Chat log from the meeting on 2022-04-12

From OpenSimulator

Revision as of 10:44, 18 May 2022 by Tampa (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search
[11:04] Cuga.Rajal Hi MrBlue
[11:04] Andrew Hellershanks: Hello, everyone.
[11:05] Meeting Chat Logger V4: Meeting chat logging has been enabled.
[11:05] Meeting Chat Logger V4: Be sure to stop and clear the log at the end of the meeting, this will create the archive @
[11:05] Gavin.Hird Hi Andrew, MB
[11:05] Meeting Chat Logger V4: You can view the log at:
[11:05] Andrew Hellershanks: Nice to be back at our usual meeting location.
[11:05] Misterblue Waves: hi all
[11:05] Gavin.Hird Hi Kayaker
[11:06] Kayaker Magic: Hello all! Dev Outreach is BACK!
[11:06] Misterblue Waves: /clap clap clap
[11:06] Andrew Hellershanks: Hello, Kayaker.
[11:07] Cuga.Rajal that was a doozey of an outage, glad its back up
[11:07] Misterblue Waves: was it an asset database problem?
[11:07] Cuga.Rajal thats what I read
[11:08] Andrew Hellershanks: Hello, Selby.
[11:08] Michael.Christopher ok hello, sorry im behind lol
[11:08] Misterblue Waves: and people think they will "solve" the asset problems to going to blockchains.... <shakes-head/>
[11:08] Gavin.Hird really
[11:09] Michael.Christopher smh
[11:09] Selby.Evans hi everyone
[11:09] Misterblue Waves: hi selby
[11:09] Gavin.Hird Hi Selby
[11:09] Michael.Christopher hey selby
[11:09] Cuga.Rajal oh MrBlue might know about the physics lsl function issue
[11:09] Vincent.Sylvester Interesting question that SetForce, it's supposed to apply a force to the root part of the parent group, which is the avatar essentially. ApplyImpulse on the other hand applies force just to the object the script is in. I used to have a landmine script somewhere that would throw you into the air, so applying force to an avatar is possible, guessing it just doesn't use the correct scope when attached
[11:10] Cuga.Rajal llPushObject works
[11:10] Cuga.Rajal but llSetForce does not
[11:10] Michael.Christopher hmm
[11:11] Cuga.Rajal same on bullet and ubode
[11:11] Cuga.Rajal that is, not working when worn as an attachment
[11:11] Vincent.Sylvester PushObject calls ApplyImpulse on the object, SetForce just applies a force to it, guess that's different
[11:12] Cuga.Rajal should I open a bug report on that?
[11:12] Vincent.Sylvester Can push object move avatars? It seems to be checking for that in the implementation
[11:13] Andrew Hellershanks: The function name says object so I wouldn't expect it to move an avatar.
[11:13] Cuga.Rajal yes
[11:13] Cuga.Rajal llPushObject on a worn object does move the avatar
[11:13] Misterblue Waves: there would be, at least, a bunch of permission checks on pushing avatars
[11:13] Cuga.Rajal but only momentarily as designed
[11:14] Cuga.Rajal that would be really useful for things like parachutes, anti-gravity etc
[11:14] Vincent.Sylvester It does AddForce on avatars
[11:14] Vincent.Sylvester Which is different from SetForce
[11:15] Vincent.Sylvester I don't have vs code open so can't check what difference that makes
[11:15] Cuga.Rajal llSetForce(<0,0,0x7FFFFFFF>, 0)  does nothing when object is attachment
[11:16] Andrew Hellershanks: The wiki page for the function says the object needs to be marked physical.
[11:16] Cuga.Rajal correct,
[11:16] Cuga.Rajal yes I made it physical and call lsl to turn on physics too
[11:17] Andrew Hellershanks: ok.
[11:17] Cuga.Rajal shall I open ticket on that?
[11:17] Vincent.Sylvester Hm
[11:19] Cuga.Rajal wanted to check here forst in case there is a known issue or I'm forgetting something stupid
[11:19] Vincent.Sylvester The only implementation of SetForce is SceneObjectPart, which I am not sure if that includes avatars directly
[11:21] Vincent.Sylvester AddForce is implemented on the physics actors so that's why that works
[11:21] Cuga.Rajal that might explain why it works on objects not worn
[11:21] Vincent.Sylvester OdeCharacter has no SetForce
[11:22] Andrew Hellershanks: Cuga, Yes, file a ticket.
[11:22] Cuga.Rajal ok, will do. Thanks for taking a look
[11:22] Vincent.Sylvester It says AddForce is meant to set a target velocity
[11:22] Vincent.Sylvester That seems odd
[11:23] Vincent.Sylvester It does just add a momentary force doesn't, just pushes you up and then you fall again
[11:24] Cuga.Rajal I've used it on worn objects in Havok and it is a constant force
[11:24] Vincent.Sylvester So PushObject works the way you need for say a parachute, just SetForce is meant to as well?
[11:25] Cuga.Rajal llPushObject is only momentary
[11:25] Cuga.Rajal a timer works but is bumpy and not server-friendly
[11:25] Vincent.Sylvester Ah okay
[11:26] Cuga.Rajal llSetForce is supposed to be the way to get constant force
[11:26] Cuga.Rajal unless someone knows of another way
[11:27] Vincent.Sylvester As in that constantly pushes you in a certain direction I presume
[11:27] Misterblue Waves: I think objects just have the attribute "velocity" so adding or setting the force mostly modifies the velocity (and direction of that velocity)
[11:27] Cuga.Rajal In my case Im only applying it in the z-direction for such things as parachutes etc
[11:28] Vincent.Sylvester Velocity should be absolute and directionless so it just tells how much you are moving, not where, think the return type of that is a float
[11:29] Vincent.Sylvester There is a SetAcceleration in here that takes a vector input
[11:29] Cuga.Rajal it seems that would make more sense with acceleration
[11:29] Misterblue Waves: absolute velocity is a float but, in the physics engines, it is a vector (direction and magnitude)
[11:30] Misterblue Waves: there is some inconsistancies with the application of all the different way to change force/acceleration/direction/...
[11:30] Misterblue Waves: so adding a Mantis entry would start the wider discussion about what really should be happening
[11:32] Michael.Christopher maybe that means we need to find someone who has time to re-evaluate the force related methods and bring them all into a more unified method.  ive still got my team rebuilding the whole world and viewer from scratch so i know they understand how all the pieces work together before i try putting them back on OS code
[11:33] Michael.Christopher once they can make a viewer able to interact successfully with a very basic test world then they can try reconnecting to hg
[11:33] Cuga.Rajal well since llPushObject is working on a scripted attachment I was hoping it wouldn't be a big jump for llSetForce()
[11:34] Cuga.Rajal but I feared it might have already been implemented unless there was a bigger issue
[11:34] Cuga.Rajal I'll open a mantis
[11:35] Cuga.Rajal would this be a physics engine issue or lsl issue?
[11:35] Andrew Hellershanks: Good question.
[11:36] Cuga.Rajal I'll mark mantis as a physics engine issue for now
[11:36] Andrew Hellershanks: Go with physics engine.
[11:36] Gavin.Hird make a lsl issue initially and I guess it can be changed if needed
[11:37] Cuga.Rajal I'll mark it as lsl issue, can be changed later
[11:39] Vincent.Sylvester I'll fiddle around with that code a bit maybe I can get some more info on what these functions do
[11:39] Andrew Hellershanks: Thanks, Vincent.
[11:41] Misterblue Waves: and the different physics engines will implement things differently so be specific on which one you're using
[11:43] Cuga.Rajal I've been testing it in the different OSG sandboxes with diff physics engines, same result
[11:44] Cuga.Rajal but my grid is using bullet
[11:44] Vincent.Sylvester Hm ubode implements a SetAcceleration, Bullet does not weird
[11:46] Misterblue Waves: for whatever reasons, physics implementation is spread between the simulator, the c# physics code, and the underlying C++ physics code. The different implementations of the C# and C++ physics code have always done things a little differently
[11:48] Andrew Hellershanks: :)
[11:49] Andrew Hellershanks: We are getting close to the top of the hour. Any other topic(s) for today?
[11:50] Jagga Meredith: I fixed my toilet!!
[11:50] Jagga Meredith: not particularly relevant but I live a lonely life and noplace else to brag about it
[11:50] Andrew Hellershanks: Good to have a working one to save you digging a hole in your backyard. :)
[11:52] Misterblue Waves: I hate working on toilets. Here in the US we have flappers and leaks. I would like the European solution that seems to be reliable enough to embed in the wall
[11:52] Andrew Hellershanks: Not a lot to report regarding changes to the OpenSim code. this past week. There was a fair number of commits but most of them appear to be related to code cleanup.
[11:52] Michael.Christopher lol glad its fixed jaggaa
[11:52] Andrew Hellershanks: There were some changes to the simulatorFeatures cap.
[11:52] Misterblue Waves: and I have been tweaking on PreBuild to allow it to process projects in "netstandard2_0" and "net6"
[11:52] Michael.Christopher had to do that myself a time or 2\
[11:53] Andrew Hellershanks: Another texture was added that is used by eep.
[11:53] Misterblue Waves: someday the simulator will be updated to the newer build systems
[11:54] Andrew Hellershanks: There was also one minor bug fix to osParcelSubdivide. It won't do anything if there isn't a match to the entire original parcel.
[11:55] Michael.Christopher misterblue - now youve gotme thinking of nickleback - "someday, somehow"
[11:55] Andrew Hellershanks: Not quite sure what that is meant to say.
[11:55] Michael.Christopher newer code bases maybe? or new physis engines?
[11:56] Andrew Hellershanks: Someday, somehow, somewhere reminds me of another song. Looking it up it is from West Side Story.
[11:56] Andrew Hellershanks: I'd like to see the viewer use a newer build system.
[11:56] Gavin.Hird There is also Somewhere Over the Rainbow
[11:56] Misterblue Waves: it is true that such a project is probably big enough that it will probably never happen... but I want to keep my hopes up
[11:57] Andrew Hellershanks: The main OS code base uses a simple build system by comparison to that used in the viewer.
[11:57] Misterblue Waves: I'm using the new build frameworks for addon_modules and not changing the simulator proper
[11:57] Gavin.Hird lol
[11:59] Andrew Hellershanks: Any news about the physics engine or viewer worth mentioning before we wrap up todays meeting?
[12:00] Gavin.Hird LL is working on a completely new cap for profiles
[12:00] Michael.Christopher not here
[12:00] Gavin.Hird not sure what it means for viewer compatibility quite yet
[12:01] Andrew Hellershanks: That could have an impact on the profile modules.
[12:01] Gavin.Hird I am also nearing the point with my rework where I will fork it and add back the opensim code
[12:01] Misterblue Waves is unsure about the relationship between LL and the third-party viewers
[12:01] Gavin.Hird absolutely will impact profiles
[12:01] Gavin.Hird 98% of the code in the TPV is LL code
[12:01] Gavin.Hird inseparable
[12:02] Vincent.Sylvester Can you add the pushobject method to the mantis so I got something to test against
[12:02] Vincent.Sylvester Cause I am not getting any movement out of it
[12:02] Andrew Hellershanks: Gavin, has LL said anything about what they have in mind for profiles?
[12:02] Gavin.Hird The bulk of TPV code is UI modding
[12:03] Gavin.Hird they are re-implementing the classic profiles, but served via a cap
[12:03] Andrew Hellershanks: ok, ty
[12:03] Gavin.Hird the classic profiles in the current viewers are primarily FS code
[12:04] Gavin.Hird so they are doing away with that
[12:05] Misterblue Waves: i'm off. take care all!!
[12:05] Gavin.Hird have a good one
[12:05] Andrew Hellershanks: ok, Misterblue. Thanks for being here.
[12:05] Cuga.Rajal tc mrblue gtsy
[12:05] Andrew Hellershanks: You picked a good day as there was some physics related questions
[12:05] Michael.Christopher im heading out too, got to get back on the road to the office
[12:05] Andrew Hellershanks: ok, Michael. Thanks for being here.
[12:05] Michael.Christopher hasve a great week and tty soon!\
[12:06] Andrew Hellershanks: We are just past the top of the hour. If there is nothing more for today I will wrap up todays meeting.
[12:06] Cuga.Rajal I'm heading out too.. Take care all!
[12:06] Andrew Hellershanks: ok, Cuga.
[12:06] Kayaker Magic: Yeah, RL is calling me also
[12:06] Andrew Hellershanks: With people needing to go I will call this meeting to a close. Thank you all for coming. See you again next week.
Personal tools
About This Wiki