[Confused]

Hi all

I occasionally read about complaints from various Pentax owners concerning the poor performance of the autofocussing abilities of their K10D's, especially under low lighting conditions. However, it would appear from reading the quote below that apparently the "grass is not always greener on the other side" ?

I DID eventually sell the 5D due to it missing focus in 1/4 of my shots. It was bad enough that I always had to use the center focus point due to the outer AF sensors not really working to well for me. I noticed even using the center point the focus would be off quite a bit.

http://www.stevehuffphotos.com/Steve_Huff_Photos/CANON_5D.html

Comments welcome.......

Best regards
Richard

[k100d]

yeah the 5D is also a common source of AF bitching on Canon forums. and since the 5DII uses a similar module, i think there is also bitching there.

but just because other people are having problems doesn't mean it's ok for pentax to have slow / inaccurate af w/ lenses missing ring motors

[Ash]

Quite an objective personal review of his experience with the Canon 5D and Nikon D200 and D2.

The AF technology continues to evolve - if you get your hands on a D700 or D3, though, you'll see how fast and accurate their ultrasonic focusing are, and under low light, quite superb - all for over 4 times the price of a Pentax system. Many of the consumer VR lenses, not being all that in terms of image 'pop' are still all over A$1,000 each.

I'm quite happy with the limitations of Pentax's AF in this regard...

[kenyee]

that's a known issue w/ the 5D. Only the center focus point is a cross type, so it's the most accurate. Everyone seems to use focus-recompose as a result...

Like I said, the grass is just a different color on the other side...not necessarily greener...just different...

[stevebrot]

yeah the 5D is also a common source of AF bitching on Canon forums. and since the 5DII uses a similar module, i think there is also bitching there.

but just because other people are having problems doesn't mean it's ok for pentax to have slow / inaccurate af w/ lenses missing ring motors

Huh? I think the obvious conclusion is that AF is not consistently excellent regardless of brand and that it is a rubbing point for Canon (and Nikon from what I hear) owners as well.

I have very limited exposure and experience, but I do know this:

• Canon AF is scary fast compared to my K10D.
• Pentax DA* lenses don't focus faster than body-driven lenses.
• The Nikon cameras I have handled don't seem to focus any faster and are less accurate than my K10D.
• I have seen Canon owners switching to MF in relatively good light because they could not acquire focus.
• I have seen Nikon owners switching to MF in poor light despite using the focus assist lamp.

My conclusion:

AF is much better than it used to be, but is still plenty poor regardless of system.

Steve

(Bought my K10D primarily with manual focus in mind...bought a split-image screen to be sure...)

[Confused]

Interestingly I also found this little gem about the 5D Mark II on Ken Rockwell's site:

The 5D Mark II's AF system is inferior for photographing moving kids in dim light…….most of my 5D Mark II shots made with a 50/1.4 USM just can't nail the focus

Best regards
Richard

[Clem Nichols]

And here is a quote from Popular Photography's test of the 5D Mark II:

"The 5D-era AF, however, is no longer competitive with the blazing speeds we see in most DSLRs these days. At the brightest light level in our tests, the 5D Mark II focused in 0.51 sec, while the Nikon D700 zipped along at 0.35 sec, and the Sony A900 roared to the head of the pack with 0.29 sec.

In low light, the Canon is downright sluggish, and in extremely low light (EV -1 and -2), it's inconsistent and sometimes fails to focus. But Canon rates the AF system to be effective down to only EV -0.5, about the same as a very poorly lit living room. This limits the utility of the high ISOs. Granted, you can focus manually, but in such low light, that's no mean feat.

Oddly, some older and newer Canon models have faster AF systems. Just as Nikon trickled down the D3's AF to the D700, we think Canon should have done this with the 1Ds Mark III and 5D Mark II."

CN

[Jonson PL]

The user Laurent also has the Eos 5D, he wrote in another thread that I just came by :

Pentax K10D beats Nikon D700 in low light AF [Page 2]: Pentax SLR Talk Forum: Digital Photography Review



On my own account, I had been having fun shooting with manual focus lenses on my K10, (and slower zooms in good light). So when I finally got my first AF fast prime for my K10, I was looking forward to see how it would fare compared to my film Pentax SLR.
I did some shooting in very limited lighting, shooting Iso 1250, f/1.9, and shutter times of ¼ sec, ½ sec, 0.4 sec.
I found that it did surprisingly well. It locked onto areas where I had trouble distinguishing the contrast myself. The setting was at night, with one light source set to another side of the room.
I really like that the Pentax AF system will not give up, but will keep trying till it finds the setting. Whereas other systems tend to try, but then give up if EV too low.

I think the trouble is that people come from P&S with unlimited DOF, and then get a 50/1.4 lens with extremely narrow DOF, and then expect it to lock on instantaneously on random moving targets like children moving about.

If I’m covering an event, I’m also very happy that not using P&S with flashes going off all around. Or a Rebel cam, with the bright AF white assist light, disturbing speakers, etc.

People can be nervous enough, making a speech at a wedding, or at an important reception; without the photographer putting more pressure on them. I’ve often encountered how people thought that I had hardly taken any pictures, and then be surprised to see that I had covered most. I prefer that the photographer be more invisible and just reporting what is going on, without playing a major part oneself.


Being inspired by the real Pro’s, I work to optimise equipment performance. I don’t put everything in auto, but apply my settings, maybe sometimes using spot AF, going with the appropriate WB, Iso setting etc.



Edit :
besides, regarding the Eos 5D, people buy it for FF, not blazing fast AF. With narrow DOF, people gotta be careful anyway, selected what they want in focus.
A friend of mine was using his 5D in tracking mode, and found it difficult. But most of the time he is happy, going manual with his Leica glass

[roentarre]

Despite all the whinging about Canon 5D's AF ability, I found the camera a pleasure to use. Not a big deal for me.

5D is certainly having more autofocus accuracy comparing to k10d.

[*isteve]

Despite all the whinging about Canon 5D's AF ability, I found the camera a pleasure to use. Not a big deal for me.

5D is certainly having more autofocus accuracy comparing to k10d.

My K20D is certainly having more focus accuracy than my k10D. Some need realignment of the mirror box. Its cheap and makes a huge difference.

I tried using a 5D with a 17-40 in moderately dim conditions (a well lit exhibition hall) and it couldnt lock on anything.

My K20D will lock in very low light even using off centre focus points, which if you are using a tripod is ESSENTIAL not just a "nice to have". I have used AF for 25 second night exposures with no issues, and it was accurate as well.

So personally I am very pleased with some of the aspects of the K20D AF. Could it use a faster processor? Sure, why not. As long as they keep it accurate and functional.

[Robert S Donovan]

The center focus point on my (old) 5D worked fine with lenses faster than f/2.8. The secondary focus points were rubbish. AF on slower (f/4) lenses wasn't very good at all. My K10D isn't quite as good as my 5D was but cost less than 1/4 as much.

[*isteve]

The center focus point on my (old) 5D worked fine with lenses faster than f/2.8. The secondary focus points were rubbish. AF on slower (f/4) lenses wasn't very good at all. My K10D isn't quite as good as my 5D was but cost less than 1/4 as much.

Hmm - not quite what it says on your blog though

Problem is, performance is heavily lens dependent as you correctly say. The AF system on the 5D is exactly the same as that on the 40D, which does much better in PopPhoto tests. One reason I suspect is that the kit lens on the APSC camera is the lightweight little 18-55 whereas the one on the 5D is the much heavier 24-105. I dont know, but its a possibility.

I actually thought the 70-200 F4 worked OK (on both) but the Tamron 70-200 F2.8 was utterly hopeless on the Canons but works great on the Pentax.

The Sony A200 blows all of the competition including the D3 into a cocked hat. I'd love to know how D3 owners explain that

[Robert S Donovan]

Hmm - not quite what it says on your blog though

Just to keep the record straight here's what I said about my 5D in my Pentax vs. Canon AF performance post:

"It was hard to draw many comparisons between the 5D and K10D. They really are quite different cameras- with the 5D being much more like a 35mm film camera than a high-tech digital. Plus, the clunky old 50mm prime I started out with on the 5D really wasn’t anything like the Sigma EX zoom lenses I had had on my K10D."

I did make an anecdotal comparison between a K20D and a Canon 1D MK III. The direct comparisons were made between a Canon 350D and my K10D.

I do agree that AF performance is heavily dependent on lens choice. My new FA 35mm f/2 AL blows away the DA 35mm f/2.8 Macro I reviewed this weekend in AF speed. Also, I am testing an FA 50mm f/1.4 and it does wonders indoors in very low lighting...so low that I can't accurately focus my manual 50 1.7 by eye

[*isteve]

Just to keep the record straight here's what I said about my 5D in my Pentax vs. Canon AF performance post:

I did make an anecdotal comparison between a K20D and a Canon 1D MK III. The direct comparisons were made between a Canon 350D and my K10D.

I do agree that AF performance is heavily dependent on lens choice. My new FA 35mm f/2 AL blows away the DA 35mm f/2.8 Macro I reviewed this weekend in AF speed. Also, I am testing an FA 50mm f/1.4 and it does wonders indoors in very low lighting...so low that I can't accurately focus my manual 50 1.7 by eye

Macro lenses are always slow - its that very long focus throw to allow for precision in the tiny DOF. My Sigma F105 macro is a slow as a dog but deadly accurate.

Odd though that the Rebel XSi also appears to focus much faster than a 5D according to PopPhoto. (In fact the 5D and K20D are similar). Its also faster than the 1DSmk3. That should tell you something about how complex this issue can get.

I really dont think static AFS testing is very useful and I've yet to see a decently constructed AFC test.

[OregonJim]

The AF technology continues to evolve - if you get your hands on a D700 or D3, though, you'll see how fast and accurate their ultrasonic focusing are, and under low light, quite superb - all for over 4 times the price of a Pentax system.

And to think that the Polaroid SX-70 had fast ultrasonic AF.....35 years ago!