In an old article, I mentioned that a Canon 18-55mm at high end isnâ€™t that different than the Nikon 17-70mm kit lens in focal length (reach).
My statement was:
After all, you can just shoot the Canon at 55mm @ 1.6x and then crop it down to a 6 megapixel photo and it will look close enough to the Nikon D70 at 70mm at 1.5x.
Iâ€™m surprised nobody has called me out on this statement!
Letâ€™s do the math on the 35mm equivalents: a 55mm 1.6x = 88mm and 70mm 1.5x = 105mm. How different is that? A trick I learned from my Physics graduate advisor: All useful numbers in the world are dimensionless parameters. We have to generate a dimensionless parameter to give meaning to this difference: what we need here is a ratio. How about comparing it to the 35mm format?
55mm/35mm = 1.6. Applying my â€œcropâ€ statement: 1.6*(Sqrt(8M)/Sqrt(6M)) = 1.8. The Nikon example works out to: 105mm/35mm = 3. Hmm thatâ€™s not the same at all!
An exception proves the rule
What led to my statement was the general rule of thumb: extra focal length makes a difference on the low end than the high. In this case the 70mm on the Nikon still has 50% more reach than the 50mm on the Canon.
Of course weâ€™re talking about 20mm here. When does 2mm make a big difference?
A recent Flickr thread debated between the purchase of the Sigma 10-20mm f/4-5.6 and the Tokina 12-24mm f/4.
Now if you interested in the actual answer to this question, I politely refer you to this venerable Nikonians article which does the best job of comparing them Iâ€™ve seen. How? Well after slogging through Ken Rockwellâ€™s treatise1 I can say that he came to basically an identical conclusion.
The part I want to bring up is when the discussion focused on if you can notice the difference of 2mm. Letâ€™s apply our battery of knowledge.
12mm*1.5/35mm = .51 and 10mm*1.5/35mm = .42.
Yep, that difference (20% I could have just as easily did the ration 12mm/10mm and got the same number) is noticeable. In the thread, I claimed it made â€œa huge difference at wide anglesâ€ but that statement is incorrect. It should have been â€œmakes a noticeable, but not huge difference.â€
So when does 2mm make a huge difference?
The answer is when you compare the 10.5mm Fisheye Nikkor to my 12-24mm.
The field of view on it is 180 degrees, while the 12-24mm Nikkor has 99 degrees at wide angle. Wow!
Why? Because itâ€™s a fisheye!
As I mentioned in the thread:
The 10.4mm Nikkor has defishing capabilities if you use Nikon Capture, Peter iNovaâ€™s action, LensFix or DxO Optics pro.
I mention this because if you want to go really wide, thatâ€™s probably the way to do it. You have a fisheye and a really wide angle. Youâ€™ll lose resolution near the edges and composition is going to be a pain, but a 10mm fisheye goes much wider than a 10mm rectilinear so the FoV is phenomenal.
How to get even wider
Take a bunch of shots with your 12-24mm and then stitch it together in spherical mode. If you squint a bit, you can pretend you have a fisheye. 🙂
Of course, you could do the reverse. To know what a de-fished Nikkor might shoot, here is a rectilinear stitch:
I wish I could capture the vertigo I felt taking that shot.
Iâ€™m curious how Canon was able to do this. Here is and found their optical design in Bobâ€™s review compared to Nikons and theyâ€™re not the same at all! Iâ€™m curious if this is an example of the uniqueness of the EF-S mount (the mount allows lens elements to be further back in the box. The idea is that you donâ€™t need as powerful â€œa telephoto in reverseâ€ optics in order to make the light telecentric (important for digital photography).
Does anyone know?