Medhue Simoni

I've had a few people ask me to make a video on how to UV unwrap in Blender. So, here it is.

Of course, it wasn't my idea. Really, IMHO, it is just the conclusion that any 3D artist would come to after thinking about it for awhile. This probably explains why LL can't think up these things, as they are coders with no real 3d creation skills. It seems to me that they are trying to do as little as possible. If they thought about this at all, they would have come to the same conclusion, saved the company money, and made every1 happy. Sadly, I agree that this will likely never happen, not because it can't, not because it's hard, but because LL doesn't give a crap. Just look, the person that actually came up with the cbone solution, also gave this solution, and LL did nothing, even tho he made a jira and everything. This is why I don't do jiras anymore, or at least, not LL jiras.

I'd chalk it up to how Fitted Mesh works. My guess is that the collision bone axis is causing this behavior. I rigged a full body suit, and when the avatar follows the mouse, you can see the body and the mesh are moving independently. I'll likely do a video about my final reactions to Fitted Mesh. It's not as bad as I intially thought, but IMHO, it isn't good at all. Even at this point, it a barely workable hack.

SolasNaGealai wrote: Hello medhue, thanks for the great video! I to love the way Inworldz's deformer works, i spent hours in maya rigging to the new SL bones and after about 2 tries it still needs more work. Tell me, do you know if it's best (when in SL ) if the "person" wears the default avi shape, puts on the fitted mesh, then change to the shape they want or should the fitted mesh adjust instantly to any shape when it is warn? You should be able to wear the Fitted Mesh without changing your shape at all, and (supposedly) it should form to you. Fitted Mesh is based on those shape morphs and you shouldn't need to reset your avatar for them to work.

Rhys Goode wrote: Oh, there is no doubt that it works, after a fashion, and even makes a sexy looking turtle neck. But thats because the SL avatar is pretty much fully masked out. You are seeing how the mesh of the sweater changes with sliders, not how the SL avatar changes, it would be bulging out left and right. Now, this is not really all *that* bad a thing. When I wear a top like that in RL, between my bra and the sweater, my shape is changed by the clothes. They do not look painted on. No, the bad thing is that it will only really work for cases where the clothing mesh completely covers the whole chest. It just seems that there is a fundamental conflict between Fitted mesh and decolletage. Exactly! LL's video is created to hide all the problems with Fitted Mesh. The whole process is convoluted to start with. I dare some1 to make detailed instructions on how to make a perfect set of collision bone weights, as It just can't be done. Even explaining it to some1 is more like explaining how to solve a rubic cube puzzle. This is a hack, and it is still a hack, not a functioning system to solve a problem. This is not how finish solutions are handed to creators. If LL wanted to use this system, then IMHO, the proper way would have been to create a cbone for every single morph on the avatar. Meaning the Butt cbone only controlls the Butt morph, and a Body fat cbone controlls only the Body fat morph. In this way the creators know exactly which bones they need to weight to get the right affect. Yes, this would mean many more bones, but once you got some good weights on all those bones, you could use them on most of your clothing with minor adjustments. What LL made for us, is a puzzle with no good solution.

Pamela Galli wrote: Okay scratch that, as far a definitive. It will work but apparently if you do any editing at all, the mesh becomes corrupted again. I know this is old, but I'll mention this anyways. Of course, I have no clue what might be wrong, but 1 way to kind of reset everything, would be to save the mesh as an OBJ. You can then reimport the OBJ, and the mesh will be devoid of any of the artifacts from previous editing.

Theresa Tennyson wrote: I don't have any technical experience with mesh clothing but I have a great deal of experience with real-world clothing and with putting together avatars of a wide variety of sizes, some quite non-standard. I also have experience as an end-user with the parametric mesh deformer, both when it was in beta on SL and currently on InWorldz. When the deformer was being tested it was actually clothing creators who had the most complaints about it. If you're using anything but the very simplest tight-fitting garments - for example, a suit with a lot of 3D detail- and you use it for an avatar with an unusual shape it will quckly make those garments fail in ugly, unpredictable ways. Clothing designers in InWorldz don't have to make five different sizes of clothing any more - they make SIX (five standard and a deformable version) because they don't trust the deformer as a one-size-fits-all solution. I've also tried clothing based rigged to collision bones - yes, they don't perfectly respond to the avatar's shape, but I feel they look much better than the results of the deformer for avatars with unusual shapes. There are now three design strategies for SL-type mesh clothing systems - standard sizes, liquid/fitted mesh and the parametric deformer. The first two systems were actually developed BY clothing creators, unlike the third . If I was starting a new world I wouldn't rely on any of these methods but we don't have that luxury. Well, I was just finishing up a dress with a puffy top, and skin tight bottom. I decided to upload it to InWordz to see how it performs. It was flawless. The puffy top didn't have even 1 issue. Notice how I didn't do any extra work to that dress to make it work with the deformer. When creators were complaining about the deformer, I explained to them that they need perfect weights to get perfect results. Getting near perfect weights for normal deform bones is not really that hard. So, basically, they complained about the deformer, and LL gave us this, which is 1000 times harder to weight. It's nonsensical. As I said earlier, I don't really make clothing. For the most part, I'm just rigging the clothing for other creators. My advice to them, is going to be to start selling in InWorldz, where they have a good solution. I suspect that the people that complained about the deformer were LL buddies, which is why the deformer got scrutinized to death, and Fitted Mesh saled thru without even head morphing cbones or a crotch cbone for male avatars. The difference between how LL handled both projects show the bias that LL had. It's just sad. The best solution would have been to allow creators to import morphs, or shapekeys. Implementing something like that would have be awesome for many more people and uses than just clothing. Unity3D just implemented shapekeys. You can even animate the shapekeys. LL tho, they twiddle their thumbs and respond with, "That's too hard".

Marielle Caerndow wrote: I'm also very sad that the deformer was denied, and the cBones are now the way to go (or not). My problem is not the extra work, nor that there'd be something new to learn - my problem is that it simply doesn't work. Take a pair of knee high boots. As soon as the customer hasn't sraight legs, 'Fitted Mesh' suddenly stops to fit. Like Rhys said, +/-5 ok, but every shape that's more different to default will not get proper fitting. (And the legs/knees are only one area of the problem.) After a lot of testing and considering the fact that it's not possible to rig the avatar-mesh in a way that it deforms like the in-world avatar-body does, I'm not sure if it makes sense to jump into this. Really pondering with the question if it's worth the effort, when only a small number of customers benefit from it - those whose shape is near the defaults. You know, I'm not a clothing designer at all. I'm a guy that cares very little about clothing, in RL and in SL. So, why the heck am I complaining about stuff that has to do with clothing? Well, because I feel for my fellow creators and know just how important a good mesh clothing solution is. For whatever reason, things have also progressed to the point that clothing creators are seeking out others with more experience in rigging, as this solution is far more complex than any creator would normally encounter. As a business person, I cringe everytime I here a customer, not even my own customer, complain about something they spend their hard earned money on. From a business stand point, I see the dollars flying out the window. From a creators standpoint, it hurts even more to know that you can't do anything about that customer not being satisfied, except give them a refund. When you add in all the extra effort, only to still have unsatisfied customers, it's heartbreaking to even deal with. From that business side of me, I know this can't be good for business, even tho I might have more satisfied a small percentage. It's still going backwards, not forward.

Perrie Juran wrote: looks official now Kick some A$$, Ebbe Linden!

I'm posting this as a comparison. The clothing in this video is not Fitted Mesh. I don't expect Fitted Mesh to work quite as good as this, but it should be, at the very least, close. The biggest part that annoys me is the extra work that LL is forcing us to do. With the Mesh Deformer, you just need good normal bone weights, like you would create on any avatar you would ever make for any platform possible. In my honest opinion, this Fitted Mesh solution further puts a divide between default clothing creators, or any1 new to 3D clothing creation, and those that master this extremely quirky solution.

Rhys Goode wrote: I've been playing around with the new collision bones for a month, trying to make a "one size fits all" skimpy bikini. Skimpy, because with the old SL bones, the bikini bottom can be locked to mPelvis and the bra to mChest, and you are done. Now with collision bones, the bikini bottom is easy, so I have a prototype for 1 size fits all panties. But no way fort the bra. I can make a mesh bra and have it grow and shrink with the sliders easily enough. But the "shape" is not particulary close to the SL avatar shape. And if you keep careful track of the tip of the bra, as you change sliders, you can see systematic changes in the shape of the SL avatar mesh that are simply not there for the fitted mesh. Even if it is the exact same SL avatar mesh you are trying to rig. The SL avatar mesh, rigged to the original CHEST bone, simply deforms differently with the sliders than the precise same mesh linked to collision bones. And while it does work well for some clothing styles, it just is not an easy solution if you want to make womens clothing that includes a bit of cleavage. I suspect that at the end of the day, I will be able to make a custom fitted top that allows the use of physics, but will not do a good job at all for an avatar with sliders more than about +/- 5 points from the design avatar for breast cleavage, breast size, body thickness and body fat sliders , and maybe +/- 2 points for breast bouancy. (Mesh rigged to the he orignal SL bones scaled perfectly with body thickness) There are really only 3 of the collision bones that are useful to rig a bra, and it is hard to fix 5 different sliders with only 3 levers to adjust. The people who are thinking this will be the end of alpha masks are going to be sorely disappointed. And the notion that you can "keep your own shape" while wearing the new stuff is simply delusional. You may be able to keep the same slider values, but the shape is going to be different, you will still have to alpha mask out your own shape, and go with whatever the clothing designer came up with for that particular garment. Yep, and that is really my point. This is not even close to finished. You can't even make a Hoody and have it form to your head. No offense to the coders or LL, but this is like half-assed. It would be something if it was simple for us and half-assed, but we got super time consuming and half-assed. Maybe I'm just harsh about these things, but this is the opposite of what I think of when I think of SL. To me, the beauty that is SL, is what it is because of the passion that each of us has for whatever we do or make. So, if you make clothing, you love clothing, and you will spend whatever time it takes to make that clothing perfect, and make your customers happy. If you make avatars, you study the character. You watch tons of videos on the subject. We have characters on the grid that come with incredible amounts of detail and features. I myself, when I sit down and want to make that EPIC AO, spend months on it giving the customer hundreds and hundreds of animations for them to use and play with. I don't see that kind of commitment on this Fitted Mesh solution. I'm not saying this is a terrible solution, just that it needs more work. I'll admit, I fully believe that the Mesh Deformer is a million times better, in every way possible, especially for the creators and users stand point. If Fitted Mesh is what LL wants to go with, I guess I have to play along, but at least give me something reasonable. I'll predict right now that people with larger, heavier type of avatars are going to not like Fitted Mesh at all, and many of them are going to waste a buttload of lindens on them, only to be sorely dissapointed. I mentioned that I was going to try making my pants fits the largest extreme, and uploading it to fit that, but with this Fitted Mesh solution, I don't think I can do that. It is gauging the deforms from the sliders, not the actual mesh. With the Mesh Deformer, if you had problems with very extreme avatars, you could just make the clothing to fit that extreme size. When you upload it, the clothing would still work on most other avatars, but it would fit best in that extreme size that you created it for. I seriously don't think you can do that with Fitted Mesh, as it just works totally differently. So, based on that observation, I will conclude that any1 with any kind of extreme avatar, including child avatars, will not have a good experience with most Fitted mesh clothing.

JackRipper666 wrote: Nice to see you get this far already Medhue. I also am not sure about this besides maybe making use of those physic's bones on a female avatar. But until I get a chance to try it I won't know what I think of that if it works lol. The way I see this skeleton as a use for me is only if I decide to make any clothing or armor available not only to my avatars I build but also to the second life default avatar after I adjust the parts to the default AVI then rig it. This would allow customers to scale the parts off the body and not along with the default avatar shape. Since everyone has their own shape they like to use you'd need a seperate control to keep the body from poking through if they have a large shape it would easily poke through the clothing, armor etc. But this is how I took this skeleton to be useful or used in SL since from the video it shows how creators had the body of the default avi poking through their clothing if they didn't build it away from the body enough. This new skeleton seems mostly just for adjusting that that problem giving you that control to scale it away from your stomach enough so that it fits. The physic's bones if they work will be nice. But I'd probably just add those to my current skeleton I use in Maya using the same naming convention. That works when I rig my avatars face with some extra deformers named after the attachment point bones. Can't see why it wouldn't for the new programmed in physic's bones as long as I use their naming convention. As for the collision bones I won't use them with my avatars I build. Since I build it custom to fit the shape of the characters body before it's even uploaded. Therefore when any of the sliders are adjusted the avatar's clothing/armor scales with it. Besides the sliders that don't work at all just yet. I know it could still be useful for my own custom built avatars but I thoroughly test the avatar usually before I upload to beta to see if it pokes at all. Then run a bunch fo different AO's to see if any animations show a part of the body that is poking through the clothing etc. But it could save me if I put in the time to add these into my current rig. Since even if the clothing isn't perfect you could just scale the clothing off the body pretty far to fix that using skin weights and should save me time in that way. So that's pretty much how I took fitted mesh as if it's made for taking care of the mesh clothing, armor etc that you build around the default avatar. Which would just resolve poking through depending on the shape of the default avatar giving you the flexibility you need to keep your clothes fitting nicely. As for it controlling the body fat etc. I didn't think it would be used for this. Fitted mesh that name just implys to me fitted mesh clothing if you know what i mean. Not so much take care of the body adjustment sliders that don't work on the first skeleton released. I could be completely wrong though but that's my only assumption as to why you're having problems with those sliders. I think you slightly misunderstand the purpose of Fitted Mesh. The clothing doesn't work independent of the avatar, they work along with the avatar's sliders. It would actually be much, much better if the clothing had it's own sliders, then the user could over compensate and have a little more control over the clothing and how it fits. Personally, I really don't understand why LL didn't add collision bones for the Pants and Shirt sliders. This would make a tons of sense because they actually give the users more control. I was thinking about redoing my Lycan Avatar to take advantage of sliders and customization that Fitted Mesh brings, but there is no way in the world I'm going to do that if this is the kind of results we are getting. It would almost be impossible for the clothing I made for him to work correctly, or to the point that is was useful. We'll see. I'm sure that if I get bored, I'll play around with it.

Monti Messmer wrote: Hello, i´m wondering a bit too this is released to public. For pants, there is a belly bone but didn´t get it to deform the pants front. The male upper legs, if they are bit muscular, do exactly what you said, no good deforming. The crotch area doesn´t do any at all, no bone, no weighting. My testing didn´t go beyond a trouser so i cannot tell about the other body parts. Did they just make a new rig wait a while then release or really worked with creators to improve, there where some comments on the JIra left. I stopped creating when mesh was introduced then the TOS disaster, maybe i just sit and wait a bit longer. Monti Thanks for confirming I'm not crazy. lol I hear you about the TOS disaster, and I'm not really creating much of my own stuff for SL right now because of it. What I am doing tho, is rigging and weighting clothing items for other creators. They took 1 look at the whole rigging fiasco and asked me to do it for them. I'll have many more clothing items to test with soon, but I'm trying to sort everything out before I work on their stuff.

To my astonishment, Fitted Mesh was released today. Now, either I'm missing something, or the avatar is missing a bunch of stuff. I've been playing around with Fitted Mesh since it's initial announcement. I would weight things until I couldn't weight things anymore, without making things worse, and then wait for the next update. The last time I waited, which was a month ago, I got no updates until now. I don't know what is exceptable to others, but I don't see where this release is exceptable at all. Like I said, I could be missing something, but I really don't think so. Let me explain my results on my pants. Pretty much most of the body sliders work, and aren't a problem. There is a hip width slider that doesn't seem to be deforming quite right. It seems like it should be a bone scaling, so I don't understand why it is not evenly stretching the mesh in the hip area. The Butt morph slider, I can only get to work about 75% of the slider. The leg muscle morph slider doesn't affect the calves right, which I have no other things to weight it to that can make it morph right. The biggest problem by far, is the body fat morph slider, which morphs fine width wise, but doesn't morph the front of the pants at all. Please, some1 tell me I'm missing something here, or doing something wrong. There are no other collision bones to match the head morph sliders. How exactly are we going to make mesh hair that deforms with your head? I didn't make hair, but I did make a hat. It doesn't deform to the sliders at all, other than head size, which is a scaling bone. What about the male avatar? I haven't tried anything with a male yet, but the male has another morph slider for the crotch. Where is the collision bone to morph with that? I'm hoping that I just need a completely different weighting combination, but I fear this is just another oversight. Like I said earlier, I have no idea what is exceptable to every1 else, but this is not exceptable to me. Even with the body sliders that do work, I feel like I'm still missing a collision bone that would allow me to make it all perfect. The Body Fat morph slider, I have no idea how I'm going to get that to work, other than make a mesh that starts in that shape with the slider all the way up. Yeah, as is, this solution might work for 70% of the avatars, but IMHO that's not enough, when it could easily be 100% including head gear. I really want to hear from other creators. What are you seeing? How good can you get it? Where are your problem areas?

Dresden Ceriano wrote: Medhue Simoni wrote: Innula Zenovka wrote: I'm not at all sure LL's CEO needs to have an understanding of the depth and complexity of SL, so much as an understanding that SL is a whole panoply of different virtual worlds, created by users according to their interests. I want a CEO who doesn't try to impose a particular vision on SL, as did M when he tried to hide Adult SL because he thought it was scaring off his target markets, business and education. Rather, I want someone who can get results with things like grid stability and rezzing times, as did Rodvik, and oversee general improvements to things like group chat and search. At a strategic level, I very much hope he can grow new income streams to move LL away from its overdependence on SL tier for its income. Personally, I want a CEO with the whole package. I really don't understand how some1 that only knows coding, or only knows marketing, or whatever, gets a CEO jobs without starting the company themself. IMHO, I don't think it is much to ask that they understand all, or the majority of the technical areas that is SL, or any game for that matter. I'm not saying any1 should no every group, or combat system, just all the technical areas of SL. When I bid for say a 3d animation video production job, I'm expected to know every single part of the process and produce it at a professional level. Now, I don't know how much every1 knows about creating 3d video animation, but it is no easy task, and when you throw in special affects, it get downright crazy what you need to know. Ok, yeah, most of these jobs are done by a team, but it is amazing how many people understand it all enough to produce quality. My point is, that is the level of understanding that I expect. It worries me that we now have a CEO that knows almost nothing about game engines, and even less about 3d creation. He will have to rely on the managers. The problem with this is, that we constantly hear that this or that is too complicated, or too time consuming, and then some1 like Qarl wips that very code out in a couple weeks. We don't even have more than a couple coders at LL that know anything about 3d creation, hence why we don't even have custom bones yet. I think if it is going to work out with Ebbe, he's going to have to stay engaged with us. I would rather have a CEO who knows absolutely nothing about game engines or 3d creation, but who actually knows how to interact with the SL user base and provide the sort of customer service which has been sorely lacking for ages. LL can make all the improvements in code and usability they want, it's the customer service that truly counts and, until they get that right, no other improvements will ever really matter. ...Dres My ulitimate point is, that CEOs get paid a crapload of money, at least in successful ventures. It's 1 thing for a person to build a company and be the CEO. It's another for a company to hire 1. When I bid for a job, or get interviewed, they want to understand what I know, and how I can help them. The more I know the more I'll get paid. It seems, that with CEOs, it doesn't work this way at all. I'd think, that if you are going to pay some1 a boatload of money, they should know a boatload. I don't really give marketeers all that much worth, because if a CEO doesn't everything else right, the marketing will happen all on it's own, especially for a world like SL. Heck, we've had over 10k signups every day since SL got the tiniest of promotions back in 2008. We don't need a marketeer, we need an engineer. IMHO

Perrie Juran wrote: When Oz announced that they were now 'officially' working on Mesh & Bones, etc, he made a comment to the effect of, "Don't get your hopes up too high, this is not as simple as it looks," I really had to wonder. If InWorlds has it working, what's so hard about it for SL? I do agree with Dres that customer service is the number one thing that needs improved. We could point to specific issues but they have been rehashed here so often it has gotten to be a joke. WTF! When did Oz say that? Don't get our hopes up? Why start if they aren't gonna finish? It might not be easy, but not impossible. He said something similar, or was it Nix, about the mesh deformer, which essentially took a month for 1 guy to make. InWorldz has it working? I think you mean CP. Perrie, you totally freaked me out with this posting. When did all this happen? We fricken need custom bones. It's ridiculous that we don't have them. It's 2014, and we have pathfinding, but we don't have good moving NPCs. Come on now.

SL only allows for bvh animation, and another file format called anim, which can only be created in Avastar, in Blender. So, basically, you need to make the animation a bvh.

I'll try to help. What Mistah is talking about is how dense the mesh is. Every mesh is made up of vertices. The more vertices you use, the more data the mesh is. In Marvelous Designer, the models come out with very high vertice counts, meaning they are a ton of data. To get wrinkle and folds to look right, you kind of need alot of vertices, but it can be a bit much for SL to handle, or rather, for your pc to handle. In Blender, there are tools, like the decimate tool, that will help you lower your vertice count. Or you could retopologize the mesh, which means redoing the mesh with fewer vertices and using the shape of the mesh as your guide. This is a bit more complex to do, but this is how most game characters are made. They first make a high polygon model, then retopologize for the game. A polygon, in simple terms is a set of 3,4 vertices, and can also be referred to as faces, not to be confused with SL faces. Blender is by far the best program to use to make SL clothing. You can use Marvelous Designer, but it won't get you all the way done. It will only allow you to make a mesh. You will have to use some other program to rig the clothing. I'd also suggest buying Avastar, which is an Addon for Blender. Avastar was specifically made to rig SL clothing, and to animate SL characters. Avastar makes the process of rigging much more simple, especially if you want to make Fitted Mesh. I don't actually make clothing, but I am an animator, which means I likely know a bit about rigging and weighting. Rigging basically means attaching your mesh to a skeleton, and weighting is the process of setting weights to vertices for a bone, meaning each vertice on the mesh is weighting to a bone, like a leg bone or an arm bone. Currently, I'm working with a clothing creator that uses Marvelous Designer. She gives me the mesh, and I bring the mesh into Blender and finish up the mesh clothing to work in SL.

Innula Zenovka wrote: I'm not at all sure LL's CEO needs to have an understanding of the depth and complexity of SL, so much as an understanding that SL is a whole panoply of different virtual worlds, created by users according to their interests. I want a CEO who doesn't try to impose a particular vision on SL, as did M when he tried to hide Adult SL because he thought it was scaring off his target markets, business and education. Rather, I want someone who can get results with things like grid stability and rezzing times, as did Rodvik, and oversee general improvements to things like group chat and search. At a strategic level, I very much hope he can grow new income streams to move LL away from its overdependence on SL tier for its income. Personally, I want a CEO with the whole package. I really don't understand how some1 that only knows coding, or only knows marketing, or whatever, gets a CEO jobs without starting the company themself. IMHO, I don't think it is much to ask that they understand all, or the majority of the technical areas that is SL, or any game for that matter. I'm not saying any1 should no every group, or combat system, just all the technical areas of SL. When I bid for say a 3d animation video production job, I'm expected to know every single part of the process and produce it at a professional level. Now, I don't know how much every1 knows about creating 3d video animation, but it is no easy task, and when you throw in special affects, it get downright crazy what you need to know. Ok, yeah, most of these jobs are done by a team, but it is amazing how many people understand it all enough to produce quality. My point is, that is the level of understanding that I expect. It worries me that we now have a CEO that knows almost nothing about game engines, and even less about 3d creation. He will have to rely on the managers. The problem with this is, that we constantly hear that this or that is too complicated, or too time consuming, and then some1 like Qarl wips that very code out in a couple weeks. We don't even have more than a couple coders at LL that know anything about 3d creation, hence why we don't even have custom bones yet. I think if it is going to work out with Ebbe, he's going to have to stay engaged with us.

I'm always thinking business, and hate to waste time. I welcomed him, but then went directly to mentioning the TOS debacle. :matte-motes-sunglasses-3:

Wow! His twitter page is hit up with hundreds of tweets from us all. I bet he didn't expect that.

Most SL users, even the old timers, don't know the history of the inworld search engine. The 2 big things that drew me to SL was the creative aspect and the build your own business aspect. Very early on, I saw just how important the inworld search engine was. Because of this, I read all that I could about it. Originally, or soon after SL started, a single SL resident with awesome coding skills took up the task of creating a search engine specifically made for SL. The amount of land played a big role in it's ranking. This encouraged every merchant to own land, and to own as much as they could afford. Yes, the engine could be gamed, but it took real effort to game the engine. This is exactly what you want. Making a search engine that can't be gamed is almost impossible, and brings more negative consequences than just allowing the gaming. When it comes to discouraging gaming of an engine, the goal should always be to make it alot of work for the person gaming. They will quickly see that their time is being wasted and just stop. If you could go back and take a snapshot of the search engine in 2009, you'd have seen an almost perfect search engine. No matter what you searched for, the location would be ranked according to their success in SL, for the most part. The more words you searched with, the more refined the results were. If you searched for "hammer animation" you were pretty much guaranteed to find a hammer animation in minutes. You could even teleport directly to the hammer animation on the parcel. No searching around at all. Then, LL started listening to the complainers saying that the search engine was bias toward the larger merchants. They remove Picks from the results algorythm. It wasn't 6 months later that LL implemented the GSA engine. With this, their goal was to make the search engine fair. They didn't understand that fair meant unusable for anything. They didn't understand how the competitive engine was better for users, merchants, and business. Their goal was to socialize the search engine, so that every1 had a chance. What they failed to understand is that every1 already had a chance with the old engine, they just couldn't be some newb merchant and dominate from the start. They had to work their way up.

Personally, I've been finding that the Avastar addon works out for me in many different ways. As an animator, I think it is obvious why I would like Avastar. That said, I don't think many understand all the different benefits. Almost, no matter what I make today, beit animation or mesh, I use the avatar in Avastar in some way or another. When you add in the ability to alter the avatar, just like you do in SL, now you have a system to see how your products work on large or smaller avatars. 1 really good example is a mesh wheelchair that I made. Originally, I made it in 3ds Max, as this was before I seriously started using Blender. I only brought the wheelchair into Blender when I actually made the animations for it. In Blender, with the Avastar figure, It was glaringly obvious that I had to make some changes to the wheelchair to make it fit all avatars better. Plus, once I made the changes, it was easy for me to create 2 different sets of animations to fit a wider range of users. I still need to make new animations to fit my bigwheel, and I know Avastar is going to make it super easy to adjust my bigwheel to the right size avatars. Heck, I might even make adult bigwheels. Sorry, I just had to plug Avastar again, as I know Gaia is too modest.

I don't think Rod had any intension at all to stay longer than a few years. That's how these things work. It is rare for any employee to stay with any company for any amount of time, at least today. Really, I think it comes with the large paychecks these people are getting. If the compensation was directly linked to actual results, then CEOs would stick around longer. Today, somehow, the CEOs are shielded from any kind of responsibility. To me tho, Rod's reign serves as comfirmation to me that all the big names in the industry are just guys that got lucky. It is their fanboys in the gaming media that keep propping them up as geniuses. The same goes for Will Wright. He is gone also. See, SL is a challenge, and neither of the the 2 of them showed that they could even understand the basics. Yes, there are some subjective areas that can't be pinpointed exactly, but there are also very scientifically straightforward things that both of these "legends" missed or screwed up completely. Seriously, any average successful business person could have figured this stuff out, but again the "legends" missed them completely. Just look at the mesh clothing issue. LL probably lost millions of dollars on that, and LL totally ignored it for years. It was not a difficult issue to begin with. As we saw, it was a couple months of work, no matter which way you went. Today, years after the initial release of mesh, we still sit here with no finished solution. It's f....ing ridiculous, especailly when you analyze what brings average people into SL, and what keeps them here. This fiasco shows a serious lack of understanding of not just the customers, but also the technology. The clothing issue is not even the biggest glaringly obvious missed step from Rod and his team. Any1 that has listened to me rant about the failures of our leaders should know where I'm going. How does any person walk into a virtual world and not notice, right away, how useless the inworld search engine is? How does any1 not understand that if your inworld search engine doesn't give good results, your virtual world will suffer greatly. It's OBVIOUS, yet it's been useless for years now. Today, I literally only open it to search for a person that messaged me, so that I can reply to them in IM. I stopped even checking where my store ranked in the search results years ago. This is directly linked to the loss of sims, which is why I keep screaming about it. For virtual worlds, the inworld search engine plays a big role in the economy. It replaces many things that are normal benchmarks in a real life economy. When sim owners aren't rewarded with higher rankings, which equates to more visitors and more income, because of their efforts, than why own a sim at all. I have always contended that the loss of sims is directly linked to the inworld search engine, and it is proven in the statistics. The date of the initial decline in sim ownership corresponds with the first major change to the inworld search engine. These are not the only major missed steps, but the 2 that stand out to me the most. Overall, I'm dumbfounded how a company of over 200 employees can get so little done. I watch other developers on many other platforms and programs work circles around what the Lab gets done. Maybe it's a matter of spending too much time debating, and not enough time actually doing. I also think you have to hire people that love your product. FAR toooooo many Lindens to don't use SL at all. IMHO, this is exactly why crap doesn't get fixed, because they don't care. It's a job, and they get paid well. They don't attract people that are passionate about the platform, they attract people that want a big paycheck and a stepping stone. If I'm CEO, I'd mandate work time to f..ck around in SL. If they still weren't passionate about it, I'd boot them out the door. I remember when 1 Linden talked about how they shot each other with nerf guns. I immediately thought, why the f.ck aren't they shooting each other in SL? Heck, If I was CEO, I make the whole company battle it out on a sim every single week, have teams, and keep score. I'd make them play with the things their customers make in SL.

Ultimately, I have no idea what is going on. I'd actually have to have the bvh to know for sure. I can speculate tho. I'm guessing that the animation optimization is screwing your animation up. The reason I'm guessing this is because you mentioned subtle movements. Basically, the importer looks for changes in movements from frame to frame. If those movements are too small, the importer won't pick up the movement. 1 simple way to fix this is to lower your fps. By lowering your fps, each frames movement will be a bit greater. I could be totally off the mark on this, but I really have no other advice without more info.