Raw Dynamic Range

ISO Invariance

A camera with a very low noise floor is able to capture a large amount of dynamic range, since it adds very little noise to the detail captured in the shadow regions of the image. This has an interesting implication: it minimizes the need to amplify the sensor's signal in order to keep it above that noise floor (which is what ISO amplification conventionally does). This provides an alternate way of working in situations that would traditionally demand higher ISO settings.

Here we've done something that may seem counter-intuitive: we've used the same aperture and shutter speed at different ISO settings to see how much difference there is between shooting at a particular ISO setting (and using hardware amplification) vs. digitally correcting the brightness, later. This has the advantage that all the shots should exhibit the same shot noise and any differences must have been contributed by the camera's circuitry.

While there is a difference between shooting the 80D at ISO 3200 vs shooting the same exposure at ISO 100 and pushing it digitally later, this difference is much smaller than it has been traditionally for Canon. The significance is that in situations traditionally demanding high ISOs, you can dial down the ISO setting - while still using the shutter speed and aperture you would've used at the higher ISO - and then brighten the resulting (dark) Raw image selectively in post, protecting highlights that are now preserved because they weren't clipped to white due to the higher amplification of an elevated ISO setting. The relatively low read noise of the camera means that when you do so, noise levels will remain relatively well controlled, and only slightly more than what you might have obtained had you shot with a higher ISO setting.

Compare the 80D with the EOS 7D Mark II and you see a rather different story: the older camera is adding significant levels of noise to shadows that limit your ability to brighten them in post. With the 7D Mark II, and most previous Canon cameras, it was more imperative to use the 'correct' ISO in-camera, which meant you were limited in your ability to protect highlights by purposefully underexposing. That said, the performance of the 80D isn't quite as good as the likes of the Nikon D7200, which we classify as almost perfectly ISO-invariant, but it's a big step forward for Canon.

Summing up

Our Raw dynamic range tests appear to confirm the success of this major change in Canon's sensor design. The results don't look quite as impressive as its Sony-based contemporaries but the reduction in noise contribution should provide an appreciable real-world benefit when it comes to protecting highlights without adding too much noise in the shadows. This should make Canon's Highlight Tone Priority more useful (it's traditionally often introduced more noise than desirable).

Here's a slightly quick-and-dirty Adobe Camera Raw conversion of this image. We're not suggesting its this capability is for 'rescuing' otherwise poorly exposed images that it valuable, but it works as an example of where you might need to balance the bright and shadow regions of an image, and the benefits of being able to do so without adding too much noise.

Canon's previous design allowed electronic noise to accumulate after the amplification stage (which we refer to as 'downstream read noise'), which had minimal impact at high ISO settings but limited the processing flexibility of low ISO files by adding noise that was visible if you tried to lift shadows or increase contrast. While the results of the 80D don't appear to quite match up to the best of its peers, this is good news for anyone committed to Canon's EF lens mount. Now that it's moved across to an on-chip column ADC architecture, we should expect to see further improvements in future generations as the design is optimized.