|
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
(1) Greyscale:
The `Greyscale=BOOL` INI setting and the optional `G` in the `+F`
command-line switch are currently documented (only) in the SDL reference
section for the obsolete `hf_gray_16` global scene setting.
I would suggest moving the documentation of these new-in-v3.7 features
to the file output options reference section, and essentially trim down
the `hf_gray_16` documentation to a reference thence.
(2) Changes to the PGM ("greyscale PPM") format:
Previously, the "greyscale PPM" (technically PGM) output file format
always used linear output. In v3.8, this will be changed to match that
of the regular (non-greyscale) PPM format (i.e. `File_Gamma` will be
honored if specified, otherwise the ITU-R BT.709 gamma transfer function
will be applied).
(3) Changes to the bit depth:
Previously, the bit depth (as specified by either the `n` in `+Fxn` or
by the `Bits_Per_Color` setting) was limited to the range 5..16; in
v3.8, this will be changed to allow values as low as 1.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Also:
(4)
In v3.8, enabling greyscale output will no longer force the bit depth to
16 bits per channel.
(5)
In v3.8, the `+F` option will allow specifying both the `G` greyscale
flag and the bit depth, using the format `+FxGn`.
(6)
In v3.8, when creating a greyscale image, the preview display _will_
show the image in greyscale.
Post a reply to this message
|
|
| |
| |
|
|
From: clipka
Subject: Re: Some stuff related to output file formats
Date: 30 Sep 2018 17:43:05
Message: <5bb14369@news.povray.org>
|
|
|
| |
| |
|
|
More:
(7) Output File Dithering:
The following dithering algorithms are supported:
+ Ordered dithering algorithms:
- `B2`..`B4`: Bayer pattern dithering using 2x2, 3x3 or 4x4 patterns,
respectively
- `BN`: {{New}} in version 3.8, blue noise pattern dithering using a
64x64 pattern.
- `BNX`: {{New}} in version 3.8, same as BN, but using the pattern's
inverse for the R and B channels.
+ Error diffusion dithering algorithms:
- `AT`: {{New}} in version 3.8, Atkinson error diffusion dithering
- `BK`: {{New}} in version 3.8, Burkes error diffusion dithering
- `D1`: Simple 1-dimensional error diffusion dithering
- `D2`: Sierra Lite error diffusion dithering
- `FS`: Floyd-Steinberg error diffusion dithering
- `JN`: {{New}} in version 3.8, Jarvis-Judice-Ninke error diffusion
dithering
- `S2`: {{New}} in version 3.8, Two-Row Sierra error diffusion dithering
- `S3`: {{New}} in version 3.8, Three-Row Sierra error diffusion dithering
- `ST`: {{New}} in version 3.8, Stucki error diffusion dithering
[Not for docs: Absence of extra memory requirements info is deliberate;
the extra memory should be negligible in all but highly pathological
cases. Also, current info is no longer accurate, as requirements for D2
and FS have changed.]
{{New}} in version 3.8, the default is `-THbn`, i.e: dithering is off,
with `BN` being the default if only `+TH` is specified.
{{New}} in version 3.8, dithering is fully gamma-aware, i.e the overall
apparent brightness of the resulting image is faithfully reproduced.
(8) Preview Window Dithering:
In v3.8, the preview window always uses blue noise dithering [this was
Bayer 4x4 in v3.7, but I can't find it documented anywhere except in the
change logs].
I intend to write up a short(-ish) dithering tutorial based on the (new,
to be pushed soon) `distribution\scenes\output\dither_showcase.pov`
sample scene, but anyone willing to pick up that glove would be welcome.
Post a reply to this message
|
|
| |
| |
|
|
From: Jim Holsenback
Subject: Re: Some stuff related to output file formats
Date: 30 Sep 2018 18:09:17
Message: <5bb1498d$1@news.povray.org>
|
|
|
| |
| |
|
|
On 9/30/18 5:43 PM, clipka wrote:
> More:
>
> (7) Output File Dithering:
>
> The following dithering algorithms are supported:
>
> + Ordered dithering algorithms:
>
> - `B2`..`B4`: Bayer pattern dithering using 2x2, 3x3 or 4x4 patterns,
> respectively
> - `BN`: {{New}} in version 3.8, blue noise pattern dithering using a
> 64x64 pattern.
> - `BNX`: {{New}} in version 3.8, same as BN, but using the pattern's
> inverse for the R and B channels.
>
> + Error diffusion dithering algorithms:
>
> - `AT`: {{New}} in version 3.8, Atkinson error diffusion dithering
> - `BK`: {{New}} in version 3.8, Burkes error diffusion dithering
> - `D1`: Simple 1-dimensional error diffusion dithering
> - `D2`: Sierra Lite error diffusion dithering
> - `FS`: Floyd-Steinberg error diffusion dithering
> - `JN`: {{New}} in version 3.8, Jarvis-Judice-Ninke error diffusion
> dithering
> - `S2`: {{New}} in version 3.8, Two-Row Sierra error diffusion dithering
> - `S3`: {{New}} in version 3.8, Three-Row Sierra error diffusion dithering
> - `ST`: {{New}} in version 3.8, Stucki error diffusion dithering
>
> [Not for docs: Absence of extra memory requirements info is deliberate;
> the extra memory should be negligible in all but highly pathological
> cases. Also, current info is no longer accurate, as requirements for D2
> and FS have changed.]
>
> {{New}} in version 3.8, the default is `-THbn`, i.e: dithering is off,
> with `BN` being the default if only `+TH` is specified.
>
> {{New}} in version 3.8, dithering is fully gamma-aware, i.e the overall
> apparent brightness of the resulting image is faithfully reproduced.
>
>
> (8) Preview Window Dithering:
>
> In v3.8, the preview window always uses blue noise dithering [this was
> Bayer 4x4 in v3.7, but I can't find it documented anywhere except in the
> change logs].
>
>
>
> I intend to write up a short(-ish) dithering tutorial based on the (new,
> to be pushed soon) `distribution\scenes\output\dither_showcase.pov`
> sample scene, but anyone willing to pick up that glove would be welcome.
>
lol...quit piling it on. i'm starting a 14 day run today at work and my
time will be spotty
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 01.10.2018 um 00:09 schrieb Jim Holsenback:
>> I intend to write up a short(-ish) dithering tutorial based on the (new,
>> to be pushed soon) `distribution\scenes\output\dither_showcase.pov`
>> sample scene, but anyone willing to pick up that glove would be welcome.
>>
> lol...quit piling it on. i'm starting a 14 day run today at work and my
> time will be spotty
If it's any consolation, I've been feeling guilty about it ever since (4) ;)
The good news is, this concludes the streak of output file format
related changes (for now at any rate). Next up are probably a couple of
changes under the hood again.
Post a reply to this message
|
|
| |
| |
|
|
From: Jim Holsenback
Subject: Re: Some stuff related to output file formats
Date: 27 Oct 2018 08:40:09
Message: <5bd45ca9$1@news.povray.org>
|
|
|
| |
| |
|
|
On 9/29/18 3:31 PM, clipka wrote:
> Also:
>
> (4)
>
> In v3.8, enabling greyscale output will no longer force the bit depth to
> 16 bits per channel.
>
> (5)
>
> In v3.8, the `+F` option will allow specifying both the `G` greyscale
> flag and the bit depth, using the format `+FxGn`.
>
> (6)
>
> In v3.8, when creating a greyscale image, the preview display _will_
> show the image in greyscale.
>
1 thru 6 are done:
http://wiki.povray.org/content/Reference:Global_Settings#HF_Gray_16
Post a reply to this message
|
|
| |
| |
|
|
From: Jim Holsenback
Subject: Re: Some stuff related to output file formats
Date: 27 Oct 2018 12:08:10
Message: <5bd48d6a$1@news.povray.org>
|
|
|
| |
| |
|
|
On 9/30/18 5:43 PM, clipka wrote:
> More:
>
> (7) Output File Dithering:
>
> The following dithering algorithms are supported:
>
> + Ordered dithering algorithms:
>
> - `B2`..`B4`: Bayer pattern dithering using 2x2, 3x3 or 4x4 patterns,
> respectively
> - `BN`: {{New}} in version 3.8, blue noise pattern dithering using a
> 64x64 pattern.
> - `BNX`: {{New}} in version 3.8, same as BN, but using the pattern's
> inverse for the R and B channels.
>
> + Error diffusion dithering algorithms:
>
> - `AT`: {{New}} in version 3.8, Atkinson error diffusion dithering
> - `BK`: {{New}} in version 3.8, Burkes error diffusion dithering
> - `D1`: Simple 1-dimensional error diffusion dithering
> - `D2`: Sierra Lite error diffusion dithering
> - `FS`: Floyd-Steinberg error diffusion dithering
> - `JN`: {{New}} in version 3.8, Jarvis-Judice-Ninke error diffusion
> dithering
> - `S2`: {{New}} in version 3.8, Two-Row Sierra error diffusion dithering
> - `S3`: {{New}} in version 3.8, Three-Row Sierra error diffusion dithering
> - `ST`: {{New}} in version 3.8, Stucki error diffusion dithering
>
> [Not for docs: Absence of extra memory requirements info is deliberate;
> the extra memory should be negligible in all but highly pathological
> cases. Also, current info is no longer accurate, as requirements for D2
> and FS have changed.]
>
> {{New}} in version 3.8, the default is `-THbn`, i.e: dithering is off,
> with `BN` being the default if only `+TH` is specified.
>
> {{New}} in version 3.8, dithering is fully gamma-aware, i.e the overall
> apparent brightness of the resulting image is faithfully reproduced.
>
>
> (8) Preview Window Dithering:
>
> In v3.8, the preview window always uses blue noise dithering [this was
> Bayer 4x4 in v3.7, but I can't find it documented anywhere except in the
> change logs].
these are done as well:
http://wiki.povray.org/content/Reference:File_Output_Options#Output_File_Dithering
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 27.10.2018 um 18:08 schrieb Jim Holsenback:
> these are done as well:
> http://wiki.povray.org/content/Reference:File_Output_Options#Output_File_Dithering
Thanks.
Found only a minor typo this time.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 27.10.2018 um 14:40 schrieb Jim Holsenback:
> On 9/29/18 3:31 PM, clipka wrote:
>> Also:
>>
>> (4)
>>
>> In v3.8, enabling greyscale output will no longer force the bit depth to
>> 16 bits per channel.
>>
>> (5)
>>
>> In v3.8, the `+F` option will allow specifying both the `G` greyscale
>> flag and the bit depth, using the format `+FxGn`.
>>
>> (6)
>>
>> In v3.8, when creating a greyscale image, the preview display _will_
>> show the image in greyscale.
>>
>
> 1 thru 6 are done:
> http://wiki.povray.org/content/Reference:Global_Settings#HF_Gray_16
Looks good to me, except the greyscale variant of the `+F`/`-F` option:
The order has to be `G` /before/ bit depth, not the other way round as
currently documented. Also, since the `g` is to be taken literally,
rather than being a placeholder for something, I think it should be
wrapped in <code> instead of <em>. (Personally I'd even lean towards
also making it uppercase for more clarity, though it might be argued
that only the actual option "keys" should be rendered in uppercase.)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|