Do all third-party lenses have sluggish AF performance?

Joined
Oct 31, 2008
Messages
1,676
Location
New Germany, N.S.
I am very happy with my Tamron 17-50mm f/2.8 and Sigma 50-150mm f/2.8 zooms, there's nothing like fast glass when you really need it, which is more often than you think you'll encounter.

I did a lot of shooting with both during the last week, I can't help but notice both seem to hunt a bit even on the D80, with its 11 AF points. During a wedding, yesterday, the Tamron didn't seem to want to focus when I wanted it to, even with the SB-800 focusing pattern being illuminated in a dark reception hall.

The Sigma can be locked on target and, all of a sudden, it goes amiss.

I guess you have to take the good with the bad as a third-party lens owner, both of my Nikkors (16-85mm VR, 70-300mm VR) don't hunt and they have slow maximum apertures.
 
Joined
Jan 25, 2009
Messages
443
Location
Portland, OR
The AF on my Sigma 10-20mm is almost instantaneous. The AF on my Sigma 150mm is horrid though.

Chad
 
Joined
Sep 10, 2008
Messages
300
Location
Calgary, Alberta, Canada
There are two separate issues here ... focus acquisition and focus tracking.

In terms of focus acquisition, nothing comes close to the pro level Nikon AF-S lenses.

I would put Sigma's HSM in second, and everything else a very very distant third.

Screw focus falls somewhere around Sigma's HSM, depending on your body and the lens.

In terms of focus tracking, I would think it is more a relationship between the camera sending the distance information to the lens, and the lens interpreting that information quickly (and correctly). It's hard to say exactly where the problem lies, because it is possible that Nikkors and third party lenses behave differently when presented with the same data. But it does not surprise me that Nikkors seem to behave better than third party lenses.
 
Joined
Dec 17, 2008
Messages
2,298
Location
Maryland USA
I am very happy with my Tamron 17-50mm f/2.8 and Sigma 50-150mm f/2.8 zooms, there's nothing like fast glass when you really need it, which is more often than you think you'll encounter.

I did a lot of shooting with both during the last week, I can't help but notice both seem to hunt a bit even on the D80, with its 11 AF points. During a wedding, yesterday, the Tamron didn't seem to want to focus when I wanted it to, even with the SB-800 focusing pattern being illuminated in a dark reception hall.

The Sigma can be locked on target and, all of a sudden, it goes amiss.

I guess you have to take the good with the bad as a third-party lens owner, both of my Nikkors (16-85mm VR, 70-300mm VR) don't hunt and they have slow maximum apertures.


Thats what you get with 3rd party lenses. My Tamron 28-300mm VC is a little iffy sometimes, but my Tamron 17-35mm focuses fast and locks on the first try all the time. Guess because it uses the in-camera focus motor. My Sigma 150mm can hunt sometimes, but overall is good.
 
Joined
Mar 21, 2008
Messages
691
Location
Orlando, Florida
My Sigma 150 is almost always very good. Got to use the limiter though.

I agree. Without the limiter, as much as I love my Sigma 150, its autofocus qualities are abysmal, in my opinion. For no reason I can think of, it will lock and just when you start to press the shutter, will change way off. None of my Nikkors do that. Even my screw drives don't hunt around that way.
 
Joined
Nov 8, 2008
Messages
425
Location
Portland
So Physically why is it slow?

I've asked this in a few forums and never got what I thought was a believiable explanation.

I know each camera manufacture has its own communicaton protocal between lense and camera. But isn't the only thing that we send from the body to the lens is drive the motor to focus and the only thing the lense sends back is "finished" and awaiting next input? Oh yes of course when you mount the lense it tells the camera aperture and focal length and other fixed stuff.

All the focus processing is done in camera and direction sent to lense. Why then do lenses hunt. I can see that pro lenses have larger/faster motors and move the optics faster. I can see SWM/AFS being quieter and mechanical being slower and noisier. It makes sense why for low light slow/smaller aperture lenses to struggle as the contrast / light isn't enough for the AF system. But I agree under certain conditions my Tamaron hunts and hunts and hunts or my 28-300 really complete locks up. I've found some quick fixes that make these none issues but still annoying!

But on my Tamrons, I agree that the BIM seems to struggle more often then not than my 70-200 or any of the other nikon AFS that I once owned. It makes no sense as all the lense does is project an image focused or not. All the processing and decision is done in the camera. Things like contrast, image sharpness etc. should effect the focus that much should it?

Just curious for a phsyically beliavble reason why 3rd party are slow. The fact they are not made by Nikon simply doesn't seem to fly but the emperical data suggest there is some special sauce missing. 3rd party engineers can't be that stuipid are they? Maybe the body knows its not a nikon lense so it throws the lense for a loop? :tongue:
 
Joined
Sep 10, 2008
Messages
300
Location
Calgary, Alberta, Canada
I have a theory based purely on my own experience and has nothing to do with any actual knowledge of how these systems work.

All lenses that have screw driven focus perform in the same way (as far as I have seen, but granted my expertise here is not that wide ranging).

When a lens has a built-in motor, the body will send distance information to the lens. The lens will then use its motor to achieve focus at that distance.

But what happens if the body cannot determine what the distance is to the focus object? This can be for a variety of reasons, but let's just say that it occurs. Lenses can respond in two ways:

1. Stay at the same focus distance. The hope here is that the user shifted the camera slightly, causing the focus object to leave the AF sensor area, and will shortly put the AF sensor area back on the object.

2. Shift the focus distance in hopes that the AF sensor will find an object in focus. The assumption here is that the user wants to focus on a different object.

To add to this, there is probably a delay such that the lens uses method 1 until X milliseconds, at which point it shifts to method 2.

Then there is the use of AF-S, AF-C, etc ...

My guess (and it's just a guess, nothing more) is that Nikkor and third party lenses weight the two options above differently, and also likely use different delays.
 
Joined
Dec 17, 2007
Messages
2,116
Location
Canada
This site "Lens Tips", tests and comments on AF accuracy--a lens quality that if often untested in most reviews. If you read many of their lens reviews, you will notice a few lenses will be 100% accurate, while others are often less inspiring. For example the Sigma 50/1.4 had a 8% AF accuracy error which they comment as moderate inaccuracy. That means one out of every 12 pictures will be out of focus!

Some lenses I have used are incredibly accurate. For example, the AFI300/2.8; AFS24-70/2.8---on the nose...without a doubt almost all the time.

AFD85/1.4--often needs refocusing to get it right---keep repeatedly pushing the AF-ON button...
Tamron 17-50/2.8 screw driven---not so hot sometimes
Sigma 50-150/2.8-- Goes hunting sometimes.


I think Nikon's AFS calibration can be really accurate in most cases, whereas a 3rd party lens like Sigma may not be as accurate--it seems the distance information may be more course and may not have fine enough increments that Nikon bodies are expecting. My sigma 50-150/2.8 would often hunt between two positions in an alternating fashion--as if the camera would like the lens to focus in between the two options the sigma lens is allowing the focus distance to be.

Anyhow, check out the lens tips for more...
 

Latest threads

Top Bottom