Darktable.

Some time ago, I used UFRaw, which worked quite well for RAW conversion, but eventually replaced it with Rawtherapee when it became stable on the Unix/Linux side. It offers a lot more than UFRaw with a very sophisticated (32-bit float) calculation engine that works well not only for RAW files, but also JPEG, PNG, and other image formats. The biggest weakness of these two are that operations are global, so all local edits needed to be done on GIMP or similar.

GIMP is a very good editor with two main weaknesses -- if has to offload RAW conversion (but then so does Photoshop), and it only had a 8-bit engine up-to and including version 2.8. Fortunately, GIMP 2.9, though labeled as a development version, is very stable and usable, and has a calculation engine that'll work with pretty much anything from 8-bit integer to 64-bit float.

When Darktable stabilized on Unix/Linux, I moved to it because it's internal calculations are 32-bit float like Rawtherapee's, but it also allows local edits. There are still some things that are a bit awkward to do in Darktable which require follow-up edits in GIMP, but those are quite rare. I almost always do all the edits I want, very quickly (usually less than a minute) in Darktable and I'm done. It's very powerful and quick.

Version 2.2 of Darktable will be released soon with functionality that'll give even less reason for a GIMP followup.

For most practical situations, it might as well be true, but I think it would be more accurate to say that it is *very, very* close in most realistic situations. Here's an example.

50mm at 10' has a view width of 7.08',
85mm gets that width at 17'.

The DoF for the 50mm at f/4 is 2.83' and for the 85mm at f/4 is 2.80'.

That's pretty close -- only about 1% difference. For something as fuzzy as DoF, I know I wouldn't be able to see that difference.

But there is a difference, and interestingly it is *because* of the perspective change you note.

So, let's see if we can make the difference a bit more apparent. Let's choose 14mm and 300mm at f/2.8, and let's choose a very specific distance -- the hyperfocal distance of the 14mm at f/2.8 -- 7.97':

14mm at 7.97' has view width of 20.38',
300mm has that width at a distance of 170.8'.

The DoF for the 14mm is infinite (since it's the hyperfocal distance), but the DOF for the 300mm at f/2.8 is only 15.98'.

I don't know if this sort of nit is important for things we're likely to do, but it is a sort-of interesting side note.

I recently ran into the exact same problem. It turned out that for 64GB devices, Nikon formats them with the exfat filesystem -- not the usual fat or vfat. I use BSD Unix and Linux, and neither by default could mount my SD card, both giving errors similar to yours. I had to find packages on each that would read/write the exfat filesystem. Once installed, all was good.

I'm not familiar with MACs, but they do use BSD Unix under their GUI layer, so there may be similarities in how you deal with it. I had to install a 'fuse' loadable kernel module, and then used the 'exfat-fuse' mount. In any case, search for 'exfat' filesystem mount capability.

I don't own one, but I think if I were in your position, I'd strongly consider the Sigma 50-100mm f/1.8 Art lens. I'm thinking that for portraits using an APS-C sensor, you probably couldn't find a better lens, offering exceptionally high quality and flexibilty -- perhaps even exceeding a 70-200mm f/2.8 on FF.

Also, since you are using a d3200, having a "dock-able" lens for fine-tune AF adjustment would be a huge plus, and this lens offers manual focus override in AF mode, so you can use it that way too.

I don't disagree with you. It doesn't look like there's much room left for sensor image quality improvement.

I don't know what the engineering compromises that led to the low-ISO DR drop on the D5 were, but I suspect it was ADC speed. From what I've seen, ADC's can be high-speed, or high-bit-depth, but getting both is hard. With the higher FPS and MPixels of the D5, the current technology just may not have been able to give the low-ISO DR that the other Sony sensors get, and perhaps that is what SensScore was dissappointed about.

The fact that the camera's ISO isn't the same as "measured" (or perhaps, "true") ISO as shown in Bill's "Measured ISO" charts:

http://www.photonstophotos.net/Charts/Measured_ISO.htm#Nikon%20D5

shows that most camera's shift the ISO in a way that causes their DR vs ISO curve to shift to the right. If the ISO plotted in was "measured" rather than the manufacturer's "provided" value, the curve would be further to the left of the Ideal FX line.

To be fair though, pretty much all cameras today, including the D4s and D5, have some ISO shift to the right that makes them, look better than they ought to compared to the Ideal FX line, leaving a bit more room for improvement than the chart suggests, right? Still, it's amazing how close today's sensors are to the ideal.

Perhaps to improve speed, the firmware commands higher speed with predictable, but different, overshoot? It still surprises me. However, if you're seeing it, I don't doubt it's real.

Wow, that surprises me. Any idea why a firmware change would do that?

If you want to test the sharpness, you'll need to use liveview mode, and/or focus manually.

Unfortunately, neither camera you mention here can be focus-fine-tuned, so sharpness using the normal PDAF focusing is luck-of-the-draw. My experience is that about 70-80% of the lenses I've purchased need to be fine-tuned. If you want to consistently see the sharpness potential from a lens, you'll need to consider the mid-range cameras (Canon 70d, 80d; Nikon d7100, d7200) the minimum acceptable bodies.

So, yes, send the d5500 back, and replace it with an 80d or d7200. Then fine-tune each of your lenses. Then compare sharpness.

Gotcha. Enjoy your trip.

You might be right, but the whole 1st picture is less sharp, including nearer the plane of focus. It looks like motion blur to me. Taking that into account, as well as the shift in brightness, to me it looks like the DoF hasn't changed.

So Chase4, have you tried repeating the experiment? As I pointed out earlier, the sticky aperture blade hypothesis can be ruled out since the DoF is the same in both photos, so I'd focus on doing a Ch burst at 1/3000 sec at any exposure-appropriate ISO and F# to test for a sticky shutter.

I suspect he was misreading the EXIF data. It contains a number of data-times, including the filesystem metadata for the actual files. I'd guess he downloaded them 42 seconds apart.

The actual image creation times show 0.20 seconds difference.

Yep.