Spline bias question in modeling

[someawfulbridge]

I might be jumping the gun on this question and will find the answer soon--

 

I'm working through the tutorials in the David Rogers book (BEST BOOK EVER, based on the first 1/3, at least--the tutorials are very instructive and logically arranged) and I just modeled my first realistically-muscled appendage (a leg).

 

To keep spline density low, the tut focuses on using spline bias manipulators to define the musculature. On the tut, however, the only cps whose bias is manipulated belong to the spline rings going around the leg. Is it standard/advisable, when modeling, to adjust bias on splines running lengthwise, as well? Is there a reason to stick with bias manipulation on only one type, directionally speaking, of spline?

 

Bias manipulators, by the way, give me a funny tickle, so beautiful a thing are they--

 

Thanks!

Mark

[luckbat]

The upside of using bias handles is that they allow you to reduce your spline count even further, while increasing your control over subtle curvatures.

 

The downside: the more custom bias values you set, the more you have to keep track of.

 

But there's no distinction between lengthwise and cross-section handles from a modeling point of view. In fact, on non-limb areas, the distinction gets quite blurry. So, use bias handles wherever you feel that they improve the curves of your model. Just be prepared to remember where you set them all...

[C-grid]

I do hope they fix the unwanted movement of a bias-handles, in realtime display and particularly at rendertime.

 

Niels.

[rusty]

In addition to the above, if you mess with the bias in areas that you intend to animate, this can cause the mesh to jump when it is moved.

 

The downside: the more custom bias values you set, the more you have to keep track of.

 

I'm curious about 'why' you'd have to 'keep track of them' -- outside of general areas where I've tweaked the bias, I don't keep track.

 

Rusty

[robcat2075]

if you mess with the bias in areas that you intend to animate, this can cause the mesh to jump when it is moved.

I do hope they fix the unwanted movement of a bias-handles, in realtime display and particularly at rendertime.

That's not supposed to be true anymore. Can you show examples of this?

[C-grid]

Can you show examples of this?

 

I could, yes.

 

Niels.

 

 

ps.

It ain't part of a character, and this particular prop isn't animated, I did make the bias-handles long and placed weird cause it's ment to be bush/foliage.

[robcat2075]

Can you show examples of this?

I could, yes.

show the example

[C-grid]

I have the model at home, I am at my parents right now and I'm about to play soccer with my nephew.

 

Niels.

 

 

ps.

It ain't unbelieve is it?

[C-grid]

Let's say the animation is 150-300 frames, and a cam is exploring the object, exactly on one frame(181) I see a spline(bias) turn negative where it should have stayed positive or vice-verce, didn't check that.

 

 

Niels.

 

 

[EDIT] I did change the model, to get rid of it, but I assume it's still there in an older backup version.

[luckbat]

I'm curious about 'why' you'd have to 'keep track of them' -- outside of general areas where I've tweaked the bias, I don't keep track.

Well, it's not a hard-and-fast rule, but if you forget which CPs you've modified, and you later make any significant alterations to your model, you may find it more difficult to wrestle with your splines. Since some of your CPs will act like default CPs, while other ones will retain their custom bias.

 

Not a big deal, just something to look out for...

[rusty]

if you mess with the bias in areas that you intend to animate, this can cause the mesh to jump when it is moved.

I do hope they fix the unwanted movement of a bias-handles, in realtime display and particularly at rendertime.

That's not supposed to be true anymore. Can you show examples of this?

 

Dog gone! I can't make it happen (spline jump). I guess they did fix it! And some how I missed the news. I'm afraid I've formed the habit of avoiding this for so long that I may not ever break the habit LOL!

 

Niels' "

...the unwanted movement...

? I've never seen it (probaly because I hardly ever tweak bias).

 

Rusty

[robcat2075]

well, i guess I can still make it happen, when one CP crosses the line between it's neighboring CPs

 

[attachmentid=17514]

 

But I was sure I saw a Hash person say this was fixed. Maybe it was something else?

[C-grid]

Hello robcat2075,

 

I looked yesterday again and concluded it wasn't the positive/negative problem like I have encountered before.

 

It was exactly on frame 185 that the spline 'moved', I tried to discover what did happen, what I discovered is that the particular spline takes the curve/shape just like the vector-drawmode, only that spline-part, the rest is drawn in curved-drawmode, like I said, I also had this in the final5p render.

 

Niels.

 

 

ps.

I also discovered how Hash can make the rendertime sometimes, half of the (render)time, this will be object/scene depended, but I have no doubt, all will at least somehow benefit from it.