View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0001171 | Spring engine | General | public | 2008-10-22 18:31 | 2009-07-25 16:11 |
| Reporter | BasiC | Assigned To | Kloot | ||
| Priority | normal | Severity | minor | Reproducibility | N/A |
| Status | resolved | Resolution | fixed | ||
| Fixed in Version | 0.79.1.0+git | ||||
| Summary | 0001171: units still cant shoot through units that have a n extremly smal lhitshpere. | ||||
| Description | If you relocate a solar's hitshperew using : collisionVolumeType=; collisionVolumeOffsets=; collisionVolumeScales=; the hitshpere relocates and changes size but units still cant shoot through the visible model. | ||||
| Tags | No tags attached. | ||||
| Checked infolog.txt for Errors | |||||
|
|
You are not supposed to leave the values blank (same effect as not adding the tags at all). Just use very small scales (collisionVolumeScales="1.5 1.5 1.5") to disable hit detection. |
|
|
reopened for new information |
|
|
Disorganized infodump: [12:21:29 PM] <[PinK]BasiC> collisionVolumeType = "sphere", [12:21:29 PM] <[PinK]BasiC> collisionVolumeOffsets = "0 6 5", [12:21:29 PM] <[PinK]BasiC> collisionVolumeScales = "20 20 20", [12:21:44 PM] <[PinK]BasiC> this makes the unit disapear from the build list [12:22:31 PM] <[SmuG]lurker> do you see "sphere" in the list? [12:22:43 PM] <[SmuG]lurker> I see "Ell" for ellipsoid [12:23:36 PM] <[PinK]BasiC> yeah i tried that [12:23:39 PM] <[PinK]BasiC> same result [12:23:40 PM] <[SmuG]lurker> and don't put it in quotes... [12:23:48 PM] <[SmuG]lurker> fbi doesn't use quotes... [12:24:11 PM] <[SmuG]lurker> for all of them [12:25:26 PM] <[PiRO]Pizzi1> and don not use the comma [12:27:18 PM] <[PinK]BasiC> yeah it works [12:27:20 PM] <[PinK]BasiC> but [12:27:31 PM] <[PinK]BasiC> even when i move the hitshpere out andto the side [12:27:38 PM] <[PinK]BasiC> an llt still cant shoto thorugh the solar [12:27:45 PM] <[PinK]BasiC> even though the solar [12:28:01 PM] <[PinK]BasiC> solar's hitshpere is small and in a completly different place.. [12:28:12 PM] <[PinK]BasiC> the solar shouldnt abstract any fire at all [12:34:30 PM] <[PinK]BasiC> if i shoot a dgun though it it explodes.. [12:34:40 PM] <[PinK]BasiC> even though the sphjere is way off to the side [2:23:46 PM] <KDR_11k> The old hitsphere is stull used for some things like AOE calculation [2:24:02 PM] <KDR_11k> No idea what avoidFriendly uses... So key things: hitsphere is moved well out of the way, but llt still can't fire over the solar, as happened with the old CollisionSphereScale tag. Possible solution: use some calculation on the hit shape to find a new size to use wherever these hit shapes aren't used. |
|
|
"hitsphere is moved well out of the way, but llt still can't fire over the solar" I know, didn't bother adjusting line-of-fire determination yet. The only thing that looks like a bug is "if i shoot a dgun though it it explodes. even though the sphjere is way off to the side" but without more information about the precise values of the tags and whether "through it" refers to the coldet volume or the rendered model there's not much I can do. |
|
|
> I know, didn't bother adjusting line-of-fire determination yet. But a very quick semi-fix would be to recalculate radius, wouldn't it? > precise values of the tags collisionVolumeType = Ell; collisionVolumeOffsets = 0 6 5; collisionVolumeScales = 20 20 20; I'm pretty sure through it means the model. |
|
|
"But a very quick semi-fix would be to recalculate radius, wouldn't it?" No, because radius is used for many other things, including damage calculations (which furthermore assume a non-offset center). The old "collisionSphereOffset" and "collisionSphereScale" tags actually changed the model radius and center for that reason, mine don't touch them since they are not restricted to spheres. In any case, I'll fix both now. |
|
|
The "shooting through the model kills the unit even though the coldet volume is offset" issue is due to the fact that explosions from projectiles that impact the ground under the model can be powerful enough to overlap with the "hitsphere" (quoted because the code uses radial checks regardless of volume type), even if it is offset. Also, <0, 6, 5> is not enough clearance by far for a BA armsolar to survive a dgun blast hitting only the model. LOF tests are now (mostly) reliable again. |
|
|
There's still a bug here. The old offset and scale tags shouldn't have been removed; the model spheres are still used in aiming. |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2008-10-22 18:31 | BasiC | New Issue | |
| 2008-10-23 20:55 | Kloot | Note Added: 0003017 | |
| 2008-10-23 20:56 | Kloot | Status | new => closed |
| 2008-10-23 20:56 | Kloot | Resolution | open => fixed |
| 2008-10-26 10:34 | tvo | Note Added: 0003034 | |
| 2008-10-26 10:34 | tvo | Status | closed => new |
| 2008-10-26 10:34 | tvo | Resolution | fixed => reopened |
| 2008-10-26 10:54 | lurker | Note Added: 0003035 | |
| 2008-10-26 13:19 | Kloot | Note Added: 0003039 | |
| 2008-10-26 13:20 | Kloot | Status | new => feedback |
| 2008-10-26 13:22 | lurker | Note Added: 0003040 | |
| 2008-10-26 13:25 | lurker | Note Edited: 0003040 | |
| 2008-10-26 14:03 | Kloot | Note Added: 0003041 | |
| 2008-10-26 14:10 | Kloot | Note Edited: 0003041 | |
| 2008-10-28 00:39 | Kloot | Note Added: 0003052 | |
| 2008-12-29 19:18 | Kloot | Status | feedback => resolved |
| 2008-12-29 19:18 | Kloot | Fixed in Version | => 0.77b5+svn |
| 2008-12-29 19:18 | Kloot | Resolution | reopened => fixed |
| 2008-12-29 19:18 | Kloot | Assigned To | => Kloot |
| 2009-01-18 17:11 | lurker | Note Added: 0003211 | |
| 2009-01-18 17:11 | lurker | Status | resolved => assigned |
| 2009-01-18 17:11 | lurker | Resolution | fixed => reopened |
| 2009-01-18 22:44 | Kloot | Relationship added | related to 0001262 |
| 2009-07-25 16:11 | Kloot | Status | assigned => resolved |
| 2009-07-25 16:11 | Kloot | Fixed in Version | 0.77b5+svn => 0.79.1.0+git |
| 2009-07-25 16:11 | Kloot | Resolution | reopened => fixed |
