MTF

MTF charts (short for Modulation Transfer Function) provide a graph analyzing a lens’ ability to resolve sharp details in very fine sets of parallel lines, and a lens’ contrast or ability to provide a sharp transfer between light and dark areas in sets of thicker parallel lines.

What is modulated is the actual image contrast. So if the contrast of the subject between say Black and White is 3-stops then after being transmitted by a lens with a MTF of 0.5 the image will only have a contrast of 2-stops. Eventually at lower MTF values the contrast will be reduced to 0-stops and no detail will be visible, just the average brightness reproduced as a grey area.

The lines direction is relative to the lens optical axis. The Sagittal lines, sometimes designated “S” on Canon’s MTF charts run from the centre of the frame outwards. At a 90° angle to these are the Meridional (or “M”) lines that are tangential to the centre of the frame.

Canon’s MTF charts give results at two apertures: wide-open, and stopped down to f/8, with the lens set to infinity focus at various distances from the centre of the frame for the M and S directions. The MTF for M and S are often not the same, particularly for zoom lenses.

So we see immediately that lens sharpness depends not only on the aperture and where in the frame the subject detail occurs, but also the orientation of the detail relative to the optical axis or centre of frame.

If we imagine a landscape for example that may be dominated by vertical detail such as grass and trees; on the horizontal centre line of the frame the Meridional MTF will be most important but on the vertical centre line the Sagittal MTF is most significant.

So care is needed is interpreting the MTF data, what is important? The average MTF; the worst case (minimum) of the Meridional and Sagittal; or some other measure? See my lens simulation page for answers on this.

If detail is of mixed orientation over the frame then the average of the Meridional and Sagittal may be a useful measure, otherwise it may be best to look at the worst case performance.

It is important to note that the Canon MTF data is computer simulation not real physical measurements. Real lenses will have production tolerances and other defects that will cause variation from the simulated performance. Generally the more complex the lens the more variation can be expected, so zooms can be expected to depart from the simulations to a greater extent than primes. Cheaper lenses with less well controlled tolerances and processes can be expected to depart from the simulation more than more expensive varieties.

However, I have compared some lens MTF data from Canon with measured data from Photodo and it seems to me the Canon data is a reasonable guide.

Another point to note is that it is not reasonable to compare full aperture MTF between lenses unless they have the same or nearly the same aperture, so in most cases the f/8 data will be of more significance.

Another consideration is format size. Finer resolution will be required for 35mm than for medium format and a 1.6X crop factor sensor will need finer resolution than a full frame 35mm sensor or film camera for comparable resolving power of subject detail.

So as far as a 1.6X crop factor sensor like the Canon 20D is compared to say, the full frame Canon 1Ds Mk2, the 20D has the advantage of only using the centre of the lens image where the image is sharper (for EF mount lenses not EF-S mount) . Conversely, the resolution in that area needs to be better for the 20D to capture the same subject detail as the 1Ds.

So in some ways the smaller format needs better resolution than the larger format, at least in the centre of the image.

When I last had to choose a lens system all that was available to go on was reviews in the photographic press. So it seems to me that as Canon have made MTF data available it should be used. Care is needed in comparing data from different manufactures, in this case I suggest using the information at the Photodo site.

Downloading from Photodo

Although Photodo has not been updated or maintained for some time. For example the MTF charts are not downloaded onto the web page due to an error in the page design.

To get at the photodo charts follow the below routine (depending on your browser, this is with Netscape):

a)    Right click on the missing plot and select "copy location"

b)    Past the location into a second browser windows address line, there will be an address of the form http://193.14.88.41/pix/lens/mtf/CAEF2028USM.gif

c)    Edit the address to have the correct website so http://www.photodo.com/pix/lens/mtf/CAEF2028USM.gif

d)    Press return and you have it.

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