Disperser Tracks

So, I’m practicing shooting birds . . . er . . . photographing birds. I’m looking forward to capturing . . . er . . . photographing Red-tailed Hawks again, but, the thing is, I’m out of practice photographing birds.  

I mean, I’m out of practice (I’m going to use “shooting” but read “photographing”) shooting birds in adverse conditions. Meaning, when it’s frigging cold out and when it’s overcast (less light) and when birds don’t stand still.

What do I mean? Well, for instance, I have the P900 that can zoom really up-close . . . but is a slow system with a small sensor. Not that it doesn’t give decent results . . . 

OK, that’s not decent. That’s less than decent. That’s slightly cropped, but pretty close to the framing of the camera. 

This is not bad . . . 

Nor are these next ones. By the way, you can click on all these photos for a larger version.

I can guarantee those would look better in better lighting conditions . . . but that’s not always available. 

Again, those don’t look too bad but realize that I’ve post-processed them. Even so, there are diminishing returns when processing a photo. 

Let me illustrate with this headshot of a heron. This was shot with the P900 at full zoom from about 30 yards away.

That’s the shot right out of the camera without any processing at all. You can click the image for a version that’s roughly twice as large.

It doesn’t look bad, but if you had access to the original (you don’t since I didn’t do a SmugMug gallery) and looked at it at the pixel level, this is what you would see (this a crop of the actual size; click on it for a slightly larger version).

Again, no one really looks at photos at that magnification unless they are photographers. 

Mind you, a lot of my photos can be viewed at that level of magnification and still look decent . . . but, not usually the P900 photos.

Anyway, I took that photo and processed it with various tools to see how much I could improve both the presentation and also the pixel-level view.

Brightness, contrast, exposure, tonality, saturation, color balance, etc.

That’s just tweaks in brightness and contrast and a few other things that can give the perception of increased sharpness.

This next mod increases the sharpness. 

Increase sharpness – Nik Tools

Increasing sharpness of a fuzzy photo usually makes the full photo look better but does nothing to improve how it looks at the pixel level (adds the equivalent of noise) 

Note: I ain’t a pro . . . my parlance is not necessarily technically correct; I’m just describing what I see and what it looks like to me. 

There are a couple of different ways you can increase sharpness but the results are pretty close to each other. If I increase sharpness, I usually also apply some amount of noise reduction (smoothing, if you will).

Increase sharpness and noise reduction – Nik’s tools

This has the effect of reducing the sharpness so you have two tools working against each other but their combined use is usually better than the original. Meaning, you shoot for having a tad more sharpness without distorting the details at the pixel level.

Almost everyone now has some sort of AI-driven tool that purports to remove noise and increase sharpness and add detail. One such tool is in Topaz studio (AI-Clear).

Remove noise and sharpen and add detail – Topaz AI-Clear

With a lot of these, it’s a matter of personal preference as to whether one tool is better than another. AI-Clear does, for me, reduce the amount of tweaking I need to do and it’s a one-click solution that most of the time gives me something I like (the photos at the beginning of the post were all processed with Nik Color Efex 4 and finished off with AI-Clear). 

I have one more version . . . 

All sorts of stuff

That was first processed through Topaz new JPG-to-RAW stand-alone program. The tool takes your JPG file and supposedly adds in stuff to turn it into a RAW file (it’s output in DNG format). 

Converting a JPG file to RAW allows the user greater flexibility in processing the photo (RAW has more information than JPG). Basically, the tool is supposed to put back the information saving the file in JPG had taken out. After the conversion, I used Nik tools and Topaz and added a few adjustments in Lightroom.

I don’t know if the end result is better than the other versions.

I’m still playing with it, but this is the 100% crop of the above . . . 

What the conversion allowed me to do is be more aggressive with the processing. Perhaps that’s good, perhaps not. 

Let me reiterate that the limiting factor here is the P900’s lack of RAW output and its poor low-light capability which is related to the fact it has a small sensor. It matters not what tool I use, I can’t create resolution that’s not there. Well, that’s not strictly true . . . I could use AI-Gigapixel and double the size of the original and then process it . . .   

Doubled in size and processed and output at 1280pixels width.

Does it look better? Does it look different?

Here is the 100% crop (you have to click on it to see the actual size) . . . 

Useful if I want to print something the size of a wall but I’m not really increasing the amount of information in the original. I’m just enlarging by adding pixels and interpolating their content from the existing pixels. Whatever I start with is what I end up with (almost — there is a bit of degradation at the pixel level because we’re making up stuff).

Again, there is a limit with what you can do with the P900 because the tool is limited. 

Let me reiterate that it’s perfectly fine for 90% of what I do because I don’t typically show the pixel-level data of the P900 photos and the photos look good enough at the resolution I do show. 

But, let me throw you (and me) a curve . . . Here’s a photo from the D7000 with the Nikon 70-200mm f/2.8 lens . . .

Notice how clean it is? I mean, yes, it’s at 1/10th the zoom magnification of the P900 but that’s how it came out of the camera. 

But, here’s the full-size pixel level with just a touch of sharpening. 

That’s at the same distance as the photo I showed at the beginning of the post; look how much I cropped the D7000 photo to get the above and still retain the detail and quality. Plus, the above has minimal processing whereas all of the P900 photos get treated to a number of adjustments.

The first photo was hardly cropped and cropping it further would only diminish the quality of the image. Here’s that photo again for a direct comparison.

Enough said.

No, wait . . . That’s the quandary I face. The P900 is fine for what it does but it lacks in many respects. The P1000 addresses some of those concerns. 

BUT . . . if I’m to consider the P1000, I’m now talking close to the same amount of money as Nikon’s excellent 200-500mm lens. If I can get the above resolution from the Nikon 80-200mm lens, what would I be missing (or gaining) if I choose the Nikon 200-500mm versus the P1000’s 125x  (3000mm equivalent) zoom? 

I don’t know. I plan to rent the lens and see if I would be happy with it. 

Here’s the gallery of the above photos:

All sorts of stuff

Doubled in size and processed and output at 1280pixels width.

That’s it. This post has ended . . . except for the stuff below.

<><><><><><><><o><><><><><><><><><o><><><><><><><>

Note: if you are not reading this blog post at DisperserTracks.com, know that it has been copied without permission, and likely is being used by someone with nefarious intention, like attracting you to a malware-infested website.  Could be they also torture small mammals.

<><><><><><><><o><><><><><><><><><o><><><><><><><>

If you’re new to this blog, it might be a good idea to read the FAQ page. If you’re considering subscribing to this blog, it’s definitively a good idea to read both the About page and the FAQ page.