In camera raw processing

Kiwi Spike

Member
Messages
18
Reaction score
0
Location
NZ
Can anyone tell me if the in camera raw processor is actually increasing the iso rating by +2 or just the brightness?..In the camera it is called iso not brightness...Not much about it in the manual, page 99...

For example I took a pic last night at 800iso last night with 1/30 shutter and the processed in camera by +2....Question...did I get a 3200 iso shot with less noise than a 1600 iso?....

thanks

Pentax k-10d......360 flash whatever.....lots of lenses....handgrip.....etc....etc....
 
Can anyone tell me if the in camera raw processor is actually
increasing the iso rating by +2 or just the brightness?..In the
camera it is called iso not brightness...Not much about it in the
manual, page 99...
For example I took a pic last night at 800iso last night with 1/30
shutter and the processed in camera by +2....Question...did I get a
3200 iso shot with less noise than a 1600 iso?....
Your question confuses me a bit. However, rest assured that a picture exposed with an ISO setting at ISO800 also is saved on the card with the data stream as it should be at ISO800. Then what happens when "developing" that data stream to a viewable JPG is a bit beyond me and up to you, the camera and the manual. (I have never done any incamera conversion.) It's also nothing I feel a need to look into but I guess you can alter brightness and so on. That doesn't make any difference to the raw material, it's still the signal from the sensor amplified to ISO800 level.

So, no you don't get an ISO3200 picture with less noise than an ISO 1600 picture.

Another way to look at it is this: If there was a way to increase the brightness/sensitivity giving the end result you ask about it would have been widely known.

I guess you found the resulting file good?

regards,

--
Jonas
 
You will never know if you never try in camera raw conversion, so why hurt yourself even answering?

I Have been using slr's for over 25 years and don't need an inconclusive unsubstantiated report like yours thanks. Thanks for nothing.
--

Pentax k-10d......360 flash whatever.....lots of lenses....handgrip.....etc....etc....
 
****
You will never know if you never try in camera raw conversion, so
why hurt yourself even answering?

I Have been using slr's for over 25 years and don't need an
inconclusive unsubstantiated report like yours thanks. Thanks for
nothing.
--
Pentax k-10d......360 flash whatever.....lots of
lenses....handgrip.....etc....etc....
 
--
All I want is a digital back for my trusty K1000 . . .
 
LOL,

I have also wondered about the Iso in camera.
The camera is quite capable of it as we know the first firmware was Iso 50-3200.
I would also like a more conclusive answer by someone in the know.
 
You will never know if you never try in camera raw conversion, so
why hurt yourself even answering?

I Have been using slr's for over 25 years and don't need an
inconclusive unsubstantiated report like yours thanks. Thanks for
nothing.
If that's typical of your reactions you'll probably have to wait another 25 years for anyone to offer you any kind of assistance.

--
John Bean [BST/GMT+1] ('British Stupid Time')

PAW 2007 Week 18:
http://waterfoot.smugmug.com/gallery/2321711/2/149696265/Large



Index page: http://waterfoot.smugmug.com
Latest walkabout (21 March 2007):
http://waterfoot.smugmug.com/gallery/2641073
 
You will never know if you never try in camera raw conversion, so
why hurt yourself even answering?
Your post had slipped down. I felt sorry about it; it wasn't very clear and then there was this question about having an ISO3200 picture with less noise than usual.
I could help you get that part straight so I told you what I know about it.
I Have been using slr's for over 25 years and don't need an
inconclusive unsubstantiated report like yours thanks. Thanks for
nothing.
I see. Is it now I should tell you I have been using SLR cameras for 30 years and I don't need questions like your one? Nah, what would that lead to?

Why not read me reply again? I gave you some basic information that is good to have and you don't need to repeat your question about a "less than ISO1600-noisy ISO3200" picture anywhere else. Hey, I think I did you a favor there.

For me not doing incamera developing... is that the part making you upset? I explained that incamera processing is not my area. That way somebody else could fill in there. Many readers of the forum have a lot of knowledge. I'm sorry that you got so upset.

I know you haven't asked about it, but here goes anyway: Don't post here if you demand answers/replies/opinions that are exactly in line with your own mood or dialect. You are talking to several houndreds of individuals around the globe.

--
Jonas
 
You will never know if you never try in camera raw conversion, so
why hurt yourself even answering?

I Have been using slr's for over 25 years and don't need an
inconclusive unsubstantiated report like yours thanks. Thanks for
nothing.
--
Pentax k-10d......360 flash whatever.....lots of
lenses....handgrip.....etc....etc....
The only thing you don't have is courtesy then.

--
Wagga [Jean]
http://www.flickr.com/gp/47418676@N00/068b39
 
You will never know if you never try in camera raw conversion, so
why hurt yourself even answering?

I Have been using slr's for over 25 years and don't need an
inconclusive unsubstantiated report like yours thanks. Thanks for
nothing.
Why don't you test it yourself then. Do the following:

Shoot a few stationary shots. First shoot properly exposed at ISO800. Then using the same shutter speed and aperture, shift the camera down to ISO200 and shoot again.

Next, using the in-camera conversion, push the ISO200 shot up three clicks. Then compare your results. Try the same thing with 1600/400.

For the tests to be meaningful, the shots should have areas with deep shadow (near black) as well as bright areas. That way you can compare how shadow detail is preserved at each ISO setting, as well as how highlights clip at each ISO.

My own guess is that the ISO800 shot is going to look better than the ISO200 shot pushed to 800, particularly in the shadow detail. But I'm not the one asking the question and I haven't taken the time to test it. I'm just the guy explaining how you might go about answering your own question, since you didn't care for the answer someone else thoughtfully provided.
 
Why all the anger?!

my guess is sexual frustration.
You will never know if you never try in camera raw conversion, so
why hurt yourself even answering?

I Have been using slr's for over 25 years and don't need an
inconclusive unsubstantiated report like yours thanks. Thanks for
nothing.
--
Pentax k-10d......360 flash whatever.....lots of
lenses....handgrip.....etc....etc....
--
Gethin Lane K 1 0 D
 
You will never know if you never try in camera raw conversion, so
why hurt yourself even answering?

I Have been using slr's for over 25 years and don't need an
inconclusive unsubstantiated report like yours thanks. Thanks for
nothing.
I am soo jealous 8-}

--
cheers!

Gunn

-- Get a big lens and get closer™.

http://www.dpreview.pentaxistDS.photoshare.co.nz
http://www.y3m.net/penwik/pmwiki.php/Main/PentaxLensWiki
FAQ: http://www.pentaxuser.org/tiki-index.php

 
First off, I do not condone that type of reaction as you displayed towards Jonas.
Can anyone tell me if the in camera raw processor is actually
increasing the iso rating by +2
Not really. Once the image is captured, it is that ISO, as Jonas said.

What one can do during RAW processing is usually called exposure compensation in other RAW converters. This means that all bytes are shifted upwards, towards "white". This is commonly called pushing (just as in film times).

For the K10D, this method is a bit special. The reason for this is that its internal ISO values are nothing more than different digital multiplications of the captured data, very similar to what happens in a RAW converter (camera or PC) when you apply exposure compensation. The K10D can do this due to its circuit design, A/D converter and processing algorithm.
For example I took a pic last night at 800iso last night with 1/30
shutter and the processed in camera by +2....Question...did I get a
3200 iso shot with less noise than a 1600 iso?....
In theory, there should be very little difference between an ISO 800 shot pushed 1 stop and an ISO 1600 shot. If you think that your ISO 800 shot pushed two stops (which indeed would equal an ISO 3200 shot) looks better than a straight 1600 shot, I guess we would welcome some sample shots so that we can compare.

--
'I only trust those photos I have faked myself.' (Me, 2007)
http://www.jensroesner.de/
--=! Condemning proprietary batteries since 1976 !=--
 
Hi Jens, you are just toooo nice ;)

I wouldn't have bothered after seeing what happened here ...

--
Frank
GMT+8hours

 
Jens, just for your further enlightenment, there would be a slight difference between push processing raw images and having the ISO boosted in camera, as follows:
First off, I do not condone that type of reaction as you displayed
towards Jonas.
I concur.
Can anyone tell me if the in camera raw processor is actually
increasing the iso rating by +2
Not really. Once the image is captured, it is that ISO, as Jonas said.
What one can do during RAW processing is usually called exposure
compensation in other RAW converters. This means that all bytes are
shifted upwards, towards "white". This is commonly called pushing
(just as in film times).
For the K10D, this method is a bit special. The reason for this is
that its internal ISO values are nothing more than different
digital multiplications of the captured data, very similar to what
happens in a RAW converter (camera or PC) when you apply exposure
compensation. The K10D can do this due to its circuit design, A/D
converter and processing algorithm.
When you boost a captured raw image with the K10D, you will be bit shifting or multiplying the 12 bit photosite readings that have been recorded so will have the possibility of code gaps in the resulting image, especially for large boosts. For ISO's boosted by the in-camera sensitivity setting, the camera has access to plenty of "guard bits" so that no code gaps are caused. This said, further processing after the exposure boost will tend to fill in the gaps such that it would take really extreme boosts of 4 or 5 stops or more to ever be able to see any signs of tonality banding, and that would likely be masked by the resulting boosted noise.

Regards, GordonBGood
 
responded to that dipstick given his poor behavior towards Jonas. But, as my wife always says...it takes a better man. Have a good one. Don A.
--



'Nothing could-be-finah-than-to-be-in-Carolina-in-the-morninnnnnn...'
 
Hi Gordon!

Thanks for chiming in.
When you boost a captured raw image with the K10D, you will be bit
shifting or multiplying the 12 bit photosite readings that have
been recorded so will have the possibility of code gaps in the
resulting image, especially for large boosts.
That I understand. That's why most RAW converters limit this to 3 stops or so, I assume.
For ISO's boosted by
the in-camera sensitivity setting, the camera has access to plenty
of "guard bits" so that no code gaps are caused.
Okay, this makes sense, too. This is due to the 24 bit architecture?

However, for the in-camera conversion, this cannot work, as the camera can only work with the file which does not include the "guard bits", right? So, with the K10D there would be an advantage to shoot the highest ISO you need.

Whereas on the DS, it is (for RAW) actually better to shoot ISO 800 and push in software.

Cheers
Jens

--
'I only trust those photos I have faked myself.' (Me, 2007)
http://www.jensroesner.de/
--=! Condemning proprietary batteries since 1976 !=--
 
I might be tempted to mess with just about anyone here if necessary, but not Jonas. For one, he is very knowledgeable and helpful, and secondly, he holds some power over good/bad luck. Just take my word for it, I know.
--
'This is more serious than I thought.....but it is still fun!
http://www.pbase.com/rupertdog Take a look- It's Free!
 
Whereas on the DS, it is (for RAW) actually better to shoot ISO 800
and push in software.
Jens, I recall reading this somewhere but I missed the bulk of the discussion that led to the conclusion. Have you a handy url to allow me to "fill in the gaps" so to speak?

--
John Bean [BST/GMT+1] ('British Stupid Time')

PAW 2007 Week 18:
http://waterfoot.smugmug.com/gallery/2321711/2/149696265/Large



Index page: http://waterfoot.smugmug.com
Latest walkabout (21 March 2007):
http://waterfoot.smugmug.com/gallery/2641073
 

Keyboard shortcuts

Back
Top