On Thu, 8 Nov 2012 04:43:29 -0500
Monty Montgomery xiphm...@gmail.com wrote:
BTW, if I sounded ungrateful for the comment, I didn't mean to be. I
was hoping for this kind of challenge to my specific proposal points
on the theory that if I can't defend then, I shouldn't be implementing
them.
Any reason for not using 8/16/32 bit as a fraction (0-1)? So 8 bit would
have a resolution of 1/255, 16 bit of 1/65535 etc. All operations can
be done with 4 byte integer math.
16 bits linear will be barely enough to capture an 8 bit nonlinear
space, and that's leaving out head- and toe-room.
Just an idea.
Sure :-) And I think you're right in the tactical sense... but
overall I think float is a more practical idea. It will be slower in
the ideal case, but I don't think we'll ever be within spitting
distance of ideal speed.
BTW, if I sounded ungrateful for the comment, I didn't
På Mon, 05 Nov 2012 14:27:02 +0100, skrev Monty Montgomery
xiphm...@gmail.com:
... and presenting 0.0 as black and 1.0 as white internally is
totally out of the question?
No. :-)
However, if black is 0.0, then super black would be negative,
and hilarity ensues. So we have to make up our
På Mon, 05 Nov 2012 14:27:02 +0100, skrev Monty Montgomery
xiphm...@gmail.com:
... and presenting 0.0 as black and 1.0 as white internally is
totally out of the question?
No. :-)
There is more. I can't see we have adressed what kind of black
and what kind of white.
* Aquisition medium
Hi
On Mon, Nov 5, 2012 at 7:55 PM, Monty Montgomery xiphm...@gmail.com wrote:
Linear-space float RGB with black point at 0. and white point at 1.
suits me fine, BTW. It's conceptually the simplest/most correct
option. I'm just worried about practicality issues I'm too dumb see
until I get
On Mon, 5 Nov 2012 12:55:14 -0500
Monty Montgomery xiphm...@gmail.com wrote:
Linear-space float RGB with black point at 0. and white point at 1.
suits me fine, BTW. It's conceptually the simplest/most correct
option. I'm just worried about practicality issues I'm too dumb see
until I get
På Mon, 05 Nov 2012 05:43:49 +0100, skrev Monty Montgomery
xiphm...@gmail.com:
The real rub in all of this, and the impetus for getting me thinking
to start with, is filter operations using colorspaces with
singularities at black (like HSV). When you don't know where black
is, any HSV
... and presenting 0.0 as black and 1.0 as white internally is
totally out of the question?
No. :-)
With your proposed studio swing
mapping most compositing and filtering operations have to scale
and bias to work right. E.g. black + black + black will be a
shade of gray, unless a bias is
I hope this isn't off topic, but for those of us still using SD formats,
we need YUV and not RGB. I tried RGBA for a while, but I always got
artifacts, so I went back to YUV. No artifacts.
Keith
On 11/05/2012 05:27 AM, Monty Montgomery wrote:
... and presenting 0.0 as black and 1.0 as
På Mon, 05 Nov 2012 16:54:15 +0100, skrev Keith Gudger ke...@sploids.com:
I hope this isn't off topic, but for those of us still using SD formats,
we need YUV and not RGB. I tried RGBA for a while, but I always got
artifacts, so I went back to YUV. No artifacts.
RGB8, I presume? The
På Mon, 05 Nov 2012 14:27:02 +0100, skrev Monty Montgomery
xiphm...@gmail.com:
How much is a lot? One conversion entering the pipeline, and another
exiting the pipeline?
Any color filters that expect to operate on R'G'B' rather than RGB. I
think in most cases, these operations only work
I suppose linear light is the native space of many (most?) common
operations. Blurring and resampling/scaling will get funny shifts
in brightness if they operate on non-linear samples, for example.
Yes, full agreement. But given that many [most] tools don't operate
in linear space, users may
På Sun, 04 Nov 2012 03:26:51 +0100, skrev Monty Montgomery
xiphm...@gmail.com:
Note that although I'm suggesting a mandatory
RGBA space, I'm also suggesting standard studio-swing levels (16-240
with head/toeroom). Full-swing imports will be mapped to the
restricted range.
In floating point
Now what about formats whose white point is far below the
max value, like some higher depth formats?
The higher-depth formats I know of (and I'm asking because I'm sure
there are ones I don't know of) still define swing in terms of the 8
bit range, and just tack on additional bits.
What is
På Sun, 04 Nov 2012 12:54:02 +0100, skrev Monty Montgomery
xiphm...@gmail.com:
Now what about formats whose white point is far below the
max value, like some higher depth formats?
The higher-depth formats I know of (and I'm asking because I'm sure
there are ones I don't know of) still
But then there are cinema-oriented formats,like Digital Intermediates and
such.
http://www.reduser.net/forum/printthread.php?t=2714pp=10page=310
That was all somewhat confusing... need equations :-) But it's clear
they're talking about something not broadcast and not nonlinear
(gamma)
Not understanding why you would want to force a clamped output at all.
Though it may properly apply
to most application, I feel certain there are many who will want a full
dynamic range. Not all my footage
will be used for broadcast purposes. If my render farm goes to all the
time and trouble to
Hi all
Thanks Monty for bringing up the topic.
The following is just from my point of view as a negative 16 mm user
A film (as not video) is nowadays (far and between unfortunately) shot
on negatives (wider dynamic range) whatever people may think.
The time of crappy reversal news reels is
On Sun, Nov 4, 2012 at 2:10 PM, Tim Copeland t...@criteriondigital.net wrote:
Not understanding why you would want to force a clamped output at all.
I don't; I was suggesting only an assumed internal space where we know
the intensity curve, black point and white point. I wasn't suggesting
Negatives need to be scan on 10 or 16 bits /pixels and no gamma. RGB 8 bits
is just not enough.
Of course not everyone has a HP 10 bits screen to grade negs.
Did you mean 'no gamma correction' or did you really mean full-linear?
Ahh, right, if it's a negative, you don't want to apply a
Hi Monty
Negatives need to be scan on 10 or 16 bits /pixels and no gamma. RGB 8 bits
is just not enough.
Of course not everyone has a HP 10 bits screen to grade negs.
Did you mean 'no gamma correction' or did you really mean full-linear?
Ahh, right, if it's a negative, you don't want to
22 matches
Mail list logo
