[#SVC-875] (Havok 4) Drop off in force applied by llPushObject seems to drop off differently from Havok 1

2. Second Life Service - SVC

(Havok 4) Drop off in force applied by llPushObject seems to drop off differently from Havok 1

Created: 25/Oct/07 09:20 AM Updated: 20/Apr/08 12:11 AM

Return to search

Component/s:

Physics

Affects Version/s:

None

Fix Version/s:

Havok4 Beta

Issue Links:

Relates

This issue Relates to:

SVC-1179 (Havok 4) llPushObject applied to avatars is inconsistent

SVC-141 Restrict Pushing disallows pushes from objects set to group

~~SVC-1648~~ Push avatar to avatar broken(Havok 4 Kick)

SVC-1968 Prims In Havok4 No longer Eject Avitars from them when Rez Around a Avitar

This issue is related to by:

~~VWR-4490~~ Havok4 - LlPushObject not functioning as expected in attachments

VWR-6256 Meta-issue: Bugs that effect C:SI Combat Samurai Island

Linden Lab Issue ID:

DEV-5276

Description

« Hide

In Havok 1, it was the case that an object colliding with an llVolumedetecting pusher object would get pushed with the correct impulse if it was near enough to the pusher that it had triggered the collision_start event in it. The 1/r^3 drop off in the impulse seemed to kick in somewhere beyond that range.

In Havok 4 the drop off seems to start just outside of a sphere inscribed inside a cubical pusher object, and so in particular, objects setting off the collision_start event are outside of the range at which the correct impulse would be given without taking into account the drop off.

All

Comments

Change History

Sort Order:

[ Permlink | « Hide ]

Andrew Linden added a comment - 01/Nov/07 06:16 PM

The problem was actually that the llPushObject() script energy consumption was off... in particular the distance^3 term was too big by a factor of about 100. Also, I had to eliminate any vertical damping when the avatar falls/jumps in order to actually get closer to the real effect of the push.

[ Permlink | « Hide ]

WarKirby Magojiro added a comment - 22/Dec/07 01:28 PM

This issue has been bulk changed to fix pending.

[ Permlink | « Hide ]

RobbyRacoon Olmstead added a comment - 25/Jan/08 12:39 PM - edited

This is still marked as "Fix Pending", but I want to point out that llPushObject() calls are still extremely underpowered when compared to havok1 sims when testing on a Havok 4 Early Adopter sim. This can be easily and consistently duplicated, for instance, with a C:SI (Combat: Samurai Island) weapon whose kick attacks are intended to knock an opponent back by about a meter or two. In the Java Island II sim, which is part of the Havok 4 Early Adopter program, the kick barely moves the opponent, perhaps 0.25 meters.

[ Permlink | « Hide ]

Seifert Surface added a comment - 25/Jan/08 01:20 PM

I would guess that this is more an issue of friction with the ground rather than the strength of the push.

Can the push be tested in freefall, as in http://jira.secondlife.com/browse/SVC-1179 ?

[ Permlink | « Hide ]

Esprite Xavier added a comment - 21/Feb/08 05:37 PM - edited

Yes, the push used in the C:SI system robby mentioned does indeed work as intended in the air. On the ground the push only works vertically.

[ Permlink | « Hide ]

Aimee Congrejo added a comment - 20/Apr/08 12:11 AM

Linked to ~~SVC-1648~~ Push avatar to avatar broken(Havok 4 Kick)