The following notes are taken from the Content Creation User Group meeting, held on Thursday, July 27th, 2017 at 13:00 SLT at the Content Creation User Group wiki page.. The meeting is chaired by Vir Linden, and agenda notes, etc, are usually available on the
Medhue Simoni live steamed the meeting to You Tube, and his video is embedded at the end of this article. These notes present the meeting in terms of topics discussed, rather than a chronological breakdown of the meeting, so the provided time stamps may appear to be out of sequence in places. All time stamps are provided as links which will open the video in a separate browser tab, allowing the discussion to be heard in full.
Note: Due to the monthly internal meeting at LL and Vir’s time on vacation, there will only be two CCUG meetings in August: Thursday, August 10th and Thursday, August 31st. Details will be posted on the wiki page.
Bakes On Mesh
Extending the current avatar baking service to allow wearable textures (skins, tattoos, clothing) to be applied directly to mesh bodies as well as system avatars. This involves server-side changes, including updating the baking service to support 1024×1024 textures. This may lead to a reduction in the complexity of mesh avatar bodies and heads.
[1:31-2:01] The project is reaching a point where internal testing at the Lab can begin, allowing the impact of the text increase to be assessed. If this proves successful, the work will start the march towards more general visibility (e.g. availability of a project viewer, probable Aditi testing, etc).
The goal of this project is to provide a means of animating rigged mesh objects using the avatar skeleton, in whole or in part, to provide things like independently moveable pets / creatures, and animated scenery features via scripted animation.
- At this point in time, this is not about adding fully functional, avatar-like non-player characters (NPCs) to Second Life
- Animated objects will not (initially):
- Have an avatar shape associated with them
- Make use of an avatar-like inventory (although individual parts can contain their own inventory such as animations and scripts)
- Make use of the server-side locomotion graph for walking, etc., and so will not use an AO
- Use the avatar baking service
- Be adjustable using the avatar shape sliders
- The project may be extended in the future.
- It will involve both back-end and viewer-side changes, likely to encompass new LSL commands to trigger and stop animations (held in the object’s contents)
- It will most likely include a new flag added to an existing rigged object type in order for the object to be given its own skeleton.
[2:04-2:20] The focus has remained on getting wire frames and right-click selections to work correctly ((e.g. when you right-click on a mesh, the object stops moving, the correct menu is displayed and the mesh is shown as a selected wire frame). Testing region crossings with animated objects has also started.
[2:21-3:26 and 4:19-4:47] Sitting avatars to animated objects: this has been part of a wider discussion on attaching avatars to animated objects and vice-versa. Vir’s view is that there is no restriction on avatars sitting on animated objects, however, the catch is that the sit point isn’t going to be animated – if it is, and as there is no relationship between the animated object’s skeletal locomotion and the avatar’s locomotion, the two will get out of sync. One suggestion for dealing with this is BUG-100864 “A means of visually rigging a sitter to an animesh skeleton bone”, and Vir indicated that the Lab is thinking along those lines, but it’s unlikely to be in the initial project viewer, when that appears.
[3:33-3:57] Animated objects as avatar attachments: unless there is an unforeseen issue, Vir is hopeful this will be possible. However, it is still awaiting work.
[5:19-5:57] Physics for animated objects: should work the same way as for non-animated objects, although this has yet to be tested.
[7:10-8:43] Scaling animated objects: this will not initially be possible using the avatar shape sliders, as animated objects will not initially have any notion of the avatar shape. However, it will likely be possible as a result of follow-on updates to the initial work.
What Vir is hoping to achieve is a method for reliably scaling an object’s skeleton based on the object’s own scale. That is: you could have three different sizes of an object (baby bear, mama bear and papa bear, say), and the skeleton will scale to whichever model is applied to it, rather than having the mesh default to the size of the skeleton (as is currently the case), which is currently defined by the joint positions.
[8:48-12:33] Shapes and Skeletons: the reason for no body shape support at present is that it makes animated mesh a much more extensive project, requiring the objects have a Current Outfit Folder, which requires them to have a dedicated, avatar-style inventory, make use of the baking service, and so on. All of this makes for a far more complicated, drawn-out project where the Lab would prefer to develop capabilities incrementally, starting with the provision of the avatar skeleton and then building from there.
This is seen as preferable to trying to incorporate everything people want to see – or believe is required – at a first pass, driving out development over a much longer period and risk developing a feature set the wider creative community in SL doesn’t want. Developing incrementally means features can be built upon and the project as a whole iterated, with deliverables presented in much shorter time frames.
[13:49-15:10] Land impact: this remains a concern for several reasons (too high, and it could stymie the use of animated objects, too low and it could thump performance for the viewer / simulator). Right now, the Lab has no clear idea of how LI will be calculated for animated objects, and Vir re-states that any LI values provided in the project viewer will be place holders which will be refined as testing and creator feedback gives more information on what the calculations should be.
[16:45-18:04] LI and scaling: concern was raised that if an animated object does not affect the bounding box, but could be scaled via animation, it could lead to the LI being game. Vir pointed out that scaling via animation is something of a hack, and for scaling with animated mesh, he is referring to the scale of the skeleton being determined by the scale of the model (which would effectively be a “static” scale), rather than having something dynamic. As such, the bounding box should reflect the object size and thus correctly influence LI calculations.
[20:14-24:41] Rigging to attachment Points: rigging to attachment points is being seen by some as the ideal means of animating attachments (e.g. twirling a gun in your hand), due to accuracy involved. Vir’s view is that problems people have encountered in uploading items rigged to attachment points is more of a bug with the LL viewer, and so the behaviour will be allowed (with the possible exception of the newer Bento bones attachment points); there is. however the concern that the ability might lead to attachment points being used as additional free-floating bones.
[37:50-39:36] Animated objects in games: could they be used for interactive elements in games, such as walls which bend when walked into, items which might interact with one another / players, traps which could physically react to an avatar (something wrapping around an avatar, for example). Short answer: yes and no. Yes, these are various interactions that are possible: trees swaying in the breeze, animals and other creature roaming and responding to avatars. No, in that animated objects will not initially have their own physics, so wrapping around an avatar, being used as some kind of clothing, etc.
[39:39-41:58] Will there be limits places on the number of animated objects in a region: there will be limits, but what these will be cannot really be determined until testing can be performed and the Lab can get better metrics on likely performance impacts animated objects have. This could also feed into how the limits are set (e.g. through the LI applied to animated objects, or limiting the number of animated objects which can be attached to an avatar or which can follow an avatar, etc), all of which might be used individually or in some combination(s) depending on the objects in question. Impact viewer-side could also be limited by having attached animated objects impact the avatar’s rendering cost.
[42:00-42:21] Imposters are also likely to be extended to apply to animated objects, although work hasn’t started on this a yet.
[0:44-1:21] Bento wiki information: It was mentioned in a previous meeting that some of the Bento wiki content was broken – links weren’t working expected downloads weren’t available. This should now all be fixed.
[25:04-36:15] Development kits for the default mesh avatars: In short, nothing planned on the Lab’s part at present, although due note was taken that there could be potential for such kits and the provision of better starter content for new users to help them in their understanding of what might be possible in SL with content creation.
The idea behind the initial question being to help give those new to mesh content creation the means to better understand what can / cannot be done with mesh in-world, get to grips with some basics of mesh development and modelling. This quickly expanded into discussions on “good” and “bad” content, broadening the availability of of content guides / best practices through to more formalised attempts at education those coming into mesh content creation An argument against this is that it could lead to misunderstands and the creation of poor content in SL, with the suggestion that more extensive best practices guidelines would be better.
[43:20-49:29] Why can’t animators replace default facial expressions in the same way they can replace walk animations? Because AOs affecting walks, sits, etc., all interact with the server-side locomotion graph which has a notion / manages these things. Facial expressions, etc., are not recognised by the locomotion graph, but are enacted viewer-side and the results effectively “passed through” the simulator (which is aware an animation – smile, frown, whatever – is being played, just not what the animation is actually doing). There is the potential to change this by extending the animation system, but outside of supplemental animation, there is no current commitment to doing this at present.
This discussion extends out into a discussion of the system avatar morph capability and sliders / limitations, which runs through until 53:11.
[54:13-1:03:00] Adjustable walk / run speeds: the ability to adjust / scale walk and run speeds to be in accordance with the size of an avatar, etc., has been a common request (see: feature requests BUG-7006 and SVC-7824 for example). Vir points out that currently, the speeds are set simulator-side and that adjusting them of any on-the-fly changes could be problematic as it involves an array of simulator and viewer changes. As such, scripted capabilities which adjust the viewer-side animation speeds might be an easier solution (Tapple Gao already supplies an AO for avatar creators which allows for some degree of speed control in their products, but something that is more generally usable is seen as ideal).