Olympus E-P3: AF speed and responsiveness

Olympus E-P3 was introduced on 30th of June. As name says it is the third version in E-P series which I use. I am interested because Olympus promises now to correct the slow AF which hampered previous models. All in all the responsivity of this camera should be at a totally new level.

I created a simple test arrangement to get a real world understanding on E-P3´s AF speed. A sat at a table with my elbows leaning on table. On the others side of table I had a newspaper (at 1m distance) and directly behind there´s a plant across the room at 5m distance. I set both E-P3 and E-P2 to do single-AF at a press of shutter button AND shutter is activated only after focusing is done. Other settings: Program AE, ISO 400, single shot mode, medium JPG, only center focusing spot active. 

Now I aimed focusing spot at newspaper and pressed shutter button until camera focuses and shutter goes, aimed focusing spot at plant and pressed shutter button until camera focuses and shutter goes, aimed focusing spot towards newspaper and pressed… I went on doing this until I counted 75 shots. The number 75 is not important. The point was to make enough shots so that slight differences between each single shots become averaged when shooting with both cameras. I repeated the same drill for four lenses:

  1)  The new Olympus m4/3 kit zoom lens 14-42mm 1:3.5-5.6 II R. Tested at 42mm focal length.

  2)  Olympus m4/3 zoom lens 40-150mm 1:4-5.6. Tested at 50mm focal length.

  3)  Olympus 4/3 zoom lens 12-60mm 1:2.8-4 with Novoflex m4/3 adapter. Tested at 50mm focal length.

  4)  Panasonic Lumix G 20mm 1:1.7

Then I looked into EXIF data of first and last shot of each series to see the time difference between them. I divided this time by number of shots to get the average time for aiming camera and focusing and exposing each shot.

These are my results:

        E-P3     E-P2        E-P3 is faster by

1)    0,82s    1,41s            42%

2)    0,89s    1,46s            39%

3)    1,96s    2,26s            13%

4)    0,87s    1,54s            44%

I have no way to know how long it took to aim the camera between each shot but it was not much. Just turned the camera a few degrees up or down. Olympus gives a shutter response time of ONLY 60ms for E-P3. My conclusion is that E-P3 is in practise twice as responsive camera as E-P2. With older 4/3 system DSLR lenses the lens is the restricting factor. I would guess that these lenses would benefit of a new firmware for m4/3 bodies.  

So how responsive E-P3 is then? Olympus claims it to be the fastest focusing camera of all system cameras. Lets see!

I repeated the same test with Nikon D3x and three Nikon lenses:

  5)   AF-S Micro Nikkor 105mm 1:2.8 G

  6)   AF Nikkor 85mm 1:1.4 D

  7)   AF-S Nikkor 14-24mm 1:2.8 G. Tested at 24mm focal length.

Results:

D3x

5) 0,97s

6) 1,15s

7) 0,70s

Results for differences between these three lenses are logical and should be familiar to everyone who has used them. Wide angle zoom 14-24mm is fast because there the distance to move focusing element is so short. Other than that look at E-P3! E-P3 is faster (with three consumer grade lenses) than Nikon. It´s responsiveness is up (or should I say down) there with a super fast professional DSLR with 10 fold price tag! 

UPDATE: 

Same test with the new Olympus M. Zuiko 12mm 1:2.0 lens. This lens is Olympus´ first lens that which they consider as belonging to the top level in their three level lens grading.

Test result with E-P3 0,74s.  E-P2 was not tested because speed difference is already known.

Already while testing the new kit zoom lens I was somewhat distracted by Live View refresh rate. Live View goes off and back on with each image, as sensor is reseted before and after exposure. (I did not have record view on, just viewfinder image). With this 12mm lens, it is quite obvious that now the restricting factor is not focusing but other functions of camera. Mere focusing back and forth is a lot faster, also faster than Nikon 14-24mm. With this lens focusing is done immediately as you press the button. How ever, this result tells how fast you can take two consecutive images with varying focusing distances. Focusing is very fast but the responsivity of a Live View camera like E-P3 is not yet quite there with a professional reflex camera like D3x.

-p-

Nikon D3x, resolution king?

Everyone is driving yesterdays car and shooting with yesterdays camera. This is how it is always from technolygy point of view. I have never thought that new models make my old cameras any worse and kept shooting with them as they fit into my needs. However, from professional point of view, staying put means falling back in a competitive environment. That´s why I have been really interested in Nikon D3x since it´s launch. All the specs and other information gave me the impression that Nikon has pushed the limits of sensor technology as far as is possible today. It looks like that low price was not a target for Nikon. I wrote some thoughts on D3x launch earlier in this blog.

Speaking of technological advances, it must be said that progress does not always go straight forward. A good example is Leaf Volare back, which was the epitome of quality long past what you could read from a spec sheet of many later launched one shot digital cameras. How ever good, Volare was the last of it´s kind because there was no market for that kind of backs.

After a period of shooting

I have now used D3x (actually two bodies) and Nikon lenses (ranging from 14 to 200mm) for a month and half. I´ve been shooting both for my own pleasure and professionally. After a background of nearly 10 years with Canon DSLRs, it took me a while before Nikon logic started to become sensible. Before that D3x seemed like a mystical computer with wrong UI. When I had managed to configure bodies according to my shooting style and my fingers started to find their way to buttons and wheels, using D3x started to be pure delight.

After this period of shooting I am assured to say that D3x´s image quality is the best in any DSLR ever made. And with DSLR I´m referring to 35mm category. If D3x is the best camera for my personal shooting or not is another matter. My present choice Canon 5D MkII has the edge in some areas of usability. Also if I were now to buy a DSLR for my personal use with my own money, I think my choice would be 5D MkII because of price difference. If I had to make the same choice today as a studio shooter, I would not hesitate to pick D3x instead of any Canon.

A Lappland shot with Nikon D3x. Magnifient scenery like this really benefits from as much resolution as ever possible.

Resolution test

The rest of this essay concentrates solely on my resolution test with D3x. I´ll write about other issues later. I have been using the same test setup for digital cameras for a decade. This setup tells about resolution and color. I´m not interested in any absolute values but comparing cameras against each other with my background experience of using the same equipment in practise. My comparison setup is easily reproduced, it gives always the same result and most importantly it answers to the right question. Test group was now D3x, 5D MkII and PhaseOne P45+ digital back. I have tested earlier Canon´s 5D MkII and 1DsMkIII, so I know that their resolutions at ISO 100 are very similar. They are not exactly the same but they are so close to each other (depending on RAW converter in use) that any critics on comparing dissimilar DSLR bodies is pointless here. Besides my experience of shooting with D3x told that it is meaningless to do any more pixel peeping between these two Canon bodies. 

It´s not possible for me to compare DSLR bodies or digital backs directly, with out optics. And I wouldn´t care because I`m a photographer. Lens is very decisive in a test like this. That´s why I included more than one lens from Canon and Nikon. In suitable focal length range I had three Nikon lenses: AF Nikkor 85mm 1.4D, PC-E Micro Nikkor 85mm 2.8D and AF-S Micro Nikkor 105mm 2.8G ED. As corresponding Canon lenses I chose my own great and proven EF 85mm 1.2L II USM plus from Potkastudios lens cabinet 3pc TS-E 90mm 2.8 and 2pc EF 100mm 2.8 Macro USM. So Canon was given three extra tries. Moreover, after shooting with it I had noticed that 105mm Micro Nikkor can´t be okay. It is a demo lens from Nikon Nordic, which according to my location shots seemed to have something wrong with possibly VR unit. My resolution chart (VR off) confirmed that this lens needs to be sent to repair shop. It was so much inferior to all other lenses. That left Nikon with 2 chances against Canon´s 6.

This case with Micro Nikkor shows again how important lenses are with a body like D3x. Camera body´s extra resolution shows only with top lenses. Perhaps Micro Nikkor´s faultiness might have passed my notice with a body like D3? At least nobody had noticed it before...

I shot tethered into CaptureOne Pro 4.7 so that files had same exposure in every case. After shooting, files were converted in C1. After that I imported RAW images also into Adobe Lightroom and matched tonal ranges to be as close to C1 files as possible. Canon files were opened also in Canon DPP software and Nikon files in Nikon Capture NX2, and again developed to match C1 files. This test period is my first experience with NX2. Matching files there succeeded only okay. I need more practice. The other converters are familiar to me since they were born, and I have been a beta (and even alpha) tester for C1 and DPP.

Both Nikon and Canon files showed one lens as better than others. Nikon´s both lenses were so close to each other that in an ordinary photograph it would have been hard to tell any difference. Anyway PC-E 85mm was slightly better. In Canon camp my favorite lens 85mm 1.2L II beat all others. Beating is of course an overstatement but it was easy to choose. Note that test chart was only in the center of image. I was looking for the best possible resolution in camera, not comparing lenses.

As a side product this test gave a comparison of both companies´ still life tilt/shift lens with their best resolution bodies. Canon, this is pathetic, hurry up with a re-designed 90mm TS!!!

Nikon advertises that D3x rivals with digital backs. Also in that context it was interesting to see a difference between 24,5 MP DSLR and 39 MP digital back. P45+ back was attached to Mamiya AFDII camera, and lens was Mamiya´s excellent 120mm Macro.

You can find comparison shots of these three cameras (the best lens of each) here. To see them at 1:1 they should be downloaded to your computer. Also use your own choice of interpolation to upscale Canon and Nikon files to match PhaseOne. Only then you can see the real differencies.

Conclusions

My conclusions:

• 1. P45+, 2. D3x, 3. 5D MkII - clearly
• D3x is surprisingly much better than Canon and surprisingly less worse than P45+. Surprisingly refers to my expectations when considering AA-filters (or not) and linear resolutions. (P45+ suffers here slightly because shooting distances were defined by long edge of frame).
• My early expectations on D3x was proven right: Nikon has managed to maximize 35mm resolution without ruining other aspects of shooting pleasure and quality. D3s might be expensive but you get what you pay for. From what I see, besides of sensor, also AA filter, microlenses and other image quality related parts are chosen and designed with image quality in mind. Canon really has to run faster now when they are developing their new flagship model.
• D3x is a match to any digital back with MP count starting with 2. Beyond that it´s only natural that size matters as P45+ shows.
• D3x is a very, very interesting choice for a studio thinking about how to invest between DSLRs and digital backs. Studios with Canon gear, like we at Potkastudios, must invest heavily in both.
• Concerning RAW converters, C1 Pro is very good for all these cameras, Lightroom shows again it´s typical mushiness and shows this 5D MkII moiré which is absent with DPP. I´ll come back with NX2 after some more studying.

High image quality is much more than resolution. My opinion on D3x regarding those issues in next blog. 

-p-