Jump to content
Linden Lab

Project Bento Feedback Thread

Recommended Posts


Cathy Foil wrote:


 
OH I have no doubt that the best animations are animations that both translate the bone and rotate the bones.  Animations that just use translations are probably not quite as good but way better than animations that are just made with rotations only.

 

What I am saying is rotation only animations for the face were surprisingly good to me.  The more subtle the expression the less noticeable the limitations of animating only with rotations are.  

 

I'm still testing, but I think it is possible to use some bones only with rotation, keeping all slider changes, and then use some bones with both rotation and translation, to be more expressive, all in the same animation. Sorry, if I might have confused some people. It's easier to show than to tell. My video is taking a bit longer than I thought, as I work out how it all could work.

 


Cathy Foil wrote:


 

Personally I think I am going to prefer an AO where my avatar repeats randomly a set of fixed subtle facial expression that play over and over again over an AO which makes my avatar face express intense or exaggerated facial expressions over and over again.

I have had the same sort of experience chatting with people whose AO makes their avatar walk around and rock back and fourth.  It makes their avtar's body language give me the impression that they are annoyed or impatient.  While I know it is just an AO and not reflective of the true mental state of the person I am chatting with it still sends mixed signals to me and I end up leaving the conversation and not wishing to chat with the person next time I see them.

I can definitely see this happening with AOs which while the animations may be wonderful with great very expressive facial animations showing a lot of emotions using animations with translations and rotations there will be this same disconnect between what the person I am chatting with is conveying in their typing or voice and what their avatar's face is expressing.

 

I think more subtle facial expressions will have way less of a disconnect but at the same time still giving the impression of the avatar being more alive than what we have had up until now.

 

Well, I don't think anyone would really make the idle face animations that expressive. Whether using rotation or translation, I create a set of around 5 very subtle animations for idle faces, which play randomly. 1 of those usually includes a blink. Generally speaking, these subtle animations can look fairly good only using rotation. Expressions tho, are triggered by the users, when they want them to trigger, using a number of different ways. The most obvious is using a hud. Less obvious is having the expressions trigger when a specific word is said in local chat. So, if I say happy, then a smile animation will play.

 


Cathy Foil wrote:

 

What would be awesome is to have a library of facial expressions animations and your web cam aimed at your face.  A program reads your facial expression and then plays in SL the animation associated with that expression.  This would be much easier than a Face Rig type system.

This way there wouldn't be a disconnect between what you are saying or typing and what your avatar's face showing.

That would be awesome, but that would require the rig to be set up completely for bone translation, not rotation.

Share this post


Link to post
Share on other sites

Cathy Foil wrote:

 

What would be awesome is to have a library of facial expressions animations and your web cam aimed at your face.  A program reads your facial expression and then plays in SL the animation associated with that expression.  This would be much easier than a Face Rig type system.

This way there wouldn't be a disconnect between what you are saying or typing and what your avatar's face showing.

Medhue Simoni wrote:
 

That would be awesome, but that would require the rig to be set up completely for bone translation, not rotation.


I probably am not explaining my self well enough.  The program that is setup to use your web cam watches your face for expression and when it recognizes an expression say like when you smile it recognizes that you are smiling and sends a signal to the viewer you are logged in with to play a pre-recorded smile animation.  So it wouldn't matter how big your smile was in real life it simply play the animation named "Smile".

 

I mean you could have several variations animations for smiling.  Mouth closed, toothy smile with mouth closed, open mouth smile.  The program could recognize the different types of smiles the camera picks up.  You could do the same thing for blinking or winking or raised eyebrows.

Depending on how good the program was at detecting small facial expressions and how many pre-made animations are made and uploaded to SL you could probably mimic a real time face rig pretty good.

If someone was really motivated an individual could create their own animations to more closely match their own facial expressions and replace the standard animations kinda like can do now for lip sync.

Share this post


Link to post
Share on other sites

Cathy Foil wrote:

Those making and selling animations will want to clearly label their animations as "Rotations Only", "Rotations and Translations" or "Translations Only"...

I think that is a good suggestion. But I think also mesh head creators will have to classify their mesh by what sort of expressions the mesh will accept. Because the customer buying the mesh head will rarely understand why the particular expression will not work with the mesh, but they can learn that some expressions will go and some not. They must not even understand what a rotation and what a translation is, but when they know this mesh fails on expressions using translation than they know which expressions the customer can look at and which are not worth of spending time for looking at the demos for them.

Also, when the mesh head or fullbody uses the expression override (be it the suggestion of polysail or LL implements the feature request for it) this must be explained too, since that is a big plus in having predefined best looking expressions (since made by the mesh creator themselves) and probably simple to use new expressions made by third party.

However, there will be a load of new information falling on the head of the to the mesh wearer and I think they have to be prepared for it. So I think there must be an explaining post somewhere targeting the customers and explaining that there is a slides vs. expressions problem, that some expressions can not work on some meshes, what an expression override is (if there will be a ready to use solution) and probably a simple to understand video about what translations and rotations are, how they affect the mesh and why some mesh can not accept translations in expressions. A bold request, sorry, but I myself am not good enough to make an easy to understand explaination and having deep understanding of the topics technically.


Cathy Foil wrote:

I probably am not explaining my self well enough.  The program that is setup to use your web cam watches your face for expression and when it recognizes an expression say like when you smile it recognizes that you are smiling and sends a signal to the viewer you are logged in with to play a pre-recorded smile animation.  So it wouldn't matter how big your smile was in real life it simply play the animation named "Smile".

This can be done by expression overrider (assumed it is available.) The web cam program will just start recognized expressions and they will run overriding animations. Needs a service that sends signals from the clinent computer to the LL server running the agent. Can be done by LL or probably the web service which the expression hud worn by the avatar offers and client web cam program is registered into.

Share this post


Link to post
Share on other sites


Cathy Foil wrote:

I probably am not explaining my self well enough.  The program that is setup to use your web cam watches your face for expression and when it recognizes an expression say like when you smile it recognizes that you are smiling and sends a signal to the viewer you are logged in with to play a pre-recorded smile animation.  So it wouldn't matter how big your smile was in real life it simply play the animation named "Smile".

 

I mean you could have several variations animations for smiling.  Mouth closed, toothy smile with mouth closed, open mouth smile.  The program could recognize the different types of smiles the camera picks up.  You could do the same thing for blinking or winking or raised eyebrows.

Depending on how good the program was at detecting small facial expressions and how many pre-made animations are made and uploaded to SL you could probably mimic a real time face rig pretty good.

If someone was really motivated an individual could create their own animations to more closely match their own facial expressions and replace the standard animations kinda like can do now for lip sync.

That's actually not a bad way of doing it for a world like SL. Instead of constrantly streaming data to have the proper expressions on your face, the program is tracking your facial and just sending data to SL when it recognizes an expression. So, in SL, you'd have your idle face animations playing, until the system sees a recognizable express.

Share this post


Link to post
Share on other sites


Medhue Simoni wrote:


wherorangi wrote:


Medhue Simoni wrote:


wherorangi wrote:


1) lovely (or manly in that case) smile, 2) lower my lashes/close my eyes

 

 

IMHO, lovely smiles are not possible with only using rotations. You can make a smile, no doubt, but it will never be a lovely smile. lol

given the choice between

 

Who said there needs to be a choice? Who said you can't have both?

you did. You said that I cant have a lovely smile with only rotations

and I said that if the choice is between a)  rotations + shape tweaking, or b) translations - shape tweaking, then a) is the winner

if I can have c) translations + rotations + shape editing  then I would like that please

and as has been discussed already by lots of people, then we could have c) if there was a set of expression bones, seperate from the shaping bones   

Share this post


Link to post
Share on other sites

THERE ARE TECHNICAL LIMITATIONS TO SECOND LIFE

 

This is something a lot of you ~ and I mean a LOT of you are failing to grasp here.

Due the way that the animation system is currently handled a slider that affects a BONE POSITION will be overwritten by an animation that affects a BONE POSITION.

There is no avoiding this short of re-working the entire back-end of how bone position based ( translation ) animations are played.  Reworking the animation system may come at a later date~ but doing so is far beyond the scope of what project bento is.

So no, you CANNOT have your cake and eat it too.  At this point, it's FUNDAMENTALLY IMPOSSIBLE, stop asking for it PLEASE.

 

Every complaint that has been posted on this board so far has been "the default positions are not ideal for *MY PROJECT*.  That does not mean they are not ideal for someone else's project.  Or that they're ideal for anyone's project.  But the fact of the matter is DUE TO THE TECHNICAL CONSTRAINTS OF SECOND LIFE they are the most versatile.

I do support ( and am even willing to help with ) fixing as many of the sliders to function off of bone SCALE as possible (I've even been trying to come up with a slight re-parenting setup that will use the present "lowerTeeth" bone as a "lip root" bone) , as that will make them all as diverse and responsive as possible.  But measures beyond that, should be, at this point, completely out of the question.

  • Like 2

Share this post


Link to post
Share on other sites


polysail wrote:

THERE ARE TECHNICAL LIMITATIONS TO SECOND LIFE


polysail wrote:

THERE ARE TECHNICAL LIMITATIONS TO SECOND LIFE

This is something a lot of you ~ and I mean a LOT of you are failing to grasp here.

when faced with choices, choose a) or b) bc you cant have c) bc current technical limits and limted resources/staff to apply to the issue, then I choose a)

not sure how anyone is failing to understand this

Share this post


Link to post
Share on other sites

 


wherorangi wrote:

when faced with choices, choose a) or b) bc you cant have c) bc current technical limits and limted resources/staff to apply to the issue, then I choose a)

not sure how anyone is failing to understand this

 

What I think you're failing to understand here is that not everyone uses human avatars, or even the same style for human avatars, and as such need different things to be able to be able to make their own products work as intended. We've already had this fight over the course of development about why exactly we need to have bone translations and how it makes animating literally everything so much easier and better looking, especially for human avatars. I'd really rather not go back 5 steps just to go forward 1. The lack of sliders has never stopped creators for making unique and diverse content, and having limited sliders could only help people diversify their content more.

There's a whole lot more people than just the market you're a part of, and it'd be unfair to shoot everyone else in the foot just because you want something a little more unique from what other people have in your little section of the grid.

Share this post


Link to post
Share on other sites


Magnanimousity wrote:

  

... sliders ... 

There's a whole lot more people than just the market you're a part of, and it'd be unfair to shoot everyone else in the foot just because you want something a little more unique from what other people have in your little section of the grid.

well you need to discuss this with Vir Linden, not with me

UI slider shaping/sizing was introduced by Vir just recently as something that LL would like to see, after all of the pre-discussion and early design planning

is the only reason I got into this chat since

so now after all that pre and early stuff, there is now a conflict apparently between translations and UI shaping

if you want translations over shaping then ask Vir to take UI slider shaping off the table, if the conflicts caused by only having 1 set of bones, are such to create major issues

+

while it is on the table then I am going to keep pointing out the obvious

the obvious is that its quite a bit bigger than a little section of the grid

if do a count of avatar types then human-looking avatars totally outnumber every other avatar types combined. Human avatars are not the 5%. And the thing that the overwhelming majority of human avatars wants is UI slider shaping before anything else ever

is the first thing that everybody does when they sign in for the first time, after walking round for a bit. Right-click and Edit Appearance. This hasnt changed. Go to Social Island and watch, see what people do. Same now, same in 2003, same since forever

some will go for robots and vehicles and animals for sure. Most tho, like at least 95%, will stay in human form and start editing their shapes

No. 1 request from everybody who wears a starter mesh avatar. How do I edit my shape and face ? I tried but is not working and I dunno why ?!

+

if UI slider shaping does stay on the table, then yes it does need a whole lot more design work done if the plan is to only stick with one set of bones

if everyone was to get everything they wanted then, and to do it properly so is a solid foundation to go forward on, then is c). Two sets of bones. One set for shaping and standard rotation animation, and another set for translations

pick either when making your avatar models, and/or combine both as it suits

and if it is to be c) eventually, resources/staff scheduling permitting, (and is no reason why it cant be eventually) then which is the priority to do first ?

 

  • Like 1

Share this post


Link to post
Share on other sites

Would it be possible to connect the eyebrow bones to the eyebrow shape sliders in the hair base bodypart? As it is, there's no way to edit the shape of the eyebrows, but the bones and sliders are there.

The behavior of the head stretch slider could use a bit of fine tuning as well (if time permits, of course) the slider currently only stretches the general area around the face, the face shape. It does not affect the nose at all and there is no other slider that can lengthen/shorten the nose vertically like this:



 

This results in a strange look when somebody uses the slider to make their face longer. The face lengthens but the nose stays disproportionately vertically short.

Would it be possible to make the head stretch slider affect bento compatible heads the same way the fitted mesh bones affect mesh heads rigged to them? When utilized on a fitted mesh head, the head stretch slider squeezes/stretches the whole head, not just the area surrounding the face, which in turn preserves the proportions (long face, vertically long nose, short face, vertically short nose).

 

Share this post


Link to post
Share on other sites

this is really good Medhue. I like it (:

+

is it possible for you to show us what happens when use the sliders on the default LocRot head ?

i accept that is only a test and that the Loc ones may go a bit funny. Is really just to get idea of what the slider range might be, without us (as users) messing it all up with the sliders  

Share this post


Link to post
Share on other sites


wherorangi wrote:

this is really good Medhue. I like it (:

+

is it possible for you to show us what happens when use the sliders on the default LocRot head ?

 

I'll do another video today showing the sliders working.

Share this post


Link to post
Share on other sites

Awesome video Medhue!

Matrice, Vir and I have been working hard with some great changes to the Bento Skeleton and Slider code along with some awesome suggestions from Mel and you and everyone from the last Bento meeting.

We moved some of the bones, mainly lip bones so that rotation animations will work even better. I am sure there will still be a need for translation animations to get the absolute best facial expressions and I love your solution of combinations of translations and rotations.

The Sliders we been working hard to get to work as much as possible with scaling so that custom bone positions and animations that use translations will work as good as possible. I think Matrice has out did himself and probably put 2 or 3 times as many hours into these new improvements as I have. Keep the awesome videos coming! :D

LL is really pushing us to make these the last changes to the skeleton and sliders.  I can't say I blame them.  I am ready for Bento to be finalized and completely adopted by SL. :)

I think with these last improvements Bento is really going to make a huge difference in SL and can honestly say Vir, Matrice and myself have been following the feedback thread here and doing our best to implement or fix issues brought up here.  And I am sure Oz and a few other Lindens have been following the feedback here as well.

Keep up the awesome work Medhue! :D

  • Like 2

Share this post


Link to post
Share on other sites

We have an updated test build for the proposed skeleton and slider modifications.

Note that because there is another bone added here, you will need to use the BentoExperimental1 region on Aditi to upload models or animations that use the new bone.

Please take a look and let us know as soon as possible about any major issues; we will be locking down further changes and, if approved, moving this update into the main bento project viewer soon. At that point we will be very unlikely to make any further skeleton or slider changes.

Share this post


Link to post
Share on other sites

Ooh ~ a jaw shaper bone!!  You guys beat me to it!  I'll take this and play with it a little, but I doubt I'll get much in before the meeting today.

 

Edit: Crossed Lip Bones.  I'm not sure how I feel about crossed Lip Bones.  That seems HIGHLY counterintuitive.  While it's true that a good animation rig nullifies the need for a logical bone structure.  I feel that this may cause more confusion than is necessary:  must play with these some more.


Edit #2: Still don't like the crossing over the centerline ~ it makes for some very bizarre rotations when rotating the corners of the mouth downwards ( for example ~ while yawning.)  I need an ( actual RL ) nap before the meeting ~ but I still am not comfortable with these crossed lip corner bones.

 

Edit #3: I still don't like this pivot placement for the mFaceLipCornerLeft/Right bones, I feel like it causes more problems than it solves.  This becomes apparent when doing various animations that have the mouth kept open~ the bizarre root of the bone starts making things behave slightly weird..  ~ I really need a nap.  Maybe I'm just sleepy and doing it wrong.

Share this post


Link to post
Share on other sites

The proposed skeleton/slider updates include scaling of the original eye joints when the eye size slider is adjusted; this is intended to allow mesh eyes to be scaled consistently with system eyes. There seems to be one possible scenario where this could cause problems for existing content. The eye joints have attachment points ("Left Eyeball" and "Right Eyeball") as children. If a mesh defines a joint position for one of these attachment points, then changing the scale of the parent eye joints will cause those attachment points, and any attachments at those points, to move. So the question is, is this a real scenario? Normally you can't upload meshes that refer to these attachment points anyway, because they have spaces in the name. I know there has been a Firestorm proposal to handle this by mapping underscores into spaces, but as far as I know this has never worked in the SL viewer. So is there a way that adding these eye scales could change the appearance of an existing avatar that previously displayed correctly in the SL viewer?

Share this post


Link to post
Share on other sites

Since it would affect an attachment point. Presuming that attachment point has not been RIGGED to ~ wouldn't this be a non-issue, since the user could simply re-adjust  the attached object to a proper location, as even no-mod items can be moved~~ or am I missing something here?

  • Like 1

Share this post


Link to post
Share on other sites

@Medhue

Beautiful demonstration, Medhue thank you, this is a great analysis, I think it's a wonderful idea to combine rolations and translations in the same proportions (50/50%) only for cheeks and lips. .. but just a comment on your video (which is very well done), I seem to see that the mouth of all your shapes are the same size in thickness and width ... and in this case it works fine ( rigg if done correctly). But keep in mind that the sliders will be greatly modified by customers and if the mouth or cheeks sizes are too change, the translations will be an eyesore in emotes / entertainment ... actually I think the best solution would be to reduce the propotion of use in animation (Translation 30% and 70% rotation) ... and above all set limits of uses sliders (in customers recommandations) on the bones where translations use the + or - maximum 20 pts . It will be difficult to find standard for the use of sliders and creating animations (rotation / translation) simply becaufe type bones weight rigg on mesh also greatly influence the rendering of the overall work.

I also look to a great video with rot / scal bones default with many shapes & animations.

 

@Fallen

I completely agree with you, it would be nice to connect the brows of their bones slider (in the brow shaper of bald base), as does the translations probably will not use on these bones this does not limit the ability to make operational the independent slider in the shaper brows. it is an option that does not breed from type incompatibility with other slider or animations rotations.

 

@Cathy: I am delighted to learn that you have to work even on the placement of bone for that job rotations better ... you have done so far a great job, so thank you very much.

Also thank you to Gael, Matrice, Vir and everyone who works to improve the project.

We may be able to do great things: D

Share this post


Link to post
Share on other sites

Vir, I understand your concern! But I don't think this should be a game changer.

 

There weren't a whole lot of creators who were rigging to attachment points, and even fewer who were rigging faces or head attachments. Most creators were using attachment point rigging for tails or similar attachments which were all parented from a single bone, like mPelvis.

 

I'm pretty sure BUG-10543 has a comprehensive list of avatars using non-pelvis attachment points, or pretty close to it. I've been able to confirm for almost every avatar on that list that the eye attachment bones are not rigged. The two exceptions are the Red Remora fox by our own Tornleaf, who I'm sure wouldn't mind making a new shape or new eyelid animations, and the Altimura avatars, for which I couldn't find any testers.

 

So, there shouldn't be too much content affected by this change. But moreover... We always knew that rigging to attachment points wasn't supported, and that it might someday break. So I feel like it would be a mistake to jump through hoops to avoid causing breakage.

  • Like 1

Share this post


Link to post
Share on other sites

Any attachment I have attached to the eyeball attachment points has been set at the dead center of the attachment point with no offsets, because that attachment point is always rotating around with the eyes. Any object that isn't would rotate strangely and unevenly in the eye socket. 

Share this post


Link to post
Share on other sites

Oh, yes, I hated using a-point rigging for anything, but thankfully we've got a huge selection now! So I'll be changing up the rig as soon as I can. So long as existing content doesn't break.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...