[X3D-Public] X3D H-Anim Fun

[Joe D Williams]

> Here's the results I got with FreeWRL.

> I'm going to change the URL reference
to the texture to see if I can get FreeWRL to find it.

Great John, it got the texture, showing dark bands.
Was the head moving at all?

Looks like only emissiveColor (green) is showing, nothing from the 
DirectionalLight(s).
Octaga looked just like yours until i set global TRUE

I just found out the header should be:

#X3D V3.1 utf8
PROFILE Interactive
COMPONENT H-Anim:1

Notice that the lights are setup as:

DirectionalLight { global TRUE }
Group {[
 Figure {}
]}

apparently works in all. except Frewrl.

If changing th header didn't fix, then maybe if the lights were moved 
to be like:

Group {[
 DirectionalLight { global TRUE or FALSE  }
 Figure  {}
] }

Every other f of the four browsers listed before lit the skin when the 
lighs were moved indside the group with the skin no matter what the 
global was set to.


scene at:
www.hypermultimedia.com/x3d/hanim/JoeSkin4EX0.x3dv

Thanks Again and Best Regards,
Joe



----- Original Message ----- 
From: "John Carlson" <john.carlson3 at sbcglobal.net>
To: "Joe D Williams" <joedwil at earthlink.net>
Cc: <x3d-public at web3d.org>
Sent: Saturday, May 09, 2009 9:25 PM
Subject: Re: [X3D-Public] X3D H-Anim Fun


> Here's the results I got with FreeWRL.  Sorry if you can't read TIFF
> (perhaps email will convert it?  I'm going to change the URL 
> reference
> to the texture to see if I can get FreeWRL to find it.
>
> John
>


--------------------------------------------------------------------------------


>
> On May 9, 2009, at 8:03 PM, Joe D Williams wrote:
>
>> So I can't help it Gang, I've just gotta press on this.
>> The X3D H-anim character (attached zip and screen pic) shows this
>> one with the LOA3 skeleton posed pretty much by the spec, then the
>> single mesh skin drawn in that same space, 'working' in four 
>> browsers.
>> More advanced tests coming soon, but this is a simplist meaningful
>> test stage just to show what can happen with the skin in the
>> 'default' skeleton pose using 'default' texture mapping, then move 
>> a
>> joint of two or more, or as it stands, only a neck joint.
>>
>> I this test I am calling BSContact the only one that works
>> completely as I think I expected.
>>
>> Maybe it is fruitless to expect any 'default' texture coordinates 
>> to
>> work right, but BS makes it seem possible. More in future tests, 
>> but
>> if you look closely you can see that both Flux and Instant handle
>> the movement of the head differently, causing other area of the
>> texture to move, where the underlying skin does not. More difficult
>> to see unless running: Octaga does not get the skin connected with
>> joint animation so the skin does not move, regardless of joint
>> animation at this time.
>>
>> If something is wrong, please tell me, but the skin (not 
>> necessarily
>> the texture) is clearly animating without error in three out these
>> four, so Octaga, please fix.
>>
>> "Three out of Four is not that bad," one might say, "and there may
>> be some out there working that haven't been checked, ... ."
>> Fine. Well, it is worse than that. As you can see by its absence
>> from the attached illustration, Xj3D refuses to run the thing at 
>> all.
>> Somejavathing about somejavething running out of somejavathing.
>> Maybe, I hope it is simple and someone will show me.
>>
>> Anyway, to me I thought it was great to show this thing almost
>> working in 3 out the 5 that I am running.
>>
>> But also I was hoping I could skip all the above culling of 
>> failures
>> and get to the fun part.
>> Unzip the file to get JoeSkin4EX0.x3dv file then open it your most
>> simple text editor with line wrap turned off.
>> Now open the file with an X3D Browser and run it.
>> Hopefully the head moves and the texture gets found and wrapped.
>>
>> Scroll down the text file past the skeleton definition and the
>> skin definition into the interpolators and find the First line 
>> with:
>>
>> keyValue [ 0 0 1 0, 0 0 1 0, 0 0 1 0 ] }
>>
>> (right knee joint) and change it to:
>>
>> keyValue [ 0 0 1 0, 1 0 0 1.5, 0 0 1 0 ] }
>>
>> Save the file then run it. How much more fun could you possibly be
>> having?
>> Not enough? Try this one:
>>
>> keyValue [ 0 0 1 0, 1 1 1 1.5, 0 0 1 0 ] }
>>
>> Each set of numbers (1 0 0 1.5, for example) are,
>> from the joint's point of view, (gaze +z, +y up, +x left),
>> really simple pitch and yaw and roll and rads.
>>
>> For starters, think of using the first three numbers (p,y,r) 
>> ranging
>> from -1 to 1 and the fourth number (radians) ranging from -2pi to 
>> 2pi.
>> When more than one of the first three are not zero, the rotations
>> sort of blend together.
>>
>> If it gets too far out of line, remember you can tame it down by
>> returning to the default initial joint rotation values of:
>>
>> keyValue [ 0 0 1 0, 0 0 1 0, 0 0 1 0 ] }
>>
>> No it is not wasted time experimenting to animate this particular
>> right knne.
>> Really, if the character follows the H-Anim standard, then every H-
>> Anim right knee will work almost exactly like this one.
>>
>> When you figure out how this right knee works, then, by all means,
>> try another joint.
>>
>> More later, and Best Regards,
>> Joe
>> <
>> JoeSkin4EX0
>> .zip><BSIOF256.jpg>_______________________________________________
>> X3D-Public mailing list
>> X3D-Public at web3d.org
>> http://web3d.org/mailman/listinfo/x3d-public_web3d.org
>
>