Chat log from the meeting on 2020-09-15

Latest revision as of 12:43, 15 September 2020

[10:58] Bill Blight: Howdy Andrew
[10:58] Andrew Hellershanks: Hello, Bill.
[10:59] Andrew Hellershanks: I thought I'd surprise you by being early today. :)
[10:59] Bill Blight: HAHAHA
[11:00] Andrew Hellershanks: Hello, Ada
[11:01] Ada Radius: Hi Andrew, Hi Ubit, Bill
[11:01] Ubit Umarov: hi
[11:01] Ada Radius: Hi Kayaker
[11:01] Ubit Umarov: hi kayaker
[11:01] Bill Blight waves
[11:01] Andrew Hellershanks: Hello, Kayaker.
[11:02] Kayaker Magic: Hi Ada, Ubit, Bill, Andrew
[11:02] Kayaker Magic: That's everyone so far....
[11:02] Ada Radius: @Ubit THANK YOU for telling me last week about what the server software does to rez avatars. Turned out very helpful, when I was trying to recruit help for the viewer stuff Kayaker and I are working on.
[11:02] Ubit Umarov: so.. was a nice meeting.. see you all next week
[11:03] Ubit Umarov: ;)
[11:03] Bill Blight: :P
[11:03] Kayaker Magic: Quitting while you are ahead?
[11:03] Ubit Umarov: was a nice try u can't deny :p
[11:03] Chat Logger: Meeting chat logging has been enabled.
[11:04] Ada Radius: want to get into the log: @Ubit THANK YOU for telling me last week about what the server software does to rez avatars. Turned out very helpful, when I was trying to recruit help for the viewer stuff Kayaker and I are working on.
[11:04] Ubit Umarov: :)
[11:04] Ubit Umarov: hi gavin.Hird
[11:05] Ada Radius: Hi Gavin
[11:05] Andrew Hellershanks: Hello, Gavin.
[11:05] Gavin.Hird @grid.xmir.org:8002: Hi everyone
[11:05] Bill Blight: Howdy Gavin
[11:07] Andrew Hellershanks: Who wants to kick things off this week?
[11:07] Ubit Umarov: wel about code, there where just a few bug fixes
[11:07] Ada Radius: anyway, we recruited Iain McCracken AKA Sodovan Torok iSL, who said stuff like "interesting" and "Python is a nice little language" and "want to get my knuckles on the ground again".
[11:07] Bill Blight: I guess it depends on who we are kicking?
[11:07] Andrew Hellershanks: hehe
[11:08] Andrew Hellershanks: Bill, We could just be kicking off our shoes. ;)
[11:08] Bill Blight: Thought I smelled something
[11:08] Andrew Hellershanks: hehe
[11:08] Ada Radius: kicking a catnip ball around
[11:09] Andrew Hellershanks: Ada, You and Kayaker are going to be introducing some Python in to the viewer?
[11:09] Kayaker Magic: I'm still looking for a way to cause that heap problem with YEngine. If I put one copy of a big script out, it sits there for months with no errors. If I put out 8 copies of it, I get heap errors after a few hours. I'm looking for a smaller script that demonstrates the same problem.
[11:09] Ada Radius: no, we need a dae exporter in Blender that works better than the current Avastar
[11:09] Andrew Hellershanks: ah, ok
[11:10] Kayaker Magic: And python is the scripting language of Blender.
[11:10] Ada Radius: not ready to actually have that yet, first need to figure out what to do about teh antique Maya volume bones
[11:10] Andrew Hellershanks nods to Kayaker
[11:10] Leighton.Marjoram @youcantfixstupid.wtf: Hi everyone
[11:10] Ada Radius: lots more research.
[11:10] Ada Radius: Hi Leighton
[11:10] Andrew Hellershanks: Hello, Leighton
[11:10] Gavin.Hird @grid.xmir.org:8002: Hi Leighton
[11:10] Bill Blight: Howdy Leighton
[11:11] Object: Script running
[11:12] Andrew Hellershanks: Ada, Kayaker, good luck with your viewer related work.
[11:13] Ada Radius: ty!
[11:13] Andrew Hellershanks: I also think of Python as a nice little language. I haven't had much time to do much with it.
[11:14] Kayaker Magic: Yeah, I wish I was more proficient in Python.
[11:14] Ubit Umarov: not sure if little applies to ti
[11:14] Ubit Umarov: and it
[11:14] Gavin.Hird @grid.xmir.org:8002: not particularly little any more (if it ever was)
[11:15] Andrew Hellershanks: hehe, probably not. The language might be considered "little" but not if you include a lot of the libraries that come with it.
[11:15] Andrew Hellershanks: On top of that is the whole P vs NP thing when it comes to writing in Python.
[11:16] Gavin.Hird @grid.xmir.org:8002: viewers are still built using version 2.7 which is a bit outdated
[11:16] Andrew Hellershanks: Hello, steevithak
[11:16] Ada Radius: Hi steevithak
[11:16] Gavin.Hird @grid.xmir.org:8002: Hi steevithak
[11:17] steevithak x: howdy
[11:17] Leighton.Marjoram @youcantfixstupid.wtf: Hi steevithak
[11:18] Andrew Hellershanks: Gavin, really? Hm... I didn't know Python was involved in the build process. That needs to change soon. Distros are trying to phase out the 2.7 version of Python
[11:18] Gavin.Hird @grid.xmir.org:8002: autobuild is python
[11:18] Gavin.Hird @grid.xmir.org:8002: without which the viewer will be even more of a pint to build
[11:18] Gavin.Hird @grid.xmir.org:8002: pain
[11:18] Andrew Hellershanks: ok. There are so many different build systems are available.
[11:18] Ada Radius: I did not know that. Thank you!
[11:18] Gavin.Hird @grid.xmir.org:8002: also the viewer manager for the SL viewer is completely built in python
[11:19] Gavin.Hird @grid.xmir.org:8002: plus there are a load of support scripts like the manifest script that actually makes the installer that is python
[11:20] Gavin.Hird @grid.xmir.org:8002: there are also a number of test scripts developed for the avatar
[11:20] Ada Radius: where are they?
[11:20] Gavin.Hird @grid.xmir.org:8002: which exist in the script subfolder of the viewer repositories
[11:20] Kayaker Magic: Python is a batch language for shell scripts in many systems. Better they should use Python than (shudder) Pearl!
[11:21] Ada Radius: ah ty
[11:21] Ubit Umarov: think WoT was made on it
[11:21] Ubit Umarov: even has jit or something
[11:21] Gavin.Hird @grid.xmir.org:8002: even more in indra/python
[11:22] Andrew Hellershanks: Hey, Selby.
[11:22] Ada Radius: Hi Selby
[11:22] Leighton.Marjoram @youcantfixstupid.wtf: Hello Selby
[11:22] Gavin.Hird @grid.xmir.org:8002: https://bitbucket.org/dayturn/dayturn-viewer/src/64-bit-clean/scripts/
[11:22] Selby.Evans @grid.kitely.com:8002: hi everyone
[11:22] Ubit Umarov: wlel wot now seems C
[11:22] Gavin.Hird @grid.xmir.org:8002: https://bitbucket.org/dayturn/dayturn-viewer/src/64-bit-clean/indra/lib/python/
[11:23] Ubit Umarov: think blender also did move things from python to c
[11:23] Ubit Umarov: hi selby.Evans
[11:23] Ubit Umarov: well c++
[11:23] Kayaker Magic: It's a good progression: Develop something in a scripting language on top of a big app, then migrate the common script into the main system.
[11:26] Ada Radius: thanks Gavin
[11:26] Gavin.Hird @grid.xmir.org:8002: yw
[11:27] Gavin.Hird @grid.xmir.org:8002: there are even more in the ll source tree but many of them are used to analyze their traffic and crash data, which I don't collect
[11:28] Kayaker Magic: Gavin, what are the issues that make building the viewer a pain?
[11:28] Gavin.Hird @grid.xmir.org:8002: ha!
[11:28] Gavin.Hird @grid.xmir.org:8002: glad you asked :-))
[11:28] Ubit Umarov: :)
[11:28] Andrew Hellershanks: :)
[11:28] Gavin.Hird @grid.xmir.org:8002: you have a small detail such as all the libraries
[11:28] Gavin.Hird @grid.xmir.org:8002: minor detial
[11:28] Andrew Hellershanks nods
[11:29] Gavin.Hird @grid.xmir.org:8002: then the fact that OpenGL 2.1 is still used
[11:29] Gavin.Hird @grid.xmir.org:8002: makes the renderer almost impossible to change
[11:29] Kayaker Magic: So there are a bunch of libraries that come from many different open source projects?
[11:29] Gavin.Hird @grid.xmir.org:8002: ... and work on newer hardware
[11:30] Bill Blight: It's more of a pain to setup the build environment than actually make changes to the viewer
[11:30] Gavin.Hird @grid.xmir.org:8002: they originate there but most are modified by LL
[11:30] Gavin.Hird @grid.xmir.org:8002: some are also very old and have not been updated for ages
[11:30] Andrew Hellershanks: It took me some time before I could get the viewer to build for the first time. What helped, for me, was the information provided by Singularity for building the viewer.
[11:31] Gavin.Hird @grid.xmir.org:8002: it is symptomatic that the Cool VL viewer reverted top openjpeg 1.4
[11:31] Gavin.Hird @grid.xmir.org:8002: because it works better
[11:31] Gavin.Hird @grid.xmir.org:8002: with the LL code that is
[11:32] Gavin.Hird @grid.xmir.org:8002: right Andrew, but you probably used the libs they or ll provided?
[11:33] Andrew Hellershanks: Right. The build process downloaded a bunch of libraries/dependencies.
[11:33] Gavin.Hird @grid.xmir.org:8002: exactly
[11:33] Gavin.Hird @grid.xmir.org:8002: and autobuild sorts out that
[11:34] Gavin.Hird @grid.xmir.org:8002: most of the time, but not always haha
[11:34] Andrew Hellershanks: :)
[11:36] Kayaker Magic: I recall someone saying the Viewer requires certain versions of some build tools.
[11:36] Gavin.Hird @grid.xmir.org:8002: there is a laundry list of tools and versions needed
[11:37] Kayaker Magic: What operating system is best for building on? WindCows?
[11:37] Gavin.Hird @grid.xmir.org:8002: you have to build it on the target system
[11:38] Gavin.Hird @grid.xmir.org:8002: macOS builds only on macOS, Windows only on Windows, Linux only on Linux
[11:38] Gavin.Hird @grid.xmir.org:8002: I have two of those environments running
[11:38] Kayaker Magic: Any of those 3 easier to work with?
[11:38] Gavin.Hird @grid.xmir.org:8002: no
[11:38] Gavin.Hird @grid.xmir.org:8002: :-)
[11:38] Bill Blight: they all suck
[11:39] Bill Blight: when it comes to setting up the build environment
[11:39] Gavin.Hird @grid.xmir.org:8002: the Windows version use cygwin
[11:39] Gavin.Hird @grid.xmir.org:8002: plus the ll tools
[11:39] Bill Blight: cygwin will end up being your best friend
[11:39] Gavin.Hird @grid.xmir.org:8002: yep
[11:40] Bill Blight: piece of advice, when you get your build environment working, save it off to a removable drive, and store it in a safe ... LOL
[11:40] Andrew Hellershanks: :)
[11:40] Gavin.Hird @grid.xmir.org:8002: you got that one right
[11:41] Andrew Hellershanks: It took a while for me to get the build environment set up. Since then it has been relatively simple to build a viewer.
[11:42] Gavin.Hird @grid.xmir.org:8002: it is relatively simple when it works
[11:42] Kayaker Magic: I'm hoping Ada will find someone else (not me) to do viewer builds.
[11:42] Gavin.Hird @grid.xmir.org:8002: but then they make changes that makes it all fall over
[11:42] Ada Radius: Working on it.
[11:42] Bill Blight: when it works is the issue .. When it doesn't it can be a an adventure, or when a lib gets updated ..
[11:42] Ada Radius: If you give me a list of the stuff you don't want to do, I can shop that around and see if there's anyone who needs to buff up their resumes
[11:43] Kayaker Magic: LOL
[11:43] Gavin.Hird @grid.xmir.org:8002: like boost
[11:43] Andrew Hellershanks: I've been lucky(?). Since I got the build environment setup I haven't had it fall over after updates.
[11:43] Gavin.Hird @grid.xmir.org:8002: everything brekas
[11:43] Kayaker Magic: Please tell me everything you don't know!
[11:43] Gavin.Hird @grid.xmir.org:8002: breaks even
[11:44] Gavin.Hird @grid.xmir.org:8002: I don't think building this viewer is a bonus for anyones resume, hehe
[11:44] Gavin.Hird @grid.xmir.org:8002: the skills are almost non-transferable
[11:44] Bill Blight: for me the worst part is when FMOD gets updated, it is a pain to package that "correctly"
[11:44] Andrew Hellershanks: Could be. Shows you don't give up easily when confronted with problems. :)
[11:44] Gavin.Hird @grid.xmir.org:8002: to modern development
[11:44] Bill Blight: even when you do it right it does not want to work most the time
[11:45] Ada Radius: ya, phrased right could look good on a resume, if not very transferrable.
[11:45] Ada Radius: Kayaker and I are retired and don't care about resumes, but I keep meeting younger programmers who are getting squeezed by international competition, everyone working from home, who could be interested
[11:45] Gavin.Hird @grid.xmir.org:8002: FMOD was not impossible, but SL voice on Windows...
[11:46] Andrew Hellershanks: I've always thought it was quite a non-standard type of build system. I sometimes wondered how it came about.
[11:46] Bill Blight: I never had an issue with the voice package, FMOD was the one that always seemed to go sideways when building a 64 bit viewer
[11:47] Gavin.Hird @grid.xmir.org:8002: it comes out of building something completely non-standard for 3 different operating systems
[11:48] Gavin.Hird @grid.xmir.org:8002: ... and trying to keep some commonality
[11:49] Bill Blight: Need to bail early, see you all next week
[11:49] Gavin.Hird @grid.xmir.org:8002: cheers Bill
[11:49] Ada Radius: Bye Bill tc
[11:49] Leighton.Marjoram @youcantfixstupid.wtf: tc Bill
[11:51] Ada Radius: I'm coming at this from a much different direction, as I can't code (tho getting better at reading the human language xml, dae, python etc.)
[11:52] Ada Radius: from how it works creating the data that will be used - armature bone positions, vertex weights, topography and so forth
[11:53] Ada Radius: if it all ends up in the llm files, and the server is scraping those up from each viewer in a sim to rez out in a sim, then there's a path, maybe to fixing the mess.
[11:54] Ada Radius: long term
[11:54] Ubit Umarov: like i said last week, server has nothing to do with those llm
[11:54] Ubit Umarov: those are local to viewers
[11:55] Ada Radius: no, but it uses them to show my viewer what your viewer knows about your avatar
[11:55] Ubit Umarov: about ruth avatar server only gets a array of bytes it sends to other viewers
[11:56] Ada Radius: yes
[11:56] Ubit Umarov: the values of some sliders
[11:56] Ubit Umarov: thats all
[11:56] Ada Radius: morphs, attachments, joint edits, yada yada
[11:56] Ubit Umarov: no
[11:56] Ubit Umarov: nothing
[11:56] Ubit Umarov: all that is viewers only
[11:56] Ada Radius: there is a baking process in there I don't understand. I figure if I keep asking the question I'll get there
[11:57] Ubit Umarov: server does get also a list of textures
[11:57] Ada Radius: yes
[11:58] Ubit Umarov: those actually come from the wearables
[11:59] Ada Radius: so the viewer takes bones arranged into armatures, morphs, including live animation morphs, and topography of both the default av and attachments and their morphs, and bakes them down into the llm files that the server reads, along with the textures that come with.
[12:00] Gavin.Hird @grid.xmir.org:8002: right, because the sliders for say system pants modfies the avatar mesh
[12:00] Ada Radius: yes
[12:00] Ubit Umarov: llm are input files
[12:00] Ada Radius: input to the server
[12:00] Ada Radius: yes
[12:00] Ubit Umarov: NO
[12:00] Ada Radius: ?
[12:00] Ubit Umarov: input to the viewer
[12:00] Ada Radius: output to the viewer?
[12:00] Gavin.Hird @grid.xmir.org:8002: server is just storage more or less
[12:00] Ada Radius: hm
[12:00] Ubit Umarov: they are the base points
[12:01] Ubit Umarov: sliders deform those to the the final mesh
[12:01] Ada Radius: ah, I see what you're saying. OK. input from the Character folder in the viewer to the rest of the viewer processing
[12:02] Ubit Umarov: some silders via bones and weigths
[12:02] Ubit Umarov: other via special morphs
[12:02] Ada Radius: then the server takes the llm data, including real time animations and animation morphs, and spits it out so everyone else can see it.
[12:02] Ubit Umarov whispers: NO
[12:02] Ada Radius: then what?
[12:02] Gavin.Hird @grid.xmir.org:8002: so the input from the character folder must be the same for all viewers or you won't see the same
[12:02] Ubit Umarov: server NEVER seens any llm
[12:02] Ada Radius: then how does it show my viewer your avatar?
[12:03] Ubit Umarov: all viewers have those
[12:03] Ubit Umarov: that make the final mesh using that array of bytes
[12:03] Ubit Umarov: that contains the sliders values
[12:03] Kayaker Magic: The server sends just the shape data to the viewr, and the viewer draws the system avatar for each person
[12:03] Ada Radius: ya, which file is it looking at to get that?
[12:03] Ada Radius: files
[12:04] Ubit Umarov: those xml files
[12:04] Gavin.Hird @grid.xmir.org:8002: it takes a parameter list stored on the server for each avatar clothing and applies it to the defaults already known by the viewer
[12:04] Ada Radius: ah OK
[12:04] Gavin.Hird @grid.xmir.org:8002: so the server does not build anything for the avatar
[12:04] Gavin.Hird @grid.xmir.org:8002: the viewer does
[12:04] Ubit Umarov: mesh part is only visual parameters
[12:04] Gavin.Hird @grid.xmir.org:8002: which is why it lages with more and more cvatars in the scene
[12:04] Ubit Umarov: nothing server does nothing
[12:05] Gavin.Hird @grid.xmir.org:8002: and the capability depends on the machine the viewer runs on
[12:05] Ada Radius: I get that the server is not building the avatar. But it is rezzing it or I wouldn't see other people's viewer output
[12:05] Ubit Umarov: gets your visual parameters and sends to me
[12:05] Gavin.Hird @grid.xmir.org:8002: rezzing just defines where it is in the scene really
[12:05] Ada Radius: do you know which files in the viewer the server reads?
[12:05] Ubit Umarov: my viewer has same llm files and rules to builds it
[12:05] Ubit Umarov: NONE
[12:06] Ada Radius: that makes no sense, or we're having a language mismatch
[12:06] Ubit Umarov: even the meaning of the visual parameters is kind of a mistery
[12:06] Ubit Umarov: all viewers need to agree on them
[12:06] Gavin.Hird @grid.xmir.org:8002: server does not do any processing on the avatar but positioning in the scene and collision
[12:06] Kayaker Magic: Ada, recall when I modified the LLM files to add horns to the default avatar? I saw horns on everyone, but nobody else saw them.
[12:06] Leighton.Marjoram @youcantfixstupid.wtf: If I understand properly the viewer is doing the rendering from the data stream coming from my viewer to yours. That data stream contains data/information but not asset files
[12:07] Kayaker Magic: My viewer had horns, theirs did not. the LLM file never leaves the viewer, never seen by the server.
[12:07] Gavin.Hird @grid.xmir.org:8002: exactly Kayaker
[12:07] Gavin.Hird @grid.xmir.org:8002: whihc is why every viewer must have the same files
[12:07] Gavin.Hird @grid.xmir.org:8002: or it cannot display your avatar correctly
[12:07] Ada Radius: @Kayaker - yes. Because all the viewers have to have the same basic parameters. I want to know which files data get transferred
[12:08] Ubit Umarov: actually is a bt more complex even lol
[12:08] Ada Radius: OK
[12:08] Ubit Umarov: yr viewer gets the list of things it is wearing from COF
[12:08] Ada Radius: COF?
[12:08] Ubit Umarov: current outfit
[12:08] Ada Radius: ah yes
[12:08] Leighton.Marjoram @youcantfixstupid.wtf: I have to head off see you all next week :)
[12:09] Ubit Umarov: body contains a lot of silders
[12:09] Ubit Umarov: etc
[12:09] Ada Radius: Bye Leighton, tc
[12:09] Andrew Hellershanks: ok, Leighton. See you next week.
[12:09] Ubit Umarov: each werable has things..
[12:09] Gavin.Hird @grid.xmir.org:8002: so for the Bento project, the new files has a subset equal to the old system avatar making it possible to rneder both with the Bento and the default rigging
[12:09] Ada Radius: yes, sliders, attachments, and inworld animation morphs.
[12:09] Gavin.Hird @grid.xmir.org:8002: + animate the rigs correctly
[12:09] Ubit Umarov: physics has the bouncing boobs parameters etc
[12:09] Ada Radius: ya
[12:10] Ubit Umarov: from those, it creates that array of bytes
[12:10] Ubit Umarov: and sends to server
[12:10] Ubit Umarov: other viewers use that not the wearables
[12:10] Ada Radius: where is that array of bytes stored?
[12:10] Ubit Umarov: server
[12:10] Ubit Umarov: well temporary
[12:10] Ada Radius: (forgive me, I have a simple and literal mind)
[12:11] Ubit Umarov: your viewers sends yours to the region on each tp
[12:11] Ubit Umarov: region then sends to others
[12:11] Ada Radius: yes. the region = the server side?
[12:11] Ubit Umarov: so provcessing your avatar is diferent from the others
[12:11] Ubit Umarov: bc it does compact data a lot
[12:12] Ubit Umarov: same for textures
[12:12] Ubit Umarov: viewer gets them from the wearables
[12:12] Ada Radius: yes - some stuff lost in translation, easy to see if you're the designer
[12:12] Ubit Umarov: and using special rules creates the baked ones
[12:12] Ada Radius: in other aps, called "baking"
[12:12] Ubit Umarov: that it sends to regjon
[12:12] Ubit Umarov: and then region sends to others
[12:13] Ubit Umarov: so there are 2 diferent avatar rendering..
[12:13] Ubit Umarov: own, and others
[12:13] Ada Radius: yes
[12:13] Ada Radius: the lag between the two varies, I can see that
[12:14] Ubit Umarov: own processing is compact all information
[12:14] Ubit Umarov: in baked textures.. visual parameters.. etc
[12:15] Ada Radius: baked textures, baked topographies (attachments +sliders), textures shipped separately and recombined
[12:15] Ada Radius: no, baked in the viewer, and the total sent
[12:15] Ubit Umarov: attachments are aside, or where
[12:15] Ubit Umarov: those are objects assets
[12:16] Ada Radius: the system av is also an object asset - a topography
[12:16] Gavin.Hird @grid.xmir.org:8002: a region has two parts - one persistent across server restarts and one temporary which deals with avatars and movement which only exist as long as the rgion is running
[12:16] Ada Radius: ah OK
[12:17] Ubit Umarov: persistent parts is called inventory :p
[12:17] Gavin.Hird @grid.xmir.org:8002: yep
[12:17] Ubit Umarov: COF the main big thing now
[12:17] Ada Radius: and whatever's rezzed out on the sim?
[12:17] Gavin.Hird @grid.xmir.org:8002: yes mainly stored in the prims table
[12:17] Ubit Umarov: ( in past there where messages abotu wearables etc..
[12:17] Ada Radius: COF - so I need to figure out where the COF is stored in the viewer and how that array of bytes gets to the region
[12:18] Ubit Umarov: now viewer mostly only uses COF as far i know
[12:18] Ubit Umarov: cames from wearables, body a lot of them
[12:18] Ada Radius: and I need to know, because that profoundly affects how to build a better system avatar.
[12:18] Ubit Umarov: mapped by avatar_lad-xml or simlar
[12:19] Ada Radius: avatar_skeleton.xml, avatar_lad.xml - there are half a dozen of them
[12:19] Ubit Umarov: avatar_lad.xml is the source of all evil :p
[12:19] Ada Radius: I believe that
[12:19] Ada Radius: I stuck it all in a spreadsheet so I could see it better
[12:20] Ubit Umarov: the "visual parameters " are a subset of all those parameters
[12:20] Ada Radius: as I did with avatar_skeleton.xml, which is when I realized how sloppy LL was with the armature
[12:20] Ubit Umarov: don't remember the conditions
[12:20] Ada Radius: OK. Visual parameters = more or less baked COF
[12:21] Ubit Umarov: slider values
[12:21] Ada Radius: ya - morphs.
[12:21] Ada Radius: or textures
[12:21] Ubit Umarov: and bones..
[12:21] Ubit Umarov: no
[12:21] Ubit Umarov: textures are the baked ones
[12:21] Gavin.Hird @grid.xmir.org:8002: slider values is also bone deformations
[12:21] Ada Radius: there are texture sliders inworld - no one uses them anymore, but I remember from 2006
[12:22] Ubit Umarov: bom uses those again i guess
[12:22] Ada Radius: yes. bone deforms for the mBones, morphs for the volume bones. Something I'd like to see improved. The Bento armature does it better but the older bits were never upgraded
[12:22] Ubit Umarov: gavin did broke his hair sliders
[12:23] Ada Radius: haaa
[12:23] Gavin.Hird @grid.xmir.org:8002: did I?
[12:23] Ubit Umarov: oops.. no no.. your hair is fine *glups*
[12:24] Ada Radius: so among the things to find out: where is that array that the region reads from the viewer, and how.
[12:24] Gavin.Hird @grid.xmir.org:8002: I'd like to remove those hair sliders
[12:24] Gavin.Hird @grid.xmir.org:8002: and the hair mesh
[12:24] Ubit Umarov: :p
[12:24] Kayaker Magic: Then what would Ubit wear on his head?
[12:24] Ada Radius: I think I could make a better hair that deformed more simply - longer shorter. I know I can, I just don't know what the coding would look like
[12:25] Ada Radius: and better texturing of course, that would shift color better
[12:25] Ubit Umarov: uff and ppl tell mesh avatars are more modern
[12:25] Gavin.Hird @grid.xmir.org:8002: the problem is to do it in a way that don't break all content
[12:25] Ubit Umarov: No RUTH is a lot more complete thing
[12:25] Ubit Umarov: mesh avatars are crap
[12:25] Ada Radius: same with default skin. So people wouldn't need mesh attachments like Ruth
[12:25] Ada Radius: yes they are
[12:26] Ubit Umarov: hacks on top of hack
[12:26] Ubit Umarov: gezz
[12:26] Ada Radius: when I realized that, I finished off work on Ruth and Roth, which at least are a lot more resource efficient than the commercial bodies, and started on R3, to see if there's a better way
[12:27] Ubit Umarov: ll just fallen on that hacks on top of hacks
[12:27] Ada Radius: because as long as the system av is horrible, and it is, people will wear horrors of hack jobs to look sexy
[12:27] Ubit Umarov: to let mesh clothes to work better
[12:28] Kayaker Magic: R3 means "can we make a better system avatar" and then can we make the viewer devs accept it?
[12:28] Ubit Umarov: "oh why not the all avatar".. and the path
[12:28] Ubit Umarov: hack on top of hack
[12:28] Ubit Umarov: ofc means $L so fine for them
[12:28] Ada Radius: so you can help, by making us understand how the region reads the viewers and dumps the data out for the other viewers to see
[12:29] Ubit Umarov: i basicly already told that :)
[12:29] Gavin.Hird @grid.xmir.org:8002: :-)
[12:29] Ada Radius: I think what I need to know is: where is that data array, which files does it come from
[12:29] Ubit Umarov: aks gavin.Hird
[12:29] Ubit Umarov: lol
[12:29] Ubit Umarov: ask
[12:30] Ubit Umarov: thats all on viewer :)
[12:30] Ada Radius: so far I'm getting "it's a mystery" and "spaghetti", but I do understand a lot more than I did a few months ago
[12:30] Ubit Umarov: well it is hard to read yes
[12:30] Gavin.Hird @grid.xmir.org:8002: and it is one aspect of the viewer that is almost cast in stone
[12:30] Gavin.Hird @grid.xmir.org:8002: because if you start touching it, all existing content will break
[12:31] Ubit Umarov: but is is a high level thing
[12:31] Ada Radius: it's data.
[12:31] Gavin.Hird @grid.xmir.org:8002: unless you put a major effort into backward compatibility like they did with Bento
[12:31] Ubit Umarov: they added bento with minor changes that i remembe
[12:31] Ubit Umarov: they added things to the xml, the viewer code al ready done did it
[12:31] Gavin.Hird @grid.xmir.org:8002: it was not minor :-))
[12:31] Ubit Umarov: same now with the BOM
[12:31] Gavin.Hird @grid.xmir.org:8002: we went from some 27 bones to 148
[12:31] Kayaker Magic: I assumed there was a packet that arrived that included the same data as in a "shape". A pile of numbers.
[12:31] Ada Radius: Bento didn't do backward compatibility. They slapped more bones on and changed the avatar height
[12:31] Ubit Umarov: yes
[12:32] Gavin.Hird @grid.xmir.org:8002: of course it did
[12:32] Ubit Umarov: but most was adding things to the xml files
[12:32] Ada Radius: but the volume bones and the original bones have the same errors they always did
[12:32] Ubit Umarov: the viewer code was done very hgh level and flexible
[12:32] Gavin.Hird @grid.xmir.org:8002: how do you think youi still can see content rigged with the old rigging in the bneto viewer?
[12:32] Ubit Umarov: you add more bones on the xml files.. viewer just does them
[12:33] Megan Cline is online.
[12:34] Ada Radius: if they had hired someone, anyone who can read numbers, to do QC, they would have fixed the pos values. Or gotten rid of them entirely
[12:34] Ada Radius: and fixed the errors in the pivot numbers
[12:34] Gavin.Hird @grid.xmir.org:8002: does not occur to you ti was required to keep backward compatibility?
[12:34] Ada Radius: but no, they added an armature - not a bad one, but it doesn't fit the topology of the avatar very well - bone lengths in the face are a nightmare.
[12:35] Ubit Umarov: and other viewers need to show.. something :)
[12:35] Ubit Umarov: well bento was a big breack on that lol
[12:35] Megan Cline is offline.
[12:35] Gavin.Hird @grid.xmir.org:8002: just fixing numbers will make old stuff deform unpredictably or animate unpredictably
[12:35] Ubit Umarov: bento on a non bento viewer was .. fun to see :)
[12:35] Ada Radius: I'm more interested in future compatibilty. I know about whiners who want to use their 2005 faves.
[12:35] Gavin.Hird @grid.xmir.org:8002: good luck with that Ada
[12:36] Gavin.Hird @grid.xmir.org:8002: that was exactly waht Sansar was all about and nobody swallowed it
[12:36] Ubit Umarov: el bento was a breaking change
[12:36] Ada Radius: and I know how to bring stuff forward - Selby's avatar in OS was brought in from SL, and looks nearly the same here as there.
[12:36] Gavin.Hird @grid.xmir.org:8002: so does mine
[12:36] Ada Radius: Sansar did suk.
[12:37] Ubit Umarov: i didnt bother copy all the numbers :p
[12:37] Gavin.Hird @grid.xmir.org:8002: the shape I have here is an exact copy of my shape in SL
[12:38] Ubit Umarov: http://wiki.secondlife.com/wiki/Avatar_Appearance
[12:38] Ada Radius: Sansar was weird and creepy. The 3DCG only marginally better, but the same company philosophy of tacking things on top of things - like bringing in Marvelous Designer without addressing the obvious and inevitable disasters that that would cause
[12:38] Ubit Umarov: ( the protocol part is a bit outdated )
[12:38] Ada Radius: duh
[12:39] Ubit Umarov: but that still explains a lot
[12:39] Ada Radius: sorry, I get frustrated.
[12:39] Gavin.Hird @grid.xmir.org:8002: In any case I am not sure how much viewer development I will be able to do this fall
[12:39] Andrew Hellershanks: We have been going an hour and a half. Time to wrap up the meeting for today. The avatar discussion could be going on for a while. I need to get going.
[12:39] Ubit Umarov: seen that url ada ?
[12:40] Ada Radius: thank you for all your help! I do get a little further along each time. @Ubit, yes, I have that page pretty much memorized for years
[12:40] Gavin.Hird @grid.xmir.org:8002: as I have a medical condition that too a turn for the worse last week, so they wanted to put me in the hospital on Saturday but I said hold your horses for a few more days
[12:40] Ubit Umarov: wel it was the source of all :)
[12:40] Ada Radius: oh no Gavin.
[12:40] Ubit Umarov: they did extend that a bit
[12:40] Andrew Hellershanks: Gavin, sorry to hear that. Hope it isn't too serious and that you will recover soon.
[12:40] Ubit Umarov: but basic is there
[12:40] Ada Radius: take care of yourself, I have spent too much time in hospitals myself. No fun at all
[12:41] Andrew Hellershanks: On that note I say thank you all for coming. See you again next week.
[12:41] Gavin.Hird @grid.xmir.org:8002: it is an autoimmune thing and there is nothing that can be fundamentally done, so we shall see