Open letter to photographyblog.com (Review E-PL5; Nikon J2)

Started Oct 30, 2012 | Discussions thread
Entropius
Veteran MemberPosts: 4,167
Like?
Re: Two ponts
In reply to caissam, Nov 1, 2012

caissam wrote:

Entropius wrote:

The point is to see how ISO 12800 behaves.

yes, but you have to check this in combination with the behavior of the noise reduction. we are talking about jpgs!

It is true that jpegs test the camera's processing in combination with the sensor response. Perhaps that is an interesting thing to test (for some people it will be, I imagine); in any case it's what they've done.

and therefore you should use it under conditions where you need it or you normally use it in practice.

Not necessarily; you could also use it under conditions that give you the same information as what you'd get in practice. Nobody shoots test charts in practice, but it's possible to use pictures of test charts to judge whether a lens will be sharp in the field.

Likewise, while it's unlikely (but not unheard of) to shoot at settings like ISO 12800 f/11 1/200, doing so will give you the same information about a camera's high ISO performance as shooting at ISO 12800 f/1.4 1/200 or ISO 12800 f/11 1/3 in light that is 64 times dimmer, assuming no change in white balance. *All* that matters is the total amount of light falling on the sensor, which is the product of

(scene illumination) * (aperture) * (exposure time)

In other words, closing the iris or using a faster shutter speed affects exposure exactly the same as does turning down the lights.

Yes, in testing ISO 12800 they picked a combination of these three things that you're not likely to use in practice. But they did what they did because they didn't have a way to decrease the scene illumination without affecting things like the evenness of the lighting or the white balance, and they wanted to answer the question "How much noisier is ISO 12800 than ISO 1600?" by changing as few things as possible. Since it's easier to change the shutter speed or aperture than to make the lighting uniformly dimmer (which is hard!), they did those things for their test.

As far as diffraction, it really is an overblown thing at apertures below f/13 or so on Four Thirds; yes, it's measurable, but it's not nearly as big of an effect as people think it is.

Reply   Reply with quote   Complain
Post (hide subjects)Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark post MMy threads
Color scheme? Blue / Yellow