Gents
Please excuse my complete ignorance with regards to shifting but I don't get it! If the image circle is the image circle, then what difference does the size of the sensor make? If you can shift a crop sensor further in millimeters, why does that change anything over a larger sensor? Lets say I shift my back 20mm, see the edge of the image circle and crop back in to what is acceptable, why would it be better or worse to shift a smaller sensor further to get the same view? I am happy to accept I have no idea what i'm talking about as I don't have access to a crop sensor to test the differences, I can't help feeling though that the limitation is the image circle not the sensor size? In extremes, if I take 9 shots at full movements, rise/fall/shift with any sensor until I am hitting the edge of acceptable image quality, won't the field of view be the same?
Sorry for stupid question number 119!
Mat
Yes, you are right. If you take a small sensor and work within the image circle, and stitch a number of images together, you can approximate the results of a larger sensor. Say for example you have a sensor 2cm square. If you put four images together, its the same as having a 4cm square sensor. A few cautions tho:
- has to be back movements, not lens movements, as that will change the point of view
- requires precision and time to take multiple shots vs. 1
- there needs to be some overlaps (for the matching) so perhaps your 2cm stitch yields a 3-3.5 cm sq image
- hassle in processing. Its not just the stitching that's an issue, but corrections need to be done either before on all 4 images or after.
There is more benefit from the larger pixels in terms of flexibility, and by many folks, quality. Several reasons other than stitching for this - of which pixel count is only one. That said, you can work by stitching, and many of us with MFDB actually stitch our images to approximate even bigger sensors which are either not available or not affordable.
The general thinking is a carryover from large format days, that a larger the recording surface (film, sensor, etc.) provides a better result. Less stress on the lenses, higher quality, more latitude, etc. However, progress in smaller sensors in recent years has put a big dent in that thinking, but it still has some relevance. Also smaller sensors (with smaller pixels) are less forgiving of off-axis light - see Torger's many posts on this issue.