[drabina]

I have K10D camera with Sigma 17-70mm lens. I really like this combo but now autofocus has trouble keeping up with my two year old. I am sort of exploring options (including other camera systems) but would like to know if upgrading my screw-drive Sigma to Pentax SDM equivalent (also 17-70mm) would give me faster AF?

I did play with Canon's ultrasonic lenses and they seem very quick in focusing comparing to my K10D with Sigma lens. Though it wasn't direct comparison because Canon lens had different zoom range.

I have few lenses that I could sell to raise money for the difference in price between Sigma and Pentax but that's about it. Upgrading to K20D is out of the question for now.

Thanks.

[Marc Sabatella]

I did play with Canon's ultrasonic lenses and they seem very quick in focusing comparing to my K10D with Sigma lens. Though it wasn't direct comparison because Canon lens had different zoom range.

Um. also it wasn't a direct comparison because the Canon lenses were being focused by a Canon body. That's really the big difference here: Canon body versus Pentax body, not screw-drive versus SDM.

[joelovotti]

I bought the Sigma HSM 70-200 a while back and the focus speed on my K10D is very quick and very silent.

Faster is a relative term. I was replacing a Tamron 28-300 and the focus throw on the Tamron was half the total rotation of the Sigma so even though the Sigma is "faster" the total time to focus is about the the same.

As time passes I can see all lenses going HSM it is sooooooooo sweet.

[drabina]

Um. also it wasn't a direct comparison because the Canon lenses were being focused by a Canon body. That's really the big difference here: Canon body versus Pentax body, not screw-drive versus SDM.

That's why I have posted my question. I am trying to find out if K10D with SDM lens will perform better in AF area than my screw-drive lens. If K10D is capable of driving SDM lenses and I get such lens, then if the AF performance is what I have seen using Canon system, then I will be happy.

[Marc Sabatella]

The K10D *is* capable of driving SDM, but that won't make it as fast as the Canon, if hundreds of posters on these forums are to be believed. That is, there is near universal agree that Canon is just faster, period. Also, most people report SDM is little or no faster than screw-drive with Pentax. Much quieter, but not faster.

[nostatic]

what he said. SDM is dead quiet, but I don't notice it is any faster. This is on a K20d though, and I'm usually comparing a 50-135* to ltd primes.

[George Lama]

Yep - light hum/whir is all you hear with SDM.

I'm not sure any lens you add to any camera system should make AF purely faster as long as the body is doing the focusing. In the Pentax system they're nearly if not close to the same (no measurebation, just recall).

Something else you also need to consider: If the AF is "having trouble keeping up" with a 2 year old, is it an AF speed issue (by the time it focuses, the subject is gone), or is it actually a focus-lock issue (can't lock, no picture is taken). You could argue either way, however my personal experience with the K10 seems to indicate it just can't lock, so speed is rarely the issue :-)

If your 2 yr old is up to par with sport stars and race cars, then a higher-end Nikon or Canon would do you well. However, before you go jumping systems, do your research and you will find there can be issues with all systems - here's a case in point - AF problems with Canon's nearly $5,000 body only flagship as reported back in August:

"In our view, there's more work to be done. In bright conditions, the EOS-1D Mark III still doesn't match the EOS-1D Mark II N; at times the two are close, and at other times the EOS-1D Mark III's autofocus performance is below what we consider acceptable," Galbraith said. "The culprit that continues to cramp the EOS-1D Mark III's autofocus style is bright sunlight."

So if bright sunlight cramps up arguably one of the most used bodies in sports photography, imagine what those guys think after dropping that kind of money on their jewels .

---

Last edited by George Lama; 12-27-2008 at 12:36 AM.

[kristoffon]

It should be noticeably faster only with lenses with lots of glass. Small zooms and primes aren't really a strain on the drive motor.

[RiceHigh]

RiceHigh's Pentax Blog: Lack of Speed of the SDM

From the above controlled experiment by that user, at least the motoring speed is slower / lower with SDM even for the same Star DA lens on the K10D.

I have K10D camera with Sigma 17-70mm lens. I really like this combo but now autofocus has trouble keeping up with my two year old. I am sort of exploring options (including other camera systems) but would like to know if upgrading my screw-drive Sigma to Pentax SDM equivalent (also 17-70mm) would give me faster AF?

I did play with Canon's ultrasonic lenses and they seem very quick in focusing comparing to my K10D with Sigma lens. Though it wasn't direct comparison because Canon lens had different zoom range.

I have few lenses that I could sell to raise money for the difference in price between Sigma and Pentax but that's about it. Upgrading to K20D is out of the question for now.

Thanks.

[kenyee]

If your 2 yr old is up to par with sport stars and race cars, then a higher-end Nikon or Canon would do you well

That's a bad comparison...you're further away when taking sports photos and kids move relatively fast.
I'm not sure any system would be fast enough...a Nikon friend also complains his kids move too fast. It's best to try a system out to see if it's "fast enough" for you.
But remember it's digital...snap away and delete the bad photos...

[Zebooka]

Would SDM lens speed up K10D's AF?

No answer.

Actually:
1) Focus speed depends on body hardware + software
2) It also depends on Lens hardware
2.a) for screw-drive lens speed depends on how long is focus ring travel + gear-ratio of screw mechanism
2.b) for SDM-Only (DA 17-70/4, DA* 55/1.4) lenses it depends on gear-ratio of "Current Frequency"/"Whirl Speed" and focus ring travel
2.c) SDM+ScrewDrive lenses such as 16-50, 50-135, 60-250, 200, 300 have NOT ring SDM motor like canon L lenses do, but they (DA*) have micro SDM motor, so actually they act like screwdrive lenses.

And, there is no clear answer what will be faster - it depends on list of parameters. Test yourself, and get answer for correct configuration.

[code4code5]

I think the issue isn't so much with screw drive AF or HSM, but the lack of a focus assist in low light without using the on-board flash. I've never had trouble keeping up with my daughter when the ambient light is sufficient, but indoors it is a struggle. Often times indoors I will go back to my M50 and just use manual anywayl.

Here's an example of outdoor AF performance, taken with the SIgma 70-300 DG Macro.

[George Lama]

That's a bad comparison...you're further away when taking sports photos and kids move relatively fast.

It was of course an attempt at humor/tongue in cheekyness.

[drabina]

Something else you also need to consider: If the AF is "having trouble keeping up" with a 2 year old, is it an AF speed issue (by the time it focuses, the subject is gone), or is it actually a focus-lock issue (can't lock, no picture is taken). You could argue either way, however my personal experience with the K10 seems to indicate it just can't lock, so speed is rarely the issue :-)

It is the lock that's the problem. The only thing is that if the AF was faster, then it may lock faster and do not miss as many shots. It looks like K10D with my Sigma lens just hunts and is never able to determine when to lock the focus.

I did play with Nikon D40 at the store trying to focus on people moving and it looked like it was spot on. I had a feeling that AF knew which way to turn so I never had any back and forth AF action.

[mike.hiran]

It is the lock that's the problem. The only thing is that if the AF was faster, then it may lock faster and do not miss as many shots. It looks like K10D with my Sigma lens just hunts and is never able to determine when to lock the focus.

I did play with Nikon D40 at the store trying to focus on people moving and it looked like it was spot on. I had a feeling that AF knew which way to turn so I never had any back and forth AF action.

a couple points... most stores are lit much brighter than a standard house so I'm not sure that experience is giving you useful info. maybe bring in your camera to the store and do a comparison w/ both cams in a similar situation. I met someone at a store who I was testing a lens at so I did quite a few tests and focusing w/ the k10d in store lighting conditions was not an issue. ymmv, though...

at a family event I was at recently, I was sitting at a table w/ my cam on the couch when my nephew starting goofing off. my unlce's D70 was on the table, so I grabbed it and tried to shoot... it fired but was out of focus. of course, I've only used his camera for maybe a total of 15 minutes prior so I am not anywhere near comfortable using it so maybe I was doing something wrong.

at that same family event, my nephew began chasing my uncle around and my attempts to capture this w/ my k10d failed miserably. I was too close making the effective DOF really shallow and since they were moving, I'm not sure I had a chance. Using the DOF calculator, at f2.8 and 28mm from 4 feet...
Subject distance 4 ft

Depth of field
Near limit 3.7 ft
Far limit 4.36 ft
Total 0.66 ft

so a 6 inch movement, and I my pics were fuzzy. but if I was 10 feet away:
Subject distance 10 ft

Depth of field
Near limit 8.28 ft
Far limit 12.6 ft
Total 4.33 ft

but realistically, I'd have zoomed in, to lets say 50mm at f2.8:
Subject distance 10 ft

Depth of field
Near limit 9.39 ft
Far limit 10.7 ft
Total 1.29 ft


If this were to happen again, I'd try to back up, crank up the shutter speed and I might go w/ manual focus if not AF-C. It's a tough call though once that "event" starts happening... do I move further away to improve my chance of getting a decent shot but risk having the event be over, or instead, do I take a hail mary shot hoping things work out but knowing that they probably won't.

Thing is, w/ sports like tennis, car racing, or motorcycle racing, the k10d does great. Distance and predictability makes these activities easy... but a hyper 2 year old w/ low, uneven tungsten light in close quarters? that's a tough shot. I'd almost think that using my 12-24 would be best since at say f4 @ 14mm everything between 3 and 12 feet would be in focus so I could just shut off AF....

Sorry for the rambling, but I'm trying to work through this scenario too...