PDA

View Full Version : Character Animation Section



spec24
12-27-2004, 09:18 AM
It would be nice to see a section in this forum under "Techniques and Discussions" specifically for character animation.

pooby
12-27-2004, 12:32 PM
It Would be good, but I fear it wouldn't be used much.
I suggested the same thing but for rigging.

I'd love to see more activity in the animation area, but examples of animation work are very few and far between.

I'm inviting controversy here by suggesting ,relatively speaking, that Lightwave isn't really used by many people or studios for character animation.
Our studio does use lightwave for 'C' animation , but I know of 2 other character animation studios in England apart from us that use Lightwave.
One is phasing in XSI for animation and the other Maya.

This shouldn't be happening. Lightwave needs to get really serious in its animation abilities if it is to be more than a hobbyist bit of fun software.
It is possible to do decent stuff, but lightwave doesn't make it easy for you.

spec24
12-27-2004, 01:23 PM
hmm. I wouldn't know any better :) I'm using Lightwave to produce an animated science based childrens show that stars a couple of characters :) I learned how to use the character anim. tools in LW by reading Albee's book. Without it I'd still be struggling with those **** bones :):) I haven't even touched the IK boost tools yet? Anyone else?? How is that feature for character animating? I remember seeing the demo for it for 8 - looked way cool. Still..... haven't used it yet :( Hoping to get to it when I have the time.

What makes the other packages better for char anim???

SplineGod
12-28-2004, 02:57 AM
IkBooster is very powerful and adds many things that have been missing in LW.
I have some videos at KURV that will definately get you goiing.
http://www.kurvstudios.com/forum/forumdisplay.php?f=24

spec24
12-28-2004, 06:54 AM
ok Larry - you talked me into it. I'm purchasing right now :)

Wonderpup
12-28-2004, 12:01 PM
I believe Larry when he says IKbooster is a powerful feature, and have no probem at all with him selling his hard won knowlege of this tool to other users.
( As a long time lurker I know just how much he contributes to this community)

What does genuinely baffle me however is why there are no 'official' Newtek tutorials ( Apart from the most basic) on how to actualy use this tool. Like many other people, I played with this tool when it first arrived, got extremely frustrated and abandoned it as either being broken or worse some kind of marketing con.

Clearly this is not the case. The tool works and does indeed enhance Lightwaves character animation abilities- so why the big secret? Why not simply tell us how to and when to apply this tool? I appreciate resources are not infinate and everything takes time- but with character animation being such an important aspect of the 3D market at present, surely any advance in this area ought to be maximised- not left to fall by the wayside for want of a clear set of instructions?

SplineGod
12-28-2004, 12:24 PM
Thanks Spec! I dont think youll be disappointed. :)

Wonderpup,
I understand what you mean. I had a very difficult time wrapping my brain around IKBooster and I had to bang my head on it for awhile too. I know a lot of professional users here who also feel the same way you do. The problem is that youre always going to have end users who know how to use the software far more then the company who produces it. Newtek would basically need to have ppl on staff that use Lightwave at the same level as people working day in and day out in the business. The video demos that Newtek has posted show the basics but it realy does take ppl hammering on it daily to get it really sussed out.

Once I figured out how it works the first thing I decided to do was create a 3 hour video showing the basics of how it works. Ive had several people who are quite experienced with rigging (Jennifer Hachigian, Ken Wilder at R&H for example) who both HATED IKBooster until they went thru my video. Jens comment after watching the video was "Inos a genius". I ran into Ken Wilder at the 64 bit LW demo. He has borrowed my IKBooster CD froma coworker and as he put it "it opened my eyes". He loves IKBooster and he also told me that he takes the time to show coworkers who hated it why they shouldnt. :)

This is similar to Motion Designer in many ways. Nobody I knew at Newtek knew enough to really show what could be done with it. Whats nice about living in LA sometimes is that you get to associate with people like Jen Hachigian who knows motion designer inside and out. So again, its an end user one has to turn to to really find out how things work. Im just glad that Lightwave is around, being developed and that theres a tight knit user base that is willing to provide information. :)

pooby
12-28-2004, 02:18 PM
Well, I'm afraid I still think it's an awkward and poorly designed 'plugin' that doesn't integrate well at all with the rest of lightwave and I'm quite aware of what it can do. I own a Character animation studio that uses Lightwave as our main tool.

It represents to me a huge waste of time and energy that could have gone into designing a solidly designed series of tools that work together. Rather than the current bits and bobs that all have different rules and dont communicate properly.

I only hope to god that Newtek don't think they've 'fixed' the animation deficit in Lightwave.
And I think talk of satisfaction with lightwave's animation tools is foolhardy and helps noone.

Big Studios don't just randomly pick Maya or XSI for character animation, It's because they are superior in this area. Newtek could easily put in good solid tools with guidance from proffesionals, but letting them think they're doing just fine isn't going to encourage them to do anything.

Are there any pieces of good animation to be seen that has been achieved with IK boost? I've only seen very poor examples and I think that is no coincidence.

SplineGod
12-28-2004, 02:57 PM
Anything someone isnt intimately familar with can appear to be awkward and poorly designed. IKBooster IS powerful and quite useable. It also provides features that havent been available in Lightwave. Apparently ppl have been able to use it in Japan for quite awhile. I also know others who are successfully using it at prominent studios.

Is it a cure all? No, but certainly a big step in the right direction and Im pretty sure nobody at Newtek feels that theyve reached the peak of some mountain and plan to sit there patting themselves on the back. Nobody is blowing smoke up Newteks backside nor is anyone in Newtek or using LIghtwave under any sort of illusion that the software is perfect.

My experience out here in LA is that MANY studios pick Maya or XSI for reasons other then which is best. I dont see many places out here using XSI in great numbers. Most decisions tend to be political. I know of several places including a prominent studio out here that decided to go with XSI because it was given to them and now are having nothing but headaches with it. ALL software has its strengths and weaknesses.

IKBooster is not a replacement IK system. Its designed to integrate with LWs current pipeline. If youve seen poorly done animation with IKBooster its more likely that the person using it doesnt understand it, the person is not very experienced at animation in general or many other possible reasons. Youre also just as likely to see more bad animation done with ANY 3D application then good animation. I dont think using that as critera to judge a tool is applicable. Nobody just jumps into animation and does it well even with the best tools.

My advice to anyone who has Lightwave is to spend the time to learn the tools since you have them. Theres plenty of books and other materials that cover how to use whats there. :)

pooby
12-28-2004, 03:16 PM
Please could I have an example of these studios that are using it so I can look at the work. I really want to see it, not to mock, but to see if what you are saying is true. I have no evidence.
I have heard plenty of talk from you about it being powerful, but, as last time, I doubt you will actually say what it is that you can do with it that's so useful.

A link will do.. to anything good regarding IK boost or done with it.

I know of 3 studios doing quality work in the UK using Lightwave for character animation. 2 of them are phasing it out because it's being superceded by other packages.

They wouldn't be doing it just for fun. it's an expensive game swapping to Maya or XSI. I know the people from these studios. They are doing it because it's a struggle in Lightwave.

We are the 3rd. If we had the money and time last year, we would have switched to XSI..

but I am giving Lightwave the benefit of the doubt, in the hope that more improvements are being made in this area.
But, I can assure you. Lightwave is NOT up there with the big boys in animation.

And I hope Newtek don't give up the fight and specialise in Lightwave being a model-render package with 'some' basic animation features too.

Librarian
12-28-2004, 04:17 PM
Well, fact is, LW isnīt use much for highend character animation. How often have we seen projects where 'Rendering,Modeling was done in LW, but Animation in in Maya or something else? .
Why should theses studios use Maya for animation instead of LW(although they own it) if Maya(and XSI , Messiah) isnīt more advanced in the animation department? Itīs easy, because Maya is more advanced for animation purposes.
Itīs not just a hype or trend, bull****.
That doesnīt mean it has to be the prefered or perfect solution for everyone.

The following is a post of a proficient user(the_jaco, cgtalk moderator) and shows some of the reasons why LW isnīt the ideal solution for some studios.
Itīs not meant to put LW down.
Anyhow it shows that itīs a long long way to the animation olymp. I hope NT is going to accept the challenge.



during the demo we got negative replies to the following questions:

1) can the order of rotation be changed and animated ? (no = waaaay more complex hierarchies to prevent gimbal lock)
2) can constraints be mixed and blended ? (no)
3) is there any external storage for the animation that is humanly readable ? (no = we can't write parser to retarget and modify animation, currently I can retarget 15000 keyframes worth of animation in a XSI's eani file in split seconds and parse them to our muscle system, no go for LW
4) did the constraints set get revamped with BASIC tools like Npoints constraints (possibly with point balance) ? (no)
5) did you finally implement a link values ? (no=no TD - no simple link values, but you can always use the new expression system, using tens of slow nodes for very simple setups)
6) do you have lattices and arbitrary cages (possibly shape animatable) ? (no)
7) does the SDK finally have BASIC things like a math object and vectors/matrix manipulation ? (no=you have to manually hardcode them in ALL THE TIME)
8) are there any ghost icons and shading modes for the bones ? (no = good luck giving visual coding of bones meanings to animators)
9) can you mix animations and sets of animations and drive the weight of the influence ? (no=no way you can do anything involving combinative systems in LW, sorry mate you are stuck to shapes for facial animation all the way thru)
10) can we use any arbitrary object for enveloping/skinning so that we get something more then points pivoting around a fulcrum ? (no = tuff luck with muscle deformation of any kind, best bet is still using A LOT of additional bones, which unluckily LW is really slow at managing)
11) can we have any referenced models of any kind or would we still have to rely on a split object/scene system replacing and renaming like in 94 ? (no proper referencing with inGUI switchers)
12) can we finally make a bone a pseudoroot (possibly in an animatable way) and have the effector not affected by the last bone ? (erm... what are you talking about ?)
13) can we access weighting from the SDK so we can have some re-skinning tools instead of asking people to deal with hundreds of bones or broken weight editring ? (erm sorry, I don't really know much about weighting in the SDK (nice demo huh?!))
14) look, there's another 20 things I'd need for my pipeline here (maya, XSI and houdini) to not fall back to 96, but it seems like they won't be there either.
tell me, are people really producing any character animation for any high end movie with this ? (erm.. well there's this ghost warrior thing... )
15) why did you even bother ? (well your producer said anything that can cut costs would have been interesting... reply: and having to hire twice the staff to animate gimbal locking skellies with no muscle deformations and firing all the TDs that wouldn't have any SDK worth speaking of is cutting cots ?)

now... there's why we are not using LW to animate or rig here.
if all of the above is bell&whistles then I won't start talking about how much easier one's life is with Maya's native motion retargeting system or XSI's superior handling of shapes in the mixer and how much of that is easily connected to Scripted Operators that I can create on the fly thank to a proper SDK.

edit: btw the above is not meant to say LW is a bad tool, I'm sure it has many strong sides, but please don't be delusional and keep praising an obsolete animation architecture, either you give NT the right input or you won't see features popping into existance in your weapon of choice.

Wonderpup
12-28-2004, 05:57 PM
Thanks for the reply Larry. I take your point that often it's only when software is used in actual production that both it's potential and shortcomings can be revealed. But what I am surprised at is the lack of guidence as to the intent of the IKboost tool- it's actual application seems shrouded in mystery. Is it possible for a company to create a complex tool of this sort without having at least a clear idea what it's intended purpose is? It seems a surreal notion. Surely the people that wrote the code must have had at least some idea of what they wanted the tool to do when they finished it?

SplineGod
12-29-2004, 12:23 AM
Pooby,
Feel free to research the topic yourself. I dont mind interacting and helping people on here but I certainly dont have the time nor inclination to do research that you should be able to do yourself. As far as LW vs XSI I would suggest that if you cant figure out IKBooster to your satisfaction or feel its deficient then do what you feel you need to do to get your job done. There are other packages out there.
I know some people at Rhythm and Hues are using LW which would include the use of IKBooster. IKBooster is part of LW and in a way its like asking to see examples in a well lit scene where someone successfully used spot lights. I know that people there are using IKBooster. I would also suggest that if you have given IKBooster a thorough going thru and have issues that you create a list of what you feel is wrong or right with it and send it into Newtek.

The bottom line is that the term high end animation can mean just about anything. Its not a matter of whether LW is suitable or not for character animation because the fact is that many have used it to do character animation successfully. From a strict sense animation and rigging are not the same thing though they are related. A good animator with a good rig will create good animation in ANY package. That list of tools is only meaningful to someone who uses a package that specifically does those things. For example LW doesnt support pole vectors, instead is uses per channel IK. Two ways to acomplish similar tasks. Ive seen plenty of things that Maya requires one to do that is dead simple in LW and Im sure the reverse is true. Ive seen situations where character work couldnt be pulled off in Maya very well where LW was a much better fit. Ive seen small groups struggle to get tasks done with Maya that a group as small or smaller was able to do in LW with greater ease and speed.
Most animators I know do very little rigging and depend on others to do it. All they want is a rig that allows them to do what they need to do. This leaves the field wide open in terms of what is a suitable rig and which software packages are capable of creating a 'useable' rig.
One thing I have seen is that when someone discovers a shortcoming in LW some tend to look down their noses at it. If Maya however has shortcomings well thats ok its expected. Jen Hachigian put it the best when asked what features from Maya that she would like to see in LW. The answer: The same budgets and the same time frame.

Wonderpup,
True enough. The programmer did and does have something in mind with it. The same idea behind IKBooster can be found in Animanium and Motion Builder. I totally agree that the documentation could be better. The programmer is from Japan and thats one of the hurdles that is being overcome. As with everything else Im sure the dev team is working on improving IKBooster and all aspects of LW (including documentation). I will say that Lightwave with IKBooster is certainly FAR better then LW without it. The tools there, its powerful and quite useful. Im glad its there because it certainly gives me something to play with. Its only going to get better. The fact is that most people dont get rigging and dont get character animation. Most dont understand the rigging tools LW has without IKBooster. Thats not a bad thing because as they learn and expand their knowledge and experience the package has more to give. :)

noyce
12-29-2004, 03:00 AM
I'd love to see a character animation section myself.

Last year i had the task of choosing a character modelling package to use at our studio, and although i tried lots of different packages and did a fair bit of research i finally decided on lightwave 8. Some of the other packages have some very nice features but generally speaking i seem to find lightwave does everything i want, and as i love modelling in lightwave in made sense to try and stick to one package.

That's not to say i haven't had times where i've wanted to throw my machine out of the window :)

pooby
12-29-2004, 03:34 AM
Feel free to research the topic yourself.

I have, exhaustively.

Found nothing.

SplineGod
12-29-2004, 12:57 PM
I'd love to see a character animation section myself.

Last year i had the task of choosing a character modelling package to use at our studio, and although i tried lots of different packages and did a fair bit of research i finally decided on lightwave 8. Some of the other packages have some very nice features but generally speaking i seem to find lightwave does everything i want, and as i love modelling in lightwave in made sense to try and stick to one package.

That's not to say i haven't had times where i've wanted to throw my machine out of the window :)

Im with you on all that :)

Zarathustra
01-05-2005, 12:24 PM
Just to play Devil's Advocate, Pooby has asked, politely, for links to examples of studios using IKBooster. If a link exists, I think perhaps it would behoove you to find both the time and inclination to provide one since you've put yourself in a position of having to promote the tool since you're selling a training dvd for it.

I've Googled and found nothing.

Also, debating whether it's politics or quality of tools as to why studios use apps other then LW doesn't really address the point of the moment - IKBooster.

Finally, I vote "yes" to a character animation section.

SplineGod
01-05-2005, 01:35 PM
I will have to disagree on the whole finding the time and the inclination part. Its certainly not my job or calling to research every person or studio using one of MANY aspects of Lightwaves toolset. What studios tend to use is also to some degree irrelevant. Studios "officially" may use one or two applications yet unoffically use a great many. Individuals at those studios may also use additional applications that suit them as well. We constantly hear about Lightwave being used at some studio but not getting "credit" for being used. Thats difficult enough to get a handle on let alone which studios using Lightwave "officially" use some particular tool in it vs which artists are unofficially using particular tools.

The reason I created the IKBooster videos wasnt to promote the tool but to show people how it works and how to use it. This is no different the people producing any training materials that teachs ANY aspect of Lightwave people dont understand. People purchase training materials on lighting, texturing, rigging, animating, modeling and so forth even though these tools have been around in LW for years.

I constantly see people saying that some tool in Lightwave is sub par or that they dont know how to use it. I also see other people saying the exact opposite about the same tools. Youre going to have people that know how to use certain tools and others that dont. This is the reason people create training materials. People tend to use the tools they know how to do to get the job done and what they dont know how to do they learn by asking questions on forums, asking friends or by purchasing training materials. I cant imagine polling every studio everytime someone doesnt know how to use or doesnt like a particular tool in LW.

As I said before, I know that at least one person at Rhythm and Hues (Ken WIlder) who thinks IKBooster is awsome. He told me however, that this wasnt the case until he went thru my IKBooster video that a coworker had bought. :)

Zarathustra
01-05-2005, 02:20 PM
Well if you create a training dvd on a given tool and want it to sell, it would seem that you would need:
1) interest in the tool
2) viability of the tool
3) an ample degree of confusion on how to use the tool

The latter certainly exists and by your own admission, was very hard to understand initially. However, that confusion seems to make most react as Pooby did. So with that in mind and no clear examples anywhere of anything remarkable done with it, I would expect interest and apparent viability of IKBooster to be low. THAT's why I thought it was in your best interest to provide some kind of link to an example of it's use, to address interest and viability.

I KNOW lighting and texturing is important. There's plenty of examples out there like Jason Lee's work to make one consider training on lighting and texturing.
Great examples of character animation exist, so naturally there would be interest in training for rigging. Same is true for all those other things you mention like animating and modeling. These topics can not be compared to IKBoost.

Your last point, Larry, about people debating it's viability - I haven't heard much about anyone promoting IKBoost beyond yourself and Vaughn so I don't see the comparison to other tools or methods such as spline vs box modeling, for example.

Now don't get me wrong, I'm not saying IKBoost isn't viable or worth attention but truth is, I don't know if it is or not. All I know is it's confusing and hard to figure out, most people online comment poorly on it and as far as I know, I haven't seen anything done with it to make me want to learn any more about it. So to make someone like me, with that experience, want to learn anymore about it I'd have to AT LEAST see something decent that was done with it. That's all I'm saying.

LONG story short: If I see a decent animation and I hear IKBoost was instrumental in creating it, then I'm thinking, "Hmmm, IKBoost. I gotta learn about this and see what's what. Who's got some training material on that?"

SplineGod
01-05-2005, 03:17 PM
LONG story short: If I see a decent animation and I hear IKBoost was instrumental in creating it, then I'm thinking, "Hmmm, IKBoost. I gotta learn about this and see what's what. Who's got some training material on that?"

I understand. I just dont see it as my responsibility to "prove" whether something is viable or not. If IKBooster was an 3rd party plugin I could see the merit in spending the extra time to prove its worth. In this case IKBooster is part of Lightwave. When someone buys Lightwave there isnt that decision on their part to buy IKBooster or not Its simply there already. I then look at IKBooster like I do any other part of Lightwave: something that some get and something that some dont get. I spent the time and effort to figure it out and produced some training materials to help those that dont get it. All the feedback Ive gotten about the training material on IKBooster has been overwhelmingly positive. Like I said, Im not trying to convince anyone to buy some 3rd party plugin. IKBooster is already there so why not use it or learn how to use it? People have the choice on learning LW to its fullest potential or not. Learning to get the most of Lightwave is probably more cost effective then buying yet another application and attempting to master that. :)

Dodgy
01-05-2005, 05:50 PM
1) can the order of rotation be changed and animated ? (no = waaaay more complex hierarchies to prevent gimbal lock)

This is true.


2) can constraints be mixed and blended ? (no)

I can mix constraints. This is not true.


3) is there any external storage for the animation that is humanly readable ? (no = we can't write parser to retarget and modify animation, currently I can retarget 15000 keyframes worth of animation in a XSI's eani file in split seconds and parse them to our muscle system, no go for LW

What exactly is he on about here? Plenty of people have written motion exporters/MD readers for LW.


4) did the constraints set get revamped with BASIC tools like Npoints constraints (possibly with point balance) ? (no)

What are N point constraints? I'd done a search on google and found nothing. Is it that basic then? Does he mean being able to constrain to n points on a mesh? Or n different point constraints? I can do the latter. I've never heard of them.


5) did you finally implement a link values ? (no=no TD - no simple link values, but you can always use the new expression system, using tens of slow nodes for very simple setups)

Channel follower?


6) do you have lattices and arbitrary cages (possibly shape animatable) ? (no)

We have metalink....I know they're not the same thing, but what does he want to do with them? Bet I could emulate it....


7) does the SDK finally have BASIC things like a math object and vectors/matrix manipulation ? (no=you have to manually hardcode them in ALL THE TIME)

I have no expertise in this area.


8) are there any ghost icons and shading modes for the bones ? (no = good luck giving visual coding of bones meanings to animators)

You can nulls and give them custom shapes, or add semi transparent non-renderable objects and use those. This is of course slightly fiddly, but how fiddly is it to set up a custom bone shape in whatever package he's using?


9) can you mix animations and sets of animations and drive the weight of the influence ? (no=no way you can do anything involving combinative systems in LW, sorry mate you are stuck to shapes for facial animation all the way thru)

What does he mean here? I can layer bones/texture deformation/ morphing and any number of other animation bits and bobs.


10) can we use any arbitrary object for enveloping/skinning so that we get something more then points pivoting around a fulcrum ? (no = tuff luck with muscle deformation of any kind, best bet is still using A LOT of additional bones, which unluckily LW is really slow at managing)

I've seen very good muscle deformation done in LW, but what does enveloping/skinning do except tie the mesh to a bone/pivot? You can use all sorts of methods to deform a mesh, all with user definable weightmaps controlling them. You can have as many weight maps as you want on a mesh.


11) can we have any referenced models of any kind or would we still have to rely on a split object/scene system replacing and renaming like in 94 ? (no proper referencing with inGUI switchers)

Object Replacement? HDinstance? The instancing in bare bones LW isn't very good, but you can add to it if you need it.


12) can we finally make a bone a pseudoroot (possibly in an animatable way) and have the effector not affected by the last bone ? (erm... what are you talking about ?)

This might be cool :)


13) can we access weighting from the SDK so we can have some re-skinning tools instead of asking people to deal with hundreds of bones or broken weight editring ? (erm sorry, I don't really know much about weighting in the SDK (nice demo huh?!))

You can access weights from LScript, you don't have to reach for the SDK. I've done weighting, it's a lot easier than people make it out to be, especially since you can cut your mesh up and weight bits separately then paste them back together.


14) look, there's another 20 things I'd need for my pipeline here (maya, XSI and houdini) to not fall back to 96, but it seems like they won't be there either.
tell me, are people really producing any character animation for any high end movie with this ? (erm.. well there's this ghost warrior thing... )

If they're so good, why is he using 3 (more expensive) packages? Maya/XSI/Houdini not good enough on it's own? Each package has it's own strengths and weaknesses.


15) why did you even bother ? (well your producer said anything that can cut costs would have been interesting... reply: and having to hire twice the staff to animate gimbal locking skellies with no muscle deformations and firing all the TDs that wouldn't have any SDK worth speaking of is cutting cots ?)

now... there's why we are not using LW to animate or rig here.
if all of the above is bell&whistles then I won't start talking about how much easier one's life is with Maya's native motion retargeting system or XSI's superior handling of shapes in the mixer and how much o that is easily connected to Scripted Operators that I can create on the fly thank to a proper SDK.

edit: btw the above is not meant to say LW is a bad tool, I'm sure it has many strong sides, but please don't be delusional and keep praising an obsolete animation architecture, either you give NT the right input or you won't see features popping into existance in your weapon of choice.

I give NT LOADS of feature requests all the time, but I don't see LW as being obsolete when I see lots of problems in the 'elite' package maya everyday at work. I think LW needs a lot of cleaning up, it sure ain't elegant in animation, but I think that of other packages too in different areas. If you have the money, buy all these other packages if you need to and use them according to their strengths. I can do quite a lot of stuff nicely in LW thanks.

colkai
01-06-2005, 04:03 AM
Surely the people that wrote the code must have had at least some idea of what they wanted the tool to do when they finished it?

Hehe, that' the wierd thing about code - the end user can always find a myriad of ways to use something in a manner the programmer never even dreamt of. ;)

colkai
01-06-2005, 04:06 AM
One thing I have seen is that when someone discovers a shortcoming in LW some tend to look down their noses at it. If Maya however has shortcomings well thats ok its expected.

BANG!!!!
There you go, that is thevery thing that bugs the heck out of me, well said Larry.
Any product needs improvement, any product can produce bad results if used without knowledge.

Wonderpup
01-06-2005, 07:38 AM
Hi Colkai

I think the point I was trying to make is not so much about the way IK Booster is being used, but rather the fact that no one ( aside from Larry) seems to know how it is supposed to be used. Am I the only one who finds it strange that the only way I can discover the intended function of a flagship lightwave 8 feature is to buy a third party training DVD? I understand that the programmer is japanese, but surely there must be way to ask him the intended usage of his tool, and then translate his reply into english. As I have already said, this is not meant as an attack on Larry or anyone else selling training material- but I would argue that if a company incorporates a tool into it's product, I should not be placed in the position being forced to buy third party training in order to discover what the basic functionality of that tool is- As a registered owner of Lightwave 8 I still, to this day, have been given no guidence as to how IK Boost is intended to integrate with the existing CA toolset, and I simply don't understand why.

colkai
01-06-2005, 07:53 AM
Yeah,
Don't get me wrong, I fully understand what you mean and I'm quite convinced many of us are not exploiting LW to it's full capacity due to this lack of clarity.

I think the problem stems from Newtek not having anyone who can invest the time to fully explore these things and explain them to us, the end user.
I'm not sure how one would achieve this in a small company as the only real answer is to throw money at it and pay for people to really work on the features and then explain them, but that's quite a resource to tie up.

I do think we need some better training material from Newtek themselves, it is irritating to have to buy training materials to figure out the added features. The old LW manuals use to be much better for this and whilst Protons video tutorials went some ways, we really need some clearer explanation of the tools.

I have learned most of what I have by stumbling along as I had no associates to "feed off" but it is far from a pleasant way to learn.

Sure, if you have money, you can buy all sorts of training, but even then, you rely on the course actually telling you what you need to know. Just because a course covers a subject, does not follow it does so in a way you personally can follow and benefit from. For those on a limited budget, (hobbyists such as myself), even that avenue is blocked, so we rely more on provided documents and examples.

I only hope that when 8.2 etc.. are released, the effort is put into documenting and providing "real" examples of the new features.

pooby
01-06-2005, 09:15 AM
I don't see that it would be hard to just say.

IK boost enables you to do ....(insert useful comment)... that was previously very difficult to achieve in lightwave.

I would even buy the training DVD if it sounded good.

But I wouldn't buy a DVD on the strength of..

'IkBooster is very powerful and adds many things that have been missing in LW. '

Its like promoting a movie by providing no more information to a potential audience than. 'this film is really good.. you'll like it'

Zarathustra
01-06-2005, 09:40 AM
Am I the only one who finds it strange that the only way I can discover the intended function of a flagship lightwave 8 feature is to buy a third party training DVD?

No, you're not the only one.

In case you have any confusion, Larry, as to my intent, I don't want to prevent you from selling any dvds. I just find it odd that with apparently little interest in IK Boost, spurred by lack of any information as to how to use it, who's using it, or any examples of it being used anywhere that you don't feel any need to at least address the latter. I mean, increasing interest in the tool would mean increased interest in the dvd, wouldn't it?

Right now it's pretty much like Pooby's movie analogy, only you can add that Ken WIlder and Jen Hachigian liked it.

spec24
01-06-2005, 10:03 AM
I only hope that when 8.2 etc.. are released, the effort is put into documenting and providing "real" examples of the new features.

REAL examples, YES! What a concept :) I look in the manual or help file for a feature I've not used yet and it gives me very little. More like a programmers explanation. No real world example of what the particular tool might do, how it might be used. Only occasionaly do I see this. This kind of thing would greatly benefit us all, especially those new to LW.

SplineGod
01-06-2005, 11:39 AM
The funny thing is that apparently there IS a lot of interest in IKBooster. Ive sold a ton of the DVDs thru KURV. The DVD has even made the front page at Safe Harbour. The reason I created that video was because a lot of people seemed to be having difficulty figuring out IKB (including myself). Thats why its about 3 hours long and 25 bucks. To be honest If I never sell another IKBooster DVD it wouldnt bother me terribly. Its sold FAR more then I anticipated and doesnt show any sign of abating. The reality is that people are always going to be curious about IKBooster simply because it comes free with the rest of Lightwave. Also, like everyone else ( or what everyone should be doing) I use the software and send in suggestions and report bugs. Im pretty sure IKBooster is one of those features that will only get better with time.

I know that there are others beyond the people I mentioned who are using IKBooster. Others I know are at various stages of learning it. I had to spend a lot of time figuring out how it works (and I still dont know every single aspect of it). At that point I decided to make an inexpensive video on what it is vs LWs current IK setup and how to get started with it and make it work. If I were to do a thorough video on IKBooster it would be MANY more hours. All I can say is that I labored under the same constraints everyone else has and was able to figure out IKBooster (and yes I had to sweat blood a few times as well). :)

I agree that it would be great to have documentation with lots of tutorials to back it up. Different people want this for just about every aspect of Lightwave. I dont think its every going to happen though. Its just a problem with any software. End users always know far more about what the software can do then the people who created it. Thats why they rely on beta testers and suggestions from end users.
Ive used many applications over the years and none every had what I considered to be complete documentation or deep enough tutorials. The examples shown are typically never the examples I need. Its always how much do you show and what examples do you show and still please everyone?

As far as why IKBooster is powerful or better....
1. IK/ FK can be used interchangeably on the fly.
2. Joints can be pinned /unpinned on the fly
3. Ik chains can be altered on the fly
4. IKB can be blended with LWs native IK
5. LW supports Translation, Rotation and Scaling of an item. IKBooster is that fourth thing that can be done. IKBooster sees LWs native translations and rotations and vice versa
6. Each joint is automatically a node or goal
7. Everything can be done right in layout without having to open other panels
8. Poses can be saved and reloaded
9. Dynamics - Again not limited to bones only. Dynamics can be applied to a heirarchy or subheirarchies. The results can be animated on top of quite easily.

Theres a lot more then this but this definately qualifies as better or more powerful in my book. :)

spec24
01-06-2005, 12:06 PM
As far as why IKBooster is powerful or better....
1. IK/ FK can be used interchangeably on the fly.
2. Joints can be pinned /unpinned on the fly
3. Ik chains can be altered on the fly
4. IKB can be blended with LWs native IK
5. LW supports Translation, Rotation and Scaling of an item. IKBooster is that fourth thing that can be done. IKBooster sees LWs native translations and rotations and vice versa
6. Each joint is automatically a node or goal
7. Everything can be done right in layout without having to open other panels
8. Poses can be saved and reloaded
9. Dynamics - Again not limited to bones only. Dynamics can be applied to a heirarchy or subheirarchies. The results can be animated on top of quite easily.

Theres a lot more then this but this definately qualifies as better or more powerful in my book. :)

Don't forget Channel linking Larry - that's a nice feature.

SplineGod
01-06-2005, 02:41 PM
I dont like the channel linking except for real basic controls mainly because you cant animate on top of it. I still like selection sets for this kind of control . The other motion modifier such as channel follower also work. :)

Wonderpup
01-06-2005, 03:59 PM
I am not sure the idea that it is down to the users to discover for themselves how the tools work is viable in the current climate. If the idea is to expand your user base, you cannot afford to appeal only to the 'nerd' mentality (and I count myself among the nerds here- so no insult intended)

Many potential users will be looking at a 3D package as only being a part of their overall toolkit ( Web designers, for example) and these people will not be willing or able to spend the time to decipher how the tools work- they will simply seek out more transparent solutions.

Chuck
01-06-2005, 04:28 PM
For LightWave [8] the team managed to produce the most extensive documentation for the product that NewTek has had to date, and to provide both a reference manual and a tutorial manual that go a long way toward getting a new user started, or an experienced user oriented to the new features. That said, there are a lot of aspects of LightWave that need to be addressed with improved documentation and which are ripe for good tutorial material. NewTek's LightWave team has every intention of continuing our improvement of the documentation of existing features, providing complete and comprehensive documentation for new features as they come along, and developing comprehensive tutorial materials for LightWave users in a variety of media, from print to html to video.

Specific to the topic of this thread, our plans include improved documentation and training material for the IK Booster tool.

colkai
01-07-2005, 02:05 AM
Thanks for that Chuck,
One thing for certain, those that say Newtek aren't listening, aren't listening. ;)

I do think the way you have gone with the HTML help is the right way because as things like this crop up and more info is added, it is better than having to redo a whomping great PDF each time. Not that the PDF isn't useful as well, but gotta love that search facility in the new online docs. :)

Wonderpup
01-07-2005, 01:22 PM
Hi Chuck

I agree, the new help system is a vast improvement- I use it nearly everyday, a fantastic resource.

Glad to hear that more of the same is on the way.

faulknermano
01-07-2005, 07:53 PM
[

12) can we finally make a bone a pseudoroot (possibly in an animatable way) and have the effector not affected by the last bone ? (erm... what are you talking about ?)


tell me, dodgy, what the heck is talking about. what's cool about it, whatever it is? :)



I can mix constraints.

yes, but the constraints themselves are rather primitive.





-6) do you have lattices and arbitrary cages (possibly shape animatable) ? (no)


We have metalink....I know they're not the same thing, but what does he want to do with them? Bet I could emulate it....


lattices, in maya for example, are deformable deformers. metalink is dynamics-contolled at the top. even if you did try to bone your metalink object (e.g. havent tried it) you would still need to run the simulation before you see the effect. you'd still be running under a simulation.

lattice points can be "clustered" to provide weighted (or specific) deformation of the cage. they can be boned. they can be applied soft body dynamics as well.

it gives a certain flexibility, more than anything. while final result is dependent upon the operator, lattices in themselves are a unique kind of deformer which lw will especially benefit from, imo.

-----


my opinion regarding animation tools in lw: when pooby remarked something about parts of lw not working *together* as they should, the first thing that came to my mind was the constraint system. it is indeed SimpleConstraints. i'm thinking, up to now, where are the AdvancedConstraints? the current one functions within set rules. outside of that it is incapable of functioning as one would `expect` from constraints. it's not the case that you just have to set other parameters to get a desired effect. it's beyond the scope of the capabilities of the plugin.

its my pet peeve: constraints. maya becomes like lw modeler: what i see is what i get. no fuss. if i want my object constrained, it's constrained in plain sight. it doesnt matter what heirarchy is below it. the guy is constrained orient, aim, or point. you can do so many things with that alone.

but this is just the tip of the iceberg.

but i were to suggest anything, it will be stripping down of lw features into tools that are more universal. as bruce lee would say "daily decrease, not increase. strip away the unessentials." i think this is key from a development standpoint. when i dislike maya, it's because it is rearing its complicated head. when i like maya, it's because the *simple* elements work together so nicely that no matter how complex the set-up eventually becomes, they dont break down.

is lw complicated? modeler, no. on the contrary, modeler is `inhibited.` i think it should do more.

but layout is a bloated and complicated. why `complicated`? from an animation standpoint, there are many `rules` and `prohibitions` and many things to watch out for, and even as a matter-of-fact, it's not even apparent exactly what those `rules` are.

-----

btw: we can go both ways: we can be `immediate pragmatists` and go and find solutions (e.g. IKBooster) from our problems NOW. and we can be 'foward-looking idealists` and suggest where lw *should* and *should not* go. one shouldnt think we can only be one or another.

faulknermano
01-07-2005, 07:55 PM
It is possible to do decent stuff, but lightwave doesn't make it easy for you.

THIS is the main argument. it is not easy.

colkai
01-08-2005, 06:31 AM
is lw complicated? modeler, no. on the contrary, modeler is `inhibited.` i think it should do more.

Surely this is the same for any package - it could always "do more" yes?



but layout is a bloated and complicated.
You are joking right? bloated?



btw: we can go both ways: we can be `immediate pragmatists` and go and find solutions (e.g. IKBooster) from our problems NOW. and we can be 'foward-looking idealists` and suggest where lw *should* and *should not* go. one shouldnt think we can only be one or another.

But...lest we forget, whatever way you want it to go has little relevance on how it is being developed, unless of course, you happen to be on the development team.
I think you have to accept that Lightwave, like other packages, will always have its flaws and areas that many need improving.
The problem has been, and indeed, will continue to be, that each user has very definite ideas about which part of LW is the worst and should be improved before anything else. The problem being that, in any given group of users, no-one can seem to agree on the same feature set. ;)

faulknermano
01-08-2005, 10:03 AM
Surely this is the same for any package - it could always "do more" yes?

i meant more in relation to layout.



You are joking right? bloated?

well.... it's hard to explain. it seems "bloated" in a sense that `there's so many things in it that it gets hard to move around in`.



But...lest we forget, whatever way you want it to go has little relevance on how it is being developed, unless of course, you happen to be on the development team.

little relevance, yes. if the `way i want it to go` makes good sense, it is irrelevent what i want, simply because what i would want just makes good sense.

so the idea is NOT `what i want`, rather, `are the things that i want good ideas to begin with?`


I think you have to accept that Lightwave, like other packages, will always have its flaws and areas that many need improving.

The problem has been, and indeed, will continue to be, that each user has very definite ideas about which part of LW is the worst and should be improved before anything else. The problem being that, in any given group of users, no-one can seem to agree on the same feature set. ;)

sure. a person who primarily lights will probably think the renderer is in worse condition than modeler. but that's because he's thinking out of ignorance of other parts within lw. now take that person into maya, for example, and let's see how he sees lw now.

it's just that one should never speak about what he / she doesnt know. i only speak strongly about maya vs lw in the aspect of animation (not CA, per se) because i am primarily an animator (which includes "dynamics animation") and texturer when i get on the maya cockpit, and i've been animating in lw for quite a long time (longer than in maya). other CG aspects, like modelling, i leave with just "i prefer to model in LW". that's because i dont model in maya, and have only basic knowledge of it. my comparison stops where my knowledge ends. needless to say, i'm pretty confident that i can see the differences in the animation aspect, having jumped into maya and having gone through the paces, and continually be faced with issues from both lw and maya *side by side*, an ongoing thing.

i can enumerate what i `prefer`, but i can also distinguish that from `predictable behavior`. i have nothing to bad or good to say about maya's dynamics system because i only have minimal time on it. i `prefer` PFX's straightforward `layman` approach - it's just `my` style. whatever lacking things, like bloated scenes of numerous clones, i bear with it and press on because i feel that the advantages (e.g. on-the-fly particle cache) outweigh the disadvantages.... sometimes. i prefer endomorphs than blendShapes, also. i prefer the `non-commital` skinning system of lw better, too. i prefer maya's expressions, because they are more flexible (e.g. not single-line, able to reference other UDFs, etc).

but one cannot prefer a bad constraint system over a robust one on the basis of a constraint system (here i go again with constraints).

my overall problem with CG apps is how you describe them being not perfect: there are good mixture of pros and cons for every app, which pisses me off because i hate jumping from app to app especially when the crossover part involves some major form of interdependency (e.g. skinning to animation). and the reason why i bother voicing out is because there are things in lw that i cant live without (e.g. modeler, endomorophs, PFX, HVs, renderer, etc) and to have a less capable aspect of lw (in my view, the animation aspect) brings me out of my chair to load maya up, wishing that i could be comfortable doing everything in one cozy little app. of course, this is just my `preference.` ;)

Dodgy
01-08-2005, 04:43 PM
[
tell me, dodgy, what the heck is talking about. what's cool about it, whatever it is? :)

The ability to treat any object in the chain as the parent temporarily, so the rest of the hierarchy rotates around it :)


yes, but the constraints themselves are rather primitive.

I'm always confused when people say advanced. What features would you like to see in constraints, and couldn't a plugin be written to include these features? I've looked at the maya constraints, and the scale/move/rotate constraints are pretty much the same, except the weighting is a bit nicer. I think the constraints could be fixed to do this quite readily. I might even have a go myself. Of course, people are going to say yet another plugin, but my reaction is so what :)


lattices, in maya for example, are deformable deformers. metalink is dynamics-contolled at the top. even if you did try to bone your metalink object (e.g. havent tried it) you would still need to run the simulation before you see the effect. you'd still be running under a simulation.

lattice points can be "clustered" to provide weighted (or specific) deformation of the cage. they can be boned. they can be applied soft body dynamics as well.

it gives a certain flexibility, more than anything. while final result is dependent upon the operator, lattices in themselves are a unique kind of deformer which lw will especially benefit from, imo.


Yeah, lattices are a bit simpler, but metalink parents can be boned, or weighted, and this all translates through to the high poly object. So it's almost there, if it was just in real time instead of having to run the simulation.

-----



my opinion regarding animation tools in lw: when pooby remarked something about parts of lw not working *together* as they should, the first thing that came to my mind was the constraint system. it is indeed SimpleConstraints. i'm thinking, up to now, where are the AdvancedConstraints? the current one functions within set rules. outside of that it is incapable of functioning as one would `expect` from constraints. it's not the case that you just have to set other parameters to get a desired effect. it's beyond the scope of the capabilities of the plugin.

its my pet peeve: constraints. maya becomes like lw modeler: what i see is what i get. no fuss. if i want my object constrained, it's constrained in plain sight. it doesnt matter what heirarchy is below it. the guy is constrained orient, aim, or point. you can do so many things with that alone.

but this is just the tip of the iceberg.

but i were to suggest anything, it will be stripping down of lw features into tools that are more universal. as bruce lee would say "daily decrease, not increase. strip away the unessentials." i think this is key from a development standpoint. when i dislike maya, it's because it is rearing its complicated head. when i like maya, it's because the *simple* elements work together so nicely that no matter how complex the set-up eventually becomes, they dont break down.

I agree wholeheartedly with this. There are a lot of overlapping tools in LW which could be amalgamated into fewer 'super tools'. I think the team will work towards this though, they just need time....


is lw complicated? modeler, no. on the contrary, modeler is `inhibited.` i think it should do more.

but layout is a bloated and complicated. why `complicated`? from an animation standpoint, there are many `rules` and `prohibitions` and many things to watch out for, and even as a matter-of-fact, it's not even apparent exactly what those `rules` are.

Again, agree completely. It would help if some of the key assignments were consolidated, made more universal across interfaces. After all, all the timeline based interfaces have different keys for panning the views, scaling them, moving items etc, and this just makes LW more awkard to use when it doesn't have to be.

-----



btw: we can go both ways: we can be `immediate pragmatists` and go and find solutions (e.g. IKBooster) from our problems NOW. and we can be 'foward-looking idealists` and suggest where lw *should* and *should not* go. one shouldnt think we can only be one or another.

I agree, but I took issue with the claims that LW is incapable of doing good character animation, not that it's not a bit more clumsy. There's a lot of good things in LW now, it just needs someone to look at reorganizing it into a coherent whole.

faulknermano
01-08-2005, 05:08 PM
The ability to treat any object in the chain as the parent temporarily, so the rest of the hierarchy rotates around it :)

ah..hmmm.. <rubs chin>


I'm always confused when people say advanced. What features would you like to see in constraints, and couldn't a plugin be written to include these features?

if there was a plugin that was better than SimpleConstraints then it ought to be in lw. if a plugin writer could make a constraint plugin that works better, and works at every instance, then that's all we are asking for.

but regarding my complaint per se: it's not features. but the solidity of the constraint. perfect example: wouldnt anyone here of us agree that orient constraints ought to act like Match Goal Orientation? or even better, we should have the option / capability of this being so? yet we cant. add a parent here and an ik thingybob here and bam! simpleConstraints goes whack. (sadly, even the 3rd party plugin MatchGoalOrientation has some limits like two MGOs in the same IK chain needs the native LW MGO turned on- which, if memory serves, was thoughtfully documented)


I agree, but I took issue with the claims that LW is incapable of doing good character animation, not that it's not a bit more clumsy.

personally, i never for once believed good animation depended on the software, CA or otherwise, but CA most especially. it just helps it along. (in fact, i'm doing a personal CA-driven project all in LW at the moment. it is most likely from this that CA / rigging issues are still fresh in my mind, hence my loud-mouthedness and my long-windedness in this thread. :D)

Dodgy
01-08-2005, 05:17 PM
if there was a plugin that was better than SimpleConstraints then it ought to be in lw. if a plugin writer could make a constraint plugin that works better, and works at every instance, then that's all we are asking for.


I agree, see my UVQuantize function which i wrote to replace LW's slapdash effort :)



but regarding my complaint per se: it's not features. but the solidity of the constraint. perfect example: wouldnt anyone here of us agree that orient constraints ought to act like Match Goal Orientation? or even better, we should have the option / capability of this being so? yet we cant. (sadly, even the 3rd party plugin MatchGoalOrientation has some limits like two MGOs in the same IK chain needs the native LW MGO turned on- which, if memory serves, was thoughtfully documented)


Hmm, I've used Orient constraint to give me that MGO effect for character animation baking for games, it worked for me. I'm I not using it enough to see the defects?



personally, i never for once believed good animation depended on the software, CA or otherwise, but CA most especially. it just helps it along. (in fact, i'm doing a personal CA-driven project all in LW at the moment. it is most likely from this that CA / rigging issues are still fresh in my mind, hence my loud-mouthedness and my long-windedness in this thread. :D)

Hey, I don't care if people criticize, but they have to do it constructively :)

faulknermano
01-09-2005, 12:39 AM
Hmm, I've used Orient constraint to give me that MGO effect for character animation baking for games, it worked for me. I'm I not using it enough to see the defects?

i'm sorry, but my issue was not about orientConstraints in relation to MGO. i have a problem of MGO as well pointConstraints and i'm just mixing them up. (suffering from insomnia recently and i'm thinking pretty skewed. gosh, i wonder what else i've said :p )

EDIT: my issue with general constraints is how they work with IK, for example. if, for some absurd reason you need to orient constrain a bone that's somewhere in the middle of an IK chain, the bone successfully orients itself. but the IK doesnt see that and solves as if there was no constraint put upon one of its members.

while this has more to do with IK than with orientConstraints, i'm trying to look at here from the overlooking perspective. so while orientConstraints does its job, its relation to IK, as an example, make that feature (not IK) less functional, even though it may not be the plugins fault.

pointConstraints are another matter. i have alot more problems with these. but that's for another thread, i think. :D

pooby
01-09-2005, 05:26 AM
I think one of the main problems is that when lightwave calculates IK, it doesn't incorporate motion modifiers into it.
The IK system was made before most of the modifiers were invented, so the modifiers have to fit around it.
The constraints,for example, only work 'after ik' which is very limiting. It would less limiting, even if there were a 'before Ik', but what we really need is an option to have 'incorporate into IK' too, so everything is aware of what's going on.
(the weighting is a palaver too.. at least there could be a 'global' option, rather than to fix envelopes on each channel manually.)

At the moment you have to frustratingly learn the bugs and errors, then plan your rigs to avoid them.
Also, If you have 2 IK solutions, you can't have one part pointing at a goal that is attached to another Ik rig, unless these 2 rigs are joined in their calculations.
I think there ought to be an option of a 2nd IK pass. and for you to be able to choose which pass whatever chain uses to solve these issues.

I was so worried when I first played with IK boost. not so much that its awkward to use, than the fact it looks like Newtek keep sticking band-aids on to hide symptoms, rather than going into surgery and fixing the root problem.

I agree with everything faulknermano has had to say on the topic. The underlying problem is the architecture.
We are using an 'core' system for animation that hasn't changed much since I started in 1997 and it needs to be updated, and many of the modifiers need to be incorporated into the core, rather than stuck on the edge.

The problem is, the longer this goes on, the more experienced animators defect to other packages and without these people on board nagging Newtek, the more likely it is that things will just stay as they are.

I think for every Character animator nagging Newtek for changes there are probably 50 modelling and 50 rendering requests..
That's not because it needs less changes in this area, it's because there are less experienced animators using Lightwave and the number is decreasing all the time.

The danger is that it will get to a point where it wont be seen as a serious CA package at ALL after a while then it'll be --'what's the point in even trying to compete now?'

spec24
01-09-2005, 11:21 AM
I agree pooby. Very little feels integrated. I get that "band-aid" feel a lot. It seems, and maybe I'm wrong, that a lot of the people who are guru's with lightwave and have no problem with it are those who have been using it since its earliest days. So with them, each band-aid added is an improvement over what they had been working with - they understand how to use the tools because it has been a progression. Those who are new to the software take a look at all these tools seemingly strewn around and don't know where to begin. I posted a comment about the "object replacer" master plug-in and it feels very hidden, especially to a new user. It seems that something like that could be integrated into, say, null object selection in the scene editor. Maybe a right click and an option to replace nul since that seems to be its main use. This is just a simple thing and I know that these are core LW issues that aren't easily implemented without rewriting considerable code. But hey!

Noratio
01-09-2005, 01:09 PM
Those who are new to the software take a look at all these tools seemingly strewn around and don't know where to begin.

Can't nag but can confirm

pooby
01-09-2005, 04:12 PM
by the way.....

I would LOVE to see a Character animation section! I think it's important to have one..

More lightwave animation on show is a good thing..