humanModeling part II - Page 11 - Simply Maya User Community

# 151 31-12-2005 , 03:53 PM

Subscriber
Join Date: Oct 2005

Location: cologne germany

Posts: 63

goil , das is perfekt fuer die sylvesterparty auf der ich heude bin .. da hamma bei nem freund ne jungle-hoelle im keller eingerichtet mit lichtorgel und stroboskop user added image

hoer sowas zwar eigendlich ned .. aber auf so parties rockt das user added image

Quote

# 152 31-12-2005 , 03:54 PM

Lead Modeler - Framestore
Join Date: Feb 2003

Location: UK

Posts: 6,287

Indeed!!

I'd love to know whats been said

Jay

Quote

# 153 31-12-2005 , 05:16 PM

Subscriber
Join Date: Oct 2005

Location: cologne germany

Posts: 63

hehe sry
falot told me what goa is and i answered that this is the music i need for my new year party.. i dont really like it but every party needs a "jungle-hell" where you can dance to jungle, drum n bass and now goa user added image

btw i think it would be much easier for me when you all learn german user added image

just a joke

Quote

# 154 03-01-2006 , 06:47 PM

Registered User
Join Date: Jan 2005

Location: vienna

Posts: 1,095

considerations about usage of the Spinal Rig

if I where to set up a chain of 25 vertebrae joints controlled by a joint chain consisting of 4 to 5 joints - how would an animator like to work with the rig? just by rotating the single control joints, or would it be neccessary to even move some of the lowRes joints? sounds odd to me. in the alias beginners guide DVD the spine is set up like this, that´s why I´m asking. in my book "Inside Maya 5" there is this devine spine setup, I constructed. I guess lowRes joints should just be rotated, but it is possible to move them as well. what the book does not mention - should it be moved or just rotated? (alltho moving doesn´t make much sense) in Jason Schleifer´s setup those controlJoints too can be rotated and moved. there is some other setup from paul thuriot which describes a open and a broken but constrained hierarchy. but still everywhere -

joints from the controling lowRes joint chain can be both - rotated and moved. this seems to lead to the fact that animators need both - rotate and move the spinal joints.

BUT:

It seems to be the most unreasonable thing moving spinal joints in space, at least for me. (cause of the non zero value in the pointing-down-the-jointchain-axis.translation) what would you animators out there prefere? moving spinal joints or just rotate them?

Last edited by Falott; 03-01-2006 at 06:50 PM.

Quote

# 155 03-01-2006 , 08:57 PM

Lead Modeler - Framestore
Join Date: Feb 2003

Location: UK

Posts: 6,287

The idea of Jay Schleifers lo res spine is for control. The spine is given the squash,stretch and rotation constraints because thats how the spine works in reality. Dont forget they are using joints, real skeletons and then muscles rigs followed by skin. This type of rig is acting as a finer level of control for a more flexible back bone. Clusters ar still placed on this type of rig for control too

This whole set up is also explained in Chris Maraffis book on character set up, where he explains the process and you can create the rig similar to that of Jay Schleifers. This is then added to with Control curves for the twist bends and side to side motions using expression instead of sdk and has more control.

Bear in mind an animator will always look for the easiest level of control to get the work done, and if its not to their satifaction they WILL tell you if they dont like something , so you may have to set a rig up with different sets of control for different animators as no one animator will work the same

Cheers
Jay

Quote

# 156 03-01-2006 , 10:54 PM

Registered User
Join Date: Jan 2005

Location: vienna

Posts: 1,095

The spine is given the squash,stretch and rotation constraints because thats how the spine works in reality.

Originally Posted by

yes, but there´s no limitation to stretchyness like in real life, except the "strech warning color". that means if I don´t have a natural limit, I could easily screw up the spine.... there are 2 things which I quite don´t bring together.

1) if the hiRes_spine is moved by only rotating its lowRes_controljoints, there is definitly no reason for installing a stretchy spinenetwork.

2) if the hiRes_spine is moved by additionally translating its lowRes_controljoints, adding strechyness is to be considered. disadvantage hereby would be dealing with an unstable system in terms of physical limitations.

my argument is: why create the possibility to screw up a rig? or is this argument counterproductive? however, I feel like the very first noob in the world with maya right now.

I´ll do some testing now with spine setups and post updates on what I hopefully come up with later.

Quote

# 157 03-01-2006 , 11:03 PM

Lead Modeler - Framestore
Join Date: Feb 2003

Location: UK

Posts: 6,287

Of course theres limitation, you'll need to set a limit on the joint. or write an expression for the movement, it can be done as Ive done it.

There is no argument, its just how difficult or easy you want it to be. Also ask yourself what the character will be doing. Its all down to the basic needs of what you want the character to do. You may be looking to do something your character doesnt need at all.

The secondary spine network is for finer control ie stretch, but if you dont need the character to do this then dont create it. Also the another method of deducting how far the limits are on the stretch would be to include a 'stretch driven Shader'. This is most useful for the animator because if the spine is moved too far from its limit the applied shader network will cause the skeleton to Glow in open GL. Dont limit yourself to how something works aesthetically because you will be digging yourself a large hole and you wont see past solving the basic probs.

Cheers
Jay

Quote

# 158 03-01-2006 , 11:27 PM

Registered User
Join Date: Jan 2005

Location: vienna

Posts: 1,095

Dont limit yourself to how something works aesthetically

Originally Posted by

you are so right. in my head it usually has to be clear before bringing it down to life. I guess most aestetic solution right now is consulting my pillow and do a real dirty rig tomorrow morning.

thx for still keeping track of this long thread!

Quote

# 159 04-01-2006 , 08:46 PM

Registered User
Join Date: Jan 2005

Location: vienna

Posts: 1,095

how zero out translation values from joints?

made a terrible mistake. I made up the skeleton from parts. when I wanted to connect all joints to a chain I got those connected joint´s Translation chart full of non zero values, rotation is 0 of course. does that even matter or is it neccessary to only have the root joint with non zero values and all others with zeroed out except the scale axis?

(I guess with IK it doesn´t make any difference..)

Quote

# 160 04-01-2006 , 09:06 PM

Lead Modeler - Framestore
Join Date: Feb 2003

Location: UK

Posts: 6,287

Paul Thuriot explains this on his paperwork

Cheers
Jay

Quote

# 161 04-01-2006 , 10:44 PM

Registered User
Join Date: Jan 2005

Location: vienna

Posts: 1,095

I was watching the vid and reading the papers for the last 2 hours user added image

without breaking the chain there seems to be no chance. btw I have to learn MEL finally. life without mel stinks more and more..

Quote

# 162 06-01-2006 , 06:48 PM

Registered User
Join Date: Jan 2005

Location: vienna

Posts: 1,095

I´m busying with the builder file from paul thuriot. just recently discerned the power of his awesome cct stuff. as I don´t know much about mel but creating shelfbuttons, I have a real hard time with it. I added some more lines of code to addapt the file for my model which generally has some more joints than his default skeleton has. got the script running half the way down my hierarchy now after some hours, but there´s still something not right with the neckJA_1, but I´ll get it soon - I hope.

Quote

# 163 06-01-2006 , 08:54 PM

Lead Modeler - Framestore
Join Date: Feb 2003

Location: UK

Posts: 6,287

Hey Daniel

Yeah keep at it, its the best thing, even if it drives you nuts, it will all fall into place.

Jay

Quote

# 164 06-01-2006 , 09:33 PM

Registered User
Join Date: Jan 2005

Location: vienna

Posts: 1,095

got it!

start enveloping again now.

Quote

# 165 07-01-2006 , 01:07 AM

Registered User
Join Date: Jan 2005

Location: vienna

Posts: 1,095

is there any technical edge to enveloping the hiRes geometry than painting weights on the lowRes and creating a wrap deformer out of it for driving the hiRes version? I´m asking because Alex Alvarez in his skinning tut does it that way, which besidly said was a tremendous step forward for me. Mr. Thuriot does it the opposite way, enveloping the hiResversion and is just leaving the loRes geo for the animators. he is even pointing on enveloping the hiRes geo...

k, I´ve just given myself the answer. stupid me. it just doesn´t matter since the animation data can be imported from lowRes to hiRes.

I think I´m really inching towards the global contexts. maya is fun again.

good night!

Quote

Posting Rules Forum Rules

Similar Threads

Topics

Free Courses

Full Courses

VFX News