Commons:Photography critiques/November 2015

Question on technical image quality

These are images from a camera I bought last year. The camera is set to maximum resolution at maximum quality.

• 
  image 1
• 
  image 2
• 
  merged image

Does the image quality stand a chance at QI? -- KlausFoehl (talk) 13:43, 5 November 2015 (UTC)

• I'm sorry to say, but honestly, I don't think it is likely to get promoted. The main concern is that the highlights on the most interesting part of the image are blown out, and the top edge of the photo is very blurry indeed. Also, the detail is rather poor and noisy throughout. It's a shame, because the view is quite interesting. dllu (t,c) 16:30, 7 December 2015 (UTC)

Thank you for your candid answer. Actually, looking at the histogram, the image is quite nicely exposed to the right. But that is about all on the plus side. The view is nice, actually there is some technical challenge to bring a camera into that position. I may retake the image when the light is better, and with a proper camera. Thank you for that blurry, poor, noisy assessment which I share, I appreciate the independent confirmation. A pity because it was not a really cheap camera, and from a camera company with some reputation. -- KlausFoehl (talk) 22:17, 15 December 2015 (UTC)

I agree that you're probably being held back by your camera. Now, I'm a Nikon guy myself, but when image quality is actually a priority, it's always going to be hard to get good value for money from a point and shoot camera, no matter whose name is on it. Sensor size matters when it comes to landscape photography. The Coolpix S6600 has a 1/2.3" sensor – that's less than 28.5 square mm – about one thirteenth of a DX sensor and about one thirtieth of a full-frame sensor. What you paid for with the S6600, then, was not image quality, but pointless megapixels, the compact form factor and the flip-out screen.

Using a DSLR would have other benefits besides a larger sensor (and better glass in front of it). You would also be able to use filters, which might help with some of that haze and bring some detail out of that blue tint.

Equipment aside, whatever camera you use, you should always try to make the most of what you have when setting up your shot. According to the file's metadata, you had image stabilisation/vibration reduction turned on, so I'm guessing you shot handheld. For a photo like this, that will always produce a suboptimal result. For best results, use a tripod and make sure VR is off. I would not be surprised if some of that blotchiness is the result of vibration reduction. Since you're zoomed in slightly at 70 mm (in 35 mm equivalent terms), any camera shake will be exacerbated, and VR is going to have to work harder.

I don't know what kind of exposure control you have on that camera, but the point of merging exposures is usually to get highlight details from a generally underexposed shot and shadow details from a generally overexposed shot, which means that those exposures need to be made deliberately. In this case, the centre part of the castle is overexposed in both, and you don't need a histogram to tell you that – the sooner you stop looking at histograms and start looking at photos and trusting your eyes, the happier you will be. :-) —LX (talk, contribs) 19:11, 16 December 2015 (UTC)