More test results from this past week.  I already know that 14-bit isn’t any better than 12-bit at base ISO on the D700.  Well, let’s just make sure things haven’t changed with new gear.

There is (quite wisely) no “uncompressed” raw on the D7100.  But if you’re foolish enough to want to waste space, there are “losslessly compressed” and “compressed” modes, in addition to being able to select between 12 and 14 bit files.  My preference is the smallest file, since that helps the buffer dump to the card faster.  Lets see if I lose anything choosing 12-bit compressed (allegedly lossy) vs 14-bit lossless, first just comparing the overall scene before manipulation:

There’s a pretty good hot spot in between those trees.  Let’s dial down the exposure and see if one setting is capturing more in the highlights:

Can you tell the difference?  There’s no posterization.  Now you can complain that the highlights didn’t blow, however, the point is that even in the 12-bit lossy version, there’s more highlight than I need.  Unless you’re doing something wrong, you probably don’t need 14-bit for highlights with the D7100.

Now let’s look at the shadows (artificially pulled up in post):

Not seeing anything different here.

Now the point is not that there’s no difference.  The point is that for actual real world shooting, there is no appreciable difference.  Don’t waste your time or card space on 14-bit files.  If the review sites were smarter or more honest, they’d tell you the same.

Of course, if you want to come up with a reproducible test with real world repercussions, please, by all means do so.  And share your results like I have!

Here’s my original raw files if you want to play with them yourself.