Noise and ETTR

Started Feb 6, 2014 | Questions thread
Anders W Forum Pro • Posts: 21,466
Re: Noise and ETTR

cmpatti wrote:

Anders W wrote:

In addition to the link to the article you already got from GeorgianBay1939, you might want to have a look at this recipe for exposure with the E-M5 specifically:

I have a question about the procedure outlined in this post. It calls for setting the largest aperture and the slowest shutter speed that will get the image and then pushing the histogram to the right by raising ISO. It's been my impression that achieving ETTR by increasing ISO doesn't accomplish any meaningful noise reduction because the ISO increase offsets the benefits of ETTR. Therefore, I've thought that ETTR is only really useful in low contrast, well-lighted situations where you can increase exposure at base ISO by reducing shutter speed without blowing out highlights. Since those situations aren't all that common, and since noise isn't usually a major concern in those situations, I haven't really taken the trouble to incorporate ETTR in my metering routine. Instead, my practice has been "shoot at the lowest ISO at which you can get the shutter speed you need" (since I'm generally shooting aperture priority).

So my question: is there any testing available that shows that raising ISO to achieve ETTR reduces noise? In other words, that ISO 200, f/4, 1/100 (with highlights a stop from the right) will produce more noise than ISO 400, f/4, 1/100? This would be useful to know.

Yes, testing that demonstrates this has been done by me as well as others. The reason that ISO 400 is better than ISO 200, if we keep exposure the same and provided that we don't clip the highlights, is that the read noise of the sensor (as measured in electrons) is lower at ISO 400 than at ISO 200. This will make for less noise in the shadows. In the highlights, the noise level will be roughly the same regardless of which of the two ISOs you use (though still marginally better at 400).

How the read noise varies across ISOs can indirectly be assessed by looking at the DR curve in the DxOMark sensor tests. If the read noise (as measured in electrons) remains constant when you double the ISO, the expected loss in DR is one EV. If it is less than that, the read noise falls a bit.

As to the details for the E-M5 specifically, including intermediate ISOs, see here:

 Anders W's gear list:Anders W's gear list
Panasonic Lumix DMC-G1 Olympus OM-D E-M5 Olympus E-M1 Panasonic Lumix G Vario 14-45mm F3.5-5.6 ASPH OIS Panasonic Lumix G Vario 7-14mm F4 ASPH +28 more
Post (hide subjects) Posted by
(unknown member)
(unknown member)
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow