Sigma Lens Focus Algorithm Comparison
If you buy Sigma’s USB dock to reprogram their lenses, it gives you the option to not only update their lens focus firmware, but also which focus algorithm(s) to choose. It’s almost like buying a new lens.
If you have never programmed your Sigma lens using their dock, then your only focus choice is “Standard AF”. You get to choose among three different “AF Speed Settings” if you program their lens via the “Sigma Optimization Pro” program. The choices are “Fast AF Priority”, “Standard AF”, or “Smooth AF Priority”.
The Sigma lenses that support the customization have a “C1” and a “C2” switch to change modes. It’s possible to have three different settings collections, using the custom switch set to “Off”, “C1”, or “C2”. If it’s set to “Off”, then you’d get their “Standard AF” (in addition to their default optical stabilization view mode). You could set the C1/C2 switches to the same or different algorithms.
The “Smooth AF Priority” is geared towards video shooters, and the focus is slow. I’m not a video shooter, so I don’t use this customization. If you’re a videographer, however, this option might be the deciding factor in buying a lens.
The “Standard AF” is medium-speed, but purportedly more accurate for focus accuracy and repeatability than a quicker focus. I have my “C2” switch set to this focus mode.
The “Fast AF Priority” is their quickest focus mode, and it’s how I set my “C1” switch now. I didn’t leave it programmed this way when I first got Sigma’s programming dock, however.
When I first got my Sigma 150-600 Contemporary, I found that my “Fast” setting wasn’t reliable, and I was forced to leave the lens on “Standard AF” to get accurate and repeatable focus. After Sigma updated their lens firmware, I discovered that their focus was made faster AND more accurate, in both the “Standard AF” and “Fast AF Priority” modes. The “Fast” mode was now so accurate and repeatable, that I use it exclusively. I left my C2 switch with the standard AF option just in case, but I’ve never needed it.
I did some focus speed testing experiments, to see how much different their “Standard” and “Fast” modes were. I used my “Super Slow Mo” video setting on my cell phone and a very accurate timer to evaluate focus timing. I performed the tests using my Nikon D500 camera, to get an idea what a “pro-level” camera could do with this lens.
Sigma Optimization Pro Focus Programming Example
You can see what the Sigma programming dialog looks like above. They have other dialogs for programming optical stabilization, focus distance ranges, and focus calibration. It gives you feedback about how it’s already programmed, in case you forgot. I leave little notes taped onto the inside of my lens cap about how each switch is programmed.
Use a timer and super-slow-mo video to track focus timing
It was pretty straightforward to take some video (240 frames per second) to accurately time focus. I set my Sigma lens on minimum focus (about 9 feet and 600mm zoom). I used a big rubber band to attach a cell phone with a timer application onto the lens, near the focus distance window. I’d start the timer, then start video recording, and then finally press the “AF ON” button to focus. I could review the recorded video right on the phone, and swipe from frame-to-frame while reviewing the footage. I could easily see as the timer changed by hundredths of a second in the video.
I used a pair of cell phones here, since I needed to make a video of the timer display next to the lens focus-distance scale display. I made sure that the big rubber band wasn’t in a location on the lens that would interfere with the lens focus operation.
I would simply start the video recording and then press the “AF ON” button to focus on a high-contrast, distant target. I could review the video frame-by-frame (with a finger swipe) to see the timer value when the lens distance scale started moving, until it reached the infinity mark. I repeated the procedures to convince myself that the test was repeatable (it was). I had to shade the phone screen and lens focus scale to record their images well enough in the video.
The timer application I used records down to 1/100th second. I made a few videos at both the “Fast AF Priority” (C1 switch) setting and then again at the “Standard AF” (C2 switch) setting. The timing was remarkably repeatable from test-to-test, so I have good confidence in the results. I skipped trying the “Smooth” auto-focus algorithm, since I am only interested in speedy focus.
I used my Nikon D500 for this test, since it has the fastest auto-focus of my cameras, due to its Expeed 5 processor. I didn’t want a slow-focusing camera to affect my lens speed measurements.
The “Standard AF” algorithm timing was 0.55 seconds to go from minimum focus to infinity. The “Fast AF Priority” algorithm timing was 0.44 seconds for the same focus range in the same (good) outdoor lighting.
In the real world, nobody actually goes from minimum distance to infinity in a shot (unless they’re a rank amateur). More likely, focus will be changing by only several feet shot-to-shot. Using the lens like this, the focus time is more like a few hundredths of a second, as witnessed in the video footage, changing from about 30 feet to about 40 feet. It is slower, of course, when the lens has to switch focus directions.
Once you have your recording setup, you can now explore different focusing scenarios that apply to your own style of shooting. Focus speed timing only makes sense if it reflects how you’d actually use it. Different lenses with different minimum focus distances and different focal lengths can’t be directly compared. You also need a sufficient level of illumination to compare different lens/camera combinations.
I have read an article about someone comparing both the Sigma Contemporary 150-600 against the Sigma Sports 150-600. The Sports version has larger-diameter (and much heavier) lens elements in it, and actually focuses SLOWER than the Contemporary version. The focus motor is purportedly the same in both lens versions. I don’t have the Sports version to test, but these reported results make sense. Keep in mind that not all lens elements are necessarily in motion while focusing, and clever lens designers don’t move more glass than they have to.
Don’t forget to use the focus-limit switch on your lens, if it’s available. It can make an enormous difference in focus acquisition speed if your subject movements are predictable enough. The focus limit range is yet another programmable setting with Sigma lenses.
The bottom line is that the Sigma focus algorithm speed difference is 20 percent between Standard and Fast. I can’t tell any differences in accuracy between them. Without the programming dock, this speed improvement wouldn’t be available.
I wanted you to know how I performed the speed timing measurements so that you can do these tests yourself and evaluate your own equipment. You don’t need elaborate equipment; just modern cell phone technology.