Jump to content

Stuck / Out-Of-World / Collision Feedback


333 replies to this topic

#301 Serpentbane

    Member

  • PipPipPipPipPipPip
  • Overlord
  • Overlord
  • 485 posts
  • LocationVanvikan, Norway

Posted 17 April 2015 - 10:22 AM

Dont see why PGI need us to post where weapons hits invissible walls, it's like everywhere.
Fired up my DWF and spent a few minutes with my lasers to find several issues.

This is a huge problem, often you can not shoot out, but the enemy mech have no problems hitting you.
Often my left arm cant shoot past a wall on the right side, still the enemy are able to hit my CT, LT and LA. In a slow mech, like the DWF, this could easely spoil the match.

Examples:
Posted Image
Posted Image
Posted ImagePosted ImagePosted Image

#302 Nerdboard

    Member

  • PipPipPipPipPipPip
  • Shredder
  • Shredder
  • 226 posts

Posted 18 April 2015 - 04:51 AM

Hello,

I had a weird event of viridian bog (unfortunately I do not know the grid tile). It was in assault or skirmish game mode.

What happened: I tried jumping on one of the medium 'hills' (or plateaus) with my Griffin when I got stuck right at the edge of it for about a second. Then I got unstuck which resulted in an immideate big jump upwards (felt like 10-15m maybe?) which also instantly used all my remaining jumpjet fuel. It might have been a lag issue as well but I felt like I jumped a lot higher than I usually could with the mech and I had no ping issue at all.

#303 Stealth Fox

    Member

  • PipPipPipPipPipPipPip
  • Big Brother
  • Big Brother
  • 736 posts
  • LocationOff in the Desert

Posted 18 April 2015 - 04:59 AM

You know.. this wasn't such a problem back before you added your completely unfun mechanic of "Walking bipedal tanks that are suppose to be the most mobile land units in the Universe, Can't climb mountains"

Maybe if you focused on the fact that if a dude with a heavy ass backpack and Rifle can climb a 45 degree angle just fine during basic training in the silly ancient US military, A fricken mech with robo muscles and a computer that lets it walk, run, and fist-to-cuffs should be able to do the same damn thing easier.

Also fix the crappy hit boxes on the Tourmaline Desert map with its crystals already! Just ..ALL of them, the entire map, can't tell you how many times I've played on that map and been laser blocked when the laser was a good 20 or so feet above the damn terrain.

Edited by Seph MacLeod, 18 April 2015 - 04:59 AM.


#304 Telmasa

    Member

  • PipPipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 1,548 posts

Posted 18 April 2015 - 04:14 PM

Augh, I finally found this thread, so I deleted the thread I put up over in Patch Feedback...and forgot to save the imgur link. :angry:

Map: Viridian Bog

Thankfully I was able to go into a training room and partially recreate the problem - I wasn't able to get myself permantly stuck again, but I did notice alot of collision glitching and was able to get myself stuck for a longer period of time than is normal (resulting in far higher leg damage than normal...which, isn't that a glitch related to how falling damage is calculated?) and was able to single out a 'corner' as the problem:

http://i.imgur.com/aG7JaoD.jpg
http://i.imgur.com/MzrxClc.jpg
http://i.imgur.com/gyS9ndN.jpg

Maybe the reason I was able to get stuck during a match & not in training room is a result of server-ping-related stuff? No real idea, but hopefully this helps anyway.

Edited by Telmasa, 18 April 2015 - 04:15 PM.


#305 sandworm16

    Member

  • PipPipPipPipPip
  • Bad Company
  • Bad Company
  • 103 posts

Posted 18 April 2015 - 05:14 PM

Invisible pole on Emerald Taiga Posted Image

#306 Telmasa

    Member

  • PipPipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 1,548 posts

Posted 06 May 2015 - 09:14 PM

Stuck, here, on Emerald Taiga, with my Nova:Posted Image


After alot of struggling with jumpjets, twisting and rotating in all directions, I finally managed to weasel out of there...the spot seems normally inaccessible to any mech, you have to fall in with a mech that has enough JJ capacity to reach the top of the buildings in the area. (You can use the minimap to see exactly where.)

Edited by Telmasa, 06 May 2015 - 09:15 PM.


#307 darkezero

    Member

  • PipPip
  • 41 posts

Posted 07 May 2015 - 10:57 AM

  • The kind of collision issue. (What happened?)
"dynamic terrain" Walls, rocks, hills, floors not appearing correctly or at all until closer. Can see lots more things than I should see, still can't hit them because the wall and image don't match and the wall is actually there.
  • The map affected.
Canyon Network
  • The game mode used.
All, any
  • The grid tile affected.
Many Grids
  • The Mech and/or weapon used to achieve the collision issue.
Any
  • A screen or video capture to help locate the bug.
see attached album, some are of the same....object...just closer/further to see the changes

http://imgur.com/a/vbP8G

#308 Aidan

    Member

  • PipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 542 posts
  • LocationFlorida, USA

Posted 11 May 2015 - 02:00 PM

PGI,

I have some questions about the on going new map re-design project.

1. In general, will destructibility be serialized to all attached clients across the network? In other words, will all clients see the same destruction and 3D coordinate position of this destruction for all attached clients?

2. How will PGI present Dynamic Time of Day for Public PUG battles when the average real time of a Public PUG match is around 8 minutes? In the case of a sunrise to Noon scenario, the sun would need to move very quickly in the 8 minutes of real time.

3. Will PGI consider limited re-spawn Public PUG matches to increase battle real time for a more natural and realistic progression of the Dynamic Time of Day feature?

4. Will PGI support DX12 and/or Vulcan when it is released?

Regards,

Aidan :)

#309 Telmasa

    Member

  • PipPipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 1,548 posts

Posted 11 May 2015 - 03:14 PM

Aidan, this isn't the right thread for that...and I can already guess at the answers.
1. - It's somewhere in the future, of course it'll be visible on all clients. (Otherwise what's the point?)
2. - It'll likely pick a random time of day and it'll be that time of the day during the match. (You can look at various other games as an example; you're probably just misinterpreting the meaning of a dynamic day/time here.)
3. - Probably not, and personally I hope to god they never do. There's so many other issues that come with "garage"-style battles, that's really not a good trade-off for having a longer period of time to experience a better "dynamic daylight cycle".
4. - I'm sure they'll consider new technology as it comes along as much as possible, at the same time I doubt it can be expected for PGI to instantly be on/ahead of the curve, with so much else they're catching up with.

#310 Calebos

    Member

  • PipPipPipPipPipPip
  • 285 posts

Posted 24 May 2015 - 05:27 AM

The collision method in this game is really horrible. Hitreg aswell because it is related to collision system. I know it is a bit complicated thanks to structure of each mech but why it is so bad with terrain collision? On many places you can fire your weapon really enough far from obstacle but that obstacle uses from player's point of view different "method" or different topology of obstacle geometry and your projectile or laser beam is hitting "open space" besides or over the geometry of obstacle. To be honest I don't know exactly what kind of collision method you are using(i know just old methods with bitmap registering of collision with terrain), but it sucks in many, many cases. And I am not writing about hitreg problem with lattency, It is complettely different cup of tea. When you play several maps as the training ground you can see it there. It is very frustrating especially in poking style of play ...

Edited by Calebos, 24 May 2015 - 05:45 AM.


#311 Calebos

    Member

  • PipPipPipPipPipPip
  • 285 posts

Posted 24 May 2015 - 05:39 AM

View PostTelmasa, on 11 May 2015 - 03:14 PM, said:

Aidan, this isn't the right thread for that...and I can already guess at the answers.
1. - It's somewhere in the future, of course it'll be visible on all clients. (Otherwise what's the point?)
2. - It'll likely pick a random time of day and it'll be that time of the day during the match. (You can look at various other games as an example; you're probably just misinterpreting the meaning of a dynamic day/time here.)
3. - Probably not, and personally I hope to god they never do. There's so many other issues that come with "garage"-style battles, that's really not a good trade-off for having a longer period of time to experience a better "dynamic daylight cycle".
4. - I'm sure they'll consider new technology as it comes along as much as possible, at the same time I doubt it can be expected for PGI to instantly be on/ahead of the curve, with so much else they're catching up with.



1. the same conditions for all players. If your client will destroy any part of terrain/obstacle/mech different way(different shape of geometry thanks to random dynamic solution) then your oponent's client you have different play conditions in the game environment because you see through the battlefield different way than your oponent or team mate. The debris from destroyed mechs is not matter of in game mech collision(so it is not influencing your walk path) but you are not able to see "through" it.

My opinion is that it cannot be implemented well because if I am not wrong(PGI please correct me If I am) the destruction of all mechs and environment in the game is post effect managed and solved by gpu(cuda in NVidia case) right after destroying mech/part of environment. It looks a bit logically because if you, just for example, destroy Thunderbolt's 5SS left tanking arm it will be torn off and fall to the ground but it is calculated by gpu after evaluating it is destroyed. In other words: all these effects are matter of your own game client and technically and visually are not affected by host server.

Edited by Calebos, 24 May 2015 - 05:42 AM.


#312 oldhasu

    Member

  • PipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 681 posts

Posted 01 June 2015 - 04:37 AM

HPG MANIFOLD hitreg bugs:


#313 Pjwned

    Member

  • PipPipPipPipPipPipPipPipPip
  • Little Helper
  • 4,731 posts
  • LocationDancing on the grave of Energy Draw LOL

Posted 04 June 2015 - 11:17 PM

The kind of collision issue. (What happened?)

Rammed into a wall after scrambling around a bit and managed to get stuck completely

The map affected.

Viridian Bog

The game mode used.

Conquest

The grid tile affected.

Between D4 and C4, exact coordinates (as shown in the screenshot) are 1943.8, 2126.6, 136.6

The Mech and/or weapon used to achieve the collision issue.

LCT-1E

A screen or video capture to help locate the bug.

Posted Image

#314 VanguardMk1

    Member

  • PipPipPipPipPipPip
  • The Tracker
  • The Tracker
  • 220 posts
  • LocationDid you check your rear?

Posted 05 June 2015 - 01:55 AM

Was happily climbing steep places with JJ's, got stuck and couldn't move. It happened while using my single JJ to get up and over that ridge.
Map: Hellebore Springs
Location: see picture
Posted Image

#315 Anvil Dragon

    Member

  • PipPipPipPipPip
  • 134 posts
  • LocationShionoha SF Bay Area

Posted 15 June 2015 - 04:41 AM

Stuck: River City, Skirmish 2180.7, 1132.0, 76.1
Top near the U shaped bit. Moving back and forth to shoot Mechs in the slot.
No forward, reverse, turn, or jump jets. Jump jets showed as no fuel.
Basic hood ornament.

#316 Tarogato

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Civil Servant
  • Civil Servant
  • 6,557 posts
  • LocationUSA

Posted 16 June 2015 - 12:24 AM

View Postoldhasu, on 01 June 2015 - 04:37 AM, said:

HPG MANIFOLD hitreg bugs:



+1 to this video. All the assets that are used for flooring on HPG need to have the collision meshes tightened up a little bit on them. All mechs float above them by a significant amount instead of actually walking on the visible surface. Super clear at edges like shown in the video where weapons collide with the mismatched geometry.

I'd like to note that HPG has the best framerate of any map in the game as far as I can tell, having more detailed collida won't hurt performance.

Edited by Tarogato, 16 June 2015 - 12:25 AM.


#317 Mister Fusion

    Rookie

  • 7 posts
  • LocationBoston, MA

Posted 15 July 2015 - 11:49 AM

Stuck: River City, Skirmish 3289.8 4552.2 60.1

Was piloting a Locust 1E at the time. The camera was blipping in and out of the world, I could not turn my mech nor throttle up.

#318 Windschreiter

    Member

  • PipPipPipPipPipPip
  • Gunsho-ni
  • Gunsho-ni
  • 215 posts

Posted 16 July 2015 - 03:31 AM

was playing Skirmish on River City in a Pirates Bane. Was moving from D4 to D3 when I saw an enemy and backed up, slid a bit down hill and got stuck at 3280/4555/62. Was able to wiggle myself free after some 10 minutes only to find myself alone facing a bunch of enemies.

Tried to reproduce on Training Grounds but elevation didn't seem to go below 70.

It "felt" like the Mech was falling. Would allow me to accelerate to almost 300 kph without moving an inch. Mech didn't turn either.

Edited by Windschreiter, 16 July 2015 - 03:31 AM.


#319 crustydog

    Member

  • PipPipPipPipPipPipPip
  • 670 posts
  • LocationCanada

Posted 28 July 2015 - 01:05 AM

The new mech collision code is proving to be problematic. Mechs are getting stuck on other mechs, and then become trapped in some kind of loop. Neither mech seems to be able to easily disengage from the other, while both are damaging their legs.

It is quite infuriating actually. I have been killed at least twice because of this bug. It is also leading to more team killing on the battlefield as warriors are reacting to collision damage from teammates.

Collision damage is one thing - but becoming stuck to other mechs for several seconds could not have been the original intent of the changes - please fix this stuff.

#320 Threat Doc

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Bowman
  • The Bowman
  • 3,715 posts
  • LocationO'Shaughnnessy MMW Base, Devon Continent, Rochester, FedCom

Posted 28 July 2015 - 08:04 AM

Hmmm, code not working correctly or not, perhaps players should be more mindful of where they are, anyway, giving one another space to move around. The problem, here, is not only with the code, but moreso with players being inured to the idea they continue to run into one another with zero consequences. PGI, please fix the code, but do not eliminate damage from contacting other 'Mechs? Behavior's of everyone in this game need to change so people stop acting like imbeciles and start acting like units, instead.

Edited by Kay Wolf, 28 July 2015 - 08:05 AM.






1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users