Single Player bugs and crashes v35 plus (SVN) - After the 18th of August 2014

I don't see the need for this argument. Nothing's wrong with Eternity as a setting imo. It introduces its own set of challenges but it's never going to be what the mod team balances the game to consider so there are going to be times when it doesn't handle itself all that well. This, however, can show us where other imbalances lie that are more subtly affecting other gamespeeds as well. So the answer to this is simple, which is something I've been saying for over a year now - earlier law enforcement units are needed. Not hard. And fully planned out for implementation. Just can't be done overnight is all.

The fact that the property system helps the game become a better real world simulation is exactly why it's a great system. The game has always been a game model of real world and it should be made to be a more accurate model of such as we go. That's pretty much the whole goal here. Sure there are simpler ways to address these properties but there's a lot of genius in the way it's already established.

Furthermore, I would prefer to honor one of the most valuable modders we've had on the team, AIAndy, with a full embracing of the mechanism. I know for any of us if someone just came in and undid our most core contribution it would pretty much be spit in the eye for all your efforts.

But even as a gamer, it's nice that not everything is so measurable and controllable in C2C as it has always been in Civ. Not that I find controlling properties too difficult but it does make for a factor that's more difficult to master and that's really nice considering how easy a simple game becomes boring.

In what way is it scaled? If you have excess gold output you still have longer to hoard it. Modifiers and base values on building commerces have no scaling factor based on gamespeed so far as I'm aware.

 

Lets use my adjustments to the eternity gamespeed as an example:
Pretty much everything (not property changes) takes 10 times longer at eternity than normal. this means you get 10 times as many turns at hoarding and that you probably have 10 times more money in the treasury, but since buying/upgrading buildings/units cost 10 times more it adds up to that the treasury effective value is the same as on normal gamespeed.

 

So then, ultimately are we saying that the autobuilding thresholds for properties should scale with gamespeed then? That probably wouldn't be too hard to do actually.

 

No, properties doesn't reach higher values on longer gamespeed.

But going from -100 education to 200 education in 5-6 turns is okay for normal gamespeed but not on eternity, it should then take 50-60 turns.

This would be the most important parameter for that:

Spoiler :

Code:

		     <PropertyManipulators>
        		<PropertySource>
        			<PropertySourceType>PROPERTYSOURCE_DECAY</PropertySourceType>
          			<PropertyType>PROPERTY_X</PropertyType>
          			<GameObjectType>GAMEOBJECT_CITY</GameObjectType>
          			<iPercent>4</iPercent>
        		</PropertySource>

If the value was 10 at normal it would be 1 on eternity; at least the way gamespeed are scaled in my modmod.

 

I wonder if it's possible to scale that by gamespeed in xml...

 

Pit, to stop the doubters, post a save game where you have reached over 14,000 turns.

 

I don't really understand the logic behind the way gamespeed is scaled in the SVN right now:

Spoiler :

Parameter - Scaling Factor (normal→eternity)

Food needed for growth - x10.66
Hammer cost of units - x6.25
Hammer cost of buildings - x14.74
Hammers for WW & NW - x20
Beakers needed for tech - x14.71
Turns for building improvements - x8.59
Turns for an improvement to upgrade - x12.5
Turns for getting a great person - x12.5
Culture needed for expansion - 12.5
Length of anarchy - x8
Barb. spawning frequency - x20
Other factors - 14.71

I skipped a lot of them here.

The inconsistency is between all gamespeeds, not only normal→eternity

 

I think, as I stated in this previous post, that it would be best to see how this xml tag is setup in the dll. I believe AIAndy included a lot of new gamespeed scaling into this tag.

 

I agree that there's a lot of odd inconsistencies in the scaling on gamespeeds at the moment. Been hoping someone would align it rationally once more. Keep in mind though that most playing longer gamespeeds would want a little more time to play with their units in the field so if there's a discrep in unit production that would be the purpose of it.

EDIT: RE your last comment... I'll have to take a look.

 

I remember when it was discussed to make it like that in C2C, but I always thought the logic was flawed. Even if it takes 10 times longer to build the unit on eternity you would still have 10 times more turns with it out in the field than on normal.

 

Yeah, but classically unit production speed was never scaled to perfectly match the ratio of added time by gamespeeds and there's a reason for it. Longer gamespeeds aren't intended to only leave you with the same amount of time, just divided into more turns. Should be close, yes, but not on all factors exactly the same.

I'm not against a little realigning but take a look at how Vanilla handles it as I feel Firaxis did well to establish guidelines there.

 

I must have remembered that the opposite was discussed, and that I thought the logic was sound. lol

Anyway, vanilla quick, normal and epic gamespeed scaled perfectly to match the added turns with the exception of hurrycost & goldenage modifiers. The reason they scaled down unit cost for marathon was probably to make the game not seem boring due to the lack of content, many consecutive end-turns without player doing anything. This is not a real problem for C2C as there will almost always be choices to make even when everything is spaced out by a factor of 10.

We would need a gamespeed that is about 30 times longer than normal before that logic applies here.

 

Ok, I'd still suggest to go a touch lax on unit production and it also makes a lot of sense to do the same with improvement generation (work speed). Not perhaps as much as we currently have it but still those two points seem to be a good idea to consider the initial reasoning behind vanilla's Marathon speed. Last time I played a few rounds of Eternity it struck me as playing fairly similar to the original Marathon.

I used to always play Marathon so I suppose that's where I was getting the presumption from.

 

Exactly.

 

I think the generic property system AIAndy added is just over the edge for a strategy game and it's far from complete. There are easier ways to do things like flammability and the other properties. For me the sometimes really bad AI and long turn times are much more boring as playing C2C with simplified properties.

This has nothing to do with dishonouring AIAndy.

 

agreed, but if you know a better way, pls start a thread, and then lets go from there, thx . . .

Now pls stop writing here, as we can start working on the Official v36 to be release, thx ., , . SO

 

DH, need a new one of this and the other sticky's thx . . .SO

 

This is a crash that happens during end-of-turn, with no pop-up or error message. CivIV just shuts down, always at exactly the same moment. The save is right before it happens.

Gigantic map, snail speed, size matters, the current (release candidate) SVN, and I'm almost done with the Renaissance, but have had no serious performance problems up til now.

 

Official v 36 is ready for downloading, thx everyone for your patience . . .SO

 

This will take a little further testing of your actual save but apparently the crash is taking place where a property value is being set to 0 and thus being removed from the list. I've never really felt it should do this in the first place so I'm inclined to comment that portion out anyhow but I need to:
1) Ensure it does actually solve the crash point because this has happened numerous times safely on many games so I'm not entirely certain I have a full picture of what's causing the actual crash.
2) Ensure doing this doesn't actually cause further crashes.