Chat log from the meeting on 2023-03-21

From OpenSimulator

Revision as of 13:46, 21 March 2023 by Tampa (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search
[11:00 AM PT]  Orbert.Tatham @hg.zetaworlds.com: Hello, Andrew
[11:02 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: From last week the patch Andrew made was merged and some minor stuff surrounding it fixed up as well
[11:02 AM PT]  Andrew Hellershanks: Hello, Motoko
[11:03 AM PT]  Gavin.Hird @grid.xmir.org:8002: anyone knows what mesh_rez_enabled is?
[11:03 AM PT]  Motoko.Karu @login.digiworldz.com:8002: Hihi Andrew...
[11:03 AM PT]  Motoko.Karu @login.digiworldz.com:8002: Hello Everyone
[11:03 AM PT]  Gavin.Hird @grid.xmir.org:8002: Hi Moto
[11:03 AM PT]  Andrew Hellershanks: Hello, Cuga
[11:03 AM PT]  Gavin.Hird @grid.xmir.org:8002: Hi Cuga
[11:03 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Where'd you find that Gavin?
[11:03 AM PT]  Cuga.Rajal @rajal.org:9000: Hi all!
[11:03 AM PT]  Andrew Hellershanks: Vincent, some other changes snuck in that weren't supposed to be in the patch.
[11:04 AM PT]  Gavin.Hird @grid.xmir.org:8002: it is some viewer code LL is removing
[11:04 AM PT]  Gavin.Hird @grid.xmir.org:8002: I really don't understand it did anything useful, so might be some old stuff from way back
[11:05 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: What's the namespace? Context?
[11:05 AM PT]  Gavin.Hird @grid.xmir.org:8002: requestMeshRezInfo
[11:05 AM PT]  Gavin.Hird @grid.xmir.org:8002: it is in llfloaterregioninfo, so region info
[11:06 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: So region capability basically
[11:06 AM PT]  Motoko.Karu @login.digiworldz.com:8002: Hihi Gavin.
[11:06 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Whether a region serves mesh and it should be rezzed or something like that would be my guess
[11:06 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: They did very recently rework parts of regioninfo
[11:06 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Also added chat distance values to it
[11:07 AM PT]  Gavin.Hird @grid.xmir.org:8002: it could beit was actively used at a time regions did not necessarily support meshes
[11:07 AM PT]  Andrew Hellershanks: Gavin, I would agree with that based on the variable name
[11:08 AM PT]  Gavin.Hird @grid.xmir.org:8002: getRegionCapability("SimConsoleAsync")
[11:08 AM PT]  Andrew Hellershanks: I hear Kayaker beaming in
[11:09 AM PT]  Andrew Hellershanks: Hello, Kayaker.
[11:09 AM PT]  Andrew Hellershanks: I don't know if Ubit will be joining us today. I don't see him online in IRC.
[11:09 AM PT]  Kayaker Magic: Hello all! Did I miss anything yet?
[11:10 AM PT]  Andrew Hellershanks: Kayaker, nothing major. We are only just getting started.
[11:10 AM PT]  Gavin.Hird @grid.xmir.org:8002: it exist in RegionConsoleModule.cs so better keep it around for now
[11:10 AM PT]  Gavin.Hird @grid.xmir.org:8002: Hi Kayaker
[11:10 AM PT]  Kayaker Magic: Gavin! I'm glad to see you here!
[11:11 AM PT]  Gavin.Hird @grid.xmir.org:8002: Thanks. Good to see you too!
[11:12 AM PT]  Andrew Hellershanks: Ah, Ubit woke up. :)
[11:12 AM PT]  Kayaker Magic: I've let my Raspberry Pi system sit on the sidelines. I'd like to get it back up again. Any news on the ARM version of OpenSim?
[11:12 AM PT]  Andrew Hellershanks: Hello, Ubit.
[11:12 AM PT]  Motoko.Karu @login.digiworldz.com:8002: .:)
[11:12 AM PT]  Ubit Umarov: shh still awking up...
[11:12 AM PT]  Gavin.Hird @grid.xmir.org:8002: Hi Ubit
[11:12 AM PT]  Ubit Umarov: hi
[11:12 AM PT]  Gavin.Hird @grid.xmir.org:8002: I am afraid it is toast Kayaker
[11:12 AM PT]  Gavin.Hird @grid.xmir.org:8002: no support for dotnet6
[11:12 AM PT]  Gavin.Hird @grid.xmir.org:8002: unless you do a manual painful install on your own
[11:12 AM PT]  Cuga.Rajal @rajal.org:9000: Hi Ubit
[11:13 AM PT]  Ubit Umarov: hmm what has no dotnet6 ?
[11:13 AM PT]  Gavin.Hird @grid.xmir.org:8002: Rasbian
[11:13 AM PT]  Gavin.Hird @grid.xmir.org:8002: Raspbian
[11:13 AM PT]  Ubit Umarov: the rasp own linux?
[11:13 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Gavin I think that is just part of the capabilities it requests from the region, it probably was some flag at some point when mesh first came in. Should be possible to scroll back in git to see when it was added, that might give a clue
[11:13 AM PT]  Gavin.Hird @grid.xmir.org:8002: yes
[11:13 AM PT]  Gavin.Hird @grid.xmir.org:8002: Microsoft does not build any support for it at all
[11:14 AM PT]  Andrew Hellershanks: Ubit, Kayaker was asking about the state of ARM version of OpenSim
[11:14 AM PT]  Gavin.Hird @grid.xmir.org:8002: according to their own documentation
[11:14 AM PT]  Ubit Umarov: dont ppl install a another linux on it?
[11:14 AM PT]  Jamie.Jordan @grid.kitely.com:8002: Hi everybody
[11:14 AM PT]  Andrew Hellershanks: I have usually stuck with Raspbian on my Pi boards.
[11:14 AM PT]  Cuga.Rajal @rajal.org:9000: The new Pi's are 64 bit, so maybe the newer ones will work
[11:15 AM PT]  Orbert.Tatham @hg.zetaworlds.com: Pi 3 and 4 are both 64 bit - there was some delay getting a 64-bit version of Raspbian
[11:15 AM PT]  Gavin.Hird @grid.xmir.org:8002: Ubunto, but that is just a debian derivative
[11:16 AM PT]  Gavin.Hird @grid.xmir.org:8002: they won't Cuga unless you are ready to build and support it compleltey on your own
[11:16 AM PT]  Gavin.Hird @grid.xmir.org:8002: Microsoft was very clear on that
[11:16 AM PT]  Cuga.Rajal @rajal.org:9000: ah
[11:16 AM PT]  Orbert.Tatham @hg.zetaworlds.com: Is that true for all ARM architectures?
[11:16 AM PT]  Cuga.Rajal @rajal.org:9000: not true for macOS arm lol, but maybe for others
[11:16 AM PT]  Gavin.Hird @grid.xmir.org:8002: they support Apple Silicon
[11:16 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Compound issue with arm is that arm != arm, between kernel differences and the various hardware configurations out there most we can do it try to add some libraries, but building universal binaries might be quite difficult
[11:17 AM PT]  Ubit Umarov: well i i may break the topic a bit
[11:17 AM PT]  Ubit Umarov: not much code changes last week :)
[11:17 AM PT]  Cuga.Rajal @rajal.org:9000: Pi arms and macOS arms use different libs altogether
[11:17 AM PT]  Gavin.Hird @grid.xmir.org:8002: the the Apple processor has number of special instructions that Microsoft may take advantage of
[11:17 AM PT]  Orbert.Tatham @hg.zetaworlds.com: Andrew, what changes "
[11:17 AM PT]  Orbert.Tatham @hg.zetaworlds.com: snuck in" with that one
[11:18 AM PT]  Ubit Umarov: we have a new contrib from guga, think now with gaving help, on new unmanaged libs for apple silicon
[11:18 AM PT]  Cuga.Rajal @rajal.org:9000: yes
[11:18 AM PT]  Ubit Umarov: a question..  does the apple sign inlcude those ugly names?
[11:18 AM PT]  Cuga.Rajal @rajal.org:9000: there was a lot of activity onn the macOS side this week
[11:18 AM PT]  Gavin.Hird @grid.xmir.org:8002: we did a bit of testing and Cuga building
[11:18 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Pi is also just one of multiple manufacturers now offering arm boards for use, usually with major changes in kernels to account for different hardware so you almost always have to compile stuff manually down to basic things like nginx
[11:18 AM PT]  Ubit Umarov: i would like cleaner names, like we had..
[11:19 AM PT]  Cuga.Rajal @rajal.org:9000: between Gavin and I we can now test OS on all macOS versions that dotnet6 supports
[11:19 AM PT]  Cuga.Rajal @rajal.org:9000: and I built new libs that are compatible
[11:19 AM PT]  Cuga.Rajal @rajal.org:9000: with all
[11:19 AM PT]  Cuga.Rajal @rajal.org:9000: Bullet was the only universal lib being used in dotnet6 branch, si that should be updated
[11:19 AM PT]  Ubit Umarov: so can the things be renamed without breaking the sgnature?
[11:20 AM PT]  Cuga.Rajal @rajal.org:9000: good question, I think so
[11:20 AM PT]  Gavin.Hird @grid.xmir.org:8002: signature does not care about the name of the lib
[11:20 AM PT]  Cuga.Rajal @rajal.org:9000: although MrBlue favors keeping the date and universal in the name
[11:20 AM PT]  Andrew Hellershanks: The kernels will have differences when it comes to dealing with the on-board hardware of an ARM board but I wouldn't say there are major changes to the kernel itself. The bulk of the kernel needs to stay the same across platforms.
[11:20 AM PT]  Ubit Umarov: ok i had that idea.. but who knows..
[11:20 AM PT]  Ubit Umarov: now i god confused with the openjpeg ( again)
[11:20 AM PT]  Cuga.Rajal @rajal.org:9000: I also included universal libs for ubODE and openJpeg
[11:21 AM PT]  Cuga.Rajal @rajal.org:9000: Thos are not cvurrently universal libs
[11:21 AM PT]  Ubit Umarov: we currently have one libopenjpeg-arm64.dylib
[11:21 AM PT]  Cuga.Rajal @rajal.org:9000: but I propose we switch to those in dotnet6
[11:21 AM PT]  Ubit Umarov: and a few *-dotnet_* ones
[11:21 AM PT]  Cuga.Rajal @rajal.org:9000: so one file will work on all macOS CPUs and macOS versions
[11:21 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Andrew you would think so, but from the boards I have tried over the years it was always a case of having to compile most software, if I wanted something that wasn't from last year.
[11:22 AM PT]  Orbert.Tatham @hg.zetaworlds.com: I just learned that there is no such thing as a standard for a bios for ARM boards
[11:22 AM PT]  Cuga.Rajal @rajal.org:9000: I included all 3 universal libs with the mantis
[11:22 AM PT]  Cuga.Rajal @rajal.org:9000: Raspberry Pi
[11:22 AM PT]  Cuga.Rajal @rajal.org:9000: oops
[11:22 AM PT]  Cuga.Rajal @rajal.org:9000: http://opensimulator.org/mantis/view.php?id=9068
[11:22 AM PT]  Ubit Umarov: so the provided openjpeg dyling will replace what?
[11:22 AM PT]  Gavin.Hird @grid.xmir.org:8002: I suspect Microsoft use the special Intel addressing instructions Apple added to make life easier for itself.
[11:23 AM PT]  Cuga.Rajal @rajal.org:9000: to replace the CPU-specific libs that are diff files for CPU s
[11:23 AM PT]  Ubit Umarov: cuga read me! jezz
[11:23 AM PT]  Cuga.Rajal @rajal.org:9000: ?
[11:23 AM PT]  Ubit Umarov: we currently have 4 openjpeg dylibs
[11:23 AM PT]  Ubit Umarov: no idea why.. but we do
[11:23 AM PT]  Andrew Hellershanks: Vincent, they may be ABI compliant but they can also be ARM CPU based but 32-bit vs 64-bit. There are also different versions of ARM which could require a recompile. I deal with some ARM based embedded systems. Some are ARM v7, and some are ARM v8. Programs for one won't run on the other.
[11:24 AM PT]  Cuga.Rajal @rajal.org:9000: currently there is one file for macOS arm64 and another file for macOS Intel
[11:24 AM PT]  Ubit Umarov: the ones in used on *config fiel are the *-dotnet*ones
[11:24 AM PT]  Cuga.Rajal @rajal.org:9000: those are just the names given
[11:25 AM PT]  Cuga.Rajal @rajal.org:9000: nothing special in the file, they are static libs
[11:25 AM PT]  Cuga.Rajal @rajal.org:9000: the new files included with the mantis are all for dotnet6
[11:25 AM PT]  Cuga.Rajal @rajal.org:9000: they just donlt have that in the filename
[11:25 AM PT]  Gavin.Hird @grid.xmir.org:8002: Ubit the names can be collapsed since they are universal so they can be used on both Intel and ARM without any name distinction
[11:25 AM PT]  Ubit Umarov: hmm the dotnet ones are compliled on libopenmetaverse
[11:25 AM PT]  Cuga.Rajal @rajal.org:9000: they should work the same for dotnet6 and mono actually
[11:26 AM PT]  Cuga.Rajal @rajal.org:9000: well maybe openjpeg
[11:26 AM PT]  Cuga.Rajal @rajal.org:9000: well I only intended them to be used for dotnet6
[11:26 AM PT]  Cuga.Rajal @rajal.org:9000: so you can add that to the filename if you want
[11:26 AM PT]  Gavin.Hird @grid.xmir.org:8002: then they mussst be dlls Ubit?
[11:26 AM PT]  Ubit Umarov: hmm do the new ones inlcude the file dotnet.cpp on source?
[11:26 AM PT]  Andrew Hellershanks: Orbert, mostly some whitespace changes that snuck in. My editor automatically converts tabs to spaces and removes trailing whitespace from lines.
[11:27 AM PT]  Cuga.Rajal @rajal.org:9000: I compiled openjpeg from https://bitbucket.org/opensimulator/libopenmetaverse/src/master/openjpeg-dotnet/
[11:28 AM PT]  Ubit Umarov: ok then it is libopenjpeg-dotnet
[11:28 AM PT]  Cuga.Rajal @rajal.org:9000: ok, you can change the filename if that makes it better
[11:28 AM PT]  Ubit Umarov: in that case think so, to make clear it does include that dotnet.cpp thing
[11:28 AM PT]  Cuga.Rajal @rajal.org:9000: I removed -dotnet from the name because I wanted to add the date and universal, and it got too long :)
[11:29 AM PT]  Ubit Umarov: well ill do the oposite
[11:29 AM PT]  Cuga.Rajal @rajal.org:9000: so maybe I shoudl keep -dotnet on that in the future
[11:29 AM PT]  Ubit Umarov: version should coem from the metafata
[11:29 AM PT]  Ubit Umarov: metadata
[11:29 AM PT]  Ubit Umarov: or guessed ;)
[11:30 AM PT]  Cuga.Rajal @rajal.org:9000: I don't know if version numbers are gettign bumped when changes are being made :)
[11:30 AM PT]  Ubit Umarov: do you ppl remember what is the other dylib ?
[11:30 AM PT]  Gavin.Hird @grid.xmir.org:8002: I would not include version numbers as that requires updates of the config files
[11:30 AM PT]  Ubit Umarov: libopenjpeg-arm64.dylib
[11:31 AM PT]  Ubit Umarov: soemthing i created by mistake last time??
[11:31 AM PT]  Cuga.Rajal @rajal.org:9000: correct
[11:31 AM PT]  Cuga.Rajal @rajal.org:9000: that was also built for dotnet
[11:31 AM PT]  Andrew Hellershanks: brb
[11:31 AM PT]  Cuga.Rajal @rajal.org:9000: if could replace that with the universal binary and add -dotnet to the name
[11:31 AM PT]  Ubit Umarov: lwt me blame that thing
[11:32 AM PT]  Gavin.Hird @grid.xmir.org:8002: if the libs for dotnet is distinctly different from the ones for mono, I would consistently add -dotnet to the name for all platforms
[11:32 AM PT]  Ubit Umarov: yeah i did added in dec
[11:33 AM PT]  Cuga.Rajal @rajal.org:9000: yes, I remember it had dotnet in name originally, then during an update it was lost
[11:33 AM PT]  Cuga.Rajal @rajal.org:9000: probably bc I didnt include it in the filename
[11:33 AM PT]  Ubit Umarov: the dotnet are there also
[11:33 AM PT]  Ubit Umarov: well ill remove that
[11:33 AM PT]  Ubit Umarov: and apply the new ones
[11:33 AM PT]  Cuga.Rajal @rajal.org:9000: keep the newer one
[11:33 AM PT]  Ubit Umarov: they you mac ppl test
[11:33 AM PT]  Cuga.Rajal @rajal.org:9000: that has a bug fix
[11:34 AM PT]  Ubit Umarov: and tell what does blowup :)
[11:34 AM PT]  Ubit Umarov: ok ?
[11:34 AM PT]  Ubit Umarov: :)
[11:34 AM PT]  Cuga.Rajal @rajal.org:9000: do you know which one is newer? look at date in filename :)
[11:34 AM PT]  Ubit Umarov: i suposed your 3 are the new ones
[11:35 AM PT]  Cuga.Rajal @rajal.org:9000: the 3 in the mantis? yes
[11:35 AM PT]  Andrew Hellershanks: I'm back.
[11:35 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Speaking of libs, should postgres just be removed now? We don't really have anyone actively maintaining it and it now has like two breaking bugs preventing it from use unless you specifically set things up with the correct versions.
[11:35 AM PT]  Cuga.Rajal @rajal.org:9000: for others I would have to look through old commits to figure out, if no date in filename
[11:36 AM PT]  Gavin.Hird @grid.xmir.org:8002: I would not remove it from the source tree, but I suppose it can be removed from the build project for now?
[11:37 AM PT]  Cuga.Rajal @rajal.org:9000: maybe include something in Readme about versions reqd for postgres
[11:37 AM PT]  Cuga.Rajal @rajal.org:9000: if kept in
[11:39 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Not even sure it works on dotnet at all, on mono it only works with 5.x branch
[11:41 AM PT]  Cuga.Rajal @rajal.org:9000: if totally broken on dotnet6 and no plans to fix, them prolly should remove
[11:41 AM PT]  Gavin.Hird @grid.xmir.org:8002: I tried it on dotnet6 and it sortof started up,  but there was massive console spew when trying to log in, so that never worked,
[11:42 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Probably same issue as is on mono. I tried fixing that and got it booting up as well, but the one thing actually using the guid type it throws up on
[11:42 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Could remove that, would have to write some migration for that, then it should work
[11:42 AM PT]  Gavin.Hird @grid.xmir.org:8002: I would keep the source in the tree, just not bother to build that part of the project. It is most likely fixable
[11:42 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: On mono the connector didn't work. It can be updated, but then requires more dlls along with it to run properly
[11:42 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: Nuget rabbit hole again
[11:45 AM PT]  Andrew Hellershanks: We are at 15 minutes to the top of the hour. Does anyone have another question or comment for today?
[11:46 AM PT]  Cuga.Rajal @rajal.org:9000: Still working on Bullet 3.25 with MisterBlue, a few advances, snails pace
[11:47 AM PT]  Cuga.Rajal @rajal.org:9000: new wrapper and trying to fix some bugs
[11:47 AM PT]  Andrew Hellershanks: Cuga, ok. As long as you are seeing progress that is still good.
[11:47 AM PT]  Cuga.Rajal @rajal.org:9000: he has a lot of irons in the fire
[11:47 AM PT]  Andrew Hellershanks: As many of us do.
[11:47 AM PT]  Gavin.Hird @grid.xmir.org:8002: what are those bugs?
[11:48 AM PT]  Gavin.Hird @grid.xmir.org:8002: segfault when unlinking?
[11:48 AM PT]  Cuga.Rajal @rajal.org:9000: one is a problem that a physics object doesn;t come out of sleep if the surface below it moves, but is not with physics enabled
[11:48 AM PT]  Cuga.Rajal @rajal.org:9000: like for example, of a script moves it
[11:48 AM PT]  Ubit Umarov: https://bitbucket.org/opensimulator/opensim/commits/3ab2ba4f07f1be04b30c4055ee0cd704e5dcf79a
[11:48 AM PT]  Cuga.Rajal @rajal.org:9000: I have a test object to demo that bug
[11:49 AM PT]  Cuga.Rajal @rajal.org:9000: the other issue is that some of the comsole commands to set Bullet parameters are not taking hold
[11:49 AM PT]  Ubit Umarov: (hope that is it.. )
[11:49 AM PT]  Cuga.Rajal @rajal.org:9000: That should be it for now :) thanks
[11:50 AM PT]  Andrew Hellershanks: Cuga, thanks for the update.
[11:52 AM PT]  Andrew Hellershanks: Do the clocks in Europe change this coming weekend?
[11:52 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: We have these Settings items now for the EEP stuff, but there doesn't seem to be a way to download or upload them
[11:52 AM PT]  Gavin.Hird @grid.xmir.org:8002: yes they do
[11:53 AM PT]  Ubit Umarov: cuga that sleep issue is a pain
[11:53 AM PT]  Andrew Hellershanks: ok. Keep that in mind as that will mean you need to be here an hour earlier next week (if I have that right).
[11:53 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: I had a look at the actual asset yesterday and it seems just to be encoded text, so should be easy to write an uploader for it and put it in a viewer
[11:53 AM PT]  Ubit Umarov: main ocntrol code needs to periodly take it out of sleep
[11:53 AM PT]  Ubit Umarov: as ubode does
[11:53 AM PT]  Ubit Umarov: but is a way that it will just probe collisions and sleep again
[11:53 AM PT]  Vincent.Sylvester @hg.zetaworlds.com: An hour later actually, this and last week the meeting was earlier for EU, it will be back to usual times for us next week
[11:54 AM PT]  Cuga.Rajal @rajal.org:9000: MrBlue is basically fixing bugs that should have been fixed by the Bullet team
[11:54 AM PT]  Ubit Umarov: SL did had same issue
[11:54 AM PT]  Cuga.Rajal @rajal.org:9000: ubODE works better than Bullet for that test case
[11:54 AM PT]  Ubit Umarov: is still does not??
[11:54 AM PT]  Ubit Umarov: if..
[11:54 AM PT]  Cuga.Rajal @rajal.org:9000: But it still does get stuck sometimes in that test case
[11:54 AM PT]  Ubit Umarov: think they also did add a periodic awakeup call
[11:55 AM PT]  Ubit Umarov: for those not following..
[11:55 AM PT]  Cuga.Rajal @rajal.org:9000: does ubODE do that?
[11:55 AM PT]  Ubit Umarov: physics engines have a feature to speed up thing
[11:55 AM PT]  Ubit Umarov: if a object does nto colide or move, etc it is placed on sleep mode
[11:55 AM PT]  Ubit Umarov: taken out of active obejcts lists
[11:56 AM PT]  Cuga.Rajal @rajal.org:9000: there are varioius things that will wake it from sleep, like being nudges by the mouse cursor or
[11:56 AM PT]  Ubit Umarov: they will get back in as soon other colides with it for example, ( bit not if stops collidiong )
[11:56 AM PT]  Cuga.Rajal @rajal.org:9000: applying an impulse from a script
[11:56 AM PT]  Ubit Umarov: yes ubode as even more
[11:57 AM PT]  Cuga.Rajal @rajal.org:9000: but a non-physical surface below it changing angles, harder
[11:57 AM PT]  Ubit Umarov: those objects when on a static collision need to keep feeeding collision events to scripts
[11:57 AM PT]  Ubit Umarov: so yeah ubode has loops to do that
[11:57 AM PT]  Ubit Umarov: and to inject the objects bact on collisions probes every 500ms or so
[11:58 AM PT]  Ubit Umarov: we can see the lag oncasionally
[11:58 AM PT]  Cuga.Rajal @rajal.org:9000: Bullet needs to add something like that
[11:58 AM PT]  Cuga.Rajal @rajal.org:9000: it is missing some kind of detection
[11:58 AM PT]  Cuga.Rajal @rajal.org:9000: You know those wood labyrinths, tilt it to make the rolling ball go in the hole
[11:58 AM PT]  Ubit Umarov: ok top is physical
[11:59 AM PT]  Ubit Umarov: lets us let it sleep
[11:59 AM PT]  Cuga.Rajal @rajal.org:9000: I made a giant version of on eof those
[11:59 AM PT]  Cuga.Rajal @rajal.org:9000: good test case
[11:59 AM PT]  Ubit Umarov: ok guess it is sleeping by now...
[11:59 AM PT]  Ubit Umarov: noticed the little lag_
[11:59 AM PT]  Ubit Umarov: noticed ?
[11:59 AM PT]  Gavin.Hird @grid.xmir.org:8002: nope
[12:00 PM PT]  Orbert.Tatham @hg.zetaworlds.com: Not very visible
[12:00 PM PT]  Ubit Umarov: well its suposed to be as short as possible, stull having the gain of sleep
[12:01 PM PT]  Selby.Evans @grid.kitely.com:8002: Bye all
[12:01 PM PT]  Ubit Umarov: cya sleby
[12:01 PM PT]  Ubit Umarov: selby.Evans
[12:01 PM PT]  Gavin.Hird @grid.xmir.org:8002: bye
[12:01 PM PT]  Andrew Hellershanks: bye, Selby. See you again next week.
[12:01 PM PT]  Ubit Umarov: as is said SL had this issue for ages also
[12:02 PM PT]  Cuga.Rajal @rajal.org:9000: Havok has really come a long way
[12:02 PM PT]  Andrew Hellershanks: I have a cat on my lap currently making it a little difficult to type. :)
[12:02 PM PT]  Ubit Umarov: we did remoe the supportng prim, that the other would stay in air for aver
[12:02 PM PT]  Ubit Umarov: even avatars
[12:02 PM PT]  Ubit Umarov: ol going to remove the support one...
[12:02 PM PT]  Gavin.Hird @grid.xmir.org:8002: till you moved
[12:02 PM PT]  Ubit Umarov: seen it?
[12:02 PM PT]  Cuga.Rajal @rajal.org:9000: so it detected the change
[12:02 PM PT]  Gavin.Hird @grid.xmir.org:8002: remember that
[12:02 PM PT]  Ubit Umarov: a but more lag this time
[12:02 PM PT]  Orbert.Tatham @hg.zetaworlds.com: 200  msec delay
[12:03 PM PT]  Cuga.Rajal @rajal.org:9000: so it comes out of sleep momentarily to see if anything has changed?
[12:03 PM PT]  Ubit Umarov: wlel that is a know issue with physics engines
[12:03 PM PT]  Orbert.Tatham @hg.zetaworlds.com: Probably 100 msec or soo plus viewer lag
[12:03 PM PT]  Gavin.Hird @grid.xmir.org:8002: you could stand there 1000 m up even if the floor was pulled under you
[12:03 PM PT]  Ubit Umarov: a pain to work arod :)
[12:03 PM PT]  Ubit Umarov: and is same with avatars
[12:03 PM PT]  Ubit Umarov: if they stand still
[12:04 PM PT]  Gavin.Hird @grid.xmir.org:8002: yep, you did not fall till you moved
[12:04 PM PT]  Cuga.Rajal @rajal.org:9000: I hacked the Bullet code and disabled sleep altogether, and the problem went away
[12:04 PM PT]  Cuga.Rajal @rajal.org:9000: but that's not a good solution
[12:04 PM PT]  Orbert.Tatham @hg.zetaworlds.com: Mr. Wile E. Coyote would feel right at home
[12:04 PM PT]  Cuga.Rajal @rajal.org:9000: oh, you prob need to rez this outside
[12:04 PM PT]  Cuga.Rajal @rajal.org:9000: then sit on it to move
[12:05 PM PT]  Cuga.Rajal @rajal.org:9000: it's a fun game
[12:05 PM PT]  Cuga.Rajal @rajal.org:9000: maybe reset the script too
[12:05 PM PT]  Ubit Umarov: outch.. big thing :p
[12:05 PM PT]  Ubit Umarov: well a test 2 complex for now
[12:05 PM PT]  Ubit Umarov: you see mine
[12:06 PM PT]  Ubit Umarov: on bullet the top prim would stay
[12:06 PM PT]  Andrew Hellershanks: Um... Giant pin ball game.
[12:06 PM PT]  Cuga.Rajal @rajal.org:9000: try moving back out and reset the script forst
[12:06 PM PT]  Ubit Umarov: in air until something did collide with it
[12:07 PM PT]  Cuga.Rajal @rajal.org:9000: move up from ground, it tilts
[12:07 PM PT]  Cuga.Rajal @rajal.org:9000: script reset?
[12:07 PM PT]  Ubit Umarov: uff ok did reset the scripts
[12:07 PM PT]  Cuga.Rajal @rajal.org:9000: ok
[12:09 PM PT]  Ubit Umarov: yes that but sleep matters where there are many things physical
[12:10 PM PT]  Ubit Umarov: in fact bullet over uses it to try to be "fast"
[12:11 PM PT]  Cuga.Rajal @rajal.org:9000: anyone can sit on it and try
[12:11 PM PT]  Cuga.Rajal @rajal.org:9000: but only owner can nudge the ball
[12:12 PM PT]  Cuga.Rajal @rajal.org:9000: the ball gets stuck in sleep when it is in the center
[12:12 PM PT]  Jamie.Jordan @grid.kitely.com:8002: have a good week yall
[12:12 PM PT]  Orbert.Tatham @hg.zetaworlds.com: Peace, Jamie
[12:12 PM PT]  Cuga.Rajal @rajal.org:9000: bc only the tilt changes
[12:13 PM PT]  Cuga.Rajal @rajal.org:9000: in othetr places it seems not to get stuck, at least with ubODE
[12:13 PM PT]  Cuga.Rajal @rajal.org:9000: problem is much worse with Bullet
[12:13 PM PT]  Orbert.Tatham @hg.zetaworlds.com: Changing the tilt also changes the height except in the center
[12:13 PM PT]  Kayaker Magic: RL is calling, I need to go.
[12:13 PM PT]  Cuga.Rajal @rajal.org:9000: that might be why the problem happens in the center
[12:13 PM PT]  Kayaker Magic: Buy all!
[12:14 PM PT]  Cuga.Rajal @rajal.org:9000: tc Kayaker
[12:14 PM PT]  Gavin.Hird @grid.xmir.org:8002: tc
[12:14 PM PT]  Motoko.Karu @login.digiworldz.com:8002: cya kayaker
[12:14 PM PT]  Andrew Hellershanks: ok, Kayaker. See you next week.k
[12:15 PM PT]  Cuga.Rajal @rajal.org:9000: so this seemed to be a good test case for the Bullet bug #1
[12:15 PM PT]  Jagga Meredith: trondheim down.  have to go to daughters; to configure new router
[12:15 PM PT]  Andrew Hellershanks: ok, Jagga.
[12:16 PM PT]  Gavin.Hird @grid.xmir.org:8002: what is trondheim?
[12:16 PM PT]  Andrew Hellershanks: It is about time to wrap up todays meeting as we are now past the hour mark.
[12:16 PM PT]  Ubit Umarov: i added a automatic build for dotnet6 at github also
[12:16 PM PT]  Andrew Hellershanks: ty, Ubit.
[12:16 PM PT]  Ubit Umarov: i cries at irc
[12:16 PM PT]  Ubit Umarov: * osgithub (~osgithub@40.84.170.11) has joined
 opensim/opensim
 dotnet compile: success
  
* osgithub (~osgithub@40.84.170.11) has left
[12:17 PM PT]  Ubit Umarov: https://github.com/opensim/opensim/releases
[12:17 PM PT]  Andrew Hellershanks: Any final comments for today? If not I will close todays gathering.
[12:18 PM PT]  Ubit Umarov: wel i do hope it is a dotnet built etc
[12:18 PM PT]  Ubit Umarov: did not tried it ;)
[12:18 PM PT]  Andrew Hellershanks: :)
[12:19 PM PT]  Motoko.Karu @login.digiworldz.com:8002: till next time... cya all...
[12:19 PM PT]  Ubit Umarov: have fun
[12:19 PM PT]  Andrew Hellershanks: ok, Motoko.
[12:19 PM PT]  Ubit Umarov: hour will change here next sunday
[12:20 PM PT]  Ubit Umarov: so eu ppl should get meetings at the normal local hour
[12:20 PM PT]  Andrew Hellershanks: That will do it for today. Thank you all for coming. Hope to see you all again next week if the clocks changing doesn't cause us to lose a few people.
[12:20 PM PT]  Cuga.Rajal @rajal.org:9000: If they forget, they will be 1 hour early
Personal tools
General
About This Wiki