http://www.nikon-box.com/MultiX/
Much better than any normal way of shooting with the D300 !
If you're happy with the results and save 3000,- dollars because you realize
that now all you need is the D300 instead of the D3, feel free to donate me
some of all that saved money. ;-)) Maybe I can finally buy that lens...
--
Sosumi
Welcome to the party! Works with any camera.
See thread "Noise Reduction By Image Averaging"
<http://groups.google.com/group/rec.photo.digital/browse_thread/thread/b9d1df059f100e26/d6b00a2ac821298f#d6b00a2ac821298f>
--
Best regards,
John Navas
Panasonic DMC-FZ8 (and several others)
This is totally different from what I do. No software is used! All in
camera, except the NEF to JPG conversion.
It's not averaging, but multiple exposure with autogain plus bracketing.
But still, it's also a good idea. Never heard of it before.
--
Sosumi
>"John Navas" <spamf...@navasgroup.com> wrote in message
>news:k9jhq35nnc3claf9p...@4ax.com...
>> Welcome to the party! Works with any camera.
>> See thread "Noise Reduction By Image Averaging"
>> <http://groups.google.com/group/rec.photo.digital/browse_thread/thread/b9d1df059f100e26/d6b00a2ac821298f#d6b00a2ac821298f>
>
>This is totally different from what I do. No software is used! All in
>camera, except the NEF to JPG conversion.
>It's not averaging, but multiple exposure with autogain plus bracketing.
That's still image averaging, albeit in camera instead of in
post-processing.
Astrophotographers do it as a matter of course, for years now.
The software is in the camera, rather than on your
computer. It's still software that does it.
>camera, except the NEF to JPG conversion.
>It's not averaging, but multiple exposure with autogain plus bracketing.
It *is* averaging, and it is *not* bracketing.
"Autogain" means that it averages the entire set.
>But still, it's also a good idea. Never heard of it before.
Your comparison to the D3 is nonsense, because the D3
can do exactly the same thing. Except of course the D3
has an FX sensor and higher ISO with less noise... so
the results are even better than with the D300.
--
Floyd L. Davidson <http://www.apaflo.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska) fl...@apaflo.com
Not a difficult trick at all, since the technique has
been available in other cameras, or by post processing,
for years before the D300 was even a gleam in the eye of
Nikon.
True, but like everything, software is not the same. Sharpening with NX
Capture gives much better results than PS.
Besides, now all info is put directly to one single frame, so it saves time
too.
>>camera, except the NEF to JPG conversion.
>>It's not averaging, but multiple exposure with autogain plus bracketing.
>
> It *is* averaging, and it is *not* bracketing.
>
> "Autogain" means that it averages the entire set.
If it's averaging, then when I make 5 shots with multiple exposure with the
*same* exposure, I should get the same result, right? But I don't. The
bracketed shots have more detail, less noise and more range.
>>But still, it's also a good idea. Never heard of it before.
>
> Your comparison to the D3 is nonsense, because the D3
> can do exactly the same thing. Except of course the D3
> has an FX sensor and higher ISO with less noise... so
> the results are even better than with the D300.
Gee whiz! Lighten up! Do I have to put a smiley for every time I'm kidding?
--
Sosumi
>"Floyd L. Davidson" <fl...@apaflo.com> wrote in message
>news:87wspjy...@apaflo.com...
>> The software is in the camera, rather than on your
>> computer. It's still software that does it.
>
>True, but like everything, software is not the same. Sharpening with NX
>Capture gives much better results than PS.
The latest versions of PS do excellent sharpening, based on
deconvolution.
<http://en.wikipedia.org/wiki/Deconvolution#Optics_and_other_imaging>
>> It *is* averaging, and it is *not* bracketing.
>>
>> "Autogain" means that it averages the entire set.
>
>If it's averaging, then when I make 5 shots with multiple exposure with the
>*same* exposure, I should get the same result, right?
Wrong.
>But I don't. The
>bracketed shots have more detail, less noise and more range.
Sure -- that's what image averaging is all about.
Trust that there is *no* bracketing. Each and every
image is taken with exactly the same exposure. Otherwise
it wouldn't work as well. You'd average in blown highlights
and lost shadows... making the result worse than the one shot
that was properly exposed.
>If it's averaging, then when I make 5 shots with multiple exposure with the
>*same* exposure, I should get the same result, right? But I don't. The
>bracketed shots have more detail, less noise and more range.
Why do think you should get the same result?
Whatever, your observation is not correct. What you get
is just less noise. The extra detail and the extra
range are just side effects of less noise.
The reason it works is because light from the scene for
any give pixel will be the same each and every time. So
the average of a number of images is exactly the same as
any one of the images. But noise is randomly
distributed, and will have a slightly different
distribution with each image. Any pixel that has a high
noise component in one image will not have a high noise
component in all images. Hence noise will average
*lower* in the combined image than the peak noise will
be in any of the individual images.
If there is less noise, then more detail is visible.
That by definition is more dynamic range.
>>>But still, it's also a good idea. Never heard of it before.
>>
>> Your comparison to the D3 is nonsense, because the D3
>> can do exactly the same thing. Except of course the D3
>> has an FX sensor and higher ISO with less noise... so
>> the results are even better than with the D300.
>
>Gee whiz! Lighten up! Do I have to put a smiley for every time I'm kidding?
Obviously you were *not* kidding, as the multiple
responses indicate.
>Trust that there is *no* bracketing. Each and every
>image is taken with exactly the same exposure. Otherwise
>it wouldn't work as well. You'd average in blown highlights
>and lost shadows... making the result worse than the one shot
>that was properly exposed.
Are you sure the camera doesn't do that? Properly done bracketing and
combining should yield a HDR image.
>The reason it works is because light from the scene for
>any give pixel will be the same each and every time. So
>the average of a number of images is exactly the same as
>any one of the images. But noise is randomly
>distributed, and will have a slightly different
>distribution with each image. Any pixel that has a high
>noise component in one image will not have a high noise
>component in all images. Hence noise will average
>*lower* in the combined image than the peak noise will
>be in any of the individual images.
Actually that the amount of noise in a given pixel of a given image is
random, and thus will tend to decrease when values are averaged. See
<http://www.cambridgeincolour.com/tutorials/noise-reduction.htm>
>If there is less noise, then more detail is visible.
>That by definition is more dynamic range.
Yes, but not the same as HDR.
How about something really over the top? Olympus E-510, 50 frames
stacked, 1600 ISO, HQ JPEG mode. Done with "Registax"
http://www.astronomie.be/registax/
1 frame:
http://www.pbase.com/andersonrm/image/92591393/original
50 frames:
http://www.pbase.com/andersonrm/image/92591393/original
The nice thing about registax is that it also automatically aligns the
frames. So I sometimes set my camera to 5fps and take 10-15 severely
underexposed handheld shots, which I then stack. The main difficulty I
have is that (obviously) I can't see a thing through the viewfinder
(since we're talking about an effective ISO in the tens of thousands
for each shot), so composition is a bit hit and miss: I've never
gotten anything I'd like to print this way, compositionally speaking.
A tripod's better if you can use one, you can look at the lcd and
adjust if it's too dark to see.
the same way the K10D is doing it ?
It won't if the "combining" is either averaging pixel
values with "autogain" circuit or just an additive
process, which are the two options available with Nikon
cameras.
>>The reason it works is because light from the scene for
>>any give pixel will be the same each and every time. So
>>the average of a number of images is exactly the same as
>>any one of the images. But noise is randomly
>>distributed, and will have a slightly different
>>distribution with each image. Any pixel that has a high
>>noise component in one image will not have a high noise
>>component in all images. Hence noise will average
>>*lower* in the combined image than the peak noise will
>>be in any of the individual images.
>
>Actually that the amount of noise in a given pixel of a given image is
>random, and thus will tend to decrease when values are averaged. See
><http://www.cambridgeincolour.com/tutorials/noise-reduction.htm>
Not quite. The randomness that counts for averaging is
how random different images are for the noise at a given
pixel. The fact that different pixels within each image
have a random noise distribution is not particularly
significant (other than the two characteristics result
from the same cause).
>>If there is less noise, then more detail is visible.
>>That by definition is more dynamic range.
>
>Yes, but not the same as HDR.
Exactly. That's why his statement that he saw more
range was not really true. All he saw was less noise.
Agreed.
???
Metaphorically speaking, or is that retorically?
wishfully?
>> Welcome to the party! Works with any camera.
>> See thread "Noise Reduction By Image Averaging"
> This is totally different from what I do. No software is used! All in
> camera, except the NEF to JPG conversion.
Your camera runs on steam and gears?
> It's not averaging,
Nikon does (spatial) averaging, at least on longer exposures, even
with it's RAW! Astro-guys are regularly bitching and moaning,
as that blurring is of course less than perfect for fainter stars.
Google for Nikon and "Mode 3" ...
And yes, it is averaging.
-Wolfgang
Nah, the new Pentax looks nice though, the cost to replace all my glass is
more than I need to spend although I could be swayed when I retire with all
that cash to burn, what about you? The D3's do look very nice! :-)
It seems to me you're relatively new to the group, so welcome. Don't
mind rude jerks like Floyd L. Davidson (who has no sense of humor and
a major stick up his ass). Many of them are good resources, so take
what you can and ignore the rest.
Regards,
Matt Clara
I still have a mortgage. :-(
All that aside I have a sign in my office that I try to live by as much as
I can......
WANT = NEED
You guys confuse the crap out of me:
In this post you first tell us, that there is less noise, more detail, so
that is more dynamic range.
Then you say it's not HDR, OK, *higher* dynamic range then?
Then you say, I only saw less noise, not more range.
Maybe I'm nuts, but it seems at least completely contradictive!
What I can tell by searching the net is that at least the tonal range is
wider.
--
Sosumi
> Metaphorically speaking, or is that retorically?
rhe-tor'i-cal-ly / that spelling courtesy dictionary dot com :-)
> Metaphorically speaking, or is that retorically?
rhe-tor'i-cal-ly / that spelling courtesy dictionary dot com :-)
Blow-it-out-your-wahzoo........................ Spelling curtesy me.
sorry, but wrong again... I do not use yahoo :-)