Mantis Bug Tracker

View Issue Details Jump to Notes ] Issue History ] Print ]
IDProjectCategoryDate SubmittedLast Update
0002338AI War[All Projects] Bug - OtherJan 9, 2011 1:32 pmFeb 14, 2013 12:15 pm
ReporterMoonshine Fox 
Assigned Tox4000 
Severitymajor 
StatusassignedResolutionreopened 
Product Version4.064 
Fixed in Version4.065 
Summary0002338: "Bouncy" ships refuse to go through wormholes.
DescriptionOccasionally, my (or my friend's) fleet will pass a wormhole by about 5 inches, return back to the wormhole, pass it again, and keep bouncing back and forth instead of going through.

Most of the time, some of the ships will fly past the wormhole and the rest go through and the other ships will go through on their "return bounce".

Timmy will provide a savegame.
TagsNo tags attached.
Internal Weight
Attached Files? file icon For Keith and Chris.sav [^] (194,831 bytes) Jan 9, 2011 1:34 pm
? file icon pathfinding_bug.sav [^] (190,895 bytes) Sep 5, 2012 3:39 pm

- Relationships
related to 0002315resolvedx4000 Wormhole Pathfinding 
related to 0010357new Collision ejection overcompensates frequently 

-  Notes
(0008082)
kingisaaclinksr (reporter)
Jan 9, 2011 1:35 pm

Yeah, its pretty bad. I included a save with a planet that currently has a bunch of ships perma-bouncing off a wormhole.

King
(0008083)
Moonshine Fox (reporter)
Jan 9, 2011 1:36 pm

The planet in question is "CD-46 11540"
(0008093)
TechSY730 (reporter)
Jan 9, 2011 3:58 pm

I think one of the biggest ways to induce this is by auto-kiting. Ships that try to auto-kite seem to "miss" wormholes a lot; they have to go back and forth one or two times to finally go through. I've gotten to the point where I don't use auto-kiting for the offense, only for defense.
(0008094)
Moonshine Fox (reporter)
Jan 9, 2011 3:59 pm

Funnily enough, neither of us (afaik) was using Auto-kiting at this time.
(0008095)
kingisaaclinksr (reporter)
Jan 9, 2011 3:59 pm

Except neither of us were using Auto-kiting at all during the game so that's not a valid possibility.

King
(0008099)
TechSY730 (reporter)
Jan 9, 2011 4:03 pm

Well, there goes that theory. With the removal of the 150 speed cap (which was silently 300 for the players), maybe ships are starting to go too fast for the engine to handle. :D
(0008112)
Moonshine Fox (reporter)
Jan 9, 2011 5:30 pm

Good idea. Except that we're currently caught in 0002304, so all our ships are moving at one fourth the intended speed (half of what we should have on home planets due to us not getting the x2 bonus, and half again for the AI planet penalty which is applied everywhere).
(0008113)
TechSY730 (reporter)
Jan 9, 2011 5:45 pm

True, I just saw my missile frigates do this a little bit. It was just one miss instead of the repeated misses you had, but still it is annoying.
(0008114)
Moonshine Fox (reporter)
Jan 9, 2011 5:52 pm

Yeah, if the ships were moving at full speed, it would probably be barely noticeable except on hostile planets, but now it's quite frustrating. Especially if you hit the jackpot and suddenly have 400 ships perma-bouncing on some backwater planet ;).
(0008129)
KPJZKC (reporter)
Jan 10, 2011 6:53 am

I've had this happen in my current game too, bye bye expensive sentinels...
(0008185)
x4000 (administrator)
Jan 10, 2011 10:18 pm

Keith, I've got this one, FYI.
(0008190)
x4000 (administrator)
Jan 10, 2011 10:33 pm

Actually it was unrelated to auto-kiting, but a good guess.

* Ships no longer decelerate to enter into wormholes. They now just fly right in at whatever speed they are at when they arrive. Additionally, when ships are closing in on a wormhole, they now do their movement-recalculation checks a lot more frequently, which uses a bit more CPU, but which also will prevent them from overshooting their goal as they have been doing at times in the last version or two.
(0028237)
Moonshine Fox (reporter)
Sep 5, 2012 1:26 pm
edited on: Sep 5, 2012 1:35 pm

I've started to see this issue a lot in recent games. Has something changed with the way wormhole pathfinding logic is handled? It appears to be mainly missile frigates doing it, and it's fairly reproducible in a savegame I have.

I'm playing at "Insane" profile on a Core i7-3820 @ 4,6 GHz. Not sure if that helps.

(0028241)
TechSY730 (reporter)
Sep 5, 2012 1:39 pm

Could you reopen the issue then?
(0028242)
Moonshine Fox (reporter)
Sep 5, 2012 1:53 pm

I would like to do some more testing before I do that. I'll be home in about 1,5 hours to test this more thoroughly and provide a save. If I can reproduce it enough, I'll reopen it. I mostly posted to see if anyone else had seen this lately.

It could, theoretically, just be an issue with my game.
(0028244)
Moonshine Fox (reporter)
Sep 5, 2012 3:38 pm
edited on: Sep 5, 2012 3:39 pm

Yep, I've now tested this. In the attached savegame (pathfinding_bug), take control group number one (1) (on Choujo), and CTRL+right click send them through to "Home". They'll instantly travel to the rally post, and epic bouncing will take place there.

(0030777)
Moonshine Fox (reporter)
Feb 14, 2013 12:12 pm

This one is tagged as "Feedback". I've done the testing I needed to do and provided a scenario. I don't know what else you need. Let me know!
(0030779)
TechSY730 (reporter)
Feb 14, 2013 12:15 pm
edited on: Feb 14, 2013 12:17 pm

Well, you might want to rerun the test. Lot's of movement, pathfinding, collision ejection, etc has been changed since 4.065. The bug might not manifest under the same circumstances anymore, or possibly even not be there anymore.

EDIT: Humorously, commenting on an issue marked feedback (if you are not the asignee of the issue) puts the status back to assigned. So basically, by pointing out that this issue is marked feedback, you knocked it out of feedback status. :D


Issue Community Support
Only registered users can voice their support. Click here to register, or here to log in.
Supporters: No one explicitly supports this issue yet.
Opponents: No one explicitly opposes this issue yet.

- Issue History
Date Modified Username Field Change
Jan 9, 2011 1:32 pm Moonshine Fox New Issue
Jan 9, 2011 1:34 pm kingisaaclinksr File Added: For Keith and Chris.sav
Jan 9, 2011 1:35 pm kingisaaclinksr Note Added: 0008082
Jan 9, 2011 1:36 pm Moonshine Fox Note Added: 0008083
Jan 9, 2011 3:58 pm TechSY730 Note Added: 0008093
Jan 9, 2011 3:59 pm Moonshine Fox Note Added: 0008094
Jan 9, 2011 3:59 pm kingisaaclinksr Note Added: 0008095
Jan 9, 2011 4:03 pm TechSY730 Note Added: 0008099
Jan 9, 2011 5:30 pm Moonshine Fox Note Added: 0008112
Jan 9, 2011 5:45 pm TechSY730 Note Added: 0008113
Jan 9, 2011 5:52 pm Moonshine Fox Note Added: 0008114
Jan 9, 2011 11:02 pm lyravega Relationship added related to 0002315
Jan 10, 2011 6:53 am KPJZKC Note Added: 0008129
Jan 10, 2011 10:18 pm x4000 Note Added: 0008185
Jan 10, 2011 10:33 pm x4000 Note Added: 0008190
Jan 10, 2011 10:33 pm x4000 Status new => resolved
Jan 10, 2011 10:33 pm x4000 Fixed in Version => 4.065
Jan 10, 2011 10:33 pm x4000 Resolution open => fixed
Jan 10, 2011 10:33 pm x4000 Assigned To => x4000
Sep 5, 2012 1:26 pm Moonshine Fox Note Added: 0028237
Sep 5, 2012 1:35 pm Moonshine Fox Note Edited: 0028237 View Revisions
Sep 5, 2012 1:39 pm TechSY730 Note Added: 0028241
Sep 5, 2012 1:53 pm Moonshine Fox Note Added: 0028242
Sep 5, 2012 3:38 pm Moonshine Fox Note Added: 0028244
Sep 5, 2012 3:38 pm Moonshine Fox Status resolved => feedback
Sep 5, 2012 3:38 pm Moonshine Fox Resolution fixed => reopened
Sep 5, 2012 3:39 pm Moonshine Fox File Added: pathfinding_bug.sav
Sep 5, 2012 3:39 pm Moonshine Fox Note Edited: 0028244 View Revisions
Feb 14, 2013 12:07 pm TechSY730 Relationship added related to 0010357
Feb 14, 2013 12:12 pm Moonshine Fox Note Added: 0030777
Feb 14, 2013 12:12 pm Moonshine Fox Status feedback => assigned
Feb 14, 2013 12:15 pm TechSY730 Note Added: 0030779
Feb 14, 2013 12:17 pm TechSY730 Note Edited: 0030779 View Revisions


Copyright © 2000 - 2011 MantisBT Group
Powered by Mantis Bugtracker