Sigma 8mm fisheye

[ifurlong]

has anyone used the Sigma
8mm f/4.0 EX Circular Fisheye ?, if so, on a 10d is the image full frame or not, an is this a good lens? thanks, Ian.

[justme_dc]

well the canon 10D has a 22.7 x 15.1 mm CMOS sensor and the sigma 8mm 180º fisheye has a 22.08mm circle so your answer is almost. you'll have plenty of coverage on the 15.1mm vertical end but you have a lack of coverage on the horizontal end of .62mm which will require cropping the edges of your images slightly.

I have the 15mm fisheye that sigma makes. It's worthless wide open but if you stop down past f8 it does an ok job. I suspect the 8mm will be similar. If you buy one be sure to post some shots.

[Guillermo Freige]

justme:
Your caculation has a little mistake. Remember the 22.08mm is the diameter of a circle, and 22.7 is the horizontal dimension of a 22.7x15.1 rectangle, so the diagonal is larger (27.26mm) so you will have a 2.59mm lack of coverage in the corners, and only a 0.32mm lack in the middle of the rectangle at both sides, so more than a slightly crop is needed

[TomKa]

Here are some pics and a tutorial for panos with the Sigma lens and the 10D.
http://www.chem.ox.ac.uk/oxfordtour/tutorial.html

[justme_dc]

Guillermo Freige wrote:
justme:
Your caculation has a little mistake. Remember the 22.08mm is the diameter of a circle, and 22.7 is the horizontal dimension of a 22.7x15.1 rectangle, so the diagonal is larger (27.26mm) so you will have a 2.59mm lack of coverage in the corners, and only a 0.32mm lack in the middle of the rectangle at both sides, so more than a slightly crop is needed

Your calculation of 0.32mm per side is only 0.02mm in total difference from my figure. At that point you are literally splitting hairs. Further, I assumed everyone would understand that I was referencing an Image circle.The lens is a Curcular fishey afterall. As such, there will be more image missing from the corners of the rectangle than from the center of the sides. That is why I didn't feel the need to point it out. Also, I said a "slight" crop this is obviously not an exact figure, so arguing an arbitrary like that is pointless and ultimatley self serving.

If you had the answer to Ian's question why didn't you anwer him instead of just coming along to "correct" my math. Both my post and tomka's were helpful and constructive. Yours amounts to nitpicking.