Print

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

Faster than 40d, no problems

42 (26.1%)

Faster than 40d, problems

72 (44.7%)

No slower than 40d, no problems

14 (8.7%)

No slower than 40d, problems

16 (9.9%)

Slower than 40d, no problems

2 (1.2%)

Slower than 40d, problems

3 (1.9%)

Doesn't work (please explain)

12 (7.5%)

Total Members Voted: 160

[Baughn]

Am I doing something wrong? I get 5000 fps in menus and when the game is paused, but when I unpause it only gets met 30 fps 

What do you get with 40d?
And what sort of processor do you have?

[Andir]

Am I doing something wrong? I get 5000 fps in menus and when the game is paused, but when I unpause it only gets met 30 fps 

That's perfectly natural.  When the game is paused, there is no movement or processing goin on so the graphics thread can render faster.  When Toady's code kicks in, it takes away from the amount of frames that can be rendered.

[Baughn]

Worse than that, actually - the fps counter the number of simulation cycles per second, not the number of graphical updates. The latter number is fixed by G_FPS, and would only start decreasing if the simulation FPS drops below G_FPS.

So when it says 5000 FPS while paused, that means your computer is doing nothing 5000 times a second.

[Eviltyphoon]

Am I doing something wrong? I get 5000 fps in menus and when the game is paused, but when I unpause it only gets met 30 fps 

What do you get with 40d?
And what sort of processor do you have?

40d also gives me 30 fps.
Proccesor: Intel(R) Core(TM)2 Duo CPU T7250 @ 2.00GHz (2 CPUs)

[Baughn]

Well.. heavily populated fort?

You should be able to get ~100fps or more on a newly embarked fort, assuming no horribly complex features are in effect. If you don't, I think something's wrong. (Force-enabling vsync in graphic card drivers can have this sort of effect, for example)

[Eviltyphoon]

Well.. heavily populated fort?

You should be able to get ~100fps or more on a newly embarked fort, assuming no horribly complex features are in effect. If you don't, I think something's wrong. (Force-enabling vsync in graphic card drivers can have this sort of effect, for example)

95 dwarves.

[Hamster Man]

The Windows file linked in the first page wouldn't extract. Just says "Error extracting". Bad link or is it just me?

[Baughn]

Just you. Try again.

[Tenebrous]

Sorry for not reading the other 65 pages of this thread before I reply, but I'm curious:

What form of pathfinding is Toady currently using? I notice that, in play, the Dwarves take shortest path, which makes me think that it is an A* derivative, full of pre-computative good/bad-ness.

Seeing as currently, the pathfinding performance degrades sharply in proportion with increases in creatures, perhaps a different approach might be best? I recently saw a lecture on a new approach to the old hill-climbing pathfinding with dynamic look-ahead values based on a pattern database to decide the best look-ahead for the proper heuristic.

Here's the original presentation I saw on D LRTA*:
http://sites.google.com/site/cosc419/schedule/2009-02-06-COSC419.pdf?attredirects=0
Here's a page I found on Google of someone else explaining it:
http://www.oddco.ca/zeroth/zblog/2008/07/29/dynamic-pathfindingcutting-edge-research/

This method provides almost perfect optimality and much lower in-game computation, at the cost of maintaining a (possibly costly) pattern database. I think, though, that even with DF's dynamic maps, that this database would not be too difficult to maintain.

Any thoughts?

[Mel_Vixen]

Iirc it was some kind of A* but this thread was and is about the rendering-engine of DF because thats the only part that various programs of toady have in common. Its almost the only thing too that was opened to the community.

Thanks anyway. I will have a good read tomorow.

[Hamster Man]

I'm still getting extract errors. After doing it one file at a time, the following files had "Error reading file":

Thumbs.db
dwarf_fortress.cmv
jpeg.dll
fmodex.dll
Dwarf Fortress.exe
optipng.exe
msvcr90.dll

Obviously I can't play without at least a few of these. I'll try completely wiping all versions of DF on my system before copying.

[Hamster Man]

Wiping out the other versions didn't help, so I tried downloading and installing on a different computer. Worked perfectly. So obviously some sort of error on this computer.

In other news - yes, it works great, except it crashes when minimized. (Run using windowed mode, 1280x1000)

[Footkerchief]

Sorry for not reading the other 65 pages of this thread before I reply, but I'm curious:

What form of pathfinding is Toady currently using? I notice that, in play, the Dwarves take shortest path, which makes me think that it is an A* derivative, full of pre-computative good/bad-ness.

Seeing as currently, the pathfinding performance degrades sharply in proportion with increases in creatures, perhaps a different approach might be best? I recently saw a lecture on a new approach to the old hill-climbing pathfinding with dynamic look-ahead values based on a pattern database to decide the best look-ahead for the proper heuristic.

Here's the original presentation I saw on D LRTA*:
http://sites.google.com/site/cosc419/schedule/2009-02-06-COSC419.pdf?attredirects=0
Here's a page I found on Google of someone else explaining it:
http://www.oddco.ca/zeroth/zblog/2008/07/29/dynamic-pathfindingcutting-edge-research/

This method provides almost perfect optimality and much lower in-game computation, at the cost of maintaining a (possibly costly) pattern database. I think, though, that even with DF's dynamic maps, that this database would not be too difficult to maintain.

Any thoughts?

As Heph said -- it's currently A*, but pathfinding discussion really doesn't belong in here.  You could repost it in this other pathfinding thread though -- the algorithm has some similarities, such as abstracting tiles into clusters.

[Tenebrous]

Oops, my bad. I misunderstood the purpose of this topic, methinks. Thanks for the link, though, Foot.

As for my feedback on the new output code, I don't have any: It worked fine for me, no problems. No noticeable speed change, but I always run at my cap anyways.

[praguepride]

Toady explained it in this interview:

http://www.gamasutra.com/view/feature/3549/interview_the_making_of_dwarf_.php?page=8