Print

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

[Scruiser]

Man, that bugfix changelog's getting pretty tall.

I like how Toady's fixing all these smaller issues. They're quick changes most of the time, and fixing them really adds to the game's polish. I'd rather 10 small bugs be fixed than one major big bug (unless the latter bug is gamebreaking, of which not many are left).

I'm especially looking forward to elven diplomats coming back in the coming version.

Yeah the small things often stop much larger features from coming into play properly.  Example: Elven diplomat.  Now that the elves will start hating on me for cutting trees, I can finally get on board with all the anti-elf forum memes and start pointless wars.  I'm Glad Toady is taking the time to quash these

[Authority2]

The elf tree-cutting cap will have to be lowered now, since trees are bigger and less common. I think someone in another thread said something about the elves coming in and naming individual trees and forbidding the cutting of those, but I'm not sure of its veracity.

[Witty]

I too would like to applaud Toady for taking some time out to do some small polishing. I hope we have more days of bug tracker wanderings in the near future.

[Toady One]

2. Actually Ag solved that bug - I had nothing to do with it: http://www.bay12games.com/dwarves/mantisbt/view.php?id=5971

Ah, whoops.  Fixed that -- I wonder if that means I left you off another one.

[Elone]

Ooooh, waiting for a little more stability before I dive into this, but the climbing's gonna be real interesting if creatures get more reluctant to climb the higher their path is from the ground. I have read that the climb pathing costs more now. I'd imagine that the pathing gets more expensive (with each path-tile ahead of them) the higher a creature means to climb.

Do the AI-controlled climbing creatures consider the wall/cliff/whatever height for their increased climbing cost?

[thvaz]

The new stealth mechanics are integrated in some way in fortress mode?

[Dame de la Licorne]

Hiya,

<snip>  Nevermind, found the answer.

-Dame de la Licorne

[Shashu]

I also cleaned up some problems with the projectile code to move them over to a slightly more acceptable place, though there are doubtless some issues. Fixing a rounding error screwed up a lot of the other analysis, so I still had to wing it for now.

Could someone comment on the significance of this? How does archery compare now to DF2012? I heard there was some sort of wackiness in 40.01-40.03 but I'm not sure of the scope.

[Putnam]

Compared to DF2012, it should be far less overpowered. The exact bugs can be seen here:

Absurdly Powerful Bolts/Arrows
Bolts, crossbows and other ranged weapons have excessive SHOOT_FORCE values, leading to ranged combat issues
Rounding Error in Calculation of Projectile Velocities

[Mu.]

Can we expect any significant performance improvements from the next few updates? You mentioned that you fixed calendar lag, and that this should have some impact on fortress mode, which sounds good. The framerate in fortress mode quickly becomes unbearable, despite keeping worlds and region sizes as meager as possible. My current fortress has about 100 pop, is 3 years old, and has breached two cave systems. I have an average of 39 FPS. This is without much clutter: my corridors are currently free of any goblin bodyparts.

I remember similar performance issues during the first few 3D releases.

[therahedwig]

Can we expect any significant performance improvements from the next few updates? You mentioned that you fixed calendar lag, and that this should have some impact on fortress mode, which sounds good. The framerate in fortress mode quickly becomes unbearable, despite keeping worlds and region sizes as meager as possible. My current fortress has about 100 pop, is 3 years old, and has breached two cave systems. I have an average of 39 FPS. This is without much clutter: my corridors are currently free of any goblin bodyparts.

I remember similar performance issues during the first few 3D releases.

The big activated world feature of this release means that the Calender is running at all times during gameplay. The Calender being basically all AI-interactions outside of your direct game.
Any optimisations to the Calender is also an optimisation to fort mode. In fact, if I am not mistaken, fort mode's lag is precisely because of the Calender being unoptimised.

EDIT: On second thought, there may now be more pathing issues due to trees as well as birds and the new climb/jump mechanics.
It will never go fully back to previous levels, because the Calender is a huge part of Dwarf Fortress as a game, but it will definitely be optimised.

[dree12]

Can we expect any significant performance improvements from the next few updates? You mentioned that you fixed calendar lag, and that this should have some impact on fortress mode, which sounds good. The framerate in fortress mode quickly becomes unbearable, despite keeping worlds and region sizes as meager as possible. My current fortress has about 100 pop, is 3 years old, and has breached two cave systems. I have an average of 39 FPS. This is without much clutter: my corridors are currently free of any goblin bodyparts.

I remember similar performance issues during the first few 3D releases.

The big activated world feature of this release means that the Calender is running at all times during gameplay. The Calender being basically all AI-interactions outside of your direct game.
Any optimisations to the Calender is also an optimisation to fort mode. In fact, if I am not mistaken, fort mode's lag is precisely because of the Calender being unoptimised.

EDIT: On second thought, there may now be more pathing issues due to trees as well as birds and the new climb/jump mechanics.
It will never go fully back to previous levels, because the Calender is a huge part of Dwarf Fortress as a game, but it will definitely be optimised.

I think it makes sense for DF to get progressively slower over time, since hardware is a whole order of magnitude faster than when DF was first released.

Of course, optimizations are nice, especially since they allow more depth in the simulation later on.

[thvaz]

For the next version Toady fixed a bug caused by fat dwarves that if I understood correctly could bring an improvement of 10% fps to large fortresses. I think there are plenty of opportunities to improve optimization, even to leves better than the last version. However, too much optimization now is kinda useless, because the game is in development. "Good enough for now" is what Toady aims to, I think.

[Footkerchief]

Can we expect any significant performance improvements from the next few updates? You mentioned that you fixed calendar lag, and that this should have some impact on fortress mode, which sounds good. The framerate in fortress mode quickly becomes unbearable, despite keeping worlds and region sizes as meager as possible. My current fortress has about 100 pop, is 3 years old, and has breached two cave systems. I have an average of 39 FPS. This is without much clutter: my corridors are currently free of any goblin bodyparts.

I remember similar performance issues during the first few 3D releases.

There's a good chance this slowness is caused by bugs.  Some reports:
http://www.bay12games.com/dwarves/mantisbt/view.php?id=7536
http://www.bay12games.com/dwarves/mantisbt/view.php?id=6913
http://www.bay12games.com/dwarves/mantisbt/view.php?id=7119
http://www.bay12games.com/dwarves/mantisbt/view.php?id=6015
http://www.bay12games.com/dwarves/mantisbt/view.php?id=7284

[Deathworks]

Hello!

Are there plans to go into addressing the limitations and crash bugs in world gen during the current short term debugging?

Personally, I like to gen medium worlds with a history of 500 years (actually, I would love to run the default 1050 years, but this is currently impossible), and even with less than 400 years, the game often crashes, either while creating the world or finalizing sites or other things. So, I am curious whether I still have to remain careful about this.

Yours,
Deathworks