POV-Ray : Newsgroups : povray.binaries.images : new contest (66 kb) Server Time
9 Aug 2024 13:23:20 EDT (-0400)
  new contest (66 kb) (Message 11 to 14 of 14)  
<<< Previous 10 Messages Goto Initial 10 Messages
From: Jeremy M  Praay
Subject: Re: new contest (66 kb)
Date: 26 Feb 2005 00:33:03
Message: <42200a0f@news.povray.org>
"Jim Charter" <jrc### [at] msncom> wrote in message 
news:421fceb3$1@news.povray.org...
>
> Maybe that's why it looks like an overbaked brownie.  It's trying to tell 
> me something.
>

I knew I never should have made the brownie reference.  :-(

Glad to see your returning to pure CSG, and better yet, to iso-surfaces and 
leaving behind your evil meshes.  (Totally joking.  I have very dry humor, I 
know...)  ;-)

I'll attempt to be more serious now.  It looks quite a bit better than your 
previous versions.  The streaks down the sides of the "hump" add quite a bit 
of detail to it.  Other than perhaps varying the texture (colors/pigment) a 
bit, I can't think of anything which would add more detail to it.  You've 
nailed the surface quite well.  However, those brighter spots on the leg in 
shadow don't look right to me.

Some ideas, however:
HDRI?
Maybe 3-point lighting?

Well, those are things that I've been playing around with lately, and I've 
had varying degrees of success.  I think 3-point (studio) lighting might 
really work well for this, if you are interested in showcasing the boli.  It 
seems to look best when you have interestingly curved surfaces like that. 
Then you can get that render time closer to 71 hours... ;-)

-- 
Jeremy from his Michigan cabana
www.beantoad.com


Post a reply to this message

From: Jim Charter
Subject: Re: new contest (66 kb)
Date: 26 Feb 2005 10:34:23
Message: <422096ff$1@news.povray.org>
Jeremy M. Praay wrote:
> "Jim Charter" <jrc### [at] msncom> wrote in message 
> news:421fceb3$1@news.povray.org...
> 
>>Maybe that's why it looks like an overbaked brownie.  It's trying to tell 
>>me something.
>>
> 
> 
> I knew I never should have made the brownie reference.  :-(

It was perfect.  Maybe too perfect. LOL.

> 
> Glad to see your returning to pure CSG, and better yet, to iso-surfaces and 
> leaving behind your evil meshes.  

It's the nighttime what's the worst.  Theys ain't like nuttin' human. 
...you start seein' 'em everywhere,...like theys followin' ya...likes 
theys come ta get cha. You believe me, don't ya pal,...well don't 
ya??????????????


(Totally joking.  I have very dry humor, I
> know...)  ;-)

You like to wear hose, I know.

> 
> I'll attempt to be more serious now.  

Oh gawd.

It looks quite a bit better than your
> previous versions.  The streaks down the sides of the "hump" add quite a bit 
> of detail to it.  Other than perhaps varying the texture (colors/pigment) a 
> bit, I can't think of anything which would add more detail to it.  You've 
> nailed the surface quite well.  However, those brighter spots on the leg in 
> shadow don't look right to me.

Seriously, thanks.  In part it might be do to the EBlob functions in 
iso_csg.  They seem to give smoother transitions than blob primitives no 
matter how you set the threshold.  The texture is pretty much identical 
to the one I posted earlier.  I would like to also try something derived 
from granite or wrinkles.  Interestingly, combining the IC_Diplacement 
with IC_EBlob, you seem to only get results within a narrow band of 
thresholds or something. I sometimes mess with the pigment maps on the 
underlaying functions for instance, like maybe sometimes only a map like 
[0 rgb .98][1 rgb 1] will give results.  On this one I used [0 rgb .4][1 
rgb .6] for instance.  I really don't understand what is going on.

> 
> Some ideas, however:
> HDRI?

Yes, but that is another whole bog to get through and my first 
encounters with it (shoe models) did not meet the success with it that 
others seem to get.  As I was saying to Shay, the whole 
background/setting problem is significant because it really goes to 
meaning.  I don't think I'll find a good answer soon.  I was trying to 
place it on a table with maybe some foliage behind, but it's a 
meaningless slog for me.  I much more like the effect of photo product 
shots with a color screen behind where the object either hovers in space 
or makes the barest concessions to gravity and support.


> Maybe 3-point lighting?
> 
> Well, those are things that I've been playing around with lately, and I've 
> had varying degrees of success.  I think 3-point (studio) lighting might 
> really work well for this, if you are interested in showcasing the boli.  It 
> seems to look best when you have interestingly curved surfaces like that. 
> Then you can get that render time closer to 71 hours... ;-)
> 

Actually I was using something more like that. 2-points I think. This 
render was not originally intended to be shown so the bg is incomplete 
and the lighting purposely stark.  But when the render just went on and 
on...I don't know, I decided to post anyway.  I will learn more 
self-discipline as I learn to work with these long renders I am sure.


Post a reply to this message

From: Jeremy M  Praay
Subject: Re: new contest (66 kb)
Date: 28 Feb 2005 10:49:18
Message: <42233d7e$1@news.povray.org>
"Jim Charter" <jrc### [at] msncom> wrote in message 
news:422096ff$1@news.povray.org...
>>
>> Some ideas, however:
>> HDRI?
>
> Yes, but that is another whole bog to get through and my first encounters 
> with it (shoe models) did not meet the success with it that others seem to 
> get.  As I was saying to Shay, the whole background/setting problem is 
> significant because it really goes to meaning.  I don't think I'll find a 
> good answer soon.  I was trying to place it on a table with maybe some 
> foliage behind, but it's a meaningless slog for me.  I much more like the 
> effect of photo product shots with a color screen behind where the object 
> either hovers in space or makes the barest concessions to gravity and 
> support.
>

Ahhh... That makes sense.  I wasn't exactly sure where you were headed.

HDRI is simple to get good results when dealing with reflective objects, 
like glass and metal, but using it to light an object where you're used to 
specular highlights (or no highlights) can be difficult.  I tried my oval 
box in a HDRI scene and wasn't entirely happy, though it's not too bad 
(http://www.beantoad.com/newimages/ovalbox-3.jpg).  I'm sure I could get it 
looking better if I spent the time, but the "kitchen probe" isn't exactly a 
lesson in great lighting either, hence the 3-point lighting experiments 
(actually 2 points, plus rad).  And of course, I'm still trying to figure 
out how to use LightSysIV effectively.

>
>> Maybe 3-point lighting?
>>
>
> Actually I was using something more like that. 2-points I think. This 
> render was not originally intended to be shown so the bg is incomplete and 
> the lighting purposely stark.  But when the render just went on and on...I 
> don't know, I decided to post anyway.  I will learn more self-discipline 
> as I learn to work with these long renders I am sure.

Like I said, most of my comments come from what I've been learning myself. 
Suddenly, certain things stand-out to me that I never noticed before.  I'm 
often hesistant to make suggestions due to "perhaps I'm stating the 
obvious."

And never forget my rule of thumb: If it's not under 10pps, then you've done 
something wrong... ;-)

-- 
Jeremy
www.beantoad.com


Post a reply to this message

From: Jim Charter
Subject: Re: new contest (66 kb)
Date: 28 Feb 2005 11:30:36
Message: <4223472c$1@news.povray.org>
Jeremy M. Praay wrote:


> 
> HDRI is simple to get good results when dealing with reflective objects, 
> like glass and metal, but using it to light an object where you're used to 
> specular highlights (or no highlights) can be difficult.  

Yes that was exactly the situation with my leather texture.  Jaime 
pretty much told me flat out that HDRI was the wrong tack to take.  I 
figure if you get Jaime commenting on your lighting, the thing to do is 
shut up and listen.

But further, I also have philosophical reservations with HDRI in terms 
of the making of meaning.


Post a reply to this message

<<< Previous 10 Messages Goto Initial 10 Messages

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