Mesh/Archive/2012-03-19

Agenda

I know the collision box is an elipsoid shaped sphere. But, I don't understand how it related to the Collision Skeleton. What is the avatar collision skeleton found in Develop->Avatar->Show Collision Skeleton? Nalates Urriah 14:41, 17 March 2012 (PDT)
Mesh import formats roadmap and need to update import mesh formats already in use (DAE only at the moment)
Status of the usergroup
Status of the mesh clothing parametric deformer
How to cut down your prim equivalency cost? Are people supposed to cut repeating parts as separate mesh objects and then copy+link them in world? Also cut simple mesh parts that could be replaced with prims? Because this is what it looks like when i was creating a medieval tower. FadeOut Razorfen
Any mesh stats? Nalates Urriah
Will we be getting hinges back? CrystalShard Foo
When non attached rigged mesh is going to be scheduled ? FadeOut Razorfen
Thank you Nyx for getting the UG back on the UG list. Nalates Urriah

Meeting Notes

Transcript for Monday March 19, 2012

[12:01]	Nyx Linden	greetings all
[12:01]	–ü–æ—Ä–≤—É –ü—É—Ç–∏–Ω–∞ –∑–∞ –æ—á–∫–æ (fadeout.razorfen)	yea, but BVH doesn't have those bones animated
[12:01]	CrystalShard Foo	But your substring implementation looks right in a fast look
[12:01]	Yuzuru (yuzuru.jewell)	Hello, Nalates.
[12:01]	Yuzuru (yuzuru.jewell)	Hello, Nyx.
[12:02]	jr (jr4201.kruyschek)	greetings Vir
[12:02]	Koli Melune (sahkolihaa.contepomi)	Hey Vir.
[12:02]	CrystalShard Foo	Heys
[12:02]	DrFran Babcock	greetings all
[12:02]	Vir Linden	Howdy folks
[12:02]	DrFran Babcock	hope I don't sit on anyone
[12:02]	Vincent Nacon	naa you're good
[12:02]	Yuzuru (yuzuru.jewell)	Hello, Vir.
[12:03]	Nyx Linden	we don't have many topics yet today: http://wiki.secondlife.com/wiki/Content_Creation/Mesh_Import_User_Group
[12:03]	CrystalShard Foo	I'm guessing it's too late to add?
[12:03]	draconis.neurocam (draconis.neurocam)	no
[12:04]	Nyx Linden	nope, feel free to add more
[12:04]	CrystalShard Foo	What do we call the stretchy mesh thing again?
[12:04]	draconis.neurocam (draconis.neurocam)	rigged mesh
[12:04]	Vincent Nacon	some people don't bother using the wiki as much anymore, thanks to Andrew and Simon Linden for not enforcing it
[12:04]	Vincent Nacon	and Kelly I think
[12:04]	CrystalShard Foo	Not rigged mesh, the stretchy one that goes around body fat and so on
[12:04]	CrystalShard Foo	Rather than just bone
[12:04]	CrystalShard Foo	mesh deforms
[12:04]	CrystalShard Foo	smething
[12:05]	draconis.neurocam (draconis.neurocam)	the parametric deformer
[12:05]	Nyx Linden	alright let's jump into it
[12:05]	CrystalShard Foo	That'sthe one, thanks
[12:05]	Nyx Linden	first topic: "I know the collision box is an elipsoid shaped sphere. But, I don't understand how it related to the Collision Skeleton. What is the avatar collision skeleton found in Develop->Avatar->Show Collision Skeleton? Nalates Urriah 14:41, 17 March 2012 (PDT)"
[12:05]	Koli Melune (sahkolihaa.contepomi)	Hey Runitai.
[12:05]	davep (runitai.linden)	howdy
[12:05]	jr (jr4201.kruyschek)	hi Runitai
[12:05]	Yuzuru (yuzuru.jewell)	Hello Runitai.
[12:06]	Nal (nalates.urriah)	?
[12:06]	CrystalShard Foo	Nix, added!
[12:06]	draconis.neurocam (draconis.neurocam)	i always thought those were bones nal
[12:07]	Nal (nalates.urriah)	The word collision in collision skeleton throws me...
[12:07]	–ü–æ—Ä–≤—É –ü—É—Ç–∏–Ω–∞ –∑–∞ –æ—á–∫–æ (fadeout.razorfen)	oh, also question about the prim equivalency. for example if i have this tower... am i supposed to cut down its PE cost by linking repeating parts, or replacing some stuff with tubes ? because otherwise its VERY prim costy
[12:07]	Geenz Kitten (geenz.spad)	seems to me that collision skeleton would be used for inverse kinematics than anything
[12:07]	Nyx Linden	please add other topics to the agenda so we can stay on topic, thanks :)
[12:07]	davep (runitai.linden)	the shapes shown in "show collision skeleton" are used for alt-zooming on your avatar
[12:07]	davep (runitai.linden)	or right clicking on your avatar
[12:07]	Geenz Kitten (geenz.spad)	ah
[12:07]	draconis.neurocam (draconis.neurocam)	interesting
[12:08]	davep (runitai.linden)	see the raycast info display
[12:08]	draconis.neurocam (draconis.neurocam)	that makes so much sense
[12:08]	Nal (nalates.urriah)	So, they are more like pick points?
[12:08]	davep (runitai.linden)	yup
[12:08]	davep (runitai.linden)	the simulator just uses a single ellipsoid for the whole avatar
[12:08]	davep (runitai.linden)	(simulator doesn't have a skeleton for each avatar, doesn't do animation)
[12:09]	draconis.neurocam (draconis.neurocam)	it would be nice if there was an info display for that ellipsoid
[12:09]	Geenz Kitten (geenz.spad)	I would have guessed a capsule, but nifty
[12:09]	davep (runitai.linden)	could probably add that to the physics shape display "someday maybe" :-p
[12:09]	Nal (nalates.urriah)	I found Andrews old 2009 stuff where he talks about the collision box/eleipse. But, nothing much on what this skeleton thing is.
[12:09]	Nal (nalates.urriah)	collision skeleton^
[12:10]	CrystalShard Foo	I remember Jeffrey Ventrella briefly working at LL to work out a collision mesh that was skeleton shaped, but that was aborted.
[12:11]	Nal (nalates.urriah)	There is a big discussion in oneof the non-LL forums about what mesh avatars do to the collision target area for weapons.
[12:11]	davep (runitai.linden)	yeah, ventrella tried using the collision skeleton for making flexis not go through your avatar, but it never worked right
[12:11]	draconis.neurocam (draconis.neurocam)	would it have been easier if flexis were done via a shader versus their current implimentation?
[12:11]	Nyx Linden	that's a surprisingly tough problem to solve
[12:11]	davep (runitai.linden)	when the mouse cursor hover code came in, we had to replace the old pixel-perfect pick render raycast with a CPU based one for performance reasons, and the avatar collision skeleton happened to be left in the code, so it was a natural choice for raycasting
[12:11]	Nal (nalates.urriah)	So, far the best test is to drop a prim on the avatar's head.
[12:12]	√ÑlveKatt (davido.chrome)	That stuff is hard even in SFX in high budget movies...
[12:12]	CrystalShard Foo	Ahh.
[12:12]	Nyx Linden	next topic: "Mesh import formats roadmap and need to update import mesh formats already in use (DAE only at the moment)"
[12:13]	jr (jr4201.kruyschek)	r8
[12:13]	jr (jr4201.kruyschek)	the latest RL softwares are moving to newer DAE versions
[12:13]	jr (jr4201.kruyschek)	and they aren`t supported by SL
[12:13]	Nyx Linden	We will certainly review and possibly accept patches to accept other file formats, DAE works most places currently :)
[12:14]	jr (jr4201.kruyschek)	maybe we could have one or more formats
[12:14]	draconis.neurocam (draconis.neurocam)	i think she is just asking for a dae version picker or something
[12:14]	jr (jr4201.kruyschek)	some for non-rigged meshes and others for rigged meshes
[12:14]	CrystalShard Foo	The new DAE version is making things somewhat more complex.
[12:14]	jr (jr4201.kruyschek)	very
[12:14]	Geenz Kitten (geenz.spad)	most exporters should retain some degree of compatibility with 1.4.1
[12:14]	jr (jr4201.kruyschek)	but newer RL software versions too
[12:14]	Geenz Kitten (geenz.spad)	mostly because most game engines don't even support newer versions (yet)
[12:14]	davep (runitai.linden)	COLLADA is supposed to be backwards compatible, if it's not and you've got a 3D modelling tool that can't export to second life through COLLADA, please open a jira and include sample content
[12:15]	jr (jr4201.kruyschek)	1.40 is not supported by SL atm, i think
[12:15]	jr (jr4201.kruyschek)	the textures cerated in RL are not fiting the imported models in DAE 1.41
[12:15]	Geenz Kitten (geenz.spad)	1.4.0 should work with SL
[12:15]	Geenz Kitten (geenz.spad)	1.4.1 should work as well
[12:15]	jr (jr4201.kruyschek)	sorry, i wanted to say that 1.41 is not supprted by SL atm
[12:16]	jr (jr4201.kruyschek)	1.40 is
[12:16]	Geenz Kitten (geenz.spad)	they both should work, I've imported content with both of them; sounds more like a bug if it's not working for you
[12:16]	Vincent Nacon	that can be a big problem for people who just brought Max 3Ds/Maya 2012
[12:16]	jr (jr4201.kruyschek)	and there is a difference in prims countings between 1.40 and 1.41
[12:16]	Nyx Linden	if you have sample content that doesn't import correctly due to versioning, send it our way and we'll take a look
[12:16]	jr (jr4201.kruyschek)	i am having a problem with softimage 2012
[12:17]	jr (jr4201.kruyschek)	but softimage 2011 looks good exporting collada
[12:17]	Motor Loon	I use Softimage 2012, havent seen issues
[12:17]	Vincent Nacon	and I do say big problem, even though there's a work around by using FBX but.. the problem is with the Maxscripts that needed for exporting such a large project
[12:17]	jr (jr4201.kruyschek)	softimage 2011 collada exporter is V5.0
[12:17]	jr (jr4201.kruyschek)	and softimage 2012 collada exporter is v.2012.0
[12:18]	jr (jr4201.kruyschek)	Motor: and textures generated by rendermap, do they fit well?
[12:18]	Motor Loon	yep
[12:18]	jr (jr4201.kruyschek)	can you explain me later how you do it?
[12:18]	jr (jr4201.kruyschek)	i mean teh procedures
[12:18]	Nyx Linden	package up an exported collada file in a JIRA with what you get when you try to upload it, we'll take a look. We want to support collada, any other formats other devs are welcome to patch in and we'll review them :)
[12:19]	jr (jr4201.kruyschek)	Motor: can i add you?
[12:19]	Motor Loon	sure
[12:20]	Nyx Linden	next topic: "Status of the usergroup"
[12:20]	Nyx Linden	as you might see here, we're continuing the usergroup celebrate
[12:20]	jr (jr4201.kruyschek)	we are also facing recalculation of prims counts by mesh prims uploaded monthes ago, but that needs another topic and i will not add it today
[12:20]	CrystalShard Foo	:D
[12:20]	Motor Loon	yaay we're still here
[12:20]	Nyx Linden	if you have any questions about what topics are appropriate, feel free to send them my way
[12:20]	Latif Khalifa	/me wonders how hard would it be to add option to to uploader to upload ll mesh asset format? then external converters could be scripted to make it. perhaps easier than make them inside the viewer
[12:20]	jr (jr4201.kruyschek)	for now i just say that more people are finding that older uploads have now bigger prims countings
[12:21]	–ü–æ—Ä–≤—É –ü—É—Ç–∏–Ω–∞ –∑–∞ –æ—á–∫–æ (fadeout.razorfen)	i just updated the agenda and put there my topic
[12:21]	DrFran Babcock	yikes, I will have to check
[12:21]	jr (jr4201.kruyschek)	ty Motor
[12:22]	–ü–æ—Ä–≤—É –ü—É—Ç–∏–Ω–∞ –∑–∞ –æ—á–∫–æ (fadeout.razorfen)	and put the demo inworld
[12:22]	Nyx Linden	next topic: "Status of the mesh clothing parametric deformer"
[12:22]	–ü–æ—Ä–≤—É –ü—É—Ç–∏–Ω–∞ –∑–∞ –æ—á–∫–æ (fadeout.razorfen)	oh
[12:22]	jr (jr4201.kruyschek)	cool
[12:22]	Nyx Linden	Oz is the linden working with Qarl, he would have more up to date status than we would
[12:22]	Motor Loon	Dont think there's any news on the deformer yet
[12:22]	jr (jr4201.kruyschek)	there are betas around
[12:22]	jr (jr4201.kruyschek)	like Zen viewer
[12:23]	CrystalShard Foo	There was a code sample out awhile back but I haven't heard anything new since.
[12:23]	Koli Melune (sahkolihaa.contepomi)	Last Oz said was that the code didn't merge cleanly into v-d.
[12:23]	Nal (nalates.urriah)	A couple of TPV Dev said the same.
[12:23]	jr (jr4201.kruyschek)	i think taht the latest Zen and snowglobe my have a beta with mesh clotehs parametric deformer, i am not sure
[12:23]	Motor Loon	so it likely back on Karls table
[12:23]	jr (jr4201.kruyschek)	Nyran`s viewer too
[12:24]	Koli Melune (sahkolihaa.contepomi)	Oz mentioned it needs someone who knows the rendering engine to deal with it.
[12:24]	Vincent Nacon	I might be a bit out of loop on deformer deal but... is there a way or even talk about flagging a prim to use deform or not?
[12:25]	Geenz Kitten (geenz.spad)	there's been talk
[12:25]	Geenz Kitten (geenz.spad)	but that's about it
[12:25]	CrystalShard Foo	My guess is that it would be better off disabled by default.
[12:25]	Nal (nalates.urriah)	The one in Niran's is slow on the initial attach of a mesh. If you play with it give it a couple of minutes to start working before changing avatar shape to test or expecting it to render right.
[12:25]	Vincent Nacon	yeah
[12:25]	Nyx Linden	next topic: "How to cut down your prim equivalency cost? Are people supposed to cut repeating parts as separate mesh objects and then copy+link them in world? Also cut simple mesh parts that could be replaced with prims? Because this is what it looks like when i was creating a medieval tower. FadeOut Razorfen"
[12:26]	jr (jr4201.kruyschek)	good i just rezzed this jeep i am working on
[12:26]	Vincent Nacon	I'm sure people who already uploaded their rigged mesh avatar or clothes would be upset if it become too small
[12:26]	Nyx Linden	complex topic, but lets see if we can break it down a bit
[12:26]	Vincent Nacon	ok
[12:26]	jr (jr4201.kruyschek)	i reduced so far from 120 prims to 57 and still will go down
[12:26]	–ü–æ—Ä–≤—É –ü—É—Ç–∏–Ω–∞ –∑–∞ –æ—á–∫–æ (fadeout.razorfen)	here is the tower model i'm talking about
[12:26]	Nyx Linden	first of all, prims are still useful, if there are very simple parts of your object that are more easily done in prims, by all means use prims
[12:26]	–ü–æ—Ä–≤—É –ü—É—Ç–∏–Ω–∞ –∑–∞ –æ—á–∫–æ (fadeout.razorfen)	above us
[12:27]	Nyx Linden	secondly, how you break apart a mesh will definitely have an impact on the cost as well as how LODs are handled
[12:27]	Motor Loon	LOD really is the key to low LI on mesh
[12:27]	–ü–æ—Ä–≤—É –ü—É—Ç–∏–Ω–∞ –∑–∞ –æ—á–∫–æ (fadeout.razorfen)	i've did everything i could on LOD
[12:27]	–ü–æ—Ä–≤—É –ü—É—Ç–∏–Ω–∞ –∑–∞ –æ—á–∫–æ (fadeout.razorfen)	doesn't help
[12:27]	Nyx Linden	its a tradeoff - smaller objects will have lower land impact because the lower LODs will be displayed at larger distances
[12:27]	Vincent Nacon	/me cringes to see the physic weight of that tower....
[12:28]	√ÑlveKatt (davido.chrome)	Isn't it the other way around?
[12:28]	Nyx Linden	which means that if you break things up into smaller pieces, it can save on land impact cost, but it will make individual pieces change LOD at different points
[12:28]	Nyx Linden	larger ranges/areas, sorry
[12:28]	jr (jr4201.kruyschek)	i used what Runitai and more people told me
[12:28]	jr (jr4201.kruyschek)	both wheels are identical
[12:29]	jr (jr4201.kruyschek)	the left one i lowered the most i could the low and lowest lods in import
[12:29]	–ü–æ—Ä–≤—É –ü—É—Ç–∏–Ω–∞ –∑–∞ –æ—á–∫–æ (fadeout.razorfen)	oh i see, but thats how its should be done... this is quite logical
[12:29]	jr (jr4201.kruyschek)	the right one i just uploaded as it came out
[12:29]	Nyx Linden	also, land impact is whatever the highest weight is - some objects have a high streaming weight, others have high physics cost
[12:29]	Vincent Nacon	wait... you made two different textures for it?
[12:29]	jr (jr4201.kruyschek)	thee prims counts are different
[12:29]	jr (jr4201.kruyschek)	yes, a texture for each
[12:29]	√ÑlveKatt (davido.chrome)	I made a chain Link, got it down to 5 LI. But if I were to make the same thing with sculpts, and make it opt into I think it would end up at 16 something.
[12:29]	Vincent Nacon	...why?
[12:29]	Nyx Linden	so check the details floater for each object and see which wieght is causing the most problem.
[12:29]	Vincent Nacon	just flip the UV map on other one
[12:29]	jr (jr4201.kruyschek)	or the characters would look like in a mirror
[12:30]	Vincent Nacon	you still can reuse it
[12:30]	Vincent Nacon	just flip UV map for the other side
[12:30]	–ü–æ—Ä–≤—É –ü—É—Ç–∏–Ω–∞ –∑–∞ –æ—á–∫–æ (fadeout.razorfen)	display 9000? physics 240
[12:30]	Koli Melune (sahkolihaa.contepomi)	2012-03-19T19:29:33Z llrender/llrendertarget.cpp(407) : error
[12:30]	Koli Melune (sahkolihaa.contepomi)	What the hell does that mean? :/
[12:30]	√ÑlveKatt (davido.chrome)	Nyx, that chain link I mentioned, it's more efficient than a sculpt, no?
[12:31]	Nyx Linden	can you rez it off to the side? its helpful to look at actual examples
[12:31]	davep (runitai.linden)	Koli: that means you tickled the viewer in a way that it tried to use a render target that's not initialized
[12:31]	–ü–æ—Ä–≤—É –ü—É—Ç–∏–Ω–∞ –∑–∞ –æ—á–∫–æ (fadeout.razorfen)	i've rezzed it already
[12:31]	–ü–æ—Ä–≤—É –ü—É—Ç–∏–Ω–∞ –∑–∞ –æ—á–∫–æ (fadeout.razorfen)	its above us
[12:31]	Koli Melune (sahkolihaa.contepomi)	o_O
[12:31]	davep (runitai.linden)	probably some combination of deferred rendering and wireframe
[12:31]	davep (runitai.linden)	?
[12:31]	Nyx Linden	the tower for example has a complex physics cost, even though its streaming cost is fairly low
[12:31]	Koli Melune (sahkolihaa.contepomi)	Ah, I had that mesh selected and was camming around it.
[12:31]	davep (runitai.linden)	or the render target failed to initialize and we tried to use it anyway
[12:32]	√ÑlveKatt (davido.chrome)	Why did you choose such a complex physics shape?
[12:32]	Motor Loon	a fairly lowpoly cylinder should do the trick
[12:32]	Nyx Linden	keep in mind physics triangles are a lot more costly than render triangles
[12:32]	–ü–æ—Ä–≤—É –ü—É—Ç–∏–Ω–∞ –∑–∞ –æ—á–∫–æ (fadeout.razorfen)	still its not the problem, display is 9000, physics is just 240
[12:32]	Anthonys Republic	yeah i noticed that
[12:32]	Nyx Linden	linking it with an invisible cylinder to use as a physics representation would probably suffice
[12:32]	davep (runitai.linden)	display doesn't hit your land impact
[12:33]	–ü–æ—Ä–≤—É –ü—É—Ç–∏–Ω–∞ –∑–∞ –æ—á–∫–æ (fadeout.razorfen)	oh, thats great idea
[12:33]	davep (runitai.linden)	it's just a metric for reference
[12:33]	√ÑlveKatt (davido.chrome)	CAn you get the land impact down by making a physics shape of prims and and set the Mesh to none?
[12:33]	Nyx Linden	yes!
[12:33]	CrystalShard Foo	o.o
[12:33]	jr (jr4201.kruyschek)	i also would like to show an example that Talin Sands called me the attention to
[12:33]	CrystalShard Foo	/me marks the center of her bingo card
[12:33]	davep (runitai.linden)	brb, lost my caps
[12:33]	√ÑlveKatt (davido.chrome)	I made a chain Link, got it down to 5 LI. But if I were to make the same thing with sculpts, and make it opt into I think it would end up at 16 something.
[12:33]	Nyx Linden	it is absolutely fine and encouraged to use attached prims as a physics shape, and set the mesh shape to none.
[12:33]	Latif Khalifa	but then prims can count twice or 3 times as much ;)
[12:33]	√ÑlveKatt (davido.chrome)	Nyx, that chain link I mentioned, it's more efficient than a sculpt, no?
[12:34]	Anthonys Republic	is that better then a box physics nyx?
[12:34]	jr (jr4201.kruyschek)	check teh prims counts of does two objects
[12:34]	√ÑlveKatt (davido.chrome)	Latif, not if you use untortured prims.
[12:34]	jr (jr4201.kruyschek)	teh white ones
[12:34]	Anthonys Republic	say a box physics uploaded
[12:35]	Nyx Linden	comparing whether a sculptie or a mesh is more efficient is....not an easy question :)
[12:35]	Vincent Nacon	also it's depends on the side of it
[12:35]	jr (jr4201.kruyschek)	the difference between does two objects is big in prims count
[12:35]	Motor Loon	physics again
[12:35]	Vincent Nacon	but no matter how small it goes... it's gonna be 0.5
[12:35]	jr (jr4201.kruyschek)	one is 181 pme and the other is 2 pme
[12:35]	√ÑlveKatt (davido.chrome)	I guess I need to know, because this is just one link in the necklace and belt I am wearing at the moment.
[12:36]	Motor Loon	2.4 physics vs 181.4
[12:36]	jr (jr4201.kruyschek)	one is much less only because has one more side so SL can treat it as a box for physics
[12:37]	Nyx Linden	I'd really hesitate to call any chain where each link was modeled "efficient" by any stretch of the imagination, whether prim or sculptie
[12:37]	jr (jr4201.kruyschek)	so, a mesh shape should always be closed enough for SL finds a base for give it a box shape physics
[12:37]	√ÑlveKatt (davido.chrome)	Ooops.
[12:37]	Latif Khalifa	breaking a big mesh into several smaller peaces to achieve lower PE inworld just tells you how "logical" PE formulas are ;)
[12:37]	√ÑlveKatt (davido.chrome)	So much more easy to fit that way though.
[12:38]	Latif Khalifa	(streaming cost/scale one is totally out of whack)
[12:38]	Vincent Nacon	you just need to know the right scale for that piece of the chain and make another chain in the same mesh object till it reach 0.5 in that scale
[12:38]	Nyx Linden	you can probably get more efficient if you create the whole chain in your modeler, though yes it will take more tweaking to size right
[12:38]	Vincent Nacon	then build the whole set of the chain with it
[12:39]	davep (runitai.linden)	er, I'm confused
[12:39]	davep (runitai.linden)	I don't see those links on the chain you're wearing
[12:39]	TheyCallMeDick Cheney	/me coughs ... Instancing
[12:39]	Vincent Nacon	he's just showing one piece of the chain link itself
[12:39]	Vincent Nacon	it's one of those typical necklace chain kind
[12:39]	davep (runitai.linden)	what is that box around the link?
[12:39]	Vincent Nacon	not the construction chain kind
[12:40]	Vincent Nacon	I think he made it for touch
[12:40]	Anthonys Republic	yeah i see that box to hmm was wondering, looks like a poly cube with trans
[12:40]	Vincent Nacon	to make it easier ot touch when it's smaller?
[12:40]	Vincent Nacon	to*
[12:41]	Second Life	Entering god mode, level 150
[12:41]	√ÑlveKatt (davido.chrome)	So, eh, It would be more efficient to make them in sets of three, even if the LODs are the same?
[12:41]	√ÑlveKatt (davido.chrome)	For example?
[12:41]	Second Life	Now leaving god mode, level 150
[12:41]	davep (runitai.linden)	depends
[12:41]	Vincent Nacon	3 or more... till you reached 0.5 weight in normal size
[12:42]	davep (runitai.linden)	the trade off is object overhead vs instancing overhead
[12:42]	TheyCallMeDick Cheney	true
[12:42]	Nyx Linden	it should allow you to eliminate some redundant triangles that way possibly
[12:42]	√ÑlveKatt (davido.chrome)	And can someone tell me the difference between those two wheel that have different land impact, because I can't figure it out...
[12:42]	Nyx Linden	especially for the lower LODs
[12:42]	davep (runitai.linden)	instancing each link the way you have it is good for mesh streaming overhead, but it adds object overhead
[12:42]	jr (jr4201.kruyschek)	i told already but i repeat, Alve
[12:43]	√ÑlveKatt (davido.chrome)	Davep, that terminology was over my head
[12:43]	Vincent Nacon	the thing is... you're using 8 sides, best to use 5 sides when it's going to be small
[12:43]	√ÑlveKatt (davido.chrome)	jr, tahnks, and sorry!
[12:43]	jr (jr4201.kruyschek)	one i uploaded as it came out, is teh right one, the left one i lowered all way down the low and lowest LOD levels in uploading so the PME is much lower
[12:43]	Nyx Linden	since they're all the same model, you only have to download one mode, but then you have to track the position & orientation of each individual link instead of each set of 3
[12:43]	Vincent Nacon	but that's just another typical videogame making rules
[12:43]	jr (jr4201.kruyschek)	np, Alve, yw
[12:44]	CrystalShard Foo	/me updates wiki agenda~
[12:44]	Nyx Linden	alright ready for the next topic?
[12:44]	jr (jr4201.kruyschek)	sure
[12:44]	Nyx Linden	next topic: "Any mesh stats? Nalates Urriah"
[12:45]	√ÑlveKatt (davido.chrome)	jr, so you neglected to make less complex LODS for the high LI one, basically?
[12:45]	Nyx Linden	I don't have the report on hand that generated those stats, I'll ask around for next week
[12:45]	Nyx Linden	new stats: most regions have meshes. yay! :)
[12:45]	jr (jr4201.kruyschek)	no, i maintained the highest and middle lods as default
[12:46]	Nyx Linden	next topic: "Will we be getting hinges back? CrystalShard Foo"
[12:46]	jr (jr4201.kruyschek)	i just lowered the maximum it was allowed the LOW and LOWEST LODS at the upload
[12:46]	Motor Loon	in our dreams ;-)I
[12:46]	CrystalShard Foo	Shh~
[12:46]	CrystalShard Foo	I just figured, with all the recent changes to physics and the introduction of physics mesh..
[12:47]	Vincent Nacon	by hinges, you mean constraints that Havok's library source have?
[12:47]	Nyx Linden	getting them back? did we used to have hinges? searches his memory banks
[12:47]	CrystalShard Foo	Heck yea we did :D
[12:47]	Motor Loon	looong ago
[12:47]	Vincent Nacon	and you should ask Falcon or Andrew for that
[12:47]	CrystalShard Foo	Up to 2006 I think
[12:47]	CrystalShard Foo	I used to use them on swing doors i 2004.
[12:47]	CrystalShard Foo	*in
[12:47]	√ÑlveKatt (davido.chrome)	Aaah.
[12:47]	Nyx Linden	aaah ok
[12:47]	draconis.neurocam (draconis.neurocam)	they were called joints
[12:48]	Vincent Nacon	yup
[12:48]	CrystalShard Foo	Oh right, thanks
[12:48]	draconis.neurocam (draconis.neurocam)	i actually have a few in my inventory they don't work anymore
[12:48]	Vincent Nacon	and a stupid name for t
[12:48]	Anthonys Republic	hinges as in pivot points on mesh, pleaseeeee. been using a single face poly to offset the pivot
[12:48]	Vincent Nacon	it*
[12:48]	CrystalShard Foo	I have atleast one broken build that used them
[12:48]	Vincent Nacon	I doubt it would work soon as you rez it
[12:48]	Motor Loon	yeah custom pivots on mesh would be reaaal nice
[12:48]	CrystalShard Foo	Nope, the door remains static on the joint
[12:48]	√ÑlveKatt (davido.chrome)	I have been toying with the idea of just having a convex surface that I put a texture of the high LOD on for the lowest LOD.
[12:48]	CrystalShard Foo	Support was removed sim side entirely. The client function was removed about a year later.
[12:49]	Nyx Linden	we're currently working on trying to stabilize the features we already have over getting new features for the time being, so sounds like a nice feature request, but I don't think we have it on our schedules currently
[12:49]	Anthonys Republic	pivots are a big deal to me anyways lol plus when coems to physics it would be cheaper if the object was pivoted correctly
[12:50]	Nyx Linden	which brings me to my next topic: "When non attached rigged mesh is going to be scheduled ? FadeOut Razorfen"
[12:50]	√ÑlveKatt (davido.chrome)	Would I need a hidden triangle with that texture on on the other LODS?
[12:50]	Nyx Linden	I don't know when :) most of our teams are focused on fixing the things that are broken right now
[12:50]	–ü–æ—Ä–≤—É –ü—É—Ç–∏–Ω–∞ –∑–∞ –æ—á–∫–æ (fadeout.razorfen)	like, any idea
[12:50]	√ÑlveKatt (davido.chrome)	That material on I mean.
[12:50]	–ü–æ—Ä–≤—É –ü—É—Ç–∏–Ω–∞ –∑–∞ –æ—á–∫–æ (fadeout.razorfen)	so absolutely no ideas ?
[12:51]	–ü–æ—Ä–≤—É –ü—É—Ç–∏–Ω–∞ –∑–∞ –æ—á–∫–æ (fadeout.razorfen)	next year or a couple of months ?
[12:51]	Nyx Linden	its a great feature, would love to see it. I personally do not know when we will have cycles to work on it.
[12:51]	–ü–æ—Ä–≤—É –ü—É—Ç–∏–Ω–∞ –∑–∞ –æ—á–∫–æ (fadeout.razorfen)	allright
[12:51]	√ÑlveKatt (davido.chrome)	I have been toying with the idea of just having a convex surface that I put a texture of the high LOD on for the lowest LOD.
[12:51]	Motor Loon	/me adds more cycles
[12:51]	√ÑlveKatt (davido.chrome)	Would I need a hidden triangle with that texture on on the other LODS?
[12:52]	Motor Loon	yes davido
[12:52]	Nyx Linden	one more topic and 10 minutes to go, "Thank you Nyx for getting the UG back on the UG list. Nalates Urriah"
[12:52]	Vincent Nacon	yup
[12:52]	Nal (nalates.urriah)	Yes... hooray for you. :)
[12:52]	jr (jr4201.kruyschek)	UG?
[12:53]	Anthonys Republic	ugggg
[12:53]	Nal (nalates.urriah)	user group
[12:53]	√ÑlveKatt (davido.chrome)	Yeah, thanks! As you can see, we need it!
[12:53]	Anthonys Republic	oh
[12:53]	jr (jr4201.kruyschek)	cool, thanks
[12:53]	Vincent Nacon	underground, like the cool kids do!
[12:53]	Vincent Nacon	:P
[12:53]	Nal (nalates.urriah)	:p
[12:53]	Latif Khalifa	So since mesh project shipped before 2027, what other project is occupying that coveted time slot? :P
[12:53]	Anthonys Republic	can a mesh hold multiple uv sets in sl?
[12:53]	jr (jr4201.kruyschek)	npc, i think
[12:53]	√ÑlveKatt (davido.chrome)	Yes.
[12:54]	√ÑlveKatt (davido.chrome)	Up to six, is it?
[12:54]	Vincent Nacon	that would be nice but no.... it does?
[12:54]	Nyx Linden	/me was thinking the give nyx a vacation project might be a good one to work on before 2027 :D
[12:54]	Vincent Nacon	unless you're talking about materials?
[12:54]	Anthonys Republic	no i dont mean materials up to 8 that is, i mean for 1 material can it hold multiple uv sets
[12:54]	√ÑlveKatt (davido.chrome)	Oh. I am. Doh.
[12:54]	davep (runitai.linden)	Anthony: no, just one
[12:54]	Nyx Linden	but currently focused on avatar rezzing issues
[12:54]	Anthonys Republic	k
[12:55]	CrystalShard Foo	You should ask to be alloted research money for the project in order to run a localized regional experiment. Specifically, Hawaii.
[12:55]	Nal (nalates.urriah)	Any news on that front? More logs still needed?
[12:55]	Vincent Nacon	yeah, just one. although it would be nice to have more for better landscaping with repeatable textures
[12:55]	Anthonys Republic	whos in hawaii?
[12:55]	Nyx Linden	more logs are always appreciated! there's a lot of points where things can break
[12:55]	√ÑlveKatt (davido.chrome)	Oh, about that physics tests thing on Aditi for Path finding. CanI have a space for a simball court that won'
[12:55]	√ÑlveKatt (davido.chrome)	t get returned?
[12:56]	√ÑlveKatt (davido.chrome)	We really need to be able to see if it will break the game in any way.
[12:56]	Anthonys Republic	really appreciate all the hard work lindens thanks bunch :)
[12:57]	Vincent Nacon	aye
[12:57]	Anthonys Republic	get those pivots in and i may send ya a mesh chocolate cake :)
[12:58]	Nyx Linden	So I think that's going to wrap us up for this week :)
[12:58]	davep (runitai.linden)	/me runs away to fight the sculpt crash monster.
[12:58]	davep (runitai.linden)	later, folks!
[12:58]	Vincent Nacon	take care
[12:58]	Nyx Linden	thanks everyone for coming out!
[12:58]	Nal (nalates.urriah)	bye
[12:58]	Koli Melune (sahkolihaa.contepomi)	See you, Lindens.
[12:58]	jr (jr4201.kruyschek)	k, thanks all
[12:58]	Latif Khalifa	laters
[12:58]	Nal (nalates.urriah)	Thx Nyx
[12:58]	draconis.neurocam (draconis.neurocam)	thanks
[12:58]	jr (jr4201.kruyschek)	see ya guys
[12:58]	Nyx Linden	come back next week and don't forget to add new topics!
[12:58]	Vir Linden	btw, pathfinding requests should go to lorca
[12:59]	Yuzuru (yuzuru.jewell)	See you, next week.
[12:59]	Vir Linden	bye all
[12:59]	Latif Khalifa	bye
[12:59]	√ÑlveKatt (davido.chrome)	Lorca, Lorca@lindelab.com?
[12:59]	Vir Linden	yup
[12:59]	√ÑlveKatt (davido.chrome)	+n
[12:59]	√ÑlveKatt (davido.chrome)	Ok. thanks!

Generated by the 'Wikify SL Chat Log' utility by DG Products for Second Life (DGP4SL)

Mesh/Archive/2012-03-19

Contents

Agenda

Meeting Notes

Attendees List

Transcript for Monday March 19, 2012

Navigation menu

Search