From: ASAAR on
On 2 Apr 2007 12:36:31 -0700, cjcampbell wrote:

>> Interesting twist on physics ... excitement = creation ;-)
>>
>
> A good April Fool's joke should have some inaccurate physics in it.
> Plausible, but inaccurate.

But then it would have irreproducable results, not good enough to
justify submitting to a journal! :)

From: David J. Littleboy on

"ASAAR" <caught(a)22.com> wrote in message
news:nes213dcj5vf5thoi3hm6b6l91qa14gpcg(a)4ax.com...
> On 2 Apr 2007 12:36:31 -0700, cjcampbell wrote:
>
>>> Interesting twist on physics ... excitement = creation ;-)
>>>
>>
>> A good April Fool's joke should have some inaccurate physics in it.
>> Plausible, but inaccurate.
>
> But then it would have irreproducable results, not good enough to
> justify submitting to a journal! :)

There's always the Journal of Irreproducible Results.

http://www.jir.com/

David J. Littleboy
Tokyo, Japan


From: David J. Littleboy on

"acl" <achilleaslazarides(a)yahoo.co.uk> wrote:
> On Apr 2, 7:52 am, "David J. Littleboy" <davi...(a)gol.com> wrote:
>
>> Measuring three colors at each point costs you sqrt(3) in dynamic range
>> and
>> noise and makes raw files three times larger.
>
> Why do you say sqrt(3)? I do not remember seeing any data about
> saturation capacities in electrons of these detectors.

You need three charge storage devices per pixel. That means each one has
only 1/3 the capacity. It's probably somewhat worse than that, since you
need three times the circuits, and thus the space left for charge storage is
reduced even further.

David J. Littleboy
Tokyo, Japan


From: David J. Littleboy on

"John Sheehy" <JPS(a)no.komm> wrote:
> "acl" <achilleaslazarides(a)yahoo.co.uk> wrote:
>
>> On Apr 2, 7:52 am, "David J. Littleboy" <davi...(a)gol.com> wrote:
>>
>>> Measuring three colors at each point costs you sqrt(3) in dynamic
>>> range and noise and makes raw files three times larger.
>>
>> Why do you say sqrt(3)? I do not remember seeing any data about
>> saturation capacities in electrons of these detectors.
>
> What he's saying is that the shot noise in each layer in each RGB sensel
> will be higher by that factor, because only 1/3 of the photons are
> captured
> in each level.

Nope. I'm saying the well depth will be 1/3, so the maxmum number of photons
counted will be 1/3.

This is true regardless of the particular technology. (I was avoiding
talking about Foveon, since it's just a particular implementation of the
idea.)

> For the Foveon used in the SD9 and 10, that is probably an
> issue, as color has to be converted from very unsaturated RAW data.

That's a specific Foveon problem. Although I'd think it'd be easier to match
human color perception in a Bayer sort of system, since you can tune the
color of each filter to match human color vision.

David J. Littleboy
Tokyo, Japan


From: ASAAR on
On Tue, 3 Apr 2007 07:44:26 +0900, David J. Littleboy wrote:

>> But then it would have irreproducable results, not good enough to
>> justify submitting to a journal! :)
>
> There's always the Journal of Irreproducible Results.

Indeed. I didn't want to be too specific as some might have
accused me of being Ignobel or something like that. That's the
journal that Asimov should have used to publish the formal results
of "The Endochronic Properties of Resublimated Thiotimiline". :)

First  |  Prev  |  Next  |  Last
Pages: 1 2 3 4 5 6 7 8 9
Prev: low light
Next: Bayer filter removal