On 07/07/2018 19:09, dick balaska wrote:
> On 07/07/2018 01:48 PM, Stephen wrote:
> 
>>
>> And here is one I did earlier for a similar occasion. ;)
>>
>> https://i.imgur.com/JXVnX.jpg
> 
> I think you need to bump up the frame rate on that one.
> 
> 

f(frame rate) → ∞ as number of frames → 0

I can't do better than that. :)

-- 

Regards
     Stephen

Post a reply to this message

hi,

"Bald Eagle" <cre### [at] netscapenet> wrote:
> So, I finally had a chance to dabble with this a bit.
> See what you think, and if this is going in the right direction.

nice.  the colour changes are good, difficult to tell whether change is solely
due to crossing a stave, even having split the mp4 into frames (I think they're
not).

> I've got a media "staff" with some moving sphere "notes".
> The media is colored according to the x,y,z values of the bozo pattern, and the
> hue is detrmined by a set of functions that convert a value between 0-1 into a
> set of rgb functions (my set of functions to do the equivalent of the CH2RGB
> macro in colors.inc)

never used 'bozo'.  have read the docs and still find it difficult to relate to
what we're seeing.  there's also a remark about a default colour_map which only
confused things (for me) further.

in a previous post you write "..define different functions based upon the
"space" the notes are traveling through".

that sounds a good idea.  how would that work?  have an object, which has the
user_defined {} construct and is not visible, travel across various function
"spaces"?  can this be discussed without descending into maths??  :-)


thanks + regards, jr.

Post a reply to this message

"jr" <cre### [at] gmailcom> wrote:

> nice.  the colour changes are good, difficult to tell whether change is solely
> due to crossing a stave, even having split the mp4 into frames (I think they're
> not).

They're not. The color changes are solely dependent upon the pattern defined in
the media functions.
Bozo is a sort of smoothly changing noise function, AFAIK.
http://www.f-lohmueller.de/pov_tut/tex/tex_755e.htm
I tried playing around with other functions, and I saw that a real problem with
trying to change colors using RGb is that you get a lot of --- washed out
gray/cyan type colors, which is why I wanted full saturation HSV


> never used 'bozo'.  have read the docs and still find it difficult to relate to
> what we're seeing.  there's also a remark about a default colour_map which only
> confused things (for me) further.

Use POV-Ray's drop-down insert menu, go to Patterns1 and then select bozo
I rendered a test box with the default bozo pigment (attached)

> in a previous post you write "..define different functions based upon the
> "space" the notes are traveling through".
>
> that sounds a good idea.  how would that work?  have an object, which has the
> user_defined {} construct and is not visible, travel across various function
> "spaces"?  can this be discussed without descending into maths??  :-)

Yes, which is exactly what is going on in the animation - though as you point
out, the execution kinda sucks.

I'll render another quick animation to see if I can better illustrate it.

Post a reply to this message

Attachments:
Download 'bozoexample.png' (40 KB)

Preview of image 'bozoexample.png'

So, it was good that I revisited this, as there were some corrections I needed
to do, and the bozo pattern was pretty bad at showing what I wanted.

Here's an animation of a rolling sphere - the sphere travels through a long box,
and the coloration of the intersection of the sphere and box is determined by
the r, g, and b values calculated from the values of the agate pattern at each
point and converted to a [correct] HSV spectrum.

I hope you like it, and if you provide more of a description of what you're
envisioning, that would make implementing it much more straightforward.  :)

(this is at 15 frames / sec - I can do it at 5 as well, but this looks much
smoother)

Back to work tomorrow   :(   but the heat has broken (for now) and hopefully the
rest of the week goes smoothly.
Hope everything is going well for you IRL.

Post a reply to this message

Attachments:
Download 'agatesphere.mp4.mpg' (465 KB)

Here's the static box that the sphere travels through to show the full agate
pattern that the sphere translates through.

I hope that visually explains it a lot better.



I need to look back at what I did with crackle and wood to see how to define the
color map for some of the patterns, since I think I need to do the same thing
with bozo since it has a default color map.

Post a reply to this message

Attachments:
Download 'movingmedia.png' (79 KB)

Preview of image 'movingmedia.png'

On 09/07/2018 00:40, Bald Eagle wrote:
> 
> So, it was good that I revisited this, as there were some corrections I needed
> to do, and the bozo pattern was pretty bad at showing what I wanted.
> 
> Here's an animation of a rolling sphere - the sphere travels through a long box,
> and the coloration of the intersection of the sphere and box is determined by
> the r, g, and b values calculated from the values of the agate pattern at each
> point and converted to a [correct] HSV spectrum.
> 
> I hope you like it, and if you provide more of a description of what you're
> envisioning, that would make implementing it much more straightforward.  :)
> 
> (this is at 15 frames / sec - I can do it at 5 as well, but this looks much
> smoother)
> 
> Back to work tomorrow   :(   but the heat has broken (for now) and hopefully the
> rest of the week goes smoothly.
> Hope everything is going well for you IRL.
> 

Not a criticism but an observation.
What the animation looks like to me. Is that the sphere's media is in 
"world space*" as opposed to "object space". There is no indication of 
the sphere rolling.

My vision of what is required would be better shown using regular 
patterns rather than random-ish ones.
Consider a sphere with a media based on a spiral density pattern rolling 
through the box with a horizontal colour gradient. The resultant image 
would be additive or subtractive.
My best guess from your description. It looks like you are doing it the 
hard way. :)


*
"world space" & "object space" were terms used by Moray. In object space 
the material moves with the object.


-- 

Regards
     Stephen

Post a reply to this message

hi,

"Bald Eagle" <cre### [at] netscapenet> wrote:
> So, it was good that I revisited this, as there were some corrections I needed
> to do, and the bozo pattern was pretty bad at showing what I wanted.
> Here's an animation of a rolling sphere - the sphere travels through a long box,
> and the coloration of the intersection of the sphere and box is determined by
> the r, g, and b values calculated from the values of the agate pattern at each
> point and converted to a [correct] HSV spectrum.
> I hope you like it,

oh, very much.  it looks very attractive.  it is an entirely different process
than the one I've in mind, the result however .. very nice.

> and if you provide more of a description of what you're
> envisioning, that would make implementing it much more straightforward.  :)

the planned animation (notes travelling along staves, revealing hidden object as
they move "across") has been somewhat on hold these past few days because I'm
working to get 'df3 tools 0.4.0' ready for release.  priorities :-)

I'm hoping to get back to "playing" before this week is out.

> (this is at 15 frames / sec - I can do it at 5 as well, but this looks much
> smoother)
> Back to work tomorrow   :(   but the heat has broken (for now) and hopefully the
> rest of the week goes smoothly.
> Hope everything is going well for you IRL.

snowed under?!  :-)

anyway, thank you also for the image showing the whole path of the sphere --
useful.


regards, jr.

Post a reply to this message

hi,

Stephen <mca### [at] aolcom> wrote:
> ... an observation.
> Consider a sphere with a media based on a spiral density pattern rolling
> through the box with a horizontal colour gradient. The resultant image
> would be additive or subtractive.

that too sounds v interesting.  I had not thought about patterning functions at
all.  could you post an outline of the objects and the materials, to help me try
this?  thanks.


regards, jr.

Post a reply to this message

On 09/07/2018 11:07, jr wrote:
> hi,
> 
> Stephen <mca### [at] aolcom> wrote:
>> ... an observation.
>> Consider a sphere with a media based on a spiral density pattern rolling
>> through the box with a horizontal colour gradient. The resultant image
>> would be additive or subtractive.
> 
> that too sounds v interesting.  I had not thought about patterning functions at
> all.  could you post an outline of the objects and the materials, to help me try
> this?  thanks.
> 
> 
> regards, jr.
> 

Just a thought experiment. Like most things the mind's eye is better 
than reality. More thought needed.

-- 

Regards
     Stephen

Post a reply to this message

OK, so as far as I understand what you want to do:

There will be static, stationary visible staves.
There will moving, visible notes.
There will be a stationary invisible object.

The visible notes will change color when they intersect the invisible object.

optional:

Portions of the staves will not be visible when a note overlaps them.


I think I can get that work - I scribbled it out during lunch.  :)


Stephen - yes, the ball does not "roll", it translates, but even if the ball
rolled, it would look the same, since as you say, it's traveling through the
pattern, rather than the pattern being attached to it.

The media method is what jr requested, and Bill P. supplied the prototype scene
code and functions for.

Presumably normal primitives could be used with transmit and filter....

Post a reply to this message