Approve the Cookies
This website uses cookies to improve your user experience. By using this site, you agree to our use of cookies and our Privacy Policy.
OK
Index  •   • New posts  •   • RTAT  •   • 'Best of'  •   • Gallery  •   • Gear  •   • Reviews
Guest
New posts  •   • RTAT  •   • 'Best of'  •   • Gallery  •   • Gear  •   • Reviews
Register to forums    Log in

 
FORUMS Canon Cameras, Lenses & Accessories Canon EOS Digital Cameras 
Thread started 09 Apr 2009 (Thursday) 18:11
Search threadPrev/next
sponsored links
(this ad will go away when you log in as a registered member)

What is the best "standard ISO" for the 5D MkII?

 
DavidPhoto
Senior Member
777 posts
Likes: 1
Joined Mar 2007
Location: MA, USA
     
Apr 12, 2009 09:28 |  #76

Daniel Browning wrote in post #7712224 (external link)
in any case, on many cameras, you can ignore ISO, because they have the same amount of read noise at ISO 50 as they do at ISO 6400. It literally makes no difference what the ISO is set to at the time of taking the photo: the pictures will come out the same from the raw converter when digital gain is applied. Only the *exposure* (light intensity falling on the sensor) will affect the image.

It's only JPEG shooters and cameras with read noise that varies by ISO (gain) setting that makes it necessary for us to be careful with how the ISO is set.

Hope that helps,
--Daniel

I've not ignored the ISO setting when shooting and I would guess thats generally bad advice to give. I may have to try what you are espousing though just to see for myself. So to be clear, you are suggesting that the ISO setting has no effective purpose on most of our cameras and that adjusting the raw in PP will have the same or similar effect as setting the ISO at the time of capture?




  
  LOG IN TO REPLY
sponsored links
(this ad will go away when you log in as a registered member)
Daniel ­ Browning
Goldmember
1,199 posts
Likes: 4
Joined Nov 2008
Location: Vancouver, WA
     
Apr 12, 2009 12:09 |  #77

Pekka wrote in post #7714409 (external link)
You shoot a subject with 1/200 1.2 ISO 3200. The subject has high dynamic range and you capture it with this exposure so that all intensities are contained in histogram, you have shadow detail and highlight detail.

Now, you say you get same result if you shot it with 1/200 1.2 ISO 1600 and push it one stop in post. No. Noise might be the same, histogram will not be, color will not be.

A one stop linear push will result in a histogram and color will is literally the same in all respects. However, a nonlinear push will be slightly different because one whole stop of highlights are retained instead of deleted. This is why it's important to never go above the highest *useful* analog gain. Gain deletes highlights and reduces read noise. If it stops reducing read noise, then it is only deleting highlights, and that is not useful.

Pekka wrote in post #7714409 (external link)
What happens is when you essentially underexpose the frame by reducing sensor sensitivity (or whatever emulation makes it tick) the left side of histogram is lost for good, shadow detail has not been captured . Not matter how you push it in RAW converter you can not make those zeros come alive again.

That is a correct description of what happens with JPEG, but not raw.

Pekka wrote in post #7714409 (external link)
Daniel said:
"in any case, on many cameras, you can ignore ISO, because they have the same amount of read noise at ISO 50 as they do at ISO 6400. It literally makes no difference what the ISO is set to at the time of taking the photo: the pictures will come out the same from the raw converter when digital gain is applied. Only the *exposure* (light intensity falling on the sensor) will affect the image."

You see things from camera's sensor point of view. In practical photography you must capture range of light into a usable file. Shooting ISO 50 and pushing 5 stops does not give you a usable ISO 1600 equivalent file no matter how sensor or theory has it. Just try it if you don't believe it.

I have tried it. You are extrapolating your experience with your raw conversion software and applying it to all cameras and all raw conversion software.

In-camera ISO is just gain. It can be digital or analog. If the gain is digital, then all it is doing is mangling and deleting the raw data. One stop of digital gain (ISO) literally deletes one stop of highlights, with absolutely no improvement in read noise or quantization error. If the gain is metadata, then it's just a recommendation to the raw converter to apply exposure compensation.

Any camera that implements ISO as metadata (including many MFDB and digicams) will produce literally identical files no matter what the ISO is set to.

Other cameras do not have ISO as metadata, but some or all of their ISO settings are just in-camera digital pushes. On these the only difference between high ISO and low ISO is that low ISO has more highlight headroom: the noise and everything else is the same.

There are some cameras that implement true analog gain. Many times, these analog gains have less read noise at high ISO than low ISO. Sometimes, the analog gain has the same read noise as lower ISO settings. So it's beneficial to use the highest *useful* analog ISO when additional highlight headroom is not needed.

Finally, there are cameras where it's a mix: they offer some true analog gain and some in-camera digital gain. On Canon, for example, the only time you get less read noise is at ISO 200-1600. ISO 3200 is analog, but the read noise does not improve. ISO 6400-25600 are just digital and never improve image quality (they only delete highlight information for no reason).

For example, if you want to shoot an ultra low light portrait at ISO 25600. With Canon's current cameras, you have two choices:

1. Set the ISO to 25600. This causes the camera to use an analog gain of 3200, then apply a digital four stop push, deleting four stops of highlight data, blowing and clipping important details all over the scene, as well as doubling or tripling the file size.

2. Set the ISO to 1600, which is the highest useful analog gain. This is 4 stops away from the desired gain, so the exposure meter, autoexposure system, and LCD review will all be useless. In post-production, nonlinear exposure compensation may be used to apply the gain without deleting the highlights needlessly: saving all four stops of information, and keeping the filesize small as a bonus.

A third option, which used to be available on the Canon 10D, and is offered by other manufacturers, is metadata ISO:

3. Set the analog ISO to 1600 and the metadata ISO to 25600. This causes a highlight-preserving EC to be used for in-camera JPEG (or even a linear EC) while leaving the raw alone. This is the best of all worlds: it retains the highlight data, file size, AE system, and LCD review.

This feature would also unlock the possibility of getting more highlight headroom even at low ISO. Right now, if you want to shoot ISO 1600 with the highlight headroom of ISO 200, you have to give up auto exposure, metering, LCD preview, etc. and just shoot analog ISO 200. With metadata ISO, you could get all those features back by just setting the metadata ISO to 1600.

So it's not just for super high ISO, it's useful at *anything* over ISO 100 when you want more highlight headroom.

Canon had this feature in the 10D. ISO 1600 was analog ISO 800 with "+1 EC" metadata that was applied to the jpeg preview and in-camera LCD and for autoexposure metering and review, but did not mangle the raw data. All subsequent cameras had this feature removed, and instead began deleting highlights from the raw data.

Years later, Canon added Highlight Tone Priority, but limited it to one stop and disabled it for some of the ISO settings where it would be most useful (12800 and 25600), and added it to the ISO settings where it can cause a lot of noise (especially ISO 200).

Pekka wrote in post #7714409 (external link)
And yes, the difference in us using terms like exposure is exactly like Tom said in https://photography-on-the.net …hp?p=7712297&po​stcount=70 . When you ask a photographer "what was the exposure" you can not avoid taking into account the effect of ISO, really.

I like to have a word that describes the amount of light falling on the sensor. This concept used to be described by "exposure", but if enough people misuse it for enough years, the meaning is lost. Maybe it's time to give up on "exposure" and try to find a new word. I'm open to suggestions. (Aperture, irony, and literally are three other words that have lost their meaning and I've yet to find a new word to replace the original concept described by these words.)

Pekka wrote in post #7714409 (external link)
If I'd give advice "shoot in this venue with these lights with 3.5 1/250, any ISO you like because it won't matter" I'd hear about it later!

If you had a camera with metadata ISO, such as digicam or MFDB, you wouldn't hear anything later except thanks for the good advice. The files are exactly the same no matter what ISO is chosen.

If you had a Canon, that would be bad advice. If you advised "use f/3.5 1/250 ISO 25600", the results would come out with 4 stops more blown highlights than "f/3.5 1/250 ISO 1600", but the same amount of noise. Again, the correct advice is to use the highest useful ISO for the camera.

DavidPhoto wrote in post #7714599 (external link)
So to be clear, you are suggesting that the ISO setting has no effective purpose on most of our cameras and that adjusting the raw in PP will have the same or similar effect as setting the ISO at the time of capture?

Yes, for any ISO setting that does not improve read noise. On my 5D2, that's ISO 1600. On many other cameras it's ISO 100.


Daniel

  
  LOG IN TO REPLY
Panopeeper
Senior Member
774 posts
Joined May 2008
     
Apr 12, 2009 13:40 |  #78

Tom W wrote in post #7714170 (external link)
In particular, I'd like to see comparisons in situations with a greater diversity of brightness and color

Hold on, Tom. The capture below shows that segment of the ISO 800 raw image (still non-demosaiced). In the above demos I blended out the "green" and "blue" channels for the visual assessment of the "red" channel.

Following basics need to be understood; I realize, that many photographers have a hard time to accept these:

1. No highlights, not even midtones are required in the raw image for the measurement of the DR

The upper end of the DR is given by the clipping level, which depends on the camera model and often on the ISO and the raw channel. Thus only the lower end needs to be determined, as a function of the acceptable noise. (Note, that the graps do not indicate any firm number for the dynamic range: you have to determine which level of noise is acceptable, then you can read the associated dynamic range.)

2. The noise does not depend on the color; specifically, it does not depend on which channel it is measured

The paper Source of Noise (external link) demonstrates this fact in the chapter Noise and raw channels, directly after the introduction.

I did not want to cause any more confusion, so I picked only from the "red" raw channel for the above demonstration.

3. The noise has to be measured on the raw channels in order to achieve comparable results

Look at the selected patch (marked with the orangy rectangle) in the capture below.

The intensity in the red channel is -9.91 EV, i.e. it is almost ten stops from clipping; the noise is 59.7% (compared to the average pixel intensity on the selected patch).

The intensity in the gren channel is -8.65 EV, i.e. the illumination of the patch under the green filters was 1 1/3 EV higher than under the red filters. As expectable, the noise is lower, 31.2%.

The intensity in the blue channel is -7.95 EV, i.e. this almost ten stops from clipping. This is two stops brighter than the red; not surprizing, for this is the blue patch of the color checker. The noise is, of course, much lower than in the red channel, it is only 21.8%.

What you see there is the combined effect of the three channels. In other words, the visual effect of the noise in the red channel is mitigated by the much lower noise in the other channels.

One can justifiedly object, that the raw channel noise is not perceivable directly in real photography. That's right, the numbers gained by such measurements do not reflect the noise in the final image. However, these numbers are comparable between ISO settings, exposures (illuminations) and camera models.

Otherwise, one would have to deal with the countless combinations of subject colors and iilluminations. For example when capturing an object in incandescent light, the red component will be less noise than the blue. Change the illumination to fluorescent, and the red becomes the more noisy. Applying the white balance is in effect a multiplication of the color channel (RGB, not raw), which multiplies the noise as well.

The method I am using yields abstract measurements, instead of trying to create settings for lots of different object colors, exposures and illumination sources.

Note, that the measurement does not depend on the exposure either: if the exposure is higher, I pick a darker patch to measure the noise at a certain illumination level/pixel intensity. Thus, when I am asking for raw files suitable for such measurement, I always request at least three stop underexposure.

IMAGE NOT FOUND
IMAGE IS A REDIRECT OR MISSING!
MIME changed to 'text/html' | Content warning: script

Gabor

  
  LOG IN TO REPLY
Panopeeper
Senior Member
774 posts
Joined May 2008
     
Apr 12, 2009 14:43 |  #79

Daniel Browning wrote in post #7715232 (external link)
Yes, for any ISO setting that does not improve read noise. On my 5D2, that's ISO 1600. On many other cameras it's ISO 100.

I am sure this is meant like on my 5D2 the highest ISO, which improves read noise is ISO 1600.

Or, on my 5D2 that is ISO 3200 and above.


Gabor

  
  LOG IN TO REPLY
Daniel ­ Browning
Goldmember
1,199 posts
Likes: 4
Joined Nov 2008
Location: Vancouver, WA
     
Apr 12, 2009 16:47 |  #80

Panopeeper wrote in post #7715832 (external link)
I am sure this is meant like on my 5D2 the highest ISO, which improves read noise is ISO 1600.

Or, on my 5D2 that is ISO 3200 and above.

Yes, that's what I meant; thanks.


Daniel

  
  LOG IN TO REPLY
Matthew ­ Hicks ­ Photography
Goldmember
Avatar
2,552 posts
Likes: 3
Joined Nov 2007
Location: Calgary, Alberta, Canada
     
Apr 12, 2009 22:59 |  #81

I don't know what read noise is, but the idea that I seem to be getting from some participants in this thread is that RAW data is simply the amount of light hitting the sensor and that it doesn't matter what you have your ISO set to because in the end once you run it through ACR or DPP or whatever you use, you're going to end up with the same noise and dynamic range. I feel I must be interpreting that wrong, but it definitely seems to be what people are saying. And that's just plain wrong. Now, I don't understand the difference in different light sensitivities, but the fact of the matter is that your dynamic range IS going to lose out if you boost or drop your exposure normally. Try bringing up a 3 stop underexposed picture and you'll see what I mean, the shadow detail that had 0 exposure value is going to stay black unless you make it gray, and it will have no detail no matter what you do. Same with highlights. There definitely is a difference between ISO 1600 pushed to 3200 and a 5D set to 3200 originally, although this doesn't apply to expansion ISOs.


Calgary Wedding Photography by Matthew Hicks: www.matthicksphoto.com (external link)

  
  LOG IN TO REPLY
Daniel ­ Browning
Goldmember
1,199 posts
Likes: 4
Joined Nov 2008
Location: Vancouver, WA
     
Apr 12, 2009 23:45 |  #82

Trainboy wrote in post #7718257 (external link)
... the idea that I seem to be getting from some participants in this thread is that ... it doesn't matter what you have your ISO set to because in the end once you run it through ACR or DPP or whatever you use, you're going to end up with the same noise and dynamic range.

Close: it's only true for some ISO settings on all cameras and all ISO settings on some cameras.

On Canon in Manual mode, for example, it doesn't matter whether you choose ISO 1600 or 25600: the noise comes out the same.

One of the points I'm trying to make is to separate "exposure" from many other factors. To me, exposure is the light intensity falling on the sensor, which is determined by f-number, shutter speed, ND, scene brightness, etc. Other factors, such as ISO setting (gain), quantum sensitivity, ADC quantization, etc. are separate.

Trainboy wrote in post #7718257 (external link)
the fact of the matter is that your dynamic range IS going to lose out if you boost or drop your exposure normally.

I agree: if the amount of light is changed (faster shutter speed) it will necessarily result in more photon shot noise. However, it's important to understand the difference between three distinct things: changing just the ISO; changing just the exposure; changing both at the same time.

Trainboy wrote in post #7718257 (external link)
Try bringing up a 3 stop underexposed picture and you'll see what I mean,

There is no camera or setting that can make up for actual exposure changes (f-number or shutter speed). The discussion is about what happens with ISO settings.

Trainboy wrote in post #7718257 (external link)
the shadow detail that had 0 exposure value is going to stay black unless you make it gray, and it will have no detail no matter what you do.

Well designed cameras don't have anything even remotely close to a 0 ("black") in them. They only go from white to noise. Even cameras that remove information from the raw file by clipping blacks at or above the mean read noise level aren't truly "black" at the 0. For the purpose of most raw conversions, the "0" only replaced what would have been a random noise value (no useful picture information).

In raw photography, black is simply a creative choice. Most raw conversion parameters are configured to place black well above the read noise level, so that no one ever sees the random noise that is at the bottom of every raw file. But it's there for any photographer who really wants to maximize the dynamic range of the camera. (Small outputs, such as web, wallet prints, DVD, etc. can utilize lower and lower black points/tone curves, to the point where one may even dip *below* the 1:1 S/N and get lots of image detail at very low spatial frequencies. This is what I do for a lot of my astrophotography and starry timelapse.)

Trainboy wrote in post #7718257 (external link)
Same with highlights. There definitely is a difference between ISO 1600 pushed to 3200 and a 5D set to 3200 originally, although this doesn't apply to expansion ISOs.

Agreed: ISO 1600 pushed to 3200 has the same noise as ISO 3200, but 1 stop more highlights.


Daniel

  
  LOG IN TO REPLY
sponsored links
(this ad will go away when you log in as a registered member)

8,474 views & 0 likes for this thread
What is the best "standard ISO" for the 5D MkII?
FORUMS Canon Cameras, Lenses & Accessories Canon EOS Digital Cameras 
AAA
x 1600
y 1600

Jump to forum...   •  Rules   •  Index   •  New posts   •  RTAT   •  'Best of'   •  Gallery   •  Gear   •  Reviews   •  Member list   •  Polls   •  Image rules   •  Search   •  Password reset

Not a member yet?
Register to forums
Registered members may log in to forums and access all the features: full search, image upload, follow forums, own gear list and ratings, likes, more forums, private messaging, thread follow, notifications, own gallery, all settings, view hosted photos, own reviews, see more and do more... and all is free. Don't be a stranger - register now and start posting!


COOKIES DISCLAIMER: This website uses cookies to improve your user experience. By using this site, you agree to our use of cookies and to our privacy policy.
Privacy policy and cookie usage info.


POWERED BY AMASS forum software 2.1forum software
version 2.1 /
code and design
by Pekka Saarinen ©
for photography-on-the.net

Latest registered member is michaelscheuren
755 guests, 187 members online
Simultaneous users record so far is 15144, that happened on Nov 22, 2018

Photography-on-the.net Digital Photography Forums is the website for photographers and all who love great photos, camera and post processing techniques, gear talk, discussion and sharing. Professionals, hobbyists, newbies and those who don't even own a camera -- all are welcome regardless of skill, favourite brand, gear, gender or age. Registering and usage is free.