Web Analytics RiceHigh's Pentax Blog: AF-S Speed and Accuracy Test for Five DSLRs/ILDC

Thursday, August 11, 2011

AF-S Speed and Accuracy Test for Five DSLRs/ILDC

Someone at the German camera forum DigitalFotoNetz has made a post (Google translated) about the latest AF test by the German photo magazine Color Foto amongst the Pentax K-5, Nikon D7000, Canon 7D, Sony A55 and Panasonic GH1.. via DPR Post here (OP) and here (for the brief human-translated results) and also a Chinese article here (Google translated).

The Chinese article has a more detailed summary of the whole test and below is the summarised results of the test. It is said that 57 lenses (original or not) were used in the test and 1,300 shots were taken:-

Camera
Good Focus
Acceptable Focus
Blurred Focus
Average Focusing Time
Pentax K-5 62.4% 28.9% 8.6% 0.39 second
Sony A55 62.1% 16.4% 21.4% 0.33 second
Canon EOS 7D 40.2% 29.7% 30% 0.46 second
Nikon D7000 35.2% 40% 24.8% 0.44 second
Panasonic GH1
85.3%14%
0.7%
0.28 second

The lenses used in the test are also listed, except for the Pana. Do note that the lenses used are not identicial nor comparable across brands, nevertheless.

So, it must be a BIG victory for the Pentax and all the Pentax fans must be really happy! ;-p But the DSLR K-5 still lost in the race against the two-year old little Pana mirrorless, btw. :-o

Comments (9)

Loading... Logging you in...
  • Logged in as
It is a victory. I would be really interesting to see what the depth of field was on the GH1's lenses. WHat it does prove is while Canon may say its in focus, its often not.
GH1 is a mirrorless camera, so It should be in another category, but it does show that mirrorless gives better focusing performance. But now, Pentax is the best among DSLR, that to me is a relieve
This result is to be expected, because they were comparing apples with pears. The 7D, A55 and Pana are not in playing in the same league as the K-5.
Also, what about specifically testing low-light focusing?
Sorry... Don´t believe this test :o(.

I have both... D7000 and K5. And in D7000 is better AF.
3 replies · active 712 weeks ago
Mike, I don't believe it myself. Just reported it anyway.. :-)
They tested accuracy, not speed. D7000 is no good in terms of accuracy than K-5.
No any miracle that EVIL has better accurancy than DSLR. It's axiom.
What does average focusing speed mean, then?
The test is not fair, because they use different lens in the test.
They should use the same lens to do the comparison.
The super slow Tamron 70-200 2.8 are tested on the 7d and D7000, but not on the K5 and A55.
For such test it is critical to test each lens after focus calibration for given camera/lens. Without calibration test results are incomparable among different cameras. For these tests the calibration was not mentioned, at least I didn't found it – i suppose calibration was not performed....then I have not much trust to test results. I contrast you could check focus consistency test of all tested cameras in optyczne.pl. For Pentax k-5 http://www.optyczne.pl/153.3-Test_aparatu-Pentax_... (use translate.google.com for the polish site translation to english ). Especially results of K-5 are pure, Nikon D7000 is better but also do not excel here, Canon 50D is much better then these two. As explained on their web these tests are performed after camera/lens calibrations. I have more trust to these test. I personally had compared focus consistency of my Pentax k-x and Canon 450D, where Canon is considerable better, similar to test results on the optyczne.pl for these cameras. I compared test results of focus consistency on Optyczne.pl for many dslrs and found that best results for low /mid class dslrs have Canon, then Sony, Pentax is considerably worse,but also Nikon didn't impress. In contrast situation changes for high level dslrs, Nixon is considerably better than Canon. Intresting is that following very popular cameras have really bad test results: Pentax K-5, Nikon D90, Canon 5D mkII (regarding focus consisteny – the disadvantage of the Canon is its high pixel count, higher pixel count requires more precise focusing )

Post a new comment

Comments by