Polymath Snuggler

I ~ uhm. Err.. What? The internal animation file for Second Life is a LLKeyframeMotion File ~ also known as a .anim file. It's a proprietary format.

Okay ~ I want to preface this by stating ~ I have no idea what I'm actually saying here~ this is a guess. But when you load an animation on an avatar in SL ~ isn't it calculating how to move and rotate those joints based on a pre-existing keyframed dataset? There is no realtime adjustment to the actual keyframed data based on the size of the joints in the avatar itself. That's done during the animation pre-load ~ not while it's being played realtime. So wouldn't scaling bones mid-animation require an entirely new method of playing animations to be used ? Please, someone who actually knows what they're talking about correct me!

I created a JIRA post~ "Bento~ The Yawning Problem" I set the post to Private, as I don't particularly want to deal with DMCA sillyness. But it's been posted. As a summary to the people here. The file I posted is the one used to generate this animation: The Yawning Problem.  The problem itself deals with floating the corners of the mouth to solve for two expressions, a gaping yawn and the rest of the regular expressions a face will generate. The problem is~ I'm struggling to find a joint location for mFaceCornerRight and mFaceCorner Left that will animate a wide yawn~ and a smile. This is a very very specific example that highlights the problems with a fixed point rotation based rig. Not only is it tedious and difficult to use~ but the designer is forced to choose between making the character able to yawn in a convincing manner ~ or making it able to smile in a non-creepy fashion. It's a blatant either-or design choice that has to be made that shouldn't have to be made to begin with.

Both! Fix the people by teaching them! Thankfully learning about polycounts and render efficiency is just a few clicks away on youtube~ for those who actually care to look! Fix the tech by making it more obvious and user friendly! ( better feedback! ) SL is finally implementing features to raise user ( and designer ) awareness about mesh rendering complexity costs. I'm sure Sansar will have a similar type of indicator. ( Hopefully even better ones!! ) Sadly, due to the framerate expectations that Sansar has on it, it only really has one option: well designed (optimized) mesh assets. The type that every other game engine in the entire world relies upon and uses. So~~ SL people will have to learn a little bit more. Every other content designer won't even bat an eye. After all Second Life is the only game type platform without incredibly restrictive polycount rules for content submission. That's okay!~! Necessity will dictate improvement~ I have very few doubts that LL will get their builders to populate the new world; as unemployed 3D artists aren't exactly a rarity. I'm sure you'll manage to come up with some horrible gloom and doom scenario with this new-found knowledge but! I'm pretty confident that the average mesh designer on SL will adapt to these new circumstances by the time Sansar rolls around and actually becomes 'a thing'~ Anyhooo !! Glad everyone understands this stuff a little better!!~~ back to Bento testing for me!!

Ah HA! That was far before my time. Good bit of insight!! Thank you!

An excellent question!~ If you pop into wireframe mode you can see that a prim is in fact subdivided, even as a flat and basic cube. Your simplest mesh box will have 8 verticies, and 12 triangles composing it. A basic prim 56 verticies and 108 triangles.  I'm not entirely certain what the logic for subdividing the basic prim was. I'm sure it has something to do with making the prim bend and twist and taper in world~ But at the end of the day the prim is inefficient to render. Mind you, you are correct about the prim being efficient over it's mesh cousin in other regards. It's download cost is lower as your viewer has a basic prim box cached at all times, so it doesn't have to go to the server and ask the server "hey how's this box made." In that fashion~prims are very efficient, there is minimal network cost for using them. Likewise with the physics calculation~ with the prim it already knows that value ( cached ) so it doesn't have to ask the server. Whereas with the mesh box your viewer has to specifically go and ask the server "what is this mesh's physics shape??"~ when it deals with the mesh box. So again, the prim is efficient in that regard. However both download costs are network traffic costs. Which LL cares a TON about as every bit you download costs them USD~. The render cost, is an ongoing 'cost' that your computer has to deal with in order to present you a pretty picture of a box in SL once all the downloading is done. That is dependant upon how many verticies and faces comprise the cube, and that alone. That render cost is 'paid' every microsecond ~( usually 15-60 times per second ) by your graphics card in terms of time that it takes to think before it puts the picture on your screen~ So, while your graphics card is exceptionally great at drawing literally millions of triangles per second, it starts to get a little bogged down by polycounts after you give it too much of a workload, and your FPS drops!! Which is why Sansar, with it's 120 FPS target benchmark due to LL's obsession with making it compatible with the occulus doothingie, needs efficient assets. Presently SL, even with a decent graphics card, once you turn on the visual bells and whistles seems to tank at 15-40 FPS with that number falling under 10 FPS on lower systems or with avatar impostors turned down. The quickgraphics beta viewer attempts to approach this problem by auto-derendering any avatars above a certain complexity so that your computer doesn't have to work as hard, but it's a band-aid on a much larger problem. Anyhow!! Hope that explains it!! ~ EDIT: I figured while I'm screenshotting things in wireframe mode, I may as well put graphical evidence of what a bad designer does. ( okay ~ maybe bad is a bit harsh... inexperienced? )  Behold~ ! The 30,000 triangle, 15606 vertex cube. It still looks like a cube, but it's going to have a whopping download cost and an even more non-sensical render cost. Both the upload cost and the land impact of the cube will go up to reflect that Linden Labs frowns on your lack of polygon efficiency, but if that box is supposed to be sitting atop someone's head, suddenly people don't care quite so much. The problem is people do this. Not with a cube obviously. But a lot of SL designers have an obsession with adding every single little bit of string~ every hair~ every microwrinkle in their dress that they made. Lo~ and behold mesh "is the cause of all lag." Why didn't LL add in an accounting system for worn mesh objects? Beats me!! HUGE goof in my opinion~ but that's water under the bridge. Seeing how they're now addressing the problem in SL ~ and listening to Ebbe talk about Sansar efficiency etc etc~ I'm confident they won't be making that mistake again. Which is also ~ why I made the comment earlier in this thread, that "I wasn't sure all SL merchants would make the cut for Sansar" as many of the "big brand" merchants in SL still have horrendous topology. If the Sansar marketplace is ranked based on topology and render complexity, it will be a very very different scene.

I... err~ Okay~ I'm just going to try and ignore the condescension in your tone and presume you're really just uneducated on this topic. You realize you're talking to someone who's worked extensively in this field right? Worked in it for years. I learned mesh before I came to SL~ I know about optimization. The fact that mesh is superior to prims in efficiency is not a theory, it's a simple matter of numbers. I'm not going to get into the individual counts of each specific prim shape and type, but I can point out one prim cube is 9X as complex as a simple mesh cube. ( that's at it's most basic form, not including "tortured" prims ), the inefficiencies only get worse from there. To reiterate that, a box being 9x as difficult to render as it's mesh counter part is the best numeric advantage, it only gets worse from there for the other prim types. The "mesh based lag" you mentioned is because designers upload 100K- 1.2 million polygon objects to SL. It's not because mesh in of itself is "bad" it's because there's such a thing as a clueless designer. Just for a frame of reference, the content I'm describing with these polycounts can usually be done with 6-15K polys. Yes that's a factor of eighty in inefficiency. Yet you're sitting there going "jeez I wonder where all this lag is coming form?" Polycounts (number of polygons in a mesh item) must be monitored ( on SL they weren't, they're just starting to do that now with Project QuickGraphics ) on Sansar you can bet your butt they will be. You cannot render a scene composed of inefficient content at 120FPS (which is an occulus rift baseline number) on current affordable tech ( I'm not counting having 4 + crazy super graphics cards in your PC ) So that's why the limitation. It's not because LL hates you, or because they're clueless. In fact it's completely the opposite. They've finally learned from their mistakes and are doing something right. So ~ I can understand your confusion, but the reality of it is, SL is outdated a 12 year old dinosaur which can't just be "updated" into a 2016 modern era platform. You are correct that Sansar is in all likelihood LL's last grasp at staying afloat. But, while I understand you're attached to Second Life, I'm sorry, but the grim truth is LL needs to reach a wider user base, and they're not going to get that on SL no matter how much they improve it due to it's server architecture. Sansar is, as I have mentioned before, a platform. A platform ( or engine ) like Sansar will be competing with Unity, and Unreal4. It inherently is not 'shiny' itself. Much like those other engines it can make shiny things, or it can make cartoony things or if you want to, it can make an entire world comprised of just boxes~ ( yay minecraft! ) So yes, you are quite right. Sansar won't be successful without users. But Sansar isn't a virtual living space. Sansar, by it's nature will attract different customers who have different needs. So, we're back to apples and oranges. Nyoooow~ I'm going back to work! PS: Don't act all condescending towards the person with the technical knowledge you just wind up looking silly! Thankfully I like explaining and teaching things to people!!

I don't get how LL is "artificially" limiting a creator base when "skills cannot be mastered by all". Simply put creating good 3D content requires expertise. Prims were a neat idea for SL but realistically for the framerates that Sansar is attempting to run at ( 60 - 120 FPS ), they're just simply not viable. There is no substitute for knowledge on how to create efficient content. There are some really fancy shortcuts that have been created by the fore-runners of software designers ( like Pixologic ) but they're for optimizing already existing mesh assets. The industry as a whole is always looking for new ways to make content creation easier, faster and more intuitive for artists. But realistically if Walt Disney Studios hasn't developed a way to do it, with their budget, I wouldn't get your hopes up for Linden labs. There's a reason we, the content designers still are "employed". It's because making things properly is a skill. While SL is a great place to learn, eventually when you want it to run at 120 FPS, you're going to have to learn something about optimization.

Maya LT was specifically included on my list as it's got an affordable monthly fee for use!! It's 95% of the standard Maya software ( which you can get for free to learn! ) And doesn't cost 3600$ USD up front. http://store.steampowered.com/app/243580/

This is what SL IS FOR, it's not going anywhere!

My post was about learning !! I'm not passing incorrect info around!

It isn't my analogy.

That's because brick and mortar places tend to be woefully inadequate when searching for a very specific thing~ thankfully ~ through the magic of online marketplaces, I've found a solution for you! http://www.amazon.com/Gogh-Stained-Glass-Coloring-Book/dp/0486456714 ::winks:: In all seriousness though~ I learned mesh way before I came to SL ~ My cube-clunking took place years ago. For years, it was for me, what it is for many people on SL~ a wonderfully time consuming hobby. Then at the prodding of my friends, I started to take it seriously. If you want to talk developmental & incremental training though, I recommend these things: http://www.lego.com/en-us/ideasproducts Which gets people thinking spatially~ then you can graduate to https://minecraft.net/ By the time that gets boring you might consider something like https://secondlife.com or http://www.garrysmod.com/ Blender is a great way for people to start "clunking mesh cubes" so to speak. And the ENTIRE autodesk software suite is available for free on a limited 3 year student license. https://www.blender.org/ http://www.autodesk.com/education/free-software/all http://www.autodesk.com/products/maya-lt/overview http://pixologic.com/sculptris/ My point is this~ There's hundreds of thousands of us out there~ with spatial minds, who model and build in our spare time. The "barrier of entry" into Sansar is hardly the Black Gate. And even with that towering edifice~ a pair of sneaksy hobbitses found their way around it because they wanted to. If everything is as bad as you think it is, none of these communities would work: https://player-studio.daybreakgames.com/ http://steamcommunity.com/workshop/ http://www.nexusmods.com/games/? https://www.assetstore.unity3d.com/en/ https://www.unrealengine.com/marketplace And that's not even counting all the professionals. Anyhow~ since none of us really know what Sansar's market is going to be like I feel like perhaps it would almost be more productive if we discussed how a dog would wear pants.

I don't have "insider information" I just peruse every little soundbyte that is recorded from Ebbe whilst working away on my projects. I think you've misunderstood what a "platform as a service" is, so I'll do my best to clarify! ^^ https://en.wikipedia.org/wiki/Platform_as_a_service A "Platform as a Service" is an attempt to make some set of tasks "more user friendly". The expression is usually used in relation to webhosting, but Linden Labs seems intent upon making that level of usability available for Virtual Reality. I believe the term "Wordpress of VR" has been tossed around rather extensively. So !! What this means is the idea is basically to provide people the means to "create a virtual world" easily, whether what that person creates is a 'on the edge of your seat' shoot-at-things game, a sex simulator or simply something as mundane as letting prospective renters wander around a scale model of the floorplan of an apartment building, LL wants Sansar to make setting that whole thing up easier. The idea is~ no matter what the designer wishes, they can build, host and invite people to their special experience they've created with the help of the designers on Sansar and the hosting provided by Linden Labs. The degree of "flashyness" and interactivity etc etc will be entirely dependant upon the experience designer. I've often stated this when discussing Sansar. Sansar is the platform that could be used to create SL 2.0 if someone decided to undertake such a task. But Sansar, in of itself is not SL 2.0 EDIT: Muph~ replying to your 2nd essay too Darrius. Sansar will have a reduced number of merchants providing services and goods on the platform due to the slight increase in design complexity. Sansar has ( as best I can understand from what I've heard ) no Prim system, it's scripting language is C#, and everything that is set up "in world" will be made out of mesh. So designers for Sansar will need ~ at the very minimum an understanding of meshes, UV maps and some rudimentary programming. Which isn't all that different from modern era SL~~ The big difference is, you can't start off knocking two blocks together in world and "grow up" into a designer. Which is sad really ~ but, it's been mentioned that SL isn't exactly going away~ so for people who have the desire to clunk boxes together can still do so. SL is presently full of 1000's of creative people who stopped building when mesh came along because mesh was too complex, and they felt their work got overshadowed~ as their prims & sculpt box clunking never quite could contend with the sleek shiny meshes in their eyes. Unfortunately those designers won't be able to find a home on Sansar ((unless they learn mesh!! It's not that hard! )). Everyone else who's learned a little mesh, (theoretically) can. Although I personally suspect even some of the biggest brand names in SL will struggle to meet Sansar's build-requirements. Polygon & Texture efficiency seems a foreign concept to many an SL designer. :matte-motes-crying:

I can answer that for you easily. Bones cause lag, and not a little, a lot. More bones, more lag. As you might surmise where I'm going with this ~ adding bones for toes, which are appendages that express nothing, and are rarely if ever seen, and almost never are required to move much, if at all. ( as they're usually inside shoes ) is ill advised.

To a certain extent ~ I can see where this line of "animation re-use" thinking comes from. Current morph target based expressions can be integrated into AO's and customers have that expectation. I'm not entirely sure which crowd LL is trying to cater to here~ But, that really requires a meeting for us to truely understand the motive / intent. I suspect that meeting will occur after Jan. 3, 2016, as LL is on vacation until then. It is worth noting that ~ for a humanoid head a translation based algorithm will actually produce a more "reuseable" animation than a rotation based heirarchy for an average sized head. As bone rotations have to have their location set so far back away from the actual point of movement that the designer is trying to emulate a translation on that it somewhat eliminates any hope of reusablility, as the joint locations will be different for every single project.

There's a fine fine line between releasing a pre-alpha that only has "sketched" in features that vaguely work or don't work at all (thus inciting an angry mob here) and keeping your customers ( and competitors ) in the loop on what your newest greatest thing is. So they can work on it ( or beat you to it ). Speculation is fun and all ~ but until we even get a firm understanding of what Sansar is ~ I find theorizing about potential market discrepancies to be largely pointless. It's been indicated multiple times that Sansar & Second Life will likely not attract the same customer base as Sansar is largely a "platform as a service" rather than a "virtual playground". That being said ~ the sales model for Second Life will probably be completely obsolete in terms of it's useability for Sansar. In second life we have individual inventories, each purchased item is guarded against duplication and transfer to another inventory of which is owned by a different person. The way Sansar is sounding like it's shaping up to be~ they're going for more of a hosting service, which means that the vast majority of connected users will have no interest in "inventories" or "avatars" at all, they'll be there to look at their apartment floor plan or shoot their monsters or play their game ~ etc etc. The point is, it's a fairly reasonable assumption that the majority of sales will be on a per experience host basis~ rather than a per user. AKA ~ Everything you, as a creator sell is 'full perms' to be used in an 'experience'. The purchaser will gain rights to use that item in whatever experiences they create. Connected users will gain temporary use out of whatever resources the experience offers, and then will not retain that once they leave the simulation. TLDR~ I think ya'll are comparing apples to oranges here! But it's fun to think about!!

I posted some more specific feedback with a bone-set breakdown on a feature request JIRA Vir. https://jira.secondlife.com/browse/BUG-11031 It is worth mentioning that range of movement not covered by the skeleton has "traditionally" been handled by multi-mesh alpha layer swapping meshes which are absurdly high in polycount and even higher in render complexity. ANY improvements over the 1.2 million tri Catwa Heads is an improvement. But I think the community at large at this point is still exceedingly curious exactly what the nature of the problem is with the current implementation of translation based animations? Is it a latency problem? A computational issue? A security issue? Why can't we have our cake and eat it too?

The vast majority of content on the marketplace is Mesh, Sculpt, Prim and Texture work that needs little to no support at all. It's valid content that people enjoy. There is no "theft" in revenue generation off of 'abandoned property', especially when that property is digital content. There is no metric for how long someone has to be "away" from SL to 'lose the rights to their account'. I literally just talked with a builder who signed in yesterday for the first time in 6 years. She was shocked to see her build still rezzed on the sim on which it was created. I will concede that there is content on the Marketplace~ predominantly script and service related, that requires support. Someone might be able to make an argument for possibly flagging that past a certain point. But again ~ I still feel the review system provides more than enough insight for would-be customers. There is also a substantial amount of content that presently requires third party servers to function { *cough* THE MESH PROJECT *cough* } that will just cease working the day that entity decides they're done with SL. That's all throughout SL and there's really no warnings that any customer can have about the internal programming of any object that they buy. There's no review system warning them of the hazards. And there's no way that LL can monitor that either. The only way we~ the 'customers' can find out about it~ is if the service stops working and suddenly every mesh body sold by TMP suddenly doesn't have a skin on it for a few hours ( "whoopsi" )? The script functionality that leads to the above scenario is required for something like Caspervend to work as well. ( Which the day that Casper decides he's done with SL, will also cease functioning, not that I'm expecting that to happen any time soon ) But that's just part of ANYTHING with an online dependency. Sometimes you buy something and at a later date it breaks! At least with a review system in place people can communicate that to others before the purchase, unlike with in-world sales. TLDR; Old SL Marketplace content is no more or less disappointing that buying something from a mail order catalog only to have it arrive and not be quite what you had hoped for. It's also not any more legally actionable against than that same scenario.

Actually LL is extremely disadvantaged by this content disappearing, especially if it's valid content. It provides people something to shop for, it provides LL with pure profit~ as Lindens are purchased, transferred but never resold for USD. It provides the content creator with actual rightfully earned money ( in most cases ) should they ever choose to cash out. There's no earthly reason to remove old listings. The remaining complaints are handled by the review system and content flagging system. Edit: It's also worth mentioning that old listings very rarely show up in "the front page" of search results, as their purchases are infrequent and since the creator is not active they get pushed down in the search rankings. So, especially with new search coming down the line~ even less reason for content removal. Edit #2: I'm going to offer this hypothetical situation as well ( because in all likelihood with the number of terminally ill people in SL it's not a "hypothetical" at ALL) That someone creates a successful store, succumbs to their illness and their next of kin will log in and take the money but never ever ever touch the avatar itself because of sentimental reasons.

The problem with your request is the majority of marketplace content that is posted by avtars whom have left SL is vaild functional content. The people who purchase it are largely satisified and everyone goes home happy. So the problem with your request is you can't mathamatically define "usefulness". It's a subjective term dependant upon the purchaser. That's why we have a review system in place, so that people who are dissatisifed with their product or the service with the associated product can voice that they are dissatisfied and everyone else can read that prior to purchasing. If you're buying a product that has the last 10 reviews posted "Do not buy, creator not in SL for a decade" then ~ I'm going to say that's more or less your own personal negligence.

I would urge you to not make any specific declarations about what Bento can and cannot do, as we, the designer community work with Linden Labs to try and sort out what is needed and what is not. Please don't rush to conclusions on this ~ especially with a limited understanding of the material. The last thing we want to do is promise the community at large something that cannot be delivered.

Little update on this for those who may be struggling with the same problem. Once I included bones in the export to FBX file the problem with all the bones collapsing the the origin went away~ Still figuring out why the actual joint positions aren't recording. Fixed!! Removed the excess bones from the file ( mPelvis is the root of the heirarchy ) Exporting now is working properly!!  Also worth mentioning that 2013.3 FBX Converter works fine. As well as straight export out of 3ds Max 2015 & 2016 straight to DAE Though that can occasionally throw warnings/errors during the export process but it still remains functional in world.

I linked the video before, but I realized after the fact that no one is really going to understand what that video is. The voice in this video is Paul Neale, a person who's knowledge and expertise I've grown to rely upon for a substantial portion of my training. His list of credentials as a Technical Director is completely eclipses anyone else's on these forums, He's worked for Walt Disney Animation, 2K games, Rockstar Games, Elliot Interactive: You can read the full list here : http://penproductions.ca/clients.htm#Games My point is this: This man knows just about everything about setting up a character rig that there is to know. Paul Neal : The number one thing I find new Technical Directors want to do is they want to try and limit controls for animators and you know make sure for instance that the elbow doesn't rotate too far, or it doesn't rotate backwards. Animators will hate you for that without question. They want to just unlock everything. The current rig I just did, they wanted to be able to move and rotate every joint in the body even if that joint isn't supposededly able to move, rotate and scale by the way, so everything needed to be able to move, rotate and scale for them to be able to get the motion they wanted. They didn't want to run into the position where they suddenly couldn't do something. And that's you know especially true for cartoony characters. I will do my best to post some more specific examples from my own work ~ but I'm in the middle of another project right now and it's a bit of effort to get this rig into a head in a functional manner ( as I don't use the foodprocessor as my main piece of software ) I will post more later!!

Back to the Bento Project itself~ 3ds max users are having some odd sort of glitch relating to bone positions. It seems to be moving a joint and then "including joint positions" causes SL to interpert the 3DS max exported file as having all joints centered at the origin. Like this  Are blender users having this problem? ~ or is it just us on 3ds max? Basically~ Is this me or should I file a JIRA?