Pilot Archive
Thread: Ship Cruising Bug Drops to 0 speed and then back to full power
Shikkari wrote:
I have began to ignore this and always attributed it to the rubberband/latency effect, however that being said...IT HAS ANNOYED ME ENOUGH TIMES IN COMBAT THAT I WISH THEY WOULD PLEASE FIX IT!!! and I do remember logging off once because it was so annoying!
Please understand that this may not be something that SOE can fix. I myself don't have this problem, and it is very possible that there is something wrong between you and SOE which is out of SOE's control. You may want to do ping and trace route tests to see if there is high latency in the connection or high packet loss. The window that shows up in SWG that tells you your latency is not 100% accurate. To improve its accuracy, ust Ctrl-Shift-G.
+
i have also had the ships mine included speed up!! making it imposible to hit or be hit realy odd my wing man could not beleve his eyes when he was looking at me lol
lund0529 wrote:Been happening to me since before beta ended.
- I'm not on autopilot when it happens.
- It happens with every engine I use.
- It happens in my Dune Lizard and my Kimogila.
- It happens in 1st and 3rd person view.
- I warp backwards every time it happens.
- It happens in combat and outside of combat.
It looks almost exactly like rubberbanding on the ground to me.
True, it's just lag, loss of packetflow, whatever. Not a bug.
I could be wrong but I imagine it to be an issue with space partitioning. When you have a large world you cannot get good numerical efficiency by calculating all collisions and rendering all objects into memory. So what they do is they partion off the space into quadrants, octals or bsp's. If you are in a high performance single player game you will notice that there are certain places when you "load" for a second and that there are portals in specific places to get to other rooms. This is for efficiency as calculations are only performed on the objects within that specific partition. On the game such partitions are probably larger and have dedicated servers and are thus referred too as "server boundaries". (might even have their own databases for some things)
If you hit a new render quadrant, from a programming perspective you cease being a member objectof the previous quadrant and your object data is transferred into the new quadrant. When then they do this it is likely that the velocity information is notcopied over although the throttle state is.
They fixed rubberbanding on the ground with vehicles (more or less), right?
atytula wrote:
Shikkari wrote:
I have began to ignore this and always attributed it to the rubberband/latency effect, however that being said...IT HAS ANNOYED ME ENOUGH TIMES IN COMBAT THAT I WISH THEY WOULD PLEASE FIX IT!!! and I do remember logging off once because it was so annoying!
Please understand that this may not be something that SOE can fix. I myself don't have this problem, and it is very possible that there is something wrong between you and SOE which is out of SOE's control. You may want to do ping and trace route tests to see if there is high latency in the connection or high packet loss. The window that shows up in SWG that tells you your latency is not 100% accurate. To improve its accuracy, ust Ctrl-Shift-G.
Uh, no...it's something SOE can fix. I use CTRL-Shift-G up all the time since I started experiencing the problems beyond the rubberband snap-back effect...that part has existed all through JTL beta as well. But I never had any problems with the invulnerability periods/ship warping/'drunken bees' effects in JTL beta that I've had ever since it launched live, and nothing changed with my system/setup between beta and live. I consistently monitor my ping rate as well as the sent/received info in CTRL-Shift-G and everything is totally normal there. My ping rates range from 85-125ms average normally and the majority of the time more like 85-100ms. The problem is some sort of sychnronization issue between the game client and the server. I understand alot of people don't experience the effects, but we've come to the conclusion that there's no single 'commonality' between OS'es/hardware/devices/drivers/programs/etc that can be pointed to on the user's end, everyone has something different that is suffering from the more extreme symptoms. Additionally, more people have started experiencing the problems as time has gone on, not less. Some started experiencing them after certain 11.x patches that came later. No, it's something SOE can and should fix.
Trust me, I wish it were something so simple and could be easily isolated to some common problem on the user's side, I wish it wasn't such a headscratching issue along these lines....but I've yet to see one hard shred of evidence to make me think this whole thing is anything other than a synchronization issue. The reason being, these problems/symptoms always clear up immediately after the rubberband snap-back effect takes place, until such time that the client/server develop synch problems again. As someone else has pointed out, this same rubberband effect plagued the ground game for awhile, actually on at least two different occasions since SWG launch, and that definitely wasn't a end-user problem either. I'll be more than happy to see someone be able to point to something, anything user related that magically solves all these issues if such a time ever comes. Until then, I'm totally convinced that only SOE can resolve this.
this is indeed a strange problem, it does not happen when i am in a fighter but when i fly my yt-1300 it happens every time i use the booster. if say i am going at full speed and hit the booster after i let go my speed drops to zero. this never happens any other time or in any other ship, so am a bit clueless as to the cause.
I could be mistaken, but I think the booster accelleration/decelleration issue in the YT and Decimator are a separate issue. There is an issue with using the booster in the big MP ships that cause steering and control issues especially if you are turning. However the issue they have discussed here and from what I have encountered , it can occur in any ship. I notice also that it seems to occur when switching to a higher detail mode, but I also think that things may switch into a higher detail once you load into a new partitioning quadrant. I have noticed that there seems to be certain "boundaries" where if you pass and repass over the boundary you can reproduce this effect.
I think if someone is in a curcular dogfight and keep hitting this partition boundariy (sending them into different quadrants) then that could cause some real issues.
If you have a 4 by 3 grid sucha as:
XBXX
XAXX
XXXX
XXXX
If you are in location A then one technique may be to perform collisions checking algorithms and high detail rendering and meshes for everything in the "A" quadrant. Once you move into quadrant "B" all of your state data is copied over to "B" and the objects in "A" are no longer loaded in high detail and collissions are no longer calculated. However "B" will now load high deatil.
If something is in quadrant B and you are shooting it from Quadrant A you may not hit it at all. I think it is possible that your velocity information my not be persisteed between partitions although your throttle state is. This would cause you to immediately got to 0 velocity and then run back up to full throttle.
When this happens, stop and make a waypoint to the location where it occurd. Go boack and forth over that location at full throttle for about 1k or so. If the effect happens consistently in that location then we know it is likely a partition issue.
I will do some testing and post my results here. Hopefully someone else can also do this to collaborate the data I collect.
On a side note, I would also point out that pub 11.2 was said to have addressed some of these issues, and it *did* do that, because the severity of these problems *were* lessened by that patch by comparison to how bad it was before...but the fixes didn't cover all the bases.