The Economist article on mirrorless cameras

Started 6 months ago | Discussions thread
Jerry Fusselman
Regular MemberPosts: 431Gear list
Like?
Re: Answer 2
In reply to Erik Magnuson, 6 months ago

Erik Magnuson wrote:

Jerry Fusselman wrote:

I don't plan to quibble on what beats means except to say that higher is better in these cases. Win a marathon by 0.01 seconds and you still won. You still beat the other guy.

Photography is not a race. That's a specious internet argument. (Hint: do you know what is the error margin for those measurements?)

I was mainly trying to show that Sony's best sensor is *not inferior* to Canon's best in low light. It is icing on the cake for it to be higher when told it surely must be much lower (and not going to catch up even in three years), even if higher by just a smidgen.

And here we are in violent agreement.

I can't imagine this quibble causing a violent agreement.  Can we call it equal or better?  Or not worse?  I hardly think the label matters so much if we agree on the basic conclusions.

First, that 1-bit superiority in color depth is surely significant: It equals the standard improvement in color depth that you get when going from ISO 200 to ISO 100. That's plenty visible for many subjects.

For color depth? Take two shots, one at ISO 100 and one at ISO 200 and see if you can tell the difference when the EXIF is stripped.

Yeah, I think I can.  Perhaps you can state the experiment you have in mind more completely.

Second, the low-light ISO score takes into account several issues

Nope. Read the last section on this page:

http://www.dxomark.com/About/Sensor-scores/Use-Case-Scores

"An SNR value of 30dB means excellent image quality. Thus low-light ISO is the highest ISO setting for a camera that allows it to achieve an SNR of 30dB [...]."

Oh, I think I see where you are coming from here now.  I think you are right if you are saying that their low-light ISO score is not the best.  I think I could show that the three discontinuities in their definition must lead to inadmissible decisions (in Decision Theory jargon).  Yes, I agree with you if you are highlighting the arbitrariness of 30bB and if you are saying that it is better to focus on the data from the curves instead.  Do I seem to understand you better on this part now?

-- hide signature --

Jerry Fusselman

 Jerry Fusselman's gear list:Jerry Fusselman's gear list
Canon EOS 5D Mark II Canon EOS 7D Canon EOS 5D Mark III Sony Alpha NEX-6 Canon EOS 100D +40 more
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