I bought the 230 gram Sigma 24 mm f/3.5 DG DN lens recently. Nice lens in many ways.However, there can be no doubting that it has a worse focus accuracy hit-rate when tracking my toddler (typically eye detect) than any of my Sony lenses.(Because of the distances involved, this is probably harder than tracking a racing motorcycle, but still: the Sony lenses deliver.)Why is that?Focus accuracy for static subjects is high.The time to drive the focus across the full range in one sweep is impressively low despite a short min-focus distance.Sigma has officially licensed the communication protocol, so there shouldn’t be the old problem of reverse-engineered hacks not quite working either.I’m left guessing that the lens somehow can’t accelerate the focusing group up to speed quickly enough to cope with many small adjustments per second. But you tell me.

Sigma has stepper motor for AF in that lens. Sony's newer lenses use top notch ultra fast linear motors - often multiple in G/GM lenses. In short, Sony has faster motors - and often multiple of them.

someguy50 wrote:In short, Sony has faster motors - and often multiple of them.It’s interesting that these motors are insignificantly faster for the full sweep of the focus range, though. I guess they must allow higher acceleration rather than higher top drive speed. And I guess that is required for accurate focus tracking of fast-moving subjects.But I’m guessing. Has anyone come across a plausible article about this stuff?How well do Sigma’s SLR lenses, imported to E-mount, work? At least some of those have hypersonic motors.

Samuel Dilworth wrote:someguy50 wrote:In short, Sony has faster motors - and often multiple of them.It’s interesting that these motors are insignificantly faster for the full sweep of the focus range, though. I guess they must allow higher acceleration rather than higher top drive speed. And I guess that is required for accurate focus tracking of fast-moving subjects.i wonder if another factor is the slow readout with unstacked sensors, the camera has to rely on prediction a lot more, so it's not telling the lens to change focus speed as quickly as it does with a stacked sensor.How well do Sigma’s SLR lenses, imported to E-mount, work? At least some of those have hypersonic motors.as i recall it's non-linear, using motors that drive ring af... i've used lenses like the 135 art on the a9/mc-11, it worked well for sports, but of course not in the same league as sony linear voice coil af... probably a lot better than stepper motors tho.steppers are quiet, they are good for video.thx for posting feedback on the sigma 24mm dn.

MILC man wrote:i wonder if another factor is the slow readout with unstacked sensors, the camera has to rely on prediction a lot more, so it's not telling the lens to change focus speed as quickly as it does with a stacked sensor.Not a factor in this specific case. I’m comparing the Sigma lens to my Sony lenses (e.g. 40 mm f/2.5) on the same α7C camera, which has impressive autofocus performance – with the right lens – despite its fairly slow readout.steppers are quiet, they are good for video.I can see that very fast focus achievement might be unwanted for video, creating a jarring snap. Though it must be possible for a fast lens to focus slowly if so commanded.It will be interesting to see if Nikon ever achieves Sony-like focus tracking with its many Z-mount lenses with stepper motors. Perhaps they’ve designed their lenses to allow stepper motors to be more responsive than Sigma’s? An extreme focus on lightweight focus groups or something?thx for posting feedback on the sigma 24mm dn.I’ll maybe post more after more use. I’m trying to master my α7C, for which I need light lenses that I’m likely to carry out the door, hence this lightweight 24 mm. But if I’d know the Sigma didn’t track focus as well as the Sony 24 mm f/2.8 probably does, I might have paid the premium for the latter. On the other hand, this Sigma has its own appeal.

Samuel Dilworth wrote:MILC man wrote:i wonder if another factor is the slow readout with unstacked sensors, the camera has to rely on prediction a lot more, so it's not telling the lens to change focus speed as quickly as it does with a stacked sensor.Not a factor in this specific case. I’m comparing the Sigma lens to my Sony lenses (e.g. 40 mm f/2.5) on the same α7C camera, which has impressive autofocus performance – with the right lens – despite its fairly slow readout.I figured you had one of those G lenses. Those are top notch - utilizing two of Sony's latest linear motors to push very light weight glass. It's almost overkill. I evaluated several lenses and settled on the Sony 24/2.8G due to its AF performance - more in focus/keepers was worth the premium.