[lol101]

Hi,

I have the pleasure to be the owner of said lens since last week and I thought I'd report my early feelings about it and maybe ask a few questions around.

First, it's probably the best lens I have ever owned in it's range, my first tests have sold me on it's IQ withour restrictions.

Construction is excellent and I am really happy to get back the FoV of my old 80-200 f2.8 without the bulk and weight.

AF precision is top notch.

Now, for the discussion.

Having read about it's AF capabilities, I begun by keeping firmware to 1.11 in my K10 and used the lens with screw-drive AF last week.

Two impression son this period:
1/ AF is quite loud.
2/ Compared to my FA 135f2.8, the AF is slower with a little bit of hunting.

I'd say moderately fast AF.

Now I switched to v1.3:

1/ AF is near silent, except for small 'cliqueting' sound when focus is achieved: nice!
2/ Compared to my FA 135f2.8, AF is slower but with less hunting.

I'd still say faster AF this time but the gain is not on the max rotational speed which is still slower than the FA lens.

Also, keep in mind an important fact about the two lenses I am comparing: with the FA 135f2.8, AF throw for going from 1m to infinity is about 80-90°, on the DA*, it's around 100-110°, that's quite a long way to go.

While focusing from infinity to 2m, the FA lens rotates very quickly (much quicker than the SDM), passes beyond the AF point, stops, comes back a bit, eventually adjusts by tiny steps and locks (all that in around a second or so).
For the same job, the SDM rotates slower but stops at the AF point, eventually adjusts by tiny steps (my impression is that it does that much less than the FA) and locks.

Now, the impression when you just make the lens rotate from one end of it's range to the other (with lens cap on) is that the FA lens is super fast while the DA* is pretty slow but when you actually focus on an obect, the total AF time seems about the same but the "AF scenario" is different.

The FA ruches back and forth into focus while the DA glides in.

Of course, differences are only visible when you focus alternatively from close to far, in minor AF adjustments, both lenses are really fast.

To me it looks like the SDM motor is limited by the maximum rotational speed it can communicate to the lens. The in-body motor of the K10 is one of the most powerfull I have seen yet, it really jerks lenses into focus pretty loudly but it's maximum rotational speed is high. The SDM motor is apparently not capable to attain the same speed.


Now comes the question: can AF speed of SDM lenses be improved by a newer body?

When I asked a friend of mine who just switched from the Canon 10D to the 40D what were his feeling regarding AF speed with his 70-200f2.8 USM lens, he replied: "Day and night, the 40D is clearly much faster".

I also wanted to chime in in the 1D series forum on DPR with a question like: "Do you think AF speed of a 70-200f2.8 USM is the same on a 400D than on a 1DMkIII (OK, MkII might be a better idea these times ) but something kept me from doing it (fear of death maybe? ).

So I'm left wondering, with in-camera AF motor, you switch to a camera with a more powerful motor and get instantly faster AF on all your lenses.
But with SDM, is it possible to get faster AF on future bodies or is everything limitated by the motor you implanted in the lens in the first place?

[m8o]

I would have to say you're limited by the motor's speed, and the power you deliver to it. If the motor is capable of taking a higher voltage+current (i.e. power), and the camera can deliver it, and the motor design is one that a higher supplied power results in a faster speed, then there ya go. ... it's possible.

Realize, a lot of "ifs" ... all supposition. There's another idea floating in my head that the motor reacts not to DC, but to rather "pulsed" DC. If then, the pulses can increase in speed w/an updated body, the focus speed could increase. I'm confident in the EE concept of the former, but this latter idea is just sheer speculation on my part.

I also read on this forum from someone who said Pentax just installed the AF motor used in the K10D body into the DA* lenses... thus the similar focus behavior/speed ... and that Cankon uses a completely different type of 'ring-type hypersonic motor' or something like that. Just paraphrasing what I read so noone quote me on that please.

[clawhammer]

Anyone feel like taking a DA* appart? I've got a Nikon SWM from the Nikkor 18-70 to compare the motor too....

[lol101]

I did another test this evening to get a different perspective on max AF speed:

I took the FA 135 and DA* 50-135 and dug out my Z1-P
With the 135, both the K10 and Z1-P focus at approximately the same speed (ie very fast: less than one second for a complete travel from 0.7m to infinity and back).

With the DA* 50-135, even though one uses the screw driven AF, the other SDM, AF speed is , again, roughly the same.

So conclusion #1, SDM is as fast as screw driven AF. There might be small difference but it's really hair splitting.

But for the DA* lense to do one round trip (1m to infinity and back) takes about 2s, on any camera whereas the FA lens only takes 1s on both cams.

So conclusion #2: I am comparing apple to oranges .

The differences in AF speed between these two lenses have nothing to do with SDM or screw-driven AF.
It's probably all caused by different AF mechanisms designs. Probably more glass to move in a more complex manner on the DA* than on the much simpler FA 135.

So conclusion #3: 'nough testin'! Let's go shootin'!!

[clawhammer]

So conclusion #3: 'nough testin'! Let's go shootin'!!

Best conclusion I've heard in years

[kenyee]

that Cankon uses a completely different type of 'ring-type hypersonic motor' or something like that

That's because they didn't want to make their lenses backwards compatible. The Pentax SDM lenses will work w/ old screw drive bodies. The Nikon/Canon USM ones work only w/ the latest bodies w/ the extra electrical contacts...

[m8o]

Thanx kenyee.

[kb244]

That's because they didn't want to make their lenses backwards compatible. The Pentax SDM lenses will work w/ old screw drive bodies. The Nikon/Canon USM ones work only w/ the latest bodies w/ the extra electrical contacts...

Course the Canon EF/EOS bodies never had a motor for the AF in the body to start with, the motors were always in the lens from day one (which is what gave them a strong boost in the AF market back then). They just didn't make the lens backward compatible to the old Canon FD bodies since they switched to a completely electronic system back in the 80s. And the Canon lens (exception to the cropped EF-S lens) with USM or without will work with all the Canon EOS film bodies even the first ones that came out in the 80s. Canon didn't need any extra electrical contacts. Thats basically where Nikon and Canon differed.

[darthkir]

My guess is the slow AF is not because of the motor. Its the AF mechanism itself ( i dont know what to call it). Because even a non motor lens work a hell lot faster on the N/C bodies than the Pentax, it shows that AF is not all about the in-lens motor.

[blende8]

Now comes the question: can AF speed of SDM lenses be improved by a newer body?

It depends.
As already said, one part of the problem is the AF detection system itself.
If it is improved and works faster, the lens will focus faster.

The other part is that the current SDM lenses do NOT feature ring motors as the better Canon lenses, but only a simpler micro motor. Only with ring motors you can improve speed further. At least that's what I've been told. I am no expert in these matters.

[lol101]

My guess is the slow AF is not because of the motor. Its the AF mechanism itself ( i dont know what to call it). Because even a non motor lens work a hell lot faster on the N/C bodies than the Pentax, it shows that AF is not all about the in-lens motor.

This has not been my experience at all.

Actually, Pentax AF in normal light condition is absolutely not slower that the equivalent C or N bodies.

FA50 f1.4 on K10 focuses at the same speed as 50f1.4 Nikon on D200.

On the other side, FA 135f2.8 focuses faster on K10 from 1.5m to infinity than any zoom in this focal range I know of, including the USM 70-200f2.8 but DA* 50-135 seems slower, screw drive or SDM (didn't compare yet).

[kb244]

This has not been my experience at all.

Actually, Pentax AF in normal light condition is absolutely not slower that the equivalent C or N bodies.

FA50 f1.4 on K10 focuses at the same speed as 50f1.4 Nikon on D200.

On the other side, FA 135f2.8 focuses faster on K10 from 1.5m to infinity than any zoom in this focal range I know of, including the USM 70-200f2.8 but DA* 50-135 seems slower, screw drive or SDM (didn't compare yet).

I can concur with this, I think it has quite a bit to do with the speed of lens as well. Some of the Canon bodies are designed to switch into high precision mode when it detects a lens that is f/2.8 or faster, otherwise its pretty broad by comparison. Pentax seems like it has a similar method, and in using similar types of lens I didn't notice it being significantly faster or slow than the C/N counterpart. One is 'almost' as fast as a USM lens I once owned, but not quite. I imagine then that the pentax lens-based motors are going to be somewhat similar as Canon's in terms of capable speed, some models being better than others of course.

The ring-type USM works faster because the design creates less strain and resistance to change, where as the micro-USM which is a small gear driven device requires more torque to work, but is however cheaper to manufacture (I'm hoping I didn't get that backwards). And as far as I can tell the broad method of either is not exactly patented to prevent any other manufacture from following the same example, though I'm sure they did patent exact specifics.