Bay 12 Games Forum

Please login or register.

Login with username, password and session length
Advanced search  
Pages: 1 [2]

Author Topic: CPU vs Bird problem  (Read 3143 times)

Robsoie

  • Bay Watcher
  • Urist McAngry
    • View Profile
Re: CPU vs Bird problem
« Reply #15 on: August 03, 2020, 12:27:23 pm »

A good idea to avoid "flying fps death" is to use something like Notepad++ batch replacer function and replace
Code: [Select]
FLIERby nothing  (just an empty space) for all files contained in \yourDF\raw\objects\  and  \yourDF\data\saves\regionX\raw\objects\

effectively removing the token from any creatures, preventing anything that could be flying to actually fly, that should be saving a lot of your CPU so DF can fps death somewhere else.
Logged

Leonidas

  • Bay Watcher
    • View Profile
Re: CPU vs Bird problem
« Reply #16 on: August 03, 2020, 02:03:03 pm »

A good idea to avoid "flying fps death" is to use something like Notepad++ batch replacer function and replace
Code: [Select]
FLIERby nothing  (just an empty space) for all files contained in \yourDF\raw\objects\  and  \yourDF\data\saves\regionX\raw\objects\

effectively removing the token from any creatures, preventing anything that could be flying to actually fly, that should be saving a lot of your CPU so DF can fps death somewhere else.
Editing the raws seems a bit extreme. You could just use DFHack's exterminate on the flyers as they present a problem.
Logged

Robsoie

  • Bay Watcher
  • Urist McAngry
    • View Profile
Re: CPU vs Bird problem
« Reply #17 on: August 03, 2020, 04:42:40 pm »

nothing is too extreme when they are flying around your fortress.
Logged

anewaname

  • Bay Watcher
  • The mattock... My choice for problem solving.
    • View Profile
Re: CPU vs Bird problem
« Reply #18 on: August 03, 2020, 10:39:13 pm »

There must be something else involved as well, as others (including me) routinely have birds appearing without any visible issues.

In adventure mode, I'll sometimes start shooting at bird or climbers to eliminate the FPS hit. IIRC, sometimes I would kill a single creature and the FPS would shoot back up, even though there were other creatures of the same type still on the screen.
Why would one of a group of creatures cause the problem... maybe only the first or last of the group is causing it?
Logged
Quote from: dragdeler
There is something to be said about, if the stakes are as high, maybe reconsider your certitudes. One has to be aggressively allistic to feel entitled to be able to trust. But it won't happen to me, my bit doesn't count etc etc... Just saying, after my recent experiences I couldn't trust the public if I wanted to. People got their risk assessment neurons rotten and replaced with game theory. Folks walk around like fat turkeys taunting the world to slaughter them.

Robsoie

  • Bay Watcher
  • Urist McAngry
    • View Profile
Re: CPU vs Bird problem
« Reply #19 on: August 04, 2020, 06:05:59 am »

Maybe there's a creature from a group that is somehow stuck and spam failed pathfinding code insanely.
I remember it happened in fortress mode when there was a dwarf stuck on a tree or a hole, the fps was taking a hit, and if said dwarf was suddenly having a "mood" it was fps death.
Logged

Leonidas

  • Bay Watcher
    • View Profile
Re: CPU vs Bird problem
« Reply #20 on: August 04, 2020, 09:52:30 am »

Why would one of a group of creatures cause the problem... maybe only the first or last of the group is causing it?
I dunno, but it's a testable hypothesis. The next time you've got a swarm of FPS-killing birds, use exterminate to kill them one at a time. If one of them is causing the problem, you're likely to see the FPS jump up before you run out of birds.

If you want to take it a step further, you could save the game first. Then you could figure out which bird(s) are causing the problem, upload the save to DFFD, and link to it in the bug tracker.
Logged

Jundavr

  • Bay Watcher
    • View Profile
Re: CPU vs Bird problem
« Reply #21 on: August 05, 2020, 07:53:57 pm »

Loose cluster fliers are the worst of them all.

When large clusters of creatures show up, the animals normally follow each other in a line kind of way, which spares your FPS quite nicely. In these cases, only the first of the group pathfinds. The other individuals will move a bit randomly sometimes, but nothing too fancy.

However, some creatures have a token that causes each individual in a cluster to move independently, and that sucks. A pack of monkeys (climber pathfinding) can max the game's resource usage and maul your FPS into oblivion. If you have sealed off every path to your fort, kea will cause the same problem: they will fly randomly, each using their own pathfinding, instead of locking on to your valuable wheelbarrows and pathfinding to them ONCE.

Funny thing I noticed with Dralthas and Elk birds: when the first creature of an ordered cluster gets captured in a cage trap, all the others stop pathfinding altogether and instead keep wandering around where the first was captured, usually leading to them being captured as well if there are more traps around.
Logged
I was out in the rain. So exasperating!

PatrikLundell

  • Bay Watcher
    • View Profile
Re: CPU vs Bird problem
« Reply #22 on: August 06, 2020, 02:36:50 am »

It's not uncommon for animals to refuse to leave when one of their number has been caught. In the past I've had fliers that just hovered in the air when one was caught, until I shot them down with ballista bolts, although they could move a few tiles if the structure built to host the ballista was too close for comfort.
Logged

Leonidas

  • Bay Watcher
    • View Profile
Re: CPU vs Bird problem
« Reply #23 on: August 06, 2020, 07:41:53 am »

It's not uncommon for animals to refuse to leave when one of their number has been caught. In the past I've had fliers that just hovered in the air when one was caught, until I shot them down with ballista bolts, although they could move a few tiles if the structure built to host the ballista was too close for comfort.
You use ballistas? I thought those were completely bugged to the point of uselessness.
Logged

PatrikLundell

  • Bay Watcher
    • View Profile
Re: CPU vs Bird problem
« Reply #24 on: August 06, 2020, 08:02:12 am »

They're excellent for removal of cavern trees in water, as well as bugged critters locked in place (when conquering the underworld I use a ballista to get rid of the last two flame based buggers, as they refused to move). For removal of hostile critters you should obviously build it out of range.
I'm not aware of them being bugged, but their range of useful applications is severely limited (the can fire down a corridor as well, but I was disappointed at the puny effect it had on targeted critters. When the critters are stationary you can just keep firing, however).
Logged

Leonidas

  • Bay Watcher
    • View Profile
Re: CPU vs Bird problem
« Reply #25 on: August 06, 2020, 08:25:00 am »

I was thinking of bug 818, where ballista bolts are said to bounce off harmlessly most of the time. But those reports are old. Maybe it has changed. For all the time I've played this game, I've never seriously tried to use siege engines.
Logged

Schmaven

  • Bay Watcher
  • Abiding
    • View Profile
Re: CPU vs Bird problem
« Reply #26 on: August 07, 2020, 06:28:25 pm »

I've had 2 instances of a ballista bolt severing legs and tearing apart organs, so I'd say the iron tipped ballista bolts are fairly lethal at least.
Logged

waerth

  • Bay Watcher
    • View Profile
Re: CPU vs Bird problem
« Reply #27 on: August 14, 2020, 05:24:36 pm »

So I read around on other topics and then found a nice tidbit about to many z-levels above ground. So I checked my world gen for that world and yep .... I had changed the 15 to 50.

So I genned a new world and have been playing 5 years on it. No real FPS caves anymore since. Just once or so.

:p

Logged
Pages: 1 [2]