PopPhoto.com -- The online home of American Photo and Popular Photography & Imaging magazine

Free Newsletter: Camera reviews,
lens tests, photo news and more!


Popular Photography American Photo

My Photo

November 2008

Sun Mon Tue Wed Thu Fri Sat
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29

« Ask Mike | Main | Noise: The Silent Killer »


Harry Winter

Here is a paper which I posted for discussion on the “Tech Support Forum” of Popular-Photography and for which I was then suspended from the forum by one of the “Moderators”. (At least I was not made to recant the truth, as Galileo was by the inquisition.) --- Since PopPhoto claims to be the World largest Photo magazine and also the only one doing a truly “scientific” analysis of lens-performance, (Why our lens-test are the best in the field; http://mcnamara.popphoto.com/the_mcn...ocus.html#more) --- I believe that many of you here are also using the “Lens Reports” by PopPhoto when purchasing a new lens for your Canon camera. (I use to do so)

How to “READ” the Pop Photo SQF Lens tests of full frame 35mm lenses, when these are to be used with digital cameras with APS format, a 1.5/1.6 cropping-factor.

When I bought my first Digital Rebel with kit-lens some time ago, I already owned the highly regarded EF 28-135 IS lens, and some times later I also bought the new Canon 70-300 DO IS; --- both are “full-frame” lenses.
In November ’90, Popular Photography published an important article about the problem of objectively testing lenses and about their solution with the SQF method.
“SUBJECTIVE QUALITY FACTOR, What is it? How did we arrive at it? And why is it the best way to test lens performance?”--- By Larry White.
I got a copy of this article, or most of it, from Herbert Keppler himself by ordinary mail. Unfortunately his secretary forgot to copy the continuation of the article on page 174. (LOL --- This evil habit of splitting up articles just came back home.) PopPhoto never published a modern version of this article but there is a very good explanation by Bob Atkins on his web-site: http://www.bobatkins.com/photography.../mtf/mtf4.html
At the times when the article was published no thought was given of how to apply these test-results for film-planes or sensor-planes of sizesdifferent than 35mm,in particular for the smaller APS “Format”. (My very first camera in 1950, a Robot with 24X24mm had such a APS format) The reason was that interchangeable lenseswere then nearly exclusively used for 35mm film cameras. --- Fortunately, the missing pages, of my copy are not important for the issue of “format” here, because these are dealingwith thephysiological effect of print-size on the perception of sharpness. Nothing of this part of the SQF-method has changed for the use of APS format Digital Cameras, except for the MAGNIFICATION FACTOR needed to get from the small film/sensor-plane format to a certain print size. On the published coloredtest-charts in Pop-Photo magazine (such as this sample from the Atkins paper: http://www.bobatkins.com/photography...mtf/popqsf.gif ) this magnification is given on the horizontal axis in the form of print dimensions, which the article says correspond to the “magnifications” I have shown above it:

4X -------- 5X ------- 6X -------8X ------- 12X ------- 17X -------- 22X ---- = --- Magnification
3.5X5 --- 4X6 ------ 5X7 -----8X10 ---- 11X14 ---- 16X20 ---- 22X24 - = ---- Print size

The perception of sharpness of a lens (as measured by the MTF numbers) is directly related to this magnification factor needed for a certain print-size, more magnification means less resolution and contrast. (less sharpness). --- If the lens is used with a cropping type camera, either digital or film, we cannot use the magnification numbers which were computed for 35mm film format, but have to multiply each of these numbers by the cropping factor, 1.5 or 1.6. --- For instance, to get an idea for the quality obtainable for an 8X10 print, the multiplication factor needs to be 12X, and not the old 8X. I need to look at the column published for 11X14 prints to get a quality reading.
To apply the Pop-Photo full-frame lens-test data the following magnification factors for their corresponding print-sizes need to be evaluated.

6X ------- 8X ------- 9X ------- 12X -------- 18X ------- 25X ------ 33X = more Magnification
3.5X5 --- 4X6 ---- 5X7 ------8X10 ----- 11X14 ----16X20 --- 22X24 --- = ---- Print size

A “quick” way is to just look at the next larger Print-size to get a good idea of the quality of a print from a digital cropping camera with the lens in question.

Harry Winter, --- former “Member of Technical Staff” of Bell Laboratories, ret.

PS. Concerning the tests of these newer digital-only lenses, I just hope that the “format adjusted” magnification factors” for the SQF method have also been applied to these. If not, the entire annual set of tests would need to be re-done to keep the high “Professional Standing” of the Popular Photography magazine intact.
By looking into several of these tests (e.g. the Tamron 17-50 F/2. I observed that the claimed performance of these “digital-only lenses” is now exceeding the performance of several new prime-lenses, such as the $2200 Leica 50mm F/1.4. This looks very suspicious to me, because no such technical/cost breakthrough for “photographic lenses” has been announced in the trade-press.
To me it is more likely that the “old” 35mm magnification factors have been re-used erroneously, which would give the same results. --- I am certain that the truth will eventually prevail and my condemnation and gag-order by PopPhoto will be lifted after my death. ------ Hope runs eternal!


Mula W. Wangsaputra

Dear Mike,

I always enjoy comparing lens SQF data in PP. I sometimes curious to compare the Lab Test with the field test. One example, I own a cheap (120 USD) Nikkor lens, 28-100/3.5-5.6 G AF. Based on the SQF data, it's an excelent lens, especially at 100mm at most commonly use apertures, i.e. f/5.6, f/8 and f/11. It's even a bit better than a 900 USD lens, 105/2.8 G AF-S VR (PP Aug 06). Of course I'm not comparing the VR, the f/2.8 max aperture, 1:1 macro, but still the 120 USD lens is too good to be true. Last week, I ran a simple field test. I compared 3 lenses: Nikkor 28-100/3.5-5.6 G AF, 35-80/4-5.6 AF D Nikkor, 70-210/4-5.6 (non D) Nikkor AF, on D70s. I set the same conditions on every test, i.e, same aperture, same shutter speed, same camera body, same ISO, same subject, same cropping (in order to distinguish the detail differences). I found that the 70-210 is sharper than 28-100 at 100mm both at f/5.6 and f/8. I also found that 35-80 is sharper than 28-100 at 80mm both at f/5.6 and f/8. Unfortunately I don't have the SQF data of those lenses (70-210 and 35-80) in order to verify the results with the SQF data. Could you explain it? Thank you. Regards, Mula W. Wangsaputra - Bandung, Indonesia

Posted by: Mula W. Wangsaputra | November 12, 2006 at 10:15 PM

Dear Mike,

Oops, I'm sorry, I'm new to the PopPhoto.com, I think I misposted a question about SQF to a current discussion on Memory Cards.

Please redirect my post to the correct field.

Thank you, again sorry for the inconveniences.


Mula W. Wangsaputra

Sorry, I was posting this question on different topic. I hope, now I'm on the right place to post/ask this question.


Yesterday at 34:29PM
long term side effects cellcept
All the time we think about our customers! This unbelievable sales event proves it! Pharmaceutical best-sellers at discount price! They sell like hot cakes!
mycophenolate mofetil epar
Herbal medicine may be even more effective than the traditional one!
GET Cellcept (Mycophenolate Mofetil) CHEAP OVERNIGHT
cellcept 2000 mg
cellcept lupus symptoms

BUY CHEAP Cellcept (Mycophenolate Mofetil) NO PRESCRIPTION //GET GENERIC Cellcept (Mycophenolate Mofetil) CANADA
cellcept side effects mental ;;mycophenolate mofetil rheumatoid arthritis
mycophenolate mofetil usp

The comments to this entry are closed.