaries would be. In this example, the vision radius stays one level wider than the city radius.

[wheathin]

Here's one more, more like the original and Cradle sizes, but a little more regular in progression. This one might also benefit from more rapid expansion - perhaps only 8 workers per ring. That would make a size 40 city VERY large, with VERY rapid growth.

It would also let coastal cities grab more ocean tiles on water maps - having efficiencies up around 200-300% means that even if half or two-thirds of an outer-ring is blocked by other cities (friendly or hostile), that outer ring is still worth developing because it'll probably be worth more than specialists.

[Martin Gühmann]

Even with redesigning the cycles the problem of incomplete rings stays. And since I decided to redesign the resource collection and process of recourses, it would be a good idea to make it so that you need for a tile one worker or an certain amount of workers you can adjust in the CitySize.txts. So that you just need 6 workers for 6 tiles and not 8 to get the full utilization if your ring has only 6 tiles.

And this way I think the algorithms are also easier to program. Espeacilly all those estimation stuff. And of course for the player it seems more logical.

-Martin

[wheathin]

Originally posted by Martin Gühmann
Even with redesigning the cycles the problem of incomplete rings stays.

Agreed. My goal was two-fold:
1. To "fix" the city radii so that the resources per ring made sense (i.e. were more uniform), instead of progressing to a hugely powerful third ring and then woefully inadequate fourth and fifth rings.
2. To make the outer rings more productive, thus offsetting some of the impact of incomplete rings. A complete outer ring is always better than an incomplete outer ring. But a 50% incomplete outer ring does not have to be worse than specialists.

it would be a good idea to make it so that you need for a tile one worker or an certain amount of workers you can adjust in the CitySize.txts. So that you just need 6 workers for 6 tiles and not 8 to get the full utilization if your ring has only 6 tiles.
-Martin

If there's a way to adjust the city sizes (or workers-per-ring) so that each city (and each ring thereof?) has a different value equal to the number of tiles in the ring, that'd be great.

[Martin Gühmann]

Originally posted by wheathin
If there's a way to adjust the city sizes (or workers-per-ring) so that each city (and each ring thereof?) has a different value equal to the number of tiles in the ring, that'd be great.

Well, I am going to make it so that you need for each tile a fixed amount of workers, let's say one worker per tile. Maybe I will expose it to the CitySize.txts.

Then I indend to modify how the Resources are processed. I like to have something like this:

Collect all the resources from tiles.
Then process the maximal stuff from terrain.
- Applying wonder modifers.
- Applying building modifers.
- Applying government modifers.
- Applying slider modifers.
- Add resources generated by specialists. Maybe applying all the stuff that is now applyed.
- Add resources from trade routes.
- Remove resources lost due to special attacks (franchise, convertion).
- Remove resources lost due crime.
Apply the utilisation factor.

This way, I think, I can expose all the stuff for estimation in the best way, without the need of huge recallculations and other complicated algorithms.

Another think I would like to change is the collection of goods. It should be done so that they are collected if the good is in the zone of influence, instead in a ring with more thab zero workers.

-Martin