POV-Ray : Newsgroups : povray.binaries.images : Flower study WIP 01 Server Time
15 Nov 2024 19:22:18 EST (-0500)
  Flower study WIP 01 (Message 1 to 10 of 15)  
Goto Latest 10 Messages Next 5 Messages >>>
From: Rune
Subject: Flower study WIP 01
Date: 13 Apr 2006 20:59:38
Message: <443ef3fa@news.povray.org>
Here goes my first post in p.b.i for ages. It's a flower of some kind.

It's going to be a fictional flower. So far it's based a lot on various 
different kinds of lilies, but none that are quite like mine. My goal is to 
make a flower that looks dazzling, not to reproduce an actual existing 
species. Existing species work nicely as inspiration though. The stem and 
leaves of my plant will not be like a lily at all, but probably be inspired 
by completely different plants instead.

I'm rather satisfied with the petals and sepals of the flower. They have the 
shape and texture that I want.

However, I'm not so sure about the stamens (the six long thingies) and the 
pistil (the long center thingy). Somehow they just don't look right to me, 
and don't match the quality that I think the petals and sepals have. I can't 
quite put my finger on what's wrong with them though. (Of course some 
randomness should be introduced in their shapes, but that's to come.)

I don't know if they have the right color. (The right color is whatever 
color will make the flower look best.) On one hand they should be clearly 
visible. Preferably even more than on real lilies, where the stamens and 
pistil often have the same color as the petals. I want them to stand out. On 
the other hand it should look plausible and at the same time not cause a 
color clash. I have tried to use an orange yellow, which seem to work 
moderately nice. I use subsurface scattering (with scattering media) for the 
stamens and pistil, since conventional texturing made them look too solid 
and fake. (I still think they look too fake, but the SS helps a little.)

I'm also definitely not sure about the anthers (the brown thingies at the 
ends of the six stamens). Anthers seem to almost always be some shade of 
brown. I have seem a few images where they are more like red or yellow, but 
I haven't got hi-res images of those, so it's difficult to see the details. 
The anthers are what I like least about my flower. They are made with 
isosurfaces in order to be able to get that fuzzy look, but it just makes 
the lighting on them look horrible. They are in definite need of 
improvement.

All kinds of comments, criticisms and suggestions are appreciated! :)

Rune
-- 
http://runevision.com


Post a reply to this message


Attachments:
Download 'flower_study_01.jpg' (204 KB)

Preview of image 'flower_study_01.jpg'
flower_study_01.jpg


 

From: Rune
Subject: Re: Flower study WIP 01
Date: 13 Apr 2006 21:01:34
Message: <443ef46e$1@news.povray.org>
I forgot to mention: There are some bright pixel artifacts a few places in 
the image. I don't know what is causing them. :(

Rune
-- 
http://runevision.com


Post a reply to this message

From: Jim Charter
Subject: Re: Flower study WIP 01
Date: 13 Apr 2006 22:59:11
Message: <443f0fff$1@news.povray.org>
Excellent results.  I think this is a huge achievement of modelling and 
texturing, expecially the petals!

The pistil did indeed break the illusion for me.  I think is is just too 
regular in shape, the shaft in particular.  It could look that way in 
reality, closeups of organic shapes reveal some amazing realities, but 
this contradicts our expectation a bit too much?  It is, in a way, the 
combination of texture and shape.  The texture I find believable, but 
not with quite so geometric a shape for the shaft.  I think it is the 
old problem.  If this were really a photo we would say, "wow that's 
amazing how precise and geometric the pistil of a flower is.  But 
knowing it is manufactured illusion, we say instead, "hmmm, the pistil 
doesn't look 'right'" My suggestion is to give it a little more taper 
towards the top.  The texturing and irregularity of the tubular 
crossection is good.  Maybe it could be pushed a little more?  You need 
to establish the concept of 'celluous' in the viewer's mind.

The stamens are subtle and beautiful. I think the sss is very effective. 
I think the antlers carry the illusion well enough.  But there is 
probably room for more experimentation there.


Post a reply to this message

From: Stefan Viljoen
Subject: Re: Flower study WIP 01
Date: 14 Apr 2006 02:56:45
Message: <443f47ac@news.povray.org>
Rune spake:

> Here goes my first post in p.b.i for ages. It's a flower of some kind.
> 
> It's going to be a fictional flower. So far it's based a lot on various
> different kinds of lilies, but none that are quite like mine. My goal is

Yikes!

Why oh why do other ppl' WIP images -always- look better than mine?

Boo hoo...

Very nicely done, I think the petal texture is exactly right (looks pixel
for pixel like a flower I saw a while ago in my garden). Love the internal
structure you depicted (called the "stamen"?)

Will this be something you'll release later? Scrumptilicious!

-- 
Stefan Viljoen
Software Support Technician / Programmer
Polar Design Solutions


Post a reply to this message

From: Rune
Subject: Re: Flower study WIP 01
Date: 14 Apr 2006 07:18:26
Message: <443f8502@news.povray.org>
"Jim Charter" <jrc### [at] msncom> wrote in message 
news:443f0fff$1@news.povray.org...
> Excellent results.  I think this is a huge achievement of modelling and 
> texturing, expecially the petals!

Thank you!

> The texturing and irregularity of the tubular crossection is good.  Maybe 
> it could be pushed a little more?  You need to establish the concept of 
> 'celluous' in the viewer's mind.

I'm not sure what you're referring to with tubular cross section here. Is it 
still the pistil, or are you talking about the "tube" that the petals form 
together? The pistil has no irregularity nor any texturing, so...

> The stamens are subtle and beautiful. I think the sss is very effective. I 
> think the antlers carry the illusion well enough.  But there is probably 
> room for more experimentation there.

I'll keep on experimenting...

Rune
-- 
http://runevision.com


Post a reply to this message

From: Rune
Subject: Re: Flower study WIP 01
Date: 14 Apr 2006 07:31:24
Message: <443f880c$1@news.povray.org>
"Stefan Viljoen polard.com>" <spamnot@<removethis> wrote:
> Yikes!
>
> Why oh why do other ppl' WIP images -always- look better than mine?

Hey, I liked The Reality Disfunction - although that wasn't a WIP...

> Will this be something you'll release later? Scrumptilicious!

Probably not the source code for entire image that this flower will be part 
of, but I might release the code for just this flower head eventually.

Rune
-- 
http://runevision.com


Post a reply to this message

From: Bob H
Subject: Re: Flower study WIP 01
Date: 14 Apr 2006 07:34:26
Message: <443f88c2@news.povray.org>
"Rune" <new### [at] runevisioncom> wrote in message 
news:443ef46e$1@news.povray.org...
>I forgot to mention: There are some bright pixel artifacts a few places in 
>the image. I don't know what is causing them. :(

Maybe similiar to a problem when sphere_sweep has a small diameter or short 
bends between vectors?

Flower looks much like those silk and plastic fake ones; which is probably a 
good thing since I'm always confusing the fakes for real flowers, especially 
the ones with plastic(?) dew or raindrops added. But I'm getting better at 
telling the difference.

I'm thinking the petals might be better if translucent. I've got some orange 
tiger lillies here but they haven't bloomed yet or else I'd take a look at 
them to check. Just seems to me that there should be light passing through 
them.


Post a reply to this message

From: Jim Charter
Subject: Re: Flower study WIP 01
Date: 14 Apr 2006 15:42:00
Message: <443ffb08@news.povray.org>
Rune wrote:

> 
> 
> I'm not sure what you're referring to with tubular cross section here. Is it 
> still the pistil, or are you talking about the "tube" that the petals form 
> together? The pistil has no irregularity nor any texturing, so...
> 
I meant the shaft|stalk|tube of the pistil.  Could have sworn that it 
was not a perfect cylinder or cone but rather that if you took the 
cross-section it would be revealed that it is more irregular.  Could 
swear I see some faint shadows or striations running lengthwise which 
could either be texturing or actual form.  But obviously my eyes are 
decieved.  Anyway that only reinforces my original comment.  That the 
straight texture is probably accurate but combined with the plain 
geometry doesn't quite make us think this structure was the result of 
organic growth.


Post a reply to this message

From: Thomas de Groot
Subject: Re: Flower study WIP 01
Date: 15 Apr 2006 03:40:01
Message: <4440a351$1@news.povray.org>
Really excellent work, Rune!
Everything has already been said, and better than I could, so just a smal
comment on the pistil. It seems to me that there should be a tiny
indentation at the top, and maybe very fine radial striation there too? It
should also be less shiny I believe. However, I have no example at hand, so
I write from memory...

Thomas


Post a reply to this message

From: Jim Holsenback
Subject: Re: Flower study WIP 01
Date: 15 Apr 2006 07:15:02
Message: <4440d5b6@news.povray.org>
Here's a photo of a Day Lily growing (last year) in my front yard ....


Post a reply to this message


Attachments:
Download 'Day Lily Red.JPG' (228 KB)

Preview of image 'Day Lily Red.JPG'
Day Lily Red.JPG


 

Goto Latest 10 Messages Next 5 Messages >>>

Copyright 2003-2023 Persistence of Vision Raytracer Pty. Ltd.