PDA

View Full Version : Annoying Gradient bug



ingo
02-06-2004, 03:38 AM
Its strange that we can't use negative numbers for gradients, but today i realized that gradients can't even go below zero on the Y-axis, weird. I have an object that is only 15 cm's minus on the y-axis and i want a gradient to start at that point, but no matter if i use "y distance to pivot point" or "y distance to object" to an object which lies at -15 cm's, the gradient always starts at zero. I hope we can see a simple fix in LW 8............

Exper
02-06-2004, 04:06 AM
It works for me!

I replicated your scene:
add a Null,
place it where you need,
use "y distance to object" referring to the Null (LW always take in account the Pivot Point if you refer to another object)

Take a look at this shot:

ingo
02-06-2004, 04:27 AM
Hmm, strange. But thanks, works here too with a null. So i have to remove the pivot point to get it done the simple way. I want to make it to use the pivot point of the own object. Still i like it more to use negative values.

Exper
02-06-2004, 04:40 AM
When you use "y distance to pivot point" then negative values are mirrored from the positive ones.

"y distance to object" always use as origin the Pivot Point of the referring object then negative values are mirrored from the positive ones.

I've already asked for negative values...
take a look to the Thread I'm mantaining about Gradient features:
Gradient Input
http://vbulletin.newtek.com/showthread.php?s=&threadid=11098
and add your own comments! ;)

Bye.

ingo
02-06-2004, 05:40 AM
Thanks for your help, as expected it only works with a null, moving the pivot point of the object doesn't made a difference.

Exper
02-06-2004, 07:44 AM
Originally posted by ingo
moving the pivot point of the object doesn't made a difference.Yep... never tried this one cos I always use a Null!

I'll add this one to the "Gradient Input" Thread! ;)

Bye.

P.S. sent to [email protected]

Exper
07-13-2004, 08:04 AM
Bug still there in LW[8]! :(

Shloime
07-13-2004, 08:40 AM
Hi.

I don't really think it is a bug, it is more of a physical thing. Although physics don't really apply in 3D, it seems for me to be impossible for something to have a "negative" distance. See it like this: distance to (something) is somehow a radial thing, the farther away you are the more distance you have, no matter which direction. "y distance to (something)" just limits that radial distance to one axis.
So the approach with an extra null seems quite legal and wanted, IMHO.
Hope I was understandable,

Exper
07-13-2004, 09:06 AM
Shloime...
all Distance Gradients are mirrored:
+ values are mirrored to - values.

When you use "Distance to Pivot Point" it takes in account the Pivot; it become the origin for Gradient trasmission upon the surface!

All ok if you don't move the Pivot.

But if you move the Pivot (in Layout) then the Gradient doesn't change at all!

I normally use Nulls but sometimes using the Pivot is better (and faster to set up)

Take a look at the picture

toby
07-13-2004, 06:18 PM
Moving the pivot in Layout only affects animation and doesn't change the model, just like skelegons. Try moving the pivot point in Modeler

Shloime
07-14-2004, 12:58 AM
@ Exper


all Distance Gradients are mirrored:
+ values are mirrored to - values.

Well, that's what I said, in other words... :rolleyes:

Exper
07-14-2004, 03:28 AM
Deuce is aware of this bug now... he sent me an email confirming the problem!

Now we'll wait for the fix! :cool:

At the same time I'll ask for new Gradients related to "Modeler Pivot" and "Layout Pivot" resolving all kind of behaviours!

Take a look:
Gradient Input
http://vbulletin.newtek.com/showthread.php?s=&threadid=11098
Message:
http://vbulletin.newtek.com/showthread.php?s=&postid=195532#post195532

@ Toby: we're not speaking about animation/deformation but simply surfacing!

@ Shloime: yeah... sorry for the misunderstanding! ;)

toby
07-14-2004, 02:27 PM
@ Toby: we're not speaking about animation/deformation but simply surfacing!

I know - what I was saying is that moving the pivot point in Layout has no effect the model, only animation ~

Exper
07-15-2004, 02:47 AM
To resolve all "Gradient - Distance to Pivot Point" behaviours I proposed (http://vbulletin.newtek.com/showthread.php?s=&threadid=11098) these modifications:

X Distance to Pivot Point
Y Distance to Pivot Point
Z Distance to Pivot Point
referring to Pivot in Layout (modifiable Pivot)

X Modeler Distance to Pivot Point
Y Modeler Distance to Pivot Point
Z Modeler Distance to Pivot Point
referring to the absolute Modeler Pivot.

stib
07-17-2004, 08:20 AM
and how about
distance to pivot point X axis
distance to pivot point Y axis
distance to pivot point Z axis

this would create a gradient that was cylindrical around the specified axis. see the pic

You could also have UV gradients, while you're there.

Exper
07-19-2004, 05:26 AM
Originally posted by stib
and how about
distance to pivot point X axis
distance to pivot point Y axis
distance to pivot point Z axis

this would create a gradient that was cylindrical around the specified axis.Interesting... I'll add them to "Gradient Input" (http://vbulletin.newtek.com/showthread.php?s=&threadid=11098) Thread!