Getting stuck in another ship/station.

Find any bugs in Vega Strike? See if someone has already found it, or report them here!
Post Reply
parkel
Hunter
Hunter
Posts: 73
Joined: Fri Jun 12, 2009 2:22 pm

Getting stuck in another ship/station.

Post by parkel »

The problem: your ship will sometimes get stuck in another ship or station.

VS version: 0.5.0 stable.

How it happens: erm... you crash. and you get stuck inside the station (any sort - MiningBase, Fighter Barracks or Refineries etc) and as well as ship, be it small or big, docking or dogfighting.

ty for your time in reading.
hope to see this bug fixed soon.

:D

here are the attached stdout.txt and the stderr.txt:
stdout.txt
stderr.txt
You do not have the required permissions to view the files attached to this post.
  • Derivative + pArKeL + a bunch of luddites = ?

    :D
loki1950
The Shepherd
Posts: 5841
Joined: Fri May 13, 2005 8:37 pm
Location: Ottawa
Contact:

Re: Getting stuck in another ship/station.

Post by loki1950 »

I'm seeing lots of Python errors but no indication of a mesh collision which would be a C++ error and i don't really know either.But i will ask safenode to have a look as he wrote the collision code but he won't be around till tomorrow :(

Enjoy the Choice :)
my box::HP Envy i5-6400 @2Q70GHzx4 8 Gb ram/1 Tb(Win10 64)/3 Tb Mint 19.2/GTX745 4Gb acer S243HL K222HQL
Q8200/Asus P5QDLX/8 Gb ram/WD 2Tb 2-500 G HD/GF GT640 2Gb Mint 17.3 64 bit Win 10 32 bit acer and Lenovo ideapad 320-15ARB Win 10/Mint 19.2
parkel
Hunter
Hunter
Posts: 73
Joined: Fri Jun 12, 2009 2:22 pm

Re: Getting stuck in another ship/station.

Post by parkel »

only other problem is that the game lags (even at low quality, no shader) when i see a load of ships crowded around a wormhole.

thx for the reply btw :D
  • Derivative + pArKeL + a bunch of luddites = ?

    :D
safemode
Developer
Developer
Posts: 2150
Joined: Mon Apr 23, 2007 1:17 am
Location: Pennsylvania
Contact:

Re: Getting stuck in another ship/station.

Post by safemode »

Ok, here's the deal as far as I remember it being (if it's the same issue). What happens is that there is _always_ some amount of penetration of one model with another between the frame that translates models in space and the check for collisions. Usually we do this more and more rapidly the closer models get to eachother. Now there were two means of this bug occurring.

1. The speed of the collision is great enough (doesn't have to be all that fast in actuality) such that they have penetrated greatly before a check for a collision has been made. This can occur for multiple reasons. a. Our game uses a homegrown sphere collider to do initial collision checks. We should move to the opcode sphere collider to minimize bugs in mis-detecting the need to do the more thorough mesh collider detection. b. The speed is so great that no amount of rapid checks for collisions would be enough to detect the collision before deep penetration has occurred. In the second case, the ships should be destroyed (or at least the smaller one).

In any case the deep penetration needs to be reversed so that the ships are only touching at the beginning of the physics frame. It's the logic in this moving the penetrating ship to the surface that can result in your ship showing up on the inside of another. Incorrectly modeled units can result in not being able to know which side is out. Also, before you get an idea, you can't simply use the center of the unit and look at where that is in relation to the collided triangle to see which side is out, the center may be behind you as you could be inside a base colliding with the sides of a docking bay wall facing away from the center. Our logic with figuring out how to correct a ship and in what direction then to bounce it is at the heart of this bug in most cases, in others it's badly modeled meshes, and in fewer others, it's the sphere collider missing a possible collision event and not calling the opcode mesh collider to check.

2. What should have been long fixed by now and thus i highly doubt to be the cause of your bug is that the opcode collider logic is at fault (compiler issue maybe) and it's returning either no collision occuring until you're well on the other side of the mesh or it's returning bogus values. This is a possibility, but it's much more likely something from #1. The only way to be sure is to heavily populate the source with debugging output and be able to reproduce the problem consistently, then to look at the model and make sure it doesn't have extra "invisible" pieces or holes in it. Unfortunately, i doubt that type of troubleshooting is within the scope of any regular user.

What you can tell me is the inclination of the collisions. Are you colliding head on perpendicular to the face of the model where you penetrate? Are you colliding at an oblique angle where you skirt the model fall through?
What speed are you traveling when you penetrate vs when you bounce correctly off the same model?
Is there a particular location on the offending models that you hit usually where you fall into it vs bouncing off?
Is the game heavily lagged/stuttering just prior to the collision?
Are you in a docking bay when this happens, or otherwise facing away from the center of the offending unit yet colliding while moving away?
What ship are you using when you fall in to these models, and does it occur with any other ships (if you've got anything other than a llama)?

Finally, yet it seems obvious, are you using the sourceforge build of VS or your own? If so, what arch are you using, mac or windows? What gpu and cpu are you using? Have you tried it at both lowest quality levels and highest in the case that the game stutters when you near units?
Ed Sweetman endorses this message.
parkel
Hunter
Hunter
Posts: 73
Joined: Fri Jun 12, 2009 2:22 pm

Re: Getting stuck in another ship/station.

Post by parkel »

@safemode

here are my 6-7 years old box's specs:
512MB RAM
Windows XP Home Edition SP3
Pentium 4, 2.6 GhZ with hyper-threading technology
Remaining Disk Space = 35.1 GB out of 70.0 GB
Nvidia Geforce 5200 FX 128MB

as for the stuttering problems... i tried it with low, medium and high quality settings, since my graphic card will not attain the extreme quality setting. no shaders used btw.

and also...
the colliding part...
colliding with a gleaner, llama, ox - any ship or station (i mean ANY), will cause the bug.

normally it happens when my ship is changing course and it just floats...
i dunno wads that called but here's a brief explanation:

after turning sometimes, your ships speed increase. then, your ship will start drifting in the direction which you 1st started from. after your ship's speed reaches the maximum limit - it overshoots the limit - it goes in your desired direction again.

when i crash like that, it just gets stuck inside the other station/ship.

another situation is that when the AIs (stupid AIs!) try to ram me, i get stuck.
and say... a uln dodo...
it tractors me towards it...
BANG! i get stuck in the dodo.

that's it.

oh and btw.. my shields DO NOT decrease when i get stuck inside the station/ship.
my armor neither... even if i keep hearing the sound when u get hit play over and over again.
but no damage done.
  • Derivative + pArKeL + a bunch of luddites = ?

    :D
theGleep
Atmospheric Pilot
Atmospheric Pilot
Posts: 1
Joined: Wed Apr 07, 2010 3:34 am

Re: Getting stuck in another ship/station.

Post by theGleep »

I have the same problem hapenning right now - I was fine to dock, but whenever I launch, I'm stuck inside the fighter base - and I can't get to the exit. If I try, I crash and blow up.

This is the case with my saved game. Would it help if I uploaded a copy of the save game? If so - how would I?
Post Reply