From: John Sheehy on
Paul Rubin <http://phr.cx(a)NOSPAM.invalid> wrote in
news:7x1wjzr5zg.fsf(a)ruckus.brouhaha.com:

> I don't see how the 1Dmk3 does better given the various observations
> that the 20D's low light performance is limited basically by photon
> noise.

It is a widely held belief that current cameras are limited mainly by
photon noise, and Roger Clark's work is often quoted and referenced to
support it, but I, for one, don't believe it. I believe that photon noise
is a relatively pleasant-looking noise, and it is ruined by patterned read
noises (both blackframe offset, and scalar illumination noises), which have
much more visual power than the randomly-distibuted poisson shot noise.

Take any Canon RAW file underexposed by several stops, and push it, What
do you see? Bands of color running horizontally, sometimes vertically.
This is not shot noise that is ruining the shadows. Look how high read
noise is in electrons, at the lowest ISOs - it is incredibly high, and
trashes the excellent shadows captured in the sensor wells. Shot noise is
the least of our digital imaging problems, IMO, especially with large
sensors.

That said, Canon does claim less wasted space on the sensor (higher fill
factor) over the mkII, and greater quantum efficiency, so, ISO 50 may be
able to have full DR, unlike the mkII, *and* more photons may be collected
for the same real-world absolute exposure.


--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS(a)no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
From: David J. Littleboy on

"John Sheehy" <JPS(a)no.komm> wrote:
>
> Take any Canon RAW file underexposed by several stops, and push it, What
> do you see? Bands of color running horizontally, sometimes vertically.
> This is not shot noise that is ruining the shadows. Look how high read
> noise is in electrons, at the lowest ISOs - it is incredibly high, and
> trashes the excellent shadows captured in the sensor wells.

In the 5D and 1Dmk2, the ISO 100 read noise looks to me to be dominated by
quantization errors; the bit depth of the A/D converter is two bits shy of
what's needed, maybe three. The dynamic range at ISO 100 to 400 is simply
consistent with a 12-bit A/D converter. It's only at ISO 800 and above that
other noise sources intrude.

That's why the D200 has the same ISO 100 dynamic range as the 5D.

> Shot noise is
> the least of our digital imaging problems, IMO, especially with large
> sensors.

I'm not seeing pattern noise in pused images. Here's a 5D ISO 3200 file
pushed 3 stops. Straight from Lightroom with noise reduction (and
sharpening) turned off.

http://www.pbase.com/davidjl/image/75374090/original

This is seriously amazing stuff. At ISO 6400, the 5D is producing images
competitive with the dreck 35mm users get with Tri-X in Rodinal (ISO 400).

Here's what you get with noise reduction and a touch of sharpening.

http://www.pbase.com/davidjl/image/75359389/original

> That said, Canon does claim less wasted space on the sensor (higher fill
> factor) over the mkII, and greater quantum efficiency, so, ISO 50 may be
> able to have full DR, unlike the mkII, *and* more photons may be collected
> for the same real-world absolute exposure.

ISO 50 will remain problematic, since the well depth is inadequate.

David J. Littleboy
Tokyo, Japan


From: Paul Rubin on
"David J. Littleboy" <davidjl(a)gol.com> writes:
> I'm not seeing pattern noise in pused images. Here's a 5D ISO 3200 file
> pushed 3 stops. Straight from Lightroom with noise reduction (and
> sharpening) turned off.
>
> http://www.pbase.com/davidjl/image/75374090/original

You mean this is at 25,600? Pretty cool. I've seen TMZ pushed to
that speed and maybe the grain isn't worse, but the TMZ has no
tonality to speak of with that much pushing.
From: David J. Littleboy on

"Paul Rubin" <http://phr.cx(a)NOSPAM.invalid> wrote:
> "David J. Littleboy" <davidjl(a)gol.com> writes:
>> I'm not seeing pattern noise in pushed images. Here's a 5D ISO 3200 file
>> pushed 3 stops. Straight from Lightroom with noise reduction (and
>> sharpening) turned off.
>>
>> http://www.pbase.com/davidjl/image/75374090/original
>
> You mean this is at 25,600?

Exactly!

> Pretty cool. I've seen TMZ pushed to
> that speed and maybe the grain isn't worse, but the TMZ has no
> tonality to speak of with that much pushing.

I wonder if you are comparing apples to apples: that's a 100% crop from at
12.7MP original. Even with that much noise, that's a lot of pixels. I'd
think pushed film viewed at that resolution would be a major disaster.

David J. Littleboy
Tokyo, Japan


From: ray on
On Thu, 08 Mar 2007 13:24:03 -0800, Scott W wrote:

> On Mar 8, 7:45 am, ray <r...(a)zianet.com> wrote:
>> Since I don't have $1500 to blow on a digital SLR, I don't plan on doing
>> my own comparison tests any time soon. I will be experimenting a little
>> with my Kodak P850 to see what it's limitations are.
> The Kodak may be able to produce a good looking image but it will not
> come
> close to the ISO performance that a DSLR will have, its sensor is just
> too
> small for that. It would be a mistake to judge what a DLSR is capable
> of based on a point and shoot digital. The point and shoot cameras
> that I have are pretty much limited to ISO 100 or less.

Those are obvious points. Manual says the P850 will do iso 800 at 1.2 mp -
400 otherwise. Certainly I would not judge what a dslr costing several
times more by it. It's what I have, and happens to fit my current
requirements better than a larger camera. I've just not had time to do
much experimenting yet.

>
> To see just how much better a DSLR is first look at how the P850 does
> at ISO 400
> <http://www.dpreview.com/reviews/kodakp850/page11.asp>
> Now look at how the a number of DLSRs do at ISO 800 and 1600
> <http://www.dpreview.com/reviews/canoneos400d/page27.asp>
>
>

One thing I note is that they advise shooting raw with the P850 at 400 - I
routinely shoot raw on the P850 - something you can't do with many P&S
cameras.

>>I also have my trusty
>> old Minolta SRT202 which I still put a roll through every once in a while
>> - particularly doing wildlife shots at Yellowstone. I've not much doubt
>> that under ideal conditions digital produces shots that are quite fine
>> enough.
> The reality is that a DSLR is far better at getting the good shots
> when conditions are not idea, it is far better in low light and it is
> far better when white balance might be tricky.

That's what everyone seems to be saying.

>
>>The OP did not state (as I recall) whether the ultimate product
>> would be files for editing on the computer or prints - I suspect that
>> could easily swing the pendulum one way or the other.
> Does not matter if you are going for prints or files a DSLR will do
> way better in low
> light.
>
>> BTW - I've produced
>> shots from a 1mp Kodak DC210+ printed to 8x10 that look pretty damned good
>> - but that was, again, under the best of circumstances.
>
> I have never gotten a 8 x 10 print from a 1 MP camera that I really
> liked, they always looked really soft to me.
>
>
>> I don't often use high ISO - I hope, as I said, to do some shooting in
>> that area as time permits.- Hide quoted text -
>
> If you use either film or your point and shoot you will likely get
> frustrated pretty fast,
> unless you do B/W and really like grain, some people do.

What I do is a lot of hiking and snowshoeing. I don't relish packing ten
pounds of dslr and lenses along. For right now, the P850 seems to fill the
bill. I don't do much B/W, I don't care for a lot of grain, I fairly
frequently shoot wildlife at some distance, so will use a big zoom. Doubt
I will get much frustrated - I believe I know what to expect - I've been
using digital for a number of years - Kodak DC210+; Minolta S414; now the
P850. I'll be sure and let you know if my frustration level gets out of
control.


>
> Scott