POV-Ray : Newsgroups : povray.binaries.images : Irid vs SSLT redux (with less assumed gamma) Server Time
7 Nov 2024 11:33:47 EST (-0500)
  Irid vs SSLT redux (with less assumed gamma) (Message 1 to 10 of 17)  
Goto Latest 10 Messages Next 7 Messages >>>
From: Simon J  Cambridge
Subject: Irid vs SSLT redux (with less assumed gamma)
Date: 2 May 2016 08:45:01
Message: <web.57274aabda101431acfa7bc90@news.povray.org>
Since I have been told that I am a very naughty boy and must not use
assumed_gamma 2.2 nor misuse irid in any way shape or form, I have reworked one
of the images in accordance with this new theology.

Points of note. The meshes need severe sub-division, the more the merrier. I
also ran the gamut of 20 to 50 mm_per_unit for SSLT and settled on 40. Irid is
really good at highlighting the normals which SSLT tends to hide.

But ... is this necessarily a good thing?

Simon.

PS. I do have an excuse for 2.2, though! CRTs! And POV-Ray versions 1, 2, 3,
3.5, 3.6, 3.62......... And CRT's! Do NOT forget CRT's!


Post a reply to this message


Attachments:
Download 'ssltvir01.png' (1226 KB)

Preview of image 'ssltvir01.png'
ssltvir01.png


 

From: Simon J  Cambridge
Subject: Re: Irid vs SSLT redux (with less assumed gamma)
Date: 2 May 2016 08:55:01
Message: <web.57274da1aca34b78acfa7bc90@news.povray.org>
"Simon J. Cambridge" <nomail@nomail> wrote:
> Since I have been told that I am a very naughty boy and must not use
> assumed_gamma 2.2 nor misuse irid in any way shape or form, I have reworked one
> of the images in accordance with this new theology.
>
> Points of note. The meshes need severe sub-division, the more the merrier. I
> also ran the gamut of 20 to 50 mm_per_unit for SSLT and settled on 40. Irid is
> really good at highlighting the normals which SSLT tends to hide.
>
> But ... is this necessarily a good thing?
>
> Simon.
>
> PS. I do have an excuse for 2.2, though! CRTs! And POV-Ray versions 1, 2, 3,
> 3.5, 3.6, 3.62......... And CRT's! Do NOT forget CRT's!

And the second image continuing the theme.

I also changed the lighting so that you can see SSLT occurring in one of the
ears. It is probably manadatory to have SSLT occuring in at least one ear.


Post a reply to this message


Attachments:
Download 'ssltvir02.png' (1159 KB)

Preview of image 'ssltvir02.png'
ssltvir02.png


 

From: clipka
Subject: Re: Irid vs SSLT redux (with less assumed gamma)
Date: 2 May 2016 09:20:46
Message: <5727542e$1@news.povray.org>
Am 02.05.2016 um 14:40 schrieb Simon J. Cambridge:

> Points of note. The meshes need severe sub-division, the more the merrier. I
> also ran the gamut of 20 to 50 mm_per_unit for SSLT and settled on 40. Irid is
> really good at highlighting the normals which SSLT tends to hide.

The current SSLT implementation is indeed a bit lax when it comes to
normals. It does respect the normal at the point the light comes out,
but not yet the one at the point the light comes in.

Maybe I need to take a time out to work on SSLT once more.


> PS. I do have an excuse for 2.2, though! CRTs! And POV-Ray versions 1, 2, 3,
> 3.5, 3.6, 3.62......... And CRT's! Do NOT forget CRT's!

*Sigh!* Naughty boys always think they have an excuse...

Nope, you don't.
You absolutely, positively don't.
Not with 3.7 anymore.

The Display_Gamma INI setting takes care of your CRT as far as the
preview display is concerned.

POV-Ray 3.6 already supported the "assumed_gamma" statement.
Half-heartedly, but it did.

The File_Gamma INI setting (or your image viewer if you use .png, .hdr
or .exr output) takes care of your CRT as far as viewing the finished
image is conerned.

And the "srgb" keyword and automatic input file gamma detection (or
"gamma" override keyword) take care of your CRT's influence as far as
your colour picker and other programs in your scene creation toolchain
are concerned.

Not to mention that gamma is not a CRT-specific thing -- LCDs have it,
too. (Not because of the underlying technology, but for backward
compatibility with -- you guessed it -- CRT-based displays.)

And if you don't like how assumed_gamma 1.0 affects your brightness
gradients, POV-Ray 3.7.1 has the "blend_type" and "blend_gamma" keywords
in colour and pigment maps for you.

And if you don't like how midtones look in your final render, the final
POV-Ray 3.7.1 will give you full artistic tonemapping.

So what did you say? You have an _excuse_?

I'll give you a few moments to think about and rephrase that sentence... :P

Heck, even Warp hasn't said a word in ages; if _he_ has run out of
excuses, there definitely aren't any left. ;)


Post a reply to this message

From: clipka
Subject: Re: Irid vs SSLT redux (with less assumed gamma)
Date: 2 May 2016 09:33:18
Message: <5727571e$1@news.povray.org>
Am 02.05.2016 um 14:52 schrieb Simon J. Cambridge:

> I also changed the lighting so that you can see SSLT occurring in one of the
> ears. It is probably manadatory to have SSLT occuring in at least one ear.

Absolutely. For elves, that is. at least for elves. Orcs, not so much.
They usually wear it in their left ear though. Your hero is standing in
front of a mirror, right?

As far as the shoulderwear is concerned: Well, what do you expect? Looks
to me like the thingumajigs receive light, with shadows from objects
further to the right that happen to almost coincide with the tips.
Nothing surprising to me there.


Post a reply to this message

From: clipka
Subject: Re: Irid vs SSLT redux (with less assumed gamma)
Date: 2 May 2016 09:41:48
Message: <5727591c$1@news.povray.org>
Oh, and one more thing...

Am 02.05.2016 um 14:40 schrieb Simon J. Cambridge:
> Points of note. The meshes need severe sub-division, the more the merrier. I
> also ran the gamut of 20 to 50 mm_per_unit for SSLT and settled on 40.

The mm_per_unit setting is not up for debate. Whatever distance in mm
one POV-Ray unit of distance supposedly corresponds to, that's the exact
value to plug in there.

So suppose your elf is 190 units tall, then mm_per_unit must be set to
10, because we all know eleves stand 1.90 m tall. (Unless of course it's
an elf, in which case mm_per_unit must be set to 1, because we all know
elfs are 19 cm in height).

All tweaking is to be done on the translucency parameter.


Post a reply to this message

From: Simon J  Cambridge
Subject: Re: Irid vs SSLT redux (with less assumed gamma)
Date: 2 May 2016 10:50:01
Message: <web.57276880aca34b7863a98df50@news.povray.org>
clipka <ano### [at] anonymousorg> wrote:
> Am 02.05.2016 um 14:52 schrieb Simon J. Cambridge:
>
> > I also changed the lighting so that you can see SSLT occurring in one of the
> > ears. It is probably manadatory to have SSLT occuring in at least one ear.
>
> Absolutely. For elves, that is. at least for elves. Orcs, not so much.
> They usually wear it in their left ear though. Your hero is standing in
> front of a mirror, right?
>
> As far as the shoulderwear is concerned: Well, what do you expect? Looks
> to me like the thingumajigs receive light, with shadows from objects
> further to the right that happen to almost coincide with the tips.
> Nothing surprising to me there.

Actually, the being depicted is not an Elf! (And in making such a rash and
unwarranted assumption you have seriously dented your credibility in the process
- you need to read my books to find out more.)*

The 2.2 thing was an assumption on my part - me being lazy. I am just glad to
find out where I have been going wrong all this time. I will also recalibrate
for both mm_per_unit and translucency and see where I get, but I might still
have to mess with the numbers as the being depicted is roughly ten feet tall.
(You really need to read my books to find out more)**

So here is the question (throwing it out to all and sundry). What do you think
of the texturing? Does it work?

(And the anomaly on the shoulder is a bit odd in that there is no light on that
side and it seems to me that light is coming from behind, through the armour,
which it shouldn't because the texture has no translucency. I have yet to work
out what is going on there.)

Cheers,

Simon.

*An entirely innocent and fortuitous plug.

**Another entirely innocent and fortuitous plug.

www.landofthefirst.com


Post a reply to this message

From: clipka
Subject: Re: Irid vs SSLT redux (with less assumed gamma)
Date: 2 May 2016 11:20:52
Message: <57277054$1@news.povray.org>
Am 02.05.2016 um 16:47 schrieb Simon J. Cambridge:

> Actually, the being depicted is not an Elf! (And in making such a rash and
> unwarranted assumption you have seriously dented your credibility in the process
> - you need to read my books to find out more.)*

What -- and delay POV-Ray development? No way! ;)

(To be honest, I guess I'm not a book person anymore. I just don't find
the time, opportunity and patience these days.)

> have to mess with the numbers as the being depicted is roughly ten feet tall.
> (You really need to read my books to find out more)**

Must be a troll then.
A seriously skinny one. You need to feed it better. :P

random association: https://youtu.be/cITjfIMXN0g


Post a reply to this message

From: Stephen
Subject: Re: Irid vs SSLT redux (with less assumed gamma)
Date: 2 May 2016 11:33:02
Message: <5727732e$1@news.povray.org>
On 5/2/2016 4:20 PM, clipka wrote:
> (To be honest, I guess I'm not a book person anymore. I just don't find
> the time, opportunity and patience these days.)

Try audio books for a change.

-- 

Regards
     Stephen


Post a reply to this message

From: Bald Eagle
Subject: Re: Irid vs SSLT redux (with less assumed gamma)
Date: 2 May 2016 12:45:00
Message: <web.57278369aca34b78b488d9aa0@news.povray.org>
"Simon J. Cambridge" <nomail@nomail> wrote:

> (And the anomaly on the shoulder is a bit odd in that there is no light on that
> side and it seems to me that light is coming from behind, through the armour,
> which it shouldn't because the texture has no translucency. I have yet to work
> out what is going on there.)

Maybe try placing a box object there to shield the shoulder from any light.
Use the no_image, so that you can still "see through it" and see the shoulder,
but it will still cast a shadow, if light is indeed somehow coming from the
front.

(Some sort of tool to trace out light rays in the scene, with a setting so that
it would show x% of the rays would be neat, eh? )


Post a reply to this message

From: Alain
Subject: Re: Irid vs SSLT redux (with less assumed gamma)
Date: 2 May 2016 20:09:21
Message: <5727ec31@news.povray.org>

> clipka <ano### [at] anonymousorg> wrote:
>> Am 02.05.2016 um 14:52 schrieb Simon J. Cambridge:
>>
>>> I also changed the lighting so that you can see SSLT occurring in one of the
>>> ears. It is probably manadatory to have SSLT occuring in at least one ear.
>>
>> Absolutely. For elves, that is. at least for elves. Orcs, not so much.
>> They usually wear it in their left ear though. Your hero is standing in
>> front of a mirror, right?
>>
>> As far as the shoulderwear is concerned: Well, what do you expect? Looks
>> to me like the thingumajigs receive light, with shadows from objects
>> further to the right that happen to almost coincide with the tips.
>> Nothing surprising to me there.

With a mesh, you sometimes see shadowing aretefacts where areas that 
look as if they should be illuminated been actualy unlit. The reverse 
may, albey more rarely, appen : Areas that "should" be unlit may actualy 
receive light.

Try scalling your model up by 10, and reduce mm_per_unit to 4 to 
compensate, and see what appen. It may be due to some floating point 
rounding errors.

>
> Actually, the being depicted is not an Elf! (And in making such a rash and
> unwarranted assumption you have seriously dented your credibility in the process
> - you need to read my books to find out more.)*

Depending on the source, an Elf may be as short as about 2' or less, all 
the way to about 12' tall.

>
> The 2.2 thing was an assumption on my part - me being lazy. I am just glad to
> find out where I have been going wrong all this time. I will also recalibrate
> for both mm_per_unit and translucency and see where I get, but I might still
> have to mess with the numbers as the being depicted is roughly ten feet tall.
> (You really need to read my books to find out more)**

Search a bit and you may find Elfes in the 10-12' range...

>
> So here is the question (throwing it out to all and sundry). What do you think
> of the texturing? Does it work?
>
> (And the anomaly on the shoulder is a bit odd in that there is no light on that
> side and it seems to me that light is coming from behind, through the armour,
> which it shouldn't because the texture has no translucency. I have yet to work
> out what is going on there.)
>
> Cheers,
>
> Simon.
>
> *An entirely innocent and fortuitous plug.
>
> **Another entirely innocent and fortuitous plug.
>
> www.landofthefirst.com
>
>
>
>


Post a reply to this message

Goto Latest 10 Messages Next 7 Messages >>>

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