Pages: [1] 2 :: one page |
|
Author |
Thread Statistics | Show CCP posts - 2 post(s) |
flank steak
Ancient Malevolence Rage Alliance
48
|
Posted - 2011.11.03 04:31:00 -
[1] - Quote
That while you are re-skinning objects in space, you can adjust the physical parameters of the objects so they are more realistic?
One of the most aggravating experience I have in eve is when i get stuck on an invisible extension of a gate, asteroid, or the dreaded station.
I am not trying to sound whiny because it's not all that detrimental to my immersion, but why is it when I am aligned to a gate under a station I get stuck and bounce around?
Also many new players find it odd that they cannot weave through asteroids or the gaps between station sections.
Any comment as to why this happens or if it is going to be changed? |
Xercodo
Xovoni Directorate Not Usually Killing Everyone.
141
|
Posted - 2011.11.03 04:41:00 -
[2] - Quote
I'd agree that some tighter collision boxes would be MUCH appreciated.... The Drake is a Lie |
flank steak
Ancient Malevolence Rage Alliance
49
|
Posted - 2011.11.03 04:48:00 -
[3] - Quote
Xercodo wrote:I'd agree that some tighter collision boxes would be MUCH appreciated....
My good sire! a like for you untill CCP arrives! I hope it shall suffice |
Kopfy
0
|
Posted - 2011.11.03 17:04:00 -
[4] - Quote
You need some help back to front page. Getting stuck between asteroids is very anoying. |
Tristan North
The Scope
33
|
Posted - 2011.11.03 18:08:00 -
[5] - Quote
Support this. |
Corina Jarr
Spazzoid Enterprises Purpose Built
14
|
Posted - 2011.11.03 18:11:00 -
[6] - Quote
So supported. Collision boxes make for terrible manual flight when fighting in the roid field. |
Abrazzar
272
|
Posted - 2011.11.03 18:13:00 -
[7] - Quote
Yeah, nothing looks more like a rush job than sloppy hit boxes. The acceleration gates are pretty dreadful for example. Please visit your user settings to re-enable images. |
Kro0k
C.A.S. Assisted Living
0
|
Posted - 2011.11.03 19:29:00 -
[8] - Quote
Yes! |
Tanya Powers
Science and Trade Institute Caldari State
105
|
Posted - 2011.11.03 19:30:00 -
[9] - Quote
Supported |
Katrina Oniseki
Revenent Defence Corperation Ishuk-Raata Enforcement Directive
71
|
Posted - 2011.11.03 19:32:00 -
[10] - Quote
Invisible walls in submarine space. Welcome to EVE. |
|
Rixiu
North Star Networks The Kadeshi
3
|
Posted - 2011.11.03 19:46:00 -
[11] - Quote
Agreed, if we were able to compare collision models with the actual model I doubt we could even see that they are the same thing. |
Morganta
Peripheral Madness The Midget Mafia
256
|
Posted - 2011.11.03 19:49:00 -
[12] - Quote
or those lovely stations where you can burn away for like 10 minutes before you get off zero.... The American public's reaction to the change was poor and the new cola was a major marketing failure. The subsequent reintroduction of Coke's original formula, re-branded as "Coca-Cola Classic", resulted in a significant gain in sales, leading to speculation that the introduction of the New Coke formula was just a marketing ploy |
Heimdallofasgard
APEX ARDENT COALITION NEM3SIS.
26
|
Posted - 2011.11.03 20:01:00 -
[13] - Quote
Bumping for justice |
SilentSkills
Estrale Frontiers
15
|
Posted - 2011.11.03 20:04:00 -
[14] - Quote
not a bad idea |
Alexsei Volkov
Order Of Black Knights
3
|
Posted - 2011.11.03 20:06:00 -
[15] - Quote
Abrazzar wrote:Yeah, nothing looks more like a rush job than sloppy hit boxes. The acceleration gates are pretty dreadful for example.
Dear god yes this. Damn gates bounce you around for 5 min without actually pushing you far enough away to use them.... |
Barbelo Valentinian
The Scope Gallente Federation
119
|
Posted - 2011.11.03 20:28:00 -
[16] - Quote
Heimdallofasgard wrote:Bumping for justice
Ha, I see what you did there!
/signed
|
Black Dranzer
1
|
Posted - 2011.11.03 20:30:00 -
[17] - Quote
Hit boxes?
I'm pretty sure they're bounding spheres. I'm pretty sure they're ALL bounding spheres. Bounding spheres in a physics simulation being run at 1 FPS. |
mkint
253
|
Posted - 2011.11.03 20:37:00 -
[18] - Quote
Morganta wrote:or those lovely stations where you can burn away for like 10 minutes before you get off zero.... lolumad? |
|
CCP Dropbear
C C P C C P Alliance
57
|
Posted - 2011.11.03 20:40:00 -
[19] - Quote
It's not as straightforward a fix to apply as you might think. The more collision spheres you have, the large the performance hit that comes from deploying that object in space because more calculations need to be made whenever players interact with those spheres.
If that object is used in missions, then there's always many instances of it around on the server at any one point, which adds up to a lot of server load.
Acceleration gates are a pretty good example because they're absolutely bloody everywhere, not even just missions (which alone would be creating thousands of instances at any one moment). Adding just one more sphere to increase collision accuracy would effectively double the effect they have on server load. The end result could be more accurate acceleration gates, but decreased client/server performance.
All that said, I'm no engineer (far from it) but this is what I understand from having sat in a few meetings about this and listening to people who are. I remember a project around a year or two ago where content designers went in and simplified these spheres to increase server performance, so it seems they can have a significant impact (enough to warrant us going in there and changing hundreds of sites - which was no small task at the time).
The long and short of it is that we all may just have to bear it, as the alternative (increased load) is the greater of two evils.
|
|
mkint
253
|
Posted - 2011.11.03 20:44:00 -
[20] - Quote
CCP Dropbear wrote:It's not as straightforward a fix to apply as you might think. The more collision spheres you have, the large the performance hit that comes from deploying that object in space because more calculations need to be made whenever players interact with those spheres.
If that object is used in missions, then there's always many instances of it around on the server at any one point, which adds up to a lot of server load.
Acceleration gates are a pretty good example because they're absolutely bloody everywhere, not even just missions (which alone would be creating thousands of instances at any one moment). Adding just one more sphere to increase collision accuracy would effectively double the effect they have on server load. The end result could be more accurate acceleration gates, but decreased client/server performance.
All that said, I'm no engineer (far from it) but this is what I understand from having sat in a few meetings about this and listening to people who are. I remember a project around a year or two ago where content designers went in and simplified these spheres to increase server performance, so it seems they can have a significant impact (enough to warrant us going in there and changing hundreds of sites - which was no small task at the time).
The long and short of it is that we all may just have to bear it, as the alternative (increased load) is the greater of two evils.
How about changing warp-ins then? Nothing's worse than warping to an acceleration gate and landing right under the little weiner that hangs down and getting stuck on the damned thing, especially since there are occasional gates out there that won't let you cancel a warp. What if warp-in beacons were moved a little higher, and maybe tighten up the warp-in accuracy? |
|
|
CCP Dropbear
C C P C C P Alliance
59
|
Posted - 2011.11.03 20:50:00 -
[21] - Quote
We recently developed a new, more accurate type of acceleration gate for the revamped "New Player Experience" that shipped with Incarna. We used it to land players right on top of their shiny new ship, within a few hundred meters.
So at this point we're now able to substitute the new gate for old ones wherever needed. That, or change warp-in points, both solutions work.
Though it helps to know (via bug reports) where such fixes are needed. We have a lot of content out there, so keeping track of minor annoyances like these is not easy unless you guys point them out as you notice them. |
|
Circumstantial Evidence
15
|
Posted - 2011.11.03 20:56:00 -
[22] - Quote
The classic conundrum of Performance vs. level of detail and immersion. |
Vincent Athena
V.I.C.E. The Lostboys
114
|
Posted - 2011.11.03 21:06:00 -
[23] - Quote
Im wondering if in the check for collision, if the method is as efficient as it could be. For example, if two objects are far enough apart in the X direction to not collide, then there is no need to do all the math for a full collision check. So the code should first test if the distance in the X direction is above the collision threshold and if its not; do a full check. (You could also pick Y. I would not pick Z as alot of action seems to play out close to the horizontal plane).
Also you do not to this:
If R < sqrt ( dx^2 + dy^2 + dz^2 )
you do this
if R^2 < dx^2 + dy^2 + dz^2
because squaring a number is faster than taking a square root.
CCP employees should never proclaim a feature to be awesome. Only subscribers should. Subscribers can never answer a question posed to CCP. Only CCP can. |
Sadayiel
Inner Conflict
8
|
Posted - 2011.11.03 21:12:00 -
[24] - Quote
This thread just brought ecstatica memories back
|
Abrazzar
273
|
Posted - 2011.11.03 21:15:00 -
[25] - Quote
Wouldn't it be feasible to just decrease the size on some of the more bulky objects like asteroids and the acc gate? Or remove them on plain scenery objects and add object with different sized bubbles to replace the many objects' ones in a site with a general bubble to retain the obstacle. Please visit your user settings to re-enable images. |
Hungry Eyes
Ministry of War Amarr Empire
42
|
Posted - 2011.11.03 21:24:00 -
[26] - Quote
mkint wrote:
Nothing's worse than warping to an acceleration gate and landing right under the little weiner that hangs down and getting stuck on the damned thing
quoting an epic quote
|
Barbelo Valentinian
The Scope Gallente Federation
120
|
Posted - 2011.11.03 22:46:00 -
[27] - Quote
mkint wrote: How about changing warp-ins then? Nothing's worse than warping to an acceleration gate and landing right under the little weiner that hangs down and getting stuck on the damned thing, especially since there are occasional gates out there that won't let you cancel a warp. What if warp-in beacons were moved a little higher, and maybe tighten up the warp-in accuracy?
edit: new EVE meme... warping in and landing on the taint.
I like this idea, since this is probably the most annoying thing that happens to do with collisions - and it happens maybe, I dunno, 1 in 5 times or something? Especially with BSs.
|
Bienator II
24th Imperial Crusade Amarr Empire
248
|
Posted - 2011.11.03 22:51:00 -
[28] - Quote
collision spheres? Well its time to replace the space subdivision algorithm than. Even unoptimised octrees should give better results without performance impact. You fail you fail you fail you fail you fail you fail you fail to jump because you are cloaked |
madsamo
Nova Ops
0
|
Posted - 2011.11.03 22:51:00 -
[29] - Quote
flank steak wrote:That while you are re-skinning objects in space, you can adjust the physical parameters of the objects so they are more realistic?
One of the most aggravating experience I have in eve is when i get stuck on an invisible extension of a gate, asteroid, or the dreaded station.
I am not trying to sound whiny because it's not all that detrimental to my immersion, but why is it when I am aligned to a gate under a station I get stuck and bounce around?
Also many new players find it odd that they cannot weave through asteroids or the gaps between station sections.
Any comment as to why this happens or if it is going to be changed?
|
Tarikla
Projet Aurora
9
|
Posted - 2011.11.03 23:01:00 -
[30] - Quote
I remember something i LIKED a lot :
Beacon-style acceleration gates . if you bump into it , you must be a serious drunker .
Why not remplace all the acceleration gates by those types , instead of keeping the old and painful gate ?
Example of this type of gate can be found in "Patient Zero" storyline mission . |
|
|
|
|
Pages: [1] 2 :: one page |
First page | Previous page | Next page | Last page |