Pls show examples where increasing brightness in post better than increasing ISO.

Started Oct 1, 2013 | Discussions thread
ProfHankD Veteran Member • Posts: 6,424
Re: The meaning of ISO, NEX-7 vs. EOS 7D

wchutt wrote:

ProfHankD wrote:

ISO film speed is a bit of a judgement call for sensors used in digital cameras. Basically, the notion of ISO change is implemented by two things:

  • Change of the exposure settings (aperture and shutter speed)
  • Change of the sensor gain

It is physically impossible to increase or decrease the sensor gain.

All sensors operate at a single gain value. Any change in gain (electronic or digital) occurs after the shutter closes.

Yes, gain is applied after the integration period ends (with or without a shutter); I wasn't referring to something like adjusting the quantum efficiency, which is impractical if not impossible. The sensor gain is the gain (amplification) of the analog sensor outputs as they go into the ADCs.

In principle, there should be no difference between increasing brightness (not exposure) by increasing (ISO) using electronic amplification or digital multiplication during post-processing. However the actual results will be highly dependent on the specific camera being used. Besides the convenience of viewing results immediately, the primary reason to use in-camera gain is to get the most out the analog-to-digital converter. Some cameras have ADCs that require no electronic gain to accommodate the ADC technology. The Nikon D7000 and Phase One digital backs are just two examples.

That's what I believe this thread is about. The "ISO-less" sensors would then be the ones where the analog amplification going into the ADCs is essentially a constant.  Then again, the raw files do have different values for different ISOs given the same exposure settings, so something is causing that. In addition, Sony has said some things in the past about dynamically adjusting analog values on sensor to reduce noise -- perhaps compensating for differences between ADCs by incorporating some type of reference level check?

That brings up the question of why postprocessing applying the same digital gain (tone mapping) that is done in-camera would do better than the presumably well-tuned implementation in the camera.  It could just be it isn't so well tuned in some cameras.

I've decided to have my students do a little test using CHDK to compare high camera ISO to low ISO + digital gain on the raw. With CHDK, this can be done using the standard conversion to JPEG in the camera, removing any potential bias in tone mapping.  We'll see what they find....

 ProfHankD's gear list:ProfHankD's gear list
Canon PowerShot SX530 Olympus TG-860 Sony a7R II Canon EOS 5D Mark IV Sony a6500 +30 more
Post (hide subjects) Posted by
(unknown member)
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow