Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[TheCakeDwarf1326]

So I was playing Masterwork as the Humans and I embarked on a volcano with a mini mountain, so of course since I cant build IN the mountain (because of humans and cave adaption making them sick) I decided to blow the mountain, so I mined out the entire floor I wanted the base to be at and dropped the mountain like a brick shithouse. Now my fps is dying and I was wondering if there is anyway to fix this because before my fps was staying steadily around 100 now its down around 10 to 30 mostly when I look at the mountain. Help?

[NullForceOmega]

You have to realize that you just altered a HELL of a lot of tile data with that move, the engine doesn't like that, and pathing is going to die horribly due to it.  It is also unlikely that you or anyone else can fix it.

[§k]

Does the whole mound collapse? The cave-in should be instantaneous. After that, fps will gradually recover.

Actually, 10 to 30 is normal fps in my fort.

[NullForceOmega]

No, when you make massive changes to the surface map via cave-in, it never recovers, I've tried it at least a dozen times in different versions.  What it will do is start randomly spiking and dipping as it attempts to load stored tile data, then has to recalculate pathing due to the ground being several (even just one) Z lower.

[Draba]

Does manual mining (as in stripping the top levels 1z at a time) permanently murder fps, or is it just cave-ins?
In my current fort I planned on leveling the 10-15 levels on the top of the mountains to make it simpler to find invaders
+get the amethyst/zircon clusters within, but I don't want to kill my fps 10 years before it's due.

[utunnels]

I remove a hill on a 1x1 map, nothing happened.

[NullForceOmega]

It's the cave in, from observations in the past it appears that tile data can get really messed up when you use cave-ins to alter the map, especially on the surface.  It does not appear that using channeling designations has an adverse effect on the surface unless you leave a lot of random ramps around.  But don't use large-scale channeling underground, the pathing calculation hates that hard.

[Utherix]

May you could fix the pathing by placing strategic high-cost path zones in areas where mountain tiles used to be?

[NullForceOmega]

That isn't the problem, the problem is tile data.  When you generate the site  (the player loads the [A x B] 48x48 tile squares that are represented by the region tiles in adventure or fortress mode, until then it is all abstract), it appears to generate a basic pathgrid on the surface (and the caverns) that allows visitors to the map navigate, by altering the surface map (specifically by cave-in, I don't know why but DF hates surface cave-ins) you also force the engine to re-calculate the pathing for the entire surface every time it needs to check for movement on that level.

[Raphite1]

That isn't the problem, the problem is tile data.  When you generate the site  (the player loads the [A x B] 48x48 tile squares that are represented by the region tiles in adventure or fortress mode, until then it is all abstract), it appears to generate a basic pathgrid on the surface (and the caverns) that allows visitors to the map navigate, by altering the surface map (specifically by cave-in, I don't know why but DF hates surface cave-ins) you also force the engine to re-calculate the pathing for the entire surface every time it needs to check for movement on that level.

This makes it sound like even the smallest surface cave-in would have this effect, but I'm sure I've used cave-ins to plug aquifers without a noticeable fps drop (it's been awhile though... a couple versions ago).

Strange that cave-ins would cause this and not things like bridging rivers, building walls, or removing trees.

[NullForceOmega]

Even the smallest cave-in does have an effect, but it will likely be unnoticeable, the more tiles you alter through cave-in (this might have something to do with preserving the surface data i.e. grass cover etc.) the larger the effect as it requires more pathing checks.

[Verjigorm]

Guess this means the end of aquifer plugging for me.   I always liked the cave-in method for it's aesthetics, and similarity to how you build bridges with caissons to get underwater.   

[PatrikLundell]

I think stopping cave in usage for aquifer plugging and cavern securing is an over reaction. Neither of those are large scale actions, and neither affects the surface. In the aquifer case, there cannot be any pregenerated pathing that's screwed up, since the aquifers are buried in rock initially.

[utunnels]

Maybe the OP can post a save so we can test if there's a fix?

[NullForceOmega]

I'd agree with Patrik here, it isn't really necessary to stop using cave-ins altogether, you just have to be smart about it and try not to overdo things.