Not a member yet? Why not Sign up today
Create an account  

  • 1 Vote(s) - 5 Average
  • 1
  • 2
  • 3
  • 4
  • 5
 
2.2.1 [dev test only for now]

I found another bug in the grid casting.
In GridCasting.GridCast, before it starts the loop it calculates an upper bound for the number of iterations. To do this, it calculates which of the 6 faces of the cuboid it will exit through by calculating which face the path of the projectile will intersect with first, starting from the point where it enters the cuboid. However, due to rounding errors, it sometimes happens that the face it intersects with first is the face it just entered the cuboid through, which is incorrect and will result in upper bound for the loop being too low (it will do at most 3 iterations instead of however many it should do), which can result in projectiles flying through blocks. I checked it with some debug logging and it does indeed happen, but it is of course difficult to observe with actual projectiles, and even more so to differentiate this from any other bugs that may be lurking in the grid casting.

Possible fixes:
only check for intersections with 3 of the faces of the cuboid instead of all 6, and choose which faces based on the signs of the components of the local direction of the projectile. So if the x-component of the direction is positive, then check the intersection with the face parallel to the y-z-plane with the largest x, otherwise check the one with the smallest x. (I would like to state that more concretely, but all the variable names of vectors and floats get lost in the compiling and decompiling process so I don't know what all those variables are called).
Reply

I have encountered some weird issue with the newest update ( or the one before, the last time I played was last Saturday).

My tanks turns by having wheels of one side to go forward while the wheels on the opposing side to go backwards. They worked last Saturday. But today when I played, some of the tank just turns excruciatingly slow, and the problem didn't go away after I replaced all the wheels and removed the propulsion balance card , while others are normal. Not sure what happened. Blue prints attached.

M8 is behaving normally, but T14 is not. They are controlled by I,K,J and L to accelerate, decelerate, turn left and turn right.

Also some of my tanks starts to launch into the air again, like the attached M1. They didn't launch into the skies last Saturday. M1 turns very slowly as well, which wasn't the case last Saturday.


Attached Files
.blueprint   T14 Armata.blueprint (Size: 100.63 KB / Downloads: 6)
.blueprint   M1 Abrams TUSK.blueprint (Size: 94.6 KB / Downloads: 6)
.blueprint   M8AGS.blueprint (Size: 60.11 KB / Downloads: 6)
Reply

good job on this version, another benefit of the overhaul to the blueprint loading, it now loads bigger stuff. WITHOUT crashing the game trying to...
so hats off on that. (i could be a few revisions off, been away from FTD for around 3-5 months)

I also mention that the networking seems much more optimized then a few months back, actually loaded that oversized blueprint in mp with my brother, who was using the same machine that i've
upgraded from (and made that blueprint before never loading it again due to size)
he also loaded a ship of his own, that never really works well in mp. and the only stutter we got from this nasty stress test, was the game struggling with my ship's size.

we then played quest for netter for 8 hours with almost no issues at all. before we could not manage half an hour without some desync.

so yeah, keep up the good work.
Reply

I have been getting random crashes while in combat off and on for a few days while using devtest, I'm not on a potato anymore, so I don't think it's an issue of my PC.


Edit: Found the logs, just got a new computer and the logs were going to a different location than I remembered them going >_<


I'm attaching the output log, let me know if it's something I can fix on my end or just a bug
Edit 2: Forgot to attach the log


Attached Files
.txt   output_log.txt (Size: 101.68 KB / Downloads: 8)
Reply



Forum Jump:


Users browsing this thread:
1 Guest(s)