1st image at maximum zoom

[Chrissie]

Hi all,

 

I took this shot yesterday at dawn (technically: before sunrise), on a really clear and cold morning, from my homeplace patio. The foreground trees cover a hillcrest maybe 2 miles away, and the mountain peaks in the background are 70 miles (as the bird flies) away. The center peak is "Sustenhorn", 3504 m above sea level. I live at 520 m. PeakFinder view for your reference.

 

Being a technically oriented guy, I wanted to explore the limits of my gear (A9 and 100-400 GM) at maximum zoom level. Shot taken from a tripod, with wireless triggering. Focus and light setting on "Auto". I did some medium sharpening in Capture One Pro. On a scale from -100 to +100 I chose + 50.

 

This is not about artistic merit, composition and such. Just the "raw" performance. Please comment honestly - I can take it.  8-)

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[Guest Jaf-Photo]

When I click the image, it says it's private. The miniature won't reveal anything technical, so...

[Chrissie]

Thanks, Jaf, for giving this a try. Looks like I not only have a Sony Alpha learning curve ahead of me, nut also a flickr one, too.

 

Try this link, please. (Can't edit my previous post any more.)

 

Just to be sure, these are the exif data:

 

 

• JFIFVersion - 1.01
• X-Resolution - 300 dpi
• Y-Resolution - 300 dpi
• Make - SONY
• Software - Capture One 10 Macintosh
• ISO Speed - 1600
• Sensitivity Type - Recommended Exposure Index
• Recommended Exposure Index - 1600
• Exif Version - 0230
• Date and Time (Original) - 2017:11:22 07:23:35
• Date and Time (Digitized) - 2017:11:22 07:23:35
• Brightness Value - 1.8609375
• Exposure Bias - 0 EV
• Max Aperture Value - 5.6
• Metering Mode - Spot
• Light Source - Unknown
• File Source - Digital Camera
• Scene Type - Directly photographed
• Custom Rendered - Normal
• Exposure Mode - Auto
• White Balance - Auto
• Digital Zoom Ratio - 1
• Focal Length (35mm format) - 400 mm
• Scene Capture Type - Standard
• Contrast - Normal
• Saturation - Normal
• Sharpness - Normal
• Lens Info - 400mm f/5.595918691
• Lens Model - Sony FE 100-400mm F4.5-5.6 GM OSS (SEL100400GM)
• Compression - JPEG (old-style)
• Thumbnail Offset - 882
• Thumbnail Length - 1449
• Viewing Cond Illuminant - 19.6445 20.3718 16.8089
• Viewing Cond Surround - 3.92889 4.07439 3.36179
• Viewing Conditions Illuminant Type - D50
• Measurement Observer - CIE 1931
• Measurement Backing - 0 0 0
• Measurement Geometry - Unknown (0)
• Measurement Flare - 0.999%
• Measurement Illuminant - D65
• XMPToolkit - XMP Core 5.1.2
• Creator Tool - Capture One 10 Macintosh
• Make - SONY
• Flash Compensation - 0
• Camera ID - ****************************
• Camera Type - Digital SLR

[Chrissie]

Well, I'm not sure if it's good or bad style commenting one's own pictures, but this much I would still like to add:

 

actually I'm a little disappointed by the obvious lack of sharpness.  8-(

 

Next time I will try the following in hopes of improving the outcome:

• manual focus using focus magnification
• turn off steady shot in camera and lens
• try variations in ISO and (smaller) aperture.  (as the peaks don't move at any noticeable speed, I might as well take some more time ...)

 

Any further suggestions are highly welcome. 

[Guest Jaf-Photo]

Now it works. Technically, it is very good. The rendering of the trees is clean and detailed with distinct colours. Considering it's 400mm f5.6 ISO 1600, it's quite astonishing compared to what cameras used to do. That also means there's room for improvement by going f8 and ISO 100.

[Chrissie]

@Jaf, as I'm coming from low end gear (Canon EOS 550D and Tamron 18-270), I may have been expecting too much of an improvement from switching over to a Sony alpha.

Your comment gives me some relieve and also some incentive to keep optimizing. Thanks for that.

 

I still believe that mastering the craft comes before giving art a try.  ;-)

[Guest Jaf-Photo]

I've actually shot a few photos with 550D and 18-270 years ago. I remember it because it was fun to shoot with but disappointing to look at the photos afterwards. There were a lot of optical defects.

 

Cameras and optics follow the law of diminishing returns. You'll see less and less improvement for each step up in price point. And to achieve a picture free of distortion is a lot more expensive than achieving a picture tgat is almost free of distortion.

[oronet commander]

I believe it is quite hard to check sharpness in a picture like this one. First of all, I find the file quite oversharpened, specially because it's been taken a rather high ISO. Besides, when focusing a very far off object, air disturbances may affect sharpness a lot.

 

Try a closer object, with good light, and the results will improve IMHO

[Chrissie]

Thanks, oronet commander, for your feedback.

 

Frankly, I've looked at the specs, for instance 6°10'  FOV at 400mm, and I figured that equates to a width of 11.8 km at at distance of 110 km. Which at 24MPix (6000x4000) would equate to a horizontal resolution of 2m / pixel.  If it wasn't of course for a reduction in resolution by the limitations of the lense, which, according to DxO may account for a loss of sensor resolution of up to 50%.  Which I'm not expecting to fully "exploit" with a GM lense. But, let's go for the worst case: I have only 50% of my nominal resolution, which should still give me 4m / pixel at that distance.

 

That was my expectation. Which was disappointed.

 

Now where did I go wrong?

 

Thanks in advance for any further insight into this. 

[oronet commander]

I think your calculations may be valid only in a perfectly still atmosphere, but in real life, at that distances sharpness will be very limited by atmospheric conditions (dust in suspension, for example). Just look to the landscape and most day you'll see the mountains very washed out and lacking detail. That's why I suggest you testing your zoom in closer objects.

[Chrissie]

I think your calculations may be valid only in a perfectly still atmosphere, but in real life, at that distances sharpness will be very limited by atmospheric conditions (dust in suspension, for example). Just look to the landscape and most day you'll see the mountains very washed out and lacking detail. That's why I suggest you testing your zoom in closer objects.

 

 

Absolutely, and you are probably speaking from your experience in Spain, where these conditions may prevail.

That's why I chose the early morning hours on a pretty cold, late-fall morning. I wasn't expecting any dust in suspension. Otherwise, I would not have been ablate view this far. But, agreed, long distance shots do carry a lot of potential for atmospheric issues in general.

 

I will certainly exercise this lens at shorter range in the near future, and I'll keep you posted. 

[Guest Jaf-Photo]

I wasn't really sure where the original question came from. But this type of back-of-the-envelope maths really isn't going to be relevant. Obviously, there are complex optical simulations that are used in lens design and even animation. But really the only way to learn the envelope of your gear is to use it lots and study the results. I believe fachidiot is a word in German? Don't be one

[Chrissie]

Jaf, I'm trying to understand, how and why theory and praxis (is this actually an English word?) come to different results. Denouncing this as "Fachidiot" doesn't help.

 

Actually I'm trying to merge the experiences of various faculties into one unified understanding of photography as a whole.

[Guest Jaf-Photo]

Alright, I meant it as a jest. Of course it would be possible to create a mathematical model, but it would be highly complex. Apart from the sighting conditions it would have to describe how the lens treats the light at various settings and how it interacts with the sensor and the image processing hardware and software in the camera.

 

I'm not completely lost when it comes to maths but I found optical theory to be quite counter-intuitive. So I've just ignored the maths and concentrated on getting to know my lenses and how they behave on different cameras (there are differences). Also, when I studied photography, we had zero hours of optical theory but countless hours of taking and processing photos.

 

It is actually a lot easier to take a lot of photos and study them.

[Chrissie]

Jaf, no problem, I'm not fragile.  ;-)

 

And be assured: I'm not in the least inclined to simulate the lens as such. I consider it for all practical purposes a "black box". Taking the FOV of 4°10' which Sony gives for the 400mm position, it isn't really hard to figure out the width of this FOV at 110 km distance. This width in "real world terms" gets mapped to the width of the sensor, which is 6000 pixels for an A9. I'm not trying to simulate distortions, just average resolutions.

 

I know, this approach is very naïve, but I wanted to start somewhere.

 

And I fully agree with oronet commander and you, that aiming at a closer target than 110km away will definitely cut out much of the atmospheric "noise" between the lens and the target. But the general principle still applies: any "width" of the real world always gets projected onto the sensor's width. Which should allow for an estimation of the maximum resolution in [m/pixel] or even [mm/pixel].

 

Unfortunately, I'm not a professional photographer, but have a full-time indoor job which leaves me basically no daylight spare time during this time of the year in the northern hemisphere. Sadly, the practical exercises are currently heavily under-represented.