Photodo / Canon MTF Cross Validation

Note that photodo do not supply data for 30 lp/mm to compare with Canon so these data are interpolated by curve fitting the 20 and 30lp/mm photodo measurements with the following formula that models the lens as a low pass filter of spatial frequencies:

Another point is that Photodo provide mid focal length information for zooms in addition to the wide and tele that Canon provide. The Canon data is point for point linear interpolated to estimate the mid zoom performance.

Finally, Photodo to not make it clear exactly how they capture the data. From the vintage of the data digital capture is ruled out considering the >80 pixel/mm top sample rate. So the lenses may be measured on an optical bench or capture to film. If film is used then the film will contribute some loss to the MTF, particularly at the higher spatial frequencies.

Results

The summary statistics for prime and zoom lenses against focal length are presented in the below plots.

Comparing the two plots above the following generalized conclusions can be drawn:

1) The peak discrepancy between Canon prime lens data and measured performance is greater below 30mm. However the discrepancy can be better or worse than expected from the Canon data.

2) Zoom lenses show a larger discrepancy between the Canon data and measured than prime lenses except perhaps below 30mm. Again this can be better or worse.

3) On Average the Canon data and Photodo measurements are in quite close agreement, with the Photodo measurement being about 0.05+/-0.2 to 2 sigma of the Canon prediction.

The below plot separates the error statistics for the zooms at wide and tele from the mid point. Although the mid point is supplied by Photodo is has to be interpolated from the two end points for the Canon data.

These data suggest the interpolation of zoom focal length for Canon data is reasonable but possibly slightly optimistic for telephoto zooms.

Conclusions

The Canon MTF data is quite a good guide to relative sharpness. Linear interpellation between wide and tele focal lengths for zoom lenses provides a reasonable estimate of lens performance.

Peak discrepancies are larger for L zooms at all focal lengths and prime lenses below 30mm focal length. This probably reflects an increased sensitivity to production tolerances.

What is not clear from this is how much the variations between lenses is due to the lens design or just due to the individual sample of lenses tested.

Last Updated 05/06/2008

All material is supplied as is and without warranty, use at your own risk.

All opinions stated are the authors own.

All quoted information remains the copyright of the respective authors.

Any trade marks mentioned are the property of their owners.

Contact: (anti spam remove underscores)

Note this email gets spammed a lot so if you want to get your message through please prefix "Holding Tank" in the title

f_ w_ a_ r_ e_ 9_ 0_ @_l_i_n_e_o_n_e_._n_e_t