Mesh/Archive/2012-04-30

From Second Life Wiki
Jump to navigation Jump to search

Agenda

  1. JIRA VWR-6713 - 'Allow Alpha Channel of Textures to be used as a 1-bit Mask (Alpha Masking)' is now four years old and has still not been reviewed. Is there any chance this could be pushed, considering how broken the automatic alpha masking system is and how glitchy alpha textures are in deferred rendering? Sahkolihaa Contepomi 06:42, 29 April 2012 (PDT)
  2. In the log of Friday Server/Scripting group, Falcon said "We're changing streaming cost for prims to be capped at 1.0...". The example linkset did not contain mesh. Officially, mesh are a kind of prim. However, it seems unlikely that the cap will be applied to mesh prims. Can you confirm that it will not apply to mesh prims. Further, the statement is unconditional, implying that the cap will still apply to legacy prims that are in a linkset that also contains mesh prims. Can you confirm whether this is the case? Drongle McMahon
  3. A couple of weeks ago, I asked for an explicit description of the algorithm that determines the point at which the behaviour and physics weight of a mesh with physics shape type "Prim" is slently switched from that expected for physics shape type "Prim" to that expected for physics shape type "Convex Hull", as its size is shrunk. Did I miss the answer?Drongle McMahon
  4. Why do spikes appear on lower LOD rigged meshes when they don't if it's uploaded by itself?Tiberious Neruda
  5. When uploading a specific .dae for the physics shape it will get scaled to fit 'the bounding box' of the main .dae you are uploading. If you create furniture with a cylinder with a plane under it that will be the shadow, and you make the .dae for physics just the cylinder... the cylinder gets wider to fit all the shadow and doesn't keep the size it should to cover only the part that should be physical. Will there be a way to disable that feature?Fem Darcy
  6. Has the source mesh for the default convex hull been changed again? It used to be the high LOD mesh, but I was just shown an example where it appeared to be made from the lowest LOD mesh or the physics mesh. Needless to say the result was useless. (Ah ... just heard it was uploded with Phoenix, so that may be the problem).Drongle McMahon
  7. I know that if you upload an LOD, you can do so with or without weights as long as the high LOD has weights. But, if you do upload a low LOD with weights, are they ignored? Ashasekayi Ra 12:43, 30 April 2012 (PDT)

Meeting Notes

Transcript for Monday April 30, 2012

[12:08] Nyx Linden hi all
[12:08] Motor Loon hey Nyx
[12:08] Vir Linden howdy folks
[12:08] DrFran Babcock /me was wondering where Vir was
[12:08] Asha (ashasekayi.ra) Heya Nyx and Vir.
[12:08] Vincent Nacon heya
[12:08] Vir Linden sorry, we had a meeting that went long
[12:09] Yuzuru (yuzuru.jewell) Hello Nyx.
[12:09] Drongle McMahon !!?? Nyx Linden and Nyx Tester!
[12:09] Nal (nalates.urriah) Happens, np... Hi
[12:09] Nyx Linden agenda for the week is here: http://wiki.secondlife.com/wiki/Content_Creation/Mesh_Import_User_Group
[12:10] Vir Linden topic: perilous proliferation of nyxes :-)
[12:10] Asha (ashasekayi.ra) :D
[12:10] Nyx Linden shall we get started? apologies for running late
[12:10] Vincent Nacon np
[12:10] Motor Loon yes pls
[12:10] Drongle McMahon Ah, Tester evaporated.
[12:10] Asha (ashasekayi.ra) Ready
[12:10] Nyx Linden first topic: "JIRA VWR-6713 - 'Allow Alpha Channel of Textures to be used as a 1-bit Mask (Alpha Masking)' is now four years old and has still not been reviewed. Is there any chance this could be pushed, considering how broken the automatic alpha masking system is and how glitchy alpha textures are in deferred rendering? Sahkolihaa Contepomi 06:42, 29 April 2012 (PDT)"
[12:10] JIRA-helper http://jira.secondlife.com/browse/VWR-6713
[12:10] Vincent Nacon really?
[12:11] Motor Loon I personally dont really think Alphas are that bad
[12:11] Vincent Nacon they're not that bad
[12:11] Xentha Melune (questar.utu) They can be, depending on what you look at.
[12:11] Vincent Nacon in fact... better than before
[12:11] Vincent Nacon better than without mask that is
[12:11] Xentha Melune (questar.utu) The question would be, why not?
[12:11] Nyx Linden it certainly would be a nice feature to have, to allow content creators to choose between masking or blending
[12:11] Motor Loon yes, I certainly saw them improve with the new rendering
[12:12] Koli Melune (sahkolihaa.contepomi) Deferred rendering and alpha blending are a big no no.
[12:12] Motor Loon nice yes, essential no
[12:12] Vincent Nacon options would be nice but... it's best not to give them options when people aren't fully aware the difference between the two
[12:12] Nyx Linden it would require a protocol-level change to add additional flags to the prim/face definition in order to implement, which is no small task.
[12:12] Xentha Melune (questar.utu) That's pretty much like saying not to give people mesh when they have sculpts. :p
[12:12] Asha (ashasekayi.ra) heh
[12:13] Vincent Nacon believe it or not, I wasn't in favor of mesh before then
[12:13] Motor Loon mesh had a huge effect... which justified the big work putting it in
[12:13] Asha (ashasekayi.ra) Totally agree on that Motor
[12:13] Xentha Melune (questar.utu) I doubt alpha masks would be terribly difficult to put in. Aren't Linden Trees already using it?
[12:13] Vincent Nacon yeah and I still don't like how it's been dumbed down for non-professional </ran>
[12:14] Vincent Nacon </rant>*
[12:14] Asha (ashasekayi.ra) Perhaps they can address 1-bit alphas with the future material system?
[12:14] Motor Loon linden tree's arent excactly "regular content"
[12:14] Motor Loon yeah seems like a good idea Ashase
[12:14] Xentha Melune (questar.utu) Point is, the code is already there, all that's missing is the ability to use it.
[12:14] Motor Loon with a massive change like a materials system, alpha changes become just a little thing to include
[12:14] Nyx Linden we already have alpha masking on our avatars - supporting it in the engine is easy. its changing the object definitions to include extra parameters that takes more time. Either way it has neither been forgotten nor prioritized over other work. We're still aware that it would be nice to have.
[12:14] Vincent Nacon you mean Shader mapping, Asha?
[12:14] Tiberious Neruda I'm still waiting for a 'faces camera' option
[12:15] Motor Loon particles tiberious °͜°
[12:15] Tiberious Neruda particles suck
[12:15] Xentha Melune (questar.utu) Heh..
[12:15] Motor Loon well...
[12:15] Tiberious Neruda and need to be constantly refreshed
[12:15] Vincent Nacon particle doesn't suck... just limited
[12:15] Tiberious Neruda plus, they're always squares
[12:15] Motor Loon so a new kind of particles?
[12:16] Vincent Nacon well duh, most particles are in game development are square
[12:16] Tiberious Neruda a rendering option for meshes/prims
[12:16] Asha (ashasekayi.ra) I just mean they will obviously have to add these protocols if they add the ability to include normal maps, spec maps etc. So, it would make sense to deal with it at that time.
[12:16] Nyx Linden so in answer to topic one - definitely a nice feature to have, not currently a priority. Shall we move to the next topic?
[12:16] Vincent Nacon yup
[12:16] Asha (ashasekayi.ra) Sure
[12:16] Xentha Melune (questar.utu) A new particle would be a bad idea, because what he's implying is a particle without the limited lifespan. Griefing would skyrocket if particles could be generated without.
[12:16] Nyx Linden next topic: "In the log of Friday Server/Scripting group, Falcon said "We're changing streaming cost for prims to be capped at 1.0...". The example linkset did not contain mesh. Officially, mesh are a kind of prim. However, it seems unlikely that the cap will be applied to mesh prims. Can you confirm that it will not apply to mesh prims. Further, the statement is unconditional, implying that the cap will still apply to legacy prims that are in a linkset that also contains mesh prims. Can you confirm whether this is the case? Drongle McMahon"
[12:16] Tiberious Neruda any faces towards, say, 'positive X' would render always faccing the camera
[12:17] Vincent Nacon I think he meant Sprite prim
[12:17] Tiberious Neruda right
[12:17] Xentha Melune (questar.utu) Yeah, sorry, I meant the equivelant implication.
[12:17] Motor Loon Falcon hasn't YET confirmed 100%sculpties are part of "anything not mesh"... but?
[12:17] B.A.M.F. (lestermoore) English muthafucker, do you speak it!?
[12:18] Vincent Nacon knewit
[12:18] Nyx Linden the proposed streaming cost cap, if implemented and deployed (nothing is final until it is released), would affect legacy prims - streaming cost of meshes would not be capped.
[12:18] Motor Loon legacy prims = sculpties ?!
[12:18] Nyx Linden it would affect legacy prims even if they are linked to meshes (and thus fall under the new accounting system)
[12:18] Asha (ashasekayi.ra) Will the server cost change only be for legacy content as well Nyx?
[12:19] Vincent Nacon there's no difference in sculpty and prim when it come to land weight cost
[12:19] Frans Charming With cap, meaning there is a max cost of for a linkset of normal prims?
[12:19] Drongle McMahon OK. Perfect answer. Thanks.
[12:19] Nyx Linden please note that this is a proposal, and we reserve the right to change any part of it if necessary before release.
[12:19] Motor Loon /me nods
[12:19] Vincent Nacon of course
[12:19] Vincent Nacon yanking the candy from a baby after you provide it
[12:19] B.A.M.F. (lestermoore) is this a meeting of the minds or what
[12:19] Motor Loon so... sculpties = legacy prims? °͜°
[12:19] Xentha Melune (questar.utu) By technicality.
[12:20] Nyx Linden whether sculpties will be considered "legacy" or not in this context is not 100% determined yet
[12:20] Motor Loon yeah thats what I thought °͜°
[12:20] Vincent Nacon yes, Lestermoore, we are the secret group who control LL and SL
[12:20] Drongle McMahon Ah.
[12:20] Asha (ashasekayi.ra) [12:18] Asha: Will the server cost change only be for legacy content as well Nyx?
[12:20] Nyx Linden since they do require more texture data to display properly, we need to carefully consider exactly how to weight them.
[12:21] Nyx Linden the proposed server cost change would be independent of prim type.
[12:21] Asha (ashasekayi.ra) Ahh good. Thank you.
[12:21] Asha (ashasekayi.ra) I hope that goes through.
[12:22] Motor Loon The thing is... if sculpties are not included as "anything not mesh", the whole idea aren't the solution it was given to be in regard to pathbug-69
[12:22] Nyx Linden its still in the early stages so bare with us if we end up needing to make tweaks between now and release, but we wanted to let everyone know that changes would be coming.
[12:22] Nyx Linden I'll let falcon answer any questions about pathfinding bugs, that's closer to his area of expertise
[12:22] Asha (ashasekayi.ra) That's understandable Nyx.
[12:23] Motor Loon the volumedetect hack is often used on builds using sculpties - to make the sculpties silly collision's become phantom and be replace with another surface
[12:23] Frans Charming That's what I do Motor.
[12:23] Nyx Linden any other questions on that topic?
[12:23] Motor Loon it's not really a pathbug... it's a change to volumedetect BECAUSE of pathfinding
[12:23] Asha (ashasekayi.ra) Not at this time
[12:24] Tiberious Neruda what's changing about volumedetect?
[12:24] Motor Loon which then would break existing content funcionality, so Falcon proposed that as a solution
[12:24] Motor Loon see the jira tiberious
[12:25] Motor Loon solution was to turn our regular prim builds and sculptie builds onto mesh accounting to use the "none" shape to do the same as we did with the volumedetect trick
[12:25] Drongle McMahon Also see the transcript of the server/scripting group last friday.
[12:25] Motor Loon but we cried because usually using mesh accounting for non-mesh means crazy high PE
[12:25] Motor Loon so this change in mesh accounting come is
[12:25] Vincent Nacon silly people using prim with mesh.... sigh.
[12:25] Motor Loon and... useless if it does not cover scuilpties too
[12:26] Motor Loon it's not really about mixing mesh with prims Vincent
[12:26] Vincent Nacon oh I know
[12:26] Motor Loon however, that could be a possible thing to do with this new change if you really wanted to
[12:26] Tiberious Neruda oh. Here I'd thought volumedetect was getting the ability to fire land collision events.... now THAT would be useful
[12:26] Nyx Linden the new accounting system is not intended to be used exclusively with linksets that include meshes - if you use the new physics shape types, it will enable the new calculations as well.
[12:27] Drongle McMahon Its more about using None with old prims.
[12:27] Motor Loon yep
[12:28] Motor Loon I'll say its a wonderful change, and would be highly useful if implemented as proposed -.- but only if it includes sculpties into the "anything not mesh" thing.
[12:28] Motor Loon and if it leads to a little bit of "misuse/overuse" by some... so be it... it's still awesome
[12:29] Nyx Linden I don't want to speculate as to the exact status of sculpties under the new streaming cost rules, as we're still discussing it internally, but your concerns are being considered when looking at the numbers and algorithms
[12:29] B.A.M.F. (lestermoore) i agree i have no clue what you are talking about
[12:29] Motor Loon °͜°
[12:29] Vincent Nacon you don't belong here then
[12:29] Vincent Nacon :P
[12:29] Motor Loon aww
[12:29] Motor Loon Lester is welcome
[12:29] B.A.M.F. (lestermoore) i try
[12:29] Nyx Linden not true, all are welcome here
[12:29] Motor Loon Dont listen to Vincent, he's just grumpy
[12:30] Vincent Nacon or logical?
[12:30] Vincent Nacon :P
[12:30] Motor Loon °͜°
[12:30] Nyx Linden its just a pretty in-depth technical topic :)
[12:30] Vincent Nacon ;)
[12:30] Vincent Nacon aye
[12:30] Nyx Linden with that, any objection to moving to the next topic?
[12:30] Vincent Nacon nope
[12:30] Motor Loon yeah a pretty tricky subject... anything regarding mesh accounting is.... haha
[12:30] Nyx Linden next: "A couple of weeks ago, I asked for an explicit description of the algorithm that determines the point at which the behaviour and physics weight of a mesh with physics shape type "Prim" is slently switched from that expected for physics shape type "Prim" to that expected for physics shape type "Convex Hull", as its size is shrunk. Did I miss the answer?Drongle McMahon"
[12:30] B.A.M.F. (lestermoore) whats a mesh account
[12:31] Vincent Nacon anyone still think I'm wrong?
[12:31] Drongle McMahon Oh dear. That sentence is much too long!
[12:31] Motor Loon ..when physics are on Drongle?
[12:31] Motor Loon or in general?
[12:31] Drongle McMahon No, it happens without physical.
[12:31] Motor Loon I havent seen that happen
[12:32] Motor Loon you have an example?
[12:32] B.A.M.F. (lestermoore) im a plumber so if your pipes break let me know vincent your a dick
[12:32] Drongle McMahon Any mesh ... just watch the weight as you shrink it.
[12:32] DrFran Babcock /me waved to claudia
[12:32] Nyx Linden I do think there is a size constraint, but I don't recall the details immediately. This is a question that would be best answered by falcon
[12:32] Motor Loon I must not have made tiny mesh's then °͜°
[12:33] Drongle McMahon How do we extract the answer from Falcon?
[12:33] Motor Loon why would you want a little tiny mesh to have a real primshape?
[12:33] claudia222 Jewell hi all
[12:33] Nyx Linden also let's please keep the name calling to a minimum here, let's keep it friendly
[12:33] claudia222 Jewell hi drfran
[12:33] Asha (ashasekayi.ra) Hi Claudia
[12:33] Motor Loon "namecalling can only be done with love"
[12:33] Drongle McMahon It's no so tiny now, about 0.5m.
[12:33] Vincent Nacon cupcakes!
[12:33] Asha (ashasekayi.ra) /me snorts at Motor.
[12:34] Motor Loon °͜°
[12:34] Nyx Linden I'll talk to falcon later today and see if he can post on the forums or attend another usergroup meeting
[12:34] Tiberious Neruda yeah. I know a few objects where I'd love to have a smaller size collide 'properly'
[12:34] Nyx Linden I'll make a note to follow up with you Drongle, but want to get the answer into our documentation as well if possible.
[12:34] Motor Loon there's the thing about very close verticies being a huge strain on the physics engine
[12:35] Nyx Linden apologies for not having the answer in advance
[12:35] Motor Loon I belive Falcon has mentioned that before
[12:35] Vincent Nacon yup
[12:35] Drongle McMahon The physics weight wiki would be the best place, I guess.
[12:35] Motor Loon Might account for it switching to convex also
[12:36] Nyx Linden I'll follow up on this one. moving on!
[12:36] Nyx Linden next: "Why do spikes appear on lower LOD rigged meshes when they don't if it's uploaded by itself?Tiberious Neruda"
[12:36] Vincent Nacon oy..
[12:36] Drongle McMahon Yes Motor, I presume that's the reasom. It prevents accidental returns from huge physics weights.
[12:36] Motor Loon yes
[12:36] Motor Loon or too much resource use
[12:37] Nyx Linden sounds like its an issue in the rigging weights, but could also potentially be a bug on our part. Could you file a jira and send in an example dae file?
[12:37] Tiberious Neruda ........
[12:38] Motor Loon lol are those good "..." or bad "..." tiberious? °͜°
[12:38] Vincent Nacon Tib, you'd need to upload your own LOD with skinning
[12:39] Nyx Linden that certainly isn't intentional behavior, but we'll need some example content to look further into it. It could be an interaction between the skinning weights and our LOD library.
[12:39] Vincent Nacon otherwise SL's uploader would make wild guess at the skinning info
[12:39] Tiberious Neruda A) it's a bad '....', and B) it is a custom LOD with weighting
[12:39] Nyx Linden ah good, that will make it easier to hunt down
[12:40] Motor Loon Doing a jira ain't THAT terrible Tiberious °͜°
[12:40] Asha (ashasekayi.ra) I know that if you upload an LOD, you can do so with or without weights as long as the high LOD has weights. But, if you do upload a low LOD with weights, are they ignored?
[12:40] Vincent Nacon he has a fear thing about being bullied on JIRA that could get him banned from SL
[12:40] Vincent Nacon again...
[12:41] Motor Loon I'll kick their ass if they bully you T °͜°
[12:41] Nyx Linden screenshots and an example mesh would be most helpful for hunting down the issue.
[12:42] Nyx Linden anyone else seeing this behavior?
[12:42] Vincent Nacon I saw it before
[12:42] Nyx Linden (the more data points the easier it is to hunt down)
[12:42] Vincent Nacon I fixed it by skinning my own LOD meshes
[12:42] Tiberious Neruda but... this current mesh I'm wearing has the behavior. It's on the Lowest LOD
[12:43] Tiberious Neruda the other two are copies of the High
[12:43] Motor Loon does seems like a lot of work to "fix" if you need to skin 4 models for one item
[12:43] Vincent Nacon yeah...
[12:43] Vincent Nacon not fun
[12:44] Motor Loon Making good mesh is time consuming enough
[12:44] Motor Loon °͜°
[12:44] Vincent Nacon we should able to upload one DAE that has its LOD meshes inside it
[12:44] Motor Loon hasn't that been experimented with already?
[12:44] Vincent Nacon via labeling object with LOD#
[12:44] Nyx Linden anyways please forward me the bug information/report and I'll see if we can find some time to investigate.
[12:44] Nyx Linden next topic: "When uploading a specific .dae for the physics shape it will get scaled to fit 'the bounding box' of the main .dae you are uploading. If you create furniture with a cylinder with a plane under it that will be the shadow, and you make the .dae for physics just the cylinder... the cylinder gets wider to fit all the shadow and doesn't keep the size it should to cover only the part that should be physical. Will there be a way to disable that feature?Fem Darcy"
[12:44] Motor Loon I'm also still hoping for custom pivots °͜°
[12:45] Fem Darcy I will rez a visual example of what I mean over the pile of I-beams. (that's not mesh just a representation of what I see in the upload window, and will happen if uploaded)
[12:45] Vincent Nacon aye, same
[12:45] Motor Loon NICE model Ferm
[12:45] Asha (ashasekayi.ra) Amen to that Motor
[12:45] Tiberious Neruda if you drop object LOD to 0.0, and look at my av, you'll see the spikes
[12:45] Motor Loon but that is expected behavior though Ferm
[12:45] Motor Loon I belive
[12:46] Fem Darcy well expected or not there should be a way to avoid it... or you would never be able to do the model I have behind, corretly
[12:46] Motor Loon If thats changed, it'll certainly break all my physics models
[12:46] Vincent Nacon Fem, the physic mesh is based on normalized max scale of a mesh, in other words, you should break it down into parts
[12:46] Drongle McMahon There is a cumbersome work-around, as used for doors piveted at the hinge.
[12:47] Motor Loon yeah just add a invisible triangle
[12:47] Motor Loon but
[12:47] Drongle McMahon It's described in the example door in the wiki.
[12:47] Motor Loon it doubles the size or your mesh
[12:47] Motor Loon = more pe
[12:47] Motor Loon PE
[12:47] Asha (ashasekayi.ra) /me nods. "The problem with the pivot workaround is that it makes the bounding box bigger.
[12:47] Vincent Nacon and screw the physic as well
[12:47] Motor Loon Not happy about that
[12:47] Motor Loon Custom pivot should be easy
[12:48] Motor Loon heck... we used to have it in the early mesh days
[12:48] Asha (ashasekayi.ra) Prep said it was causing huge server issues though.
[12:48] Drongle McMahon But the bbounding box is already the size of the shadow bas and it doesn't need to be bigger than that.
[12:48] Motor Loon issues can be fixed
[12:48] Vincent Nacon in other words, Nyx. quit taking too long with the pivot held back
[12:48] Asha (ashasekayi.ra) Apparently not easily :P
[12:49] Motor Loon bah... it was just easier to take pivots out then to fix them
[12:49] Asha (ashasekayi.ra) /me shrugs. "Who knows.."
[12:49] Drongle McMahon I think the offset pivot problem is the physics engine has to deal with them.
[12:49] Vincent Nacon or instead of having custom pivot inside DAE mesh upload...
[12:50] Fem Darcy I hear what you say but I still don't see a optimized way to make this work, since the physical dae will be always scaled if it isn't as big as the model uploading
[12:50] Vincent Nacon why not just have ability to edit pivot in build edit window?
[12:50] Asha (ashasekayi.ra) That would be a nice feature Vincent.
[12:50] Vincent Nacon aye
[12:50] Motor Loon it would
[12:50] Motor Loon even better
[12:50] Vir Linden The change to the bounding box behavior might be worth considering, but we don't have any changes like that in the pipeline currently.
[12:50] Nal (nalates.urriah) Make lots of things easier
[12:50] Vincent Nacon also work on normal prims
[12:50] Vincent Nacon and easier to edit them on the fly
[12:51] Fem Darcy Would a Jira help with that? Should I create one?
[12:51] Motor Loon but thats a much bigger task than allowing custom pivots for mesh objects
[12:51] Tiberious Neruda my question....
[12:51] Vincent Nacon sure, go for it, Fem
[12:51] Tiberious Neruda why is the 'bounding box' enforced anyway?
[12:51] Vir Linden Yes, JIRA would be good.
[12:52] Vincent Nacon once you finished that JIRA request, slap it on Linden's forehead to remind them
[12:52] Drongle McMahon I think there may be a very old one in CTS, but a new one would add emphasis.
[12:52] Tiberious Neruda this literally means we can't have round objects using anything other than 4, 8, 12, 16 or ANY number of vertices other than one divisible by 4, onANY level of detail
[12:52] Nyx Linden please do file a jira, moving on for now
[12:53] Vincent Nacon she's on it! ( I think)
[12:53] Nyx Linden Has the source mesh for the default convex hull been changed again? It used to be the high LOD mesh, but I was just shown an example where it appeared to be made from the lowest LOD mesh or the physics mesh. Needless to say the result was useless. (Ah ... just heard it was uploded with Phoenix, so that may be the problem).Drongle McMahon
[12:53] Nyx Linden nothing has changed in that code recently
[12:54] Drongle McMahon OK. I should have waited .... it's probably just Phoenix.
[12:54] Nyx Linden I think its based off of medium or low, I can look up which it is, but that code has not changed since the official release to my knowledge
[12:54] Drongle McMahon Oh! I thought it was high. There were problems in the beta when they changed it to a lower one.
[12:55] Drongle McMahon Is it ever done from the physics mesh?
[12:55] Drongle McMahon I'll have to do some experiments again.
[12:55] Vincent Nacon yup
[12:56] Vincent Nacon you'll find out
[12:57] Nyx Linden convex hull is the physics representation, I think we use a LOD by default at first but if you specify a physics mesh and want to use a convex decomposition, you should go through the process to replace it.
[12:57] Vincent Nacon is that all the topics so far?
[12:57] Asha (ashasekayi.ra) I added a question to the agenda.
[12:57] Vincent Nacon well make it quick in last 3 minutes left
[12:59] Nyx Linden last topic: "I know that if you upload an LOD, you can do so with or without weights as long as the high LOD has weights. But, if you do upload a low LOD with weights, are they ignored? Ashasekayi Ra 12:43, 30 April 2012 (PDT)"
[12:59] Asha (ashasekayi.ra) I'm sure he will get to it when he is done answering Drongle. :D
[12:59] Vincent Nacon he seem to be on it now
[13:00] Smokie Sabretooth hi there
[13:00] Asha (ashasekayi.ra) Hi Smokie
[13:00] Vincent Nacon Sorry Smokie, you're a bit late for the meeting
[13:00] Nyx Linden I thought we loaded weight data for each LOD?
[13:01] Asha (ashasekayi.ra) You can add LODs with now weights though.
[13:01] Asha (ashasekayi.ra) *now = no
[13:01] Vincent Nacon Nyx, the skin weight is loaded from the high if other lods doesn't hav eit
[13:01] Vincent Nacon have*
[13:01] Vincent Nacon and it make a "educated" guess how it should handle the skin weight data down across the LODs
[13:02] Asha (ashasekayi.ra) I just wanted to know if you do add one with weights, do they get used or are they ignored.
[13:02] Nyx Linden even with custom lods?
[13:02] Vincent Nacon some times it works, some times doesn't when it's too complex
[13:02] Asha (ashasekayi.ra) Yes, I'm talking about all custom LODs.
[13:02] Ima Mechanique got to run, bye all
[13:02] Vincent Nacon seeya Ima
[13:02] Nyx Linden honestly I don't recall what the exact mechanism is there
[13:02] Rusalka Writer Aloha.
[13:02] Nyx Linden its been a while since I've been deep into that code :)
[13:03] Vincent Nacon dive dive dive!
[13:03] Asha (ashasekayi.ra) /me snorts
[13:03] B.A.M.F. (lestermoore) pa tu la
[13:04] Vincent Nacon and the thing is.... I wouldn't be surprised if I'm wrong, but from my experience it seem to work that way
[13:04] Nyx Linden I'll have to ask around on that one
[13:04] Asha (ashasekayi.ra) Alrighty then
[13:04] Vincent Nacon so I'd like to know more about it in case of something done wrong, that I'd like to know
[13:04] Nyx Linden thanks for coming out everyone, we'll wrap up here for the week
[13:04] Asha (ashasekayi.ra) Thanks for the meeting.
[13:04] Vincent Nacon sure, take care all
[13:05] Fem Darcy Thanks for your time =D Bye everyone!
[13:05] Rusalka Writer Aloha!
[13:05] claudia222 Jewell bye all
[13:05] claudia222 Jewell thanks
[13:05] Vaalith Jinn uh oh i messed up my timezones
[13:05] Frans Charming take care all. Thank you Nyx and Vir.
[13:05] Asha (ashasekayi.ra) Bye all :)
[13:05] Nyx Linden I'll try to get a transcript up later today if possible Vaalith
[13:05] Drongle McMahon Byeee
[13:05] Vaalith Jinn that'd be awesome, thanks :D
[13:05] Vir Linden Bye folks.
[13:05] Nyx Linden see everyone next week
Retrieved from "https://wiki.secondlife.com/w/index.php?title=Mesh/Archive/2012-04-30&oldid=1166605"