[kenyee]

Nobody made serious test of SR and AF with 1.03 and with 1.00.

Nope. Not yet.
I even provided steps on how to do the AF test w/ a videocam on multiple threads here and on DPR

[cupic]

As you can see, my first shot after the upgrade that of my 3rd shot of Raki, hand-held at 1/13th of a second still seemed to have come out okay

I'd say go for it. (The firmware upgrade, that is)

Mate come back after 6 shots


cheers

[Ratmagiclady]

Nobody made serious test of SR and AF with 1.03 and with 1.00.

I'm sure someone will. Have you tried it, yourself? The update's only been out for two days of what's a holiday weekend for most folks here. I certainly noticed a difference immediately in the AF's behavior..sounds different, moves different: sometimes it really cranks hard where it didn't before, and stops on a dime: I never had a big complaint with the AF speed to begin with: the way I shoot, it's always been adequate, not that faster isn't nice to have. (I'm much more concerned with accuracy, past a point)

I did notice that I was having an easier time with certain of my AF targets around the house, mostly with my 1.4 FA, which is far and away my most-used lens. The only other AF lens I have is the kit lens, and that seemed a lot more confident, though I haven't used it much.

Anecdotally, something's certainly different. I'm pretty sensitive to how things move and sound, (I like to be able to predict what automation will do) and the change was pretty obvious to me. Under what conditions might there be a speed difference? I'd look where there's previously been a lot of 'jiggles' of small corrections: there's usually fewer of those by a long shot: typically there'd be two or three for a lot of shots: often there's one or none, now.

It'd be interesting to see what a scientific test might show, but it does seem there's an improvement in speed. I don't really get it, but that's what I'm seeing.

[legacyb4]

I've noticed the same thing for sure; "sharper" movements is the best way I can describe it.

I certainly noticed a difference immediately in the AF's behavior..sounds different, moves different: sometimes it really cranks hard where it didn't before, and stops on a dime.

[Rener]

add me to the happy camper of the V 1.03, I noticed the difference. Seems my 16-45DA gears are subjected to a massive torque input.

cheer,

rene

[danny09]

For all the "Improved AF" folks did you see if any of the shots were Out Of Focus (OOF) ? I mean if K20D locks faster onto the target but if its OOF....whats the use of a faster AF locking?
Forgive my stupid questions..i m a newbie :-)

[Rener]

For all the "Improved AF" folks did you see if any of the shots were Out Of Focus (OOF) ? I mean if K20D locks faster onto the target but if its OOF....whats the use of a faster AF locking?
Forgive my stupid questions..i m a newbie :-)

Hello Danny, mine is right on.

cheer,

rene

[Ratmagiclady]

For all the "Improved AF" folks did you see if any of the shots were Out Of Focus (OOF) ? I mean if K20D locks faster onto the target but if its OOF....whats the use of a faster AF locking?
Forgive my stupid questions..i m a newbie :-)

Nah, it's a good question. My first concern when I noticed the change, actually. I'd gladly wait a hair longer to really nail focus, if I had to make that tradeoff. It was a great day yesterday to try out the SR, since I just wasn't very steady... not so great to evaluate sharpness, ...looking things over today, I'm not seeing conspicuous problems: the AF *will* still hunt, by the looks of things, if it thinks to need to. It just does it a lot less. Whatever causes the perceived improvement, I don't think it's some kind of 'cheat' like that. If so, the tradeoff is subtle.

My sweetie's already put up with a lot of test shots since yesterday, but I think at some point I can ask for a moving subject. Maybe some of the real action and long-lens shooters can have some input.

Just uploaded version 1.0.3 and the camera exploded in my face and ripped the skin off back to my ears. Other than that no problem, works fine...Love it!

So are you better looking now Ben?

[vinzer]

add me to the happy camper of the V 1.03, I noticed the difference. Seems my 16-45DA gears are subjected to a massive torque input.

cheer,

rene

The whirring on my Tamron lens certainly sounds as if it's working harder with the firmware update. As to quicker AF, I'm seeing it, but it's not a night and day difference, and it certainly won't give the D300's AF a run for its money, but I'm satisfied. Like Ratmagiclady, my style of shooting doesn't benefit much from faster AF, but it's good to have.

I am getting better results from the SR, though, in my eye. I have gotten down to 0.4" (is this 1/5 of a second?) at 75mm (112.5mm equivalent) with usable results. And the SR seems to lock faster, too.

Nah, it's a good question. My first concern when I noticed the change, actually. I'd gladly wait a hair longer to really nail focus, if I had to make that tradeoff. It was a great day yesterday to try out the SR, since I just wasn't very steady... not so great to evaluate sharpness, ...looking things over today, I'm not seeing conspicuous problems: the AF *will* still hunt, by the looks of things, if it thinks to need to. It just does it a lot less. Whatever causes the perceived improvement, I don't think it's some kind of 'cheat' like that. If so, the tradeoff is subtle.

My sweetie's already put up with a lot of test shots since yesterday, but I think at some point I can ask for a moving subject. Maybe some of the real action and long-lens shooters can have some input.

That's what I'm seeing with the update, too.

I don't have much time (nor resources) to do proper testing with two different K20Ds loaded with different firmwares, so I'm going by personal feel on this one. It would be interesting to see if there were changes made to improve C-AF, too.

[Entropy]

SR wouldn't affect autofocus, as the SR only moves the sensor, not the AF mechanism.

[augustmoon]

Focusing speed is about the same, maybe even a tad slower with my Voigtlander lenses so far. (reverse-placebo effect?).

[Arpe]

Just loaded 1.03. Couldn't identify any particular SR improvement, but will try something abit more scientific. It did, however, fix a problem with ISO limits in TAv mode.

Interesting I've never had a problem in TAv mode, and I use it a lot.

Wonder if the Hyper-programme bug is fixed whereby the "something I can't remember wasn't working properly so I stopped using it and have forgotten what it was"! Something like auto-ISo only working when altering aperture and not working when changing shutter.

[vinzer]

SR wouldn't affect autofocus, as the SR only moves the sensor, not the AF mechanism.

Nobody suggested it does, at least what I read thus far in this thread.

[Kameraten]

This is a deal breaker for me. Still an hour ago I could not but concede that my Sony A700 was far superior to the K20D exactly when it comes to shake reduction (made a number of tests with a number of lenses, some identical 3. party lenses for Pentax and Sony). But now – they are equal! The diffference in SR performance is quite noticeable.
Pentax very much needed to improve on this. A while ago I noticed that a borrowed Nikon D70 WITHOUT SR did as well as the K20D with it – the mirror movement is better dampened in Nikon, as it is in Sony and probably in Canon as well.
If this keeps me from having to go back to Sony for my work I'm more than happy.
Haven't had time to notice any difference in AF yet, but I suspect Hoya would have been happy to tell us if an improvement had been made. What I have noticed earlier is that SR in some instances seems to blur by moving the sensor at the wrong moment, when the camera movement has already stopped. This could make a difference.