It's funny how people keep reimplementing unsharp masking (the hard way).

The dSLR version is called Magic Lantern. http://www.magiclantern.fm

Since the camera thinks the exposure and ISO were identical on both shots, it couldn't have been a metering difference, nor is it likely there the lighting changed that much in roughly 1/6 second. The WB is on auto, but their doesn't appear to be a shift in hue, so that's also an unlikely culprit. This brightness difference shouldn't have happened.

The only thing I can think that might explain it is that the camera or lens may have a problem. Possible problems that come to mind are 1) a sticky shutter mechanism that ran slow and allowed more light than the CPU told it too, or 2) sticky aperture blades that didn't close down far enough the first time. Since the depth of field doesn't appear to have changed, the sticky aperture blades hypothesis can apparently be ruled out. So maybe a sluggish shutter mechanism. If so, it may have fixed itself just by cycling, but if it happens any more, I'd want to call Nikon and see what they think.

Your test results really surprise me given how highly thought-of the Tamron 24-70 is, but sure enough, it definitely looks soft on the right.

With this test, which looks like a pretty good one, it does depend on the lens axis being perpendicular to the building wall. It looks to me like the 24-70 is aimed more to the right than the other two. Is it possible the right hand side is just further away because the aim is slightly off perpendicular? How about re-doing the test by bracketing the aim spot a little between right and left?

That's an interesting test. Maybe we should all try it to get an objective measure of our equipment.

Here's a bit of simple and approximate math to go with it. The angular jitter on the wall is about:

(1/10 in/jitter) / (40 ft *12 in/ft) ~= 0.00021 radians/jitter

A pixel on your sensor is about 24mm/6000p ~= 0.004 mm/pixel long. So
at worst case (300mm focal length) with your camera&lens combo, a pixel will see an angle of:

(0.004 mm/p)/(300mm) ~= 0.000013 radians/pixel

So, there are about:

(0.00021 radians/jitter) / (0.000013 radians/pixel) ~= 16 pixels/jitter

and at 55mm, if would be about 1/5th that -- about 3 pixels/jitter.

I would be interesting to try lowing the jitter amplitude by trying things like hanging some weight off the tripod or even weighting the lens.

Nice experiment DerBiermeister.

I find the Godox S-Type Bowens bracket mount:

http://www.amazon.com/Godox-Bracket-Speedlite-Softbox-Honeycomb/dp/B00JS3MINC/ref=sr_1_2?ie=UTF8&qid=1457791440&sr=8-2&keywords=bowens+s-type+adapter

to be amazingly useful. They provide a very sturdy mount (unlike the shoe mounts) for the speedlight, and you can use umbrellas, softboxes that require a speedring, and Bowens mount modifiers: softboxes of all sorts, snoots, reflectors with optional barndoors, grids, and color filters, etc.

I've bought a number of these bracket mounts and modifiers with Bowens mounts, which allows me to mount any speedlight or the Winstro monolight, and attach pretty much any modifier to any speedlight regardless of brands. It's also turned out to be a good way to start small and grow as desired.

For those that like the D5x00 series, how to you deal with lenses that front or back focus? Do you only buy Sigma dockable lenses (or now, "TAP-able" Tamrons)?

Personally, though I love Nikon dSLRs that accommodate AF fine tuning, I would much prefer a mirrorless CDAF camera over a dSLR that can't quite get the focus right.

You don't say much about your setup. Is it possibly light going in through the viewfinder eyepiece?

I think this is good advise. The problem with D3000 and D5000 series bodies is that they don't allow lens-focus fine tuning. I personally don't like the idea of spending good money on lenses that front- or back-focus when using normal PDAF, and I've seen too many people disappointed when their camera won't focus at the right place with absolutely no way to deal with it other than purchase a higher-end body (D7000 or up).

Sigma's new Global Visions lenses solve that problem. They allow the fine tuning to happen in the lens so the body doesn't have to. If I had one of these non-tunable bodies, the only lenses I'd buy for it would be the tunable Sigmas. Period.

For what you are doing and with your camera body, the Sigma 17-70mm f/2.8-4 is the perfect solution.

Speaking of memories fading (which I'm as guilty as anyone), I see your message in this thread dated May 6, 12 18:47:13 says you got an email from Nikon saying UHS-I SDXC is supported at 35MB/s on the D7000.

It looks to me like the file was truncated and/or overwritten. RawTherapee shouldn't have touched the file other than reading it. Do you have something that might have accidentally opened it for writing?

--alan

You're not using Linux/Unix yet? My condolences.

Darktable. Rawtherapee and UFRaw are also good.

I'm sort of baffled by this. Floyd, you're a smart guy, and it's always worth considering what you are saying, but I've seen times when you argue with someone about something other than what the someone is talking about. Is that happening here?

Perhaps you could help clear that up with an example. Searcher posted a DOF example from the CambridgeInColour website DoF calculator:



It certainly looks to me like the only thing that has changed is the focal length on the inputs, and the DoF is certainly different -- by more than a little. This is in agreement with other DoF calculators I've seen, including the one I wrote and use. So could you post the DoF results you think should be coming out of the calculator with those inputs? Or perhaps, concretely, with a similar example, why you are arguing something different?

I'm baffled why you'd say DoF has nothing to do with focal length as the equation for DoF clearly depends on focal length and every DoF calculator (and the math behind it) I've seen requires the focal length as input.

So are you talking about the same thing the rest of us are?

It couldn't be easier. Just slap in a nodal plate such as this one from Amazon <http://www.amazon.com/SUNWAYFOTO-100mm-Compatible-DMP-100-Sunway/dp/B005SUMB7M/ref=sr_1_2?ie=UTF8&qid=1425576510&sr=8-2&keywords=arca-swiss+nodal+plate> and shoot away.