How We Test Lenses

When _Popular Photography’_s lab installed one of the first SQF(subjective quality factor) optical testing systems in 1988, it was state of the art. Based on an electronic, computer-controlled, optical bench that measures the MTF (modulation transfer function) percentage of lenses directly at each aperture, it provided a more accurate picture of a lens’ imaging performance than lines-per-millimeter resolution testing on film, which evaluates lenses at only one spatial frequency and is subject to the judgments of the individual test reader. MTF is generally recognized as the most comprehensive and accurate lens testing system in use, and SQF, which is directly based on MTF, has the further advantages of presenting differences in lens performance as quality grades for different print sizes instead of raw percentages, which have little meaning for the average photographer.

Our original EROS SQF/MTF apparatus served us admirably for over 12 years, and provided our readers with the best information available at the time, but even the best equipmentrequires calibration and servicing, and we were faced with the prospect of replacing several key mechanical components, some of which were no longer available. So, we turned to Optikos Corporation of Cambridge, MA, the leading supplier of high-end optical testing equipment in the U.S., and their staff of experts. Even with their unstinting help, and with the full financial backing of our parent company Hachette Filipacchi Magazines, Inc., it took over six months for our SQF testing facility to return to full operation. But the end result is not merely the replacement of a few worn-out components—it’s a complete system upgrade.

Tweaking the system

While we are confident that our newly revised SQF system will provide much improved reliability and enhanced accuracy, achieving these benefits entailed a slight revision of our readout system. Because we now use a pinpoint light source instead of a slit as our imaging target (to reduce mechanical errors in going from vertical to horizontal mode), and a new steep-cutoff IR filter in the light path (which allows a wider range of visible red light to reach the target), the SQF percentage numbers appearing on our test charts are slightly lower (by about 5-7%) than they were previously. In order to keep forthcoming SQF results as a consistent as possible with previously published SQF tests, we have also made slight adjustments in the color-coded quality readouts—see the accompanying charts for specifics.

SQF chart

While the changes are minimal, you may note that quality grades using the new system are a bit lower than they were with the old system, especially at high magnifications (bigger print sizes) and at marginal performance apertures for wide-angles and tele focal lengths. To give you a precise visual picture of how the recalibrated system affects the results, we’ve reproduced charts for a well-known lens we’ve tested (the Aigma 28-135mm) using both the old and new systems.

Test-table comparison of the old and new systems.

Extending the spectral distribution of the light source (which now more accurately reflects on-film performance) as well as upgrading both mechanical and optical components of our SQF apparatus will give Popular Photography readers the most accurate information available for making intelligent decisions on choosing and using lenses. That has always been the top priority of our testing program, and it always will be.