Solver's unofficial BtS 3.17 patch

[MusX]

Rivers running through deserts will result in floodplains

every desert near river? or just those near somebodys starting location?

 

[oyzar]

Rivers running through deserts will result in floodplains

every desert near river? or just those near somebodys starting location?

This sounds like fixing something that is not broken.. It is supposed to be riverside deserts in the normal game.. I have seen floodplains with 3.17 before so it is obviously not the case of readding floodplains, but instead removing natural riverside deserts...

 

[Balderstrom]

@ Oyzar, Bhruic's - as far as I know, only checked for Desert/River FloodPlains for start locations. Other areas of the map (non-start locations) might be a floodplains Desert/river or not - dependant on the map script.

Hopefully that's what has been implemented in this case.

 

[grumbler]

The more civs the game allows, the shorter civ and leader names have to be.

Just did a quick test:

1. changed MAX_CIV_PLAYERS to 50
2. compiled DLL
3. started custom game with 50 players
4. saved and loaded multiple times

Result: All leader names appeared unshortened and no errors occured.

Could it be that this "the more leaders the shorter the names have to be"-thing was probably true for an earlier version of Civ, but was fixed by Firaxis in one of the patches?

 

[Refar]

The more civs one is certainly usefull for many modders (Tho 50 ?! Besides Civ gold i have trouble finding a mod thad does actually have 50 civs...). And it's very forthcoming from solver to include it. But it definitely does cross the line between patching and modding, which might urge some "mainstream" users to forgoe the patch, which would be a pity.

I think the main branch of the patch should stay with the standard 18 civ dll.

 

[MusX]

@ Oyzar, Bhruic's - as far as I know, only checked for Desert/River FloodPlains for start locations. Other areas of the map (non-start locations) might be a floodplains Desert/river or not - dependant on the map script.

I hope it works that way

@grumbler - extend max civilization seems to be great feature, not just a bug fix. Any way would be great if Solver could change default 18 max civs to for example 36!

 

[Arian]

The more civs one is certainly usefull for many modders (Tho 50 ?! Besides Civ gold i have trouble finding a mod thad does actually have 50 civs...). And it's very forthcoming from solver to include it. But it definitely does cross the line between patching and modding, which might urge some "mainstream" users to forgoe the patch, which would be a pity.

I think the main branch of the patch should stay with the standard 18 civ dll.

Consider it icing on the cake

 

[Psyringe]

Just did a quick test:
Result: All leader names appeared unshortened and no errors occured.

Could it be that this "the more leaders the shorter the names have to be"-thing was probably true for an earlier version of Civ, but was fixed by Firaxis in one of the patches?

You tested regular random map games. In such games, there never was a limit to the leader names. The limit only affected worldbuilder saves. If you want to test whether it's still there, you need to build a scenario with lots of players and try to load it, not start a random map game.

Also, I remember somebody reporting a slight performance hit with a 34+ players dll, even when playing with just 18 civs. But that was long ago, it might have been changed, or I may misremember.

Another problem was that MAFs occurred more frequently when using bigger maps and/or a large number of players. Limiting the game to 18 civs and huge maps was Firaxis' way of playing it safe.

Personally, I *love* games with lots of civs. I didn't even start a new game with 3.17 until the 50 players dll was available. So I'm all for maintaining such a version, but further testing needs to be done before we can claim that it could just as well replace the regular dll.

 

[Balderstrom]

Well again I ask, (which was not answered?)
If there isn't a performance hit when playing w/ 18 civs & 50Civ DLL vs normal DLL
If CIV/Leader names aren't truncated -
And if the modified 50Civ DLL allows you to play from 2 - 50, and isn't locked to 50.
Why not have the DLL set for more civ's than the base 18... and have that be the standard "unofficial patch"?

It would seem like a win/win? So ... am I missing something?

 

[Wodan]

Solver, how about fixing the description of WoodsII? It bugs me every time I see it. It's not "double movement". It's "+1 movement".

Wodan

 

[Solver]

Well, the standard is 18 civs. There is a slight performance hit if you increase that number - not sure how much of a hit for 50 civs, for example, though the hit may be negligible on newer machines. Anyway, the unofficial patch is staying at 18 civs. That's why the 50 civ DLL isn't in the first post, again, I only attached it for 0.19 because I got a request to do so.

Wodan, I'll look at it when I get to fixing text errors.

 

[grumbler]

You tested regular random map games.

Well, I looked but couldn't find an exact description of the problem so I had to start somewhere.

In such games, there never was a limit to the leader names. The limit only affected worldbuilder saves. If you want to test whether it's still there, you need to build a scenario with lots of players and try to load it, not start a random map game.

So, I now

1. took my 50-civ testgame
2. saved it from WB
3. started a new game (custom scenario option), using the WBS from the previous step

Result: Still no problem (i.e. no crash, all names intact).

So, am I missing something?

Another problem was that MAFs occurred more frequently when using bigger maps and/or a large number of players.

I mainly play standard games (as in unmodded) and FFH2 on huge maps which is probably the reason I'm not seeing these very often. (Maybe once or twice since 3.13 came out.)

Personally, I *love* games with lots of civs. I didn't even start a new game with 3.17 until the 50 players dll was available. So I'm all for maintaining such a version, but further testing needs to be done before we can claim that it could just as well replace the regular dll.

I agree. Keep in mind, my quick test was really only intended to shed some light on the naming issue and was more born out of curiosity than a desire to do a thorough examination of the suitability of the 50-civ as the default variant .

 

[Psyringe]

So, I now

1. took my 50-civ testgame
2. saved it from WB
3. started a new game (custom scenario option), using the WBS from the previous step

Result: Still no problem (i.e. no crash, all names intact).

So, am I missing something?

Apparently the bug has been fixed. If you need further information to make it sure, here's a thread that explains the problem. Check posts #17 and #21 specifically.

But as far as I understand it, the test you performed above couldn't have worked unless this bug has been fixed, hence I'm concluding that this issue indeed doesn't exist anymore.

 

[jkp1187]

Solver: After playing around a little, I can say that I think you have the capitulation threshold for AI just right.

 

[Solver]

I haven't changed the capitulation threshold, for the record, it's still as 3.17, which I still find a bit too low.

 

[Caboose]

I agree. Gandhi is ridiculous with his capitulation, for example.

 

[Wolfshanze]

Since capitulation changed from v3.13 to v3.17 perhaps a happy medium could be done in the Solver patch? It seems a lot of folks are complaining that capitulation is too easy in v3.17 now... if it was "too hard" to get in v3.13, surely there must be a "medium" setting between the two that could/should be implemented in the Solver patch... harder to capitulate in Solver then in v3.17, but easier then in v3.13.

 

[Willem]

This sounds like fixing something that is not broken.. It is supposed to be riverside deserts in the normal game.. I have seen floodplains with 3.17 before so it is obviously not the case of readding floodplains, but instead removing natural riverside deserts...

It was broken, and it has nothing to do with removing Desert tiles. This fix involves the placing of a river when the game is improving a starting location. Every start has to have a source of fresh water and if there is none originally, then the game will create either a River or a Lake. The problem that happens when it creates a River is that Flooplain tiles are only placed before this stage and not after. So any of these brand new rivers that pass through Desert terrain will not be given that Feature. The fix you mention tries to address this by adding Floodplains to any created river that passes through Desert after they're added to the map.

 

[Ninja2]

The 50-civ DLL was a one-off thing I did upon request. I don't really want to maintain both, particularly because changing the number of players requires a full recompile of the DLL, which takes a good amount of time.

Six minutes is a good amount of time? I understand why you want to keep it at 18, though.

 

[Solver]

Well, I'm still having bad memories from the time when it took me about 50 minutes to fully recompile the DLL on my PC