Pentax 645z vs Nikon Z7 comparison

[MichaelEzra]

11/09/2018 UPDATED WITH ROUND-2: ADDED Z7 @F5.6 and F8.0


Having both 645z and Nikon Z7 I made a comparison using these two cameras with some select lenses at my disposal.

I made test shots of the same outdoor scene with very little wind and sunny weather.
Both cameras were on tripod, in live view with 2 sec shutter delay and Z7 with IBIS and EFCS.

645z shot @F11 and Z7 @F10



Lenses used:

60mm FF Equivalent
    smc PENTAX-FA645 45-85mm F4.5 @75mm
    smc PENTAX-FA645 75mm F2.8
    Nikkor 60.0 mm F2.8 G ED
    NIKKOR Z 24-70mm F4 S @60mm

35mm FF Equivalent
    smc PENTAX-FA645 45-85mm F4.5 @45mm
    NIKKOR Z 24-70mm F4 S @35mm
    Tamron 15.0-30.0 mm F2.8 @30mm (OS enabled)


28 mm FF Equivalent
    smc PENTAX-A645 35mm F3.5
    NIKKOR Z 24-70mm F4 S @28mm


Raw images developed in the latest ACR 11.0.
No noise reduction applied.
CA corrected in ACR. Minor purple fringing corrected for Z7.
All images sharpened identically in ACR with amount 50, radius 1.0, detail 25, masking 0.

The bottom line:
1. In this comparison 645z just kicks ass.
2. Z7 with the 24-70 S lens is solid; the new zoom's performance is unusually uniform across range.


It is interesting to see how Pentax lenses differ and how Nikon mount lenses differ. View images in full size (right click and select "view image"). Have fun!

[32BT]

IBIS and tripod?

[Fotophil]

A recent test reported in DPReview suggests that the  Nikon Z7 IBIS System does not cause blurring problems when used on a tripod. I wonder what the Nikon recommendation is for using the IBIS when mounted on a tripod?

[Bo_Dez]

Those Nikon colour look hideous. What's up with that?

[MichaelEzra]

Nikon NEF files were converted using the camera-matching Standard ACR profile as a base. This leads to specific sky coloring that can certainly be tamed with further processing, but is not part of this test.

[Telecaster]

The Pentax 35/3.5 and 75/2.8 are both very nice lenses. Add the 150/2.8 and you've got a lovely trio. They're why I've kept my 645D.

-Dave-

[MichaelEzra]

Yes, I just got the 150 F2.8. Magical!:)

[Paul2660]

From the images provided the 24-70 Z lens is impressive. Seems to hold up well in regard to overall sharpness and details.

Paul C

[chapel]

Lately I have been doing some still life with my Pentax 645z, with my Manfroto Carbon Fiber Tripod watching live view zoomed all the way in to check focus, I could see the movement stop after 7-8 seconds so I switched to the 12 second delay and it makes a huge difference in the sharpness. This is both with the 45-85 and most noticeable with the 90mm F 2.8. I always used the 2 second delay thinking it was enough but I don't think that anymore.

Greg

[Jack Hogan]

Thanks Michael, interesting test.  But why is the Z7 at f/10?  If you are shooting the 645z at f/11 you should set the Z7 at f/8 or thereabouts, per the crop ratio.  It would improve sharpness on the Z7 and also ensure about the same DOF aotbe.  Also, contrast seems to be different.  ACR/LR are famous for having different settings under the hood even though all sliders are in the same positions.  You may want to try Adobe Color/Standard or similar profile on both.

Jack

[MichaelEzra]

Greg, interesting about the delay, I will try. 645z sharpness in this test already stands on its own.

Jack, I will check the F10 vs F8 for diffraction on Z7. However, I expect that it will not be enough of a difference to reach 645z sharpness with the lenses used.

Overall, for landscapes, I find F11 to be the aperture of choice for 645z and the F10 so for FF Nikon (will reconfirm with Z7) - this is based on both sensors and the lenses, their corner and edge performance and central sharpness as well DOF for the landscape scenes. Nikon mount lenses are not as great in the extreme corners, so stopping down to F10 helps that, while still limiting the effect of diffraction.

[Aram Hăvărneanu]

Since I am thinking of buying one of these two cameras, I can't thank you enough for this test.

[MichaelEzra]

Aram, happy to hear that its helpful.
Z7 is a very, very versatile camera for both stills and video.
645z has an edge in image quality for stills. With skilled use and processing almost equal results could be obtained with Z7.

I bought Z7 as a backup for D800e and as a video camera. Clearly Z7 can serve as a backup for 645z. It is probably time for me to part with my Mamiya ZD camera:)

[chapel]

Michael,
I think all your work looks awesome. In this example the Pentax definitely shines. Your mention of the delay just reminded me of my own experience on the sharpness with the longer delay. I bought a used Blackmagic 2.5 cinema camera for video. Still learning it.
Greg

[BernardLanguillier]

F10 for the Z7 is already deep into diffraction territory. F5.6-f8 is about optimal.

Cheers,
Bernard

[MichaelEzra]

UPDATE - I updated original post with Z7 at apertures 5.6 and 8. Lighting & weather is different, but that's the best available now in NYC:)

Bernard, you are correct, sharpness on Z7 improves dramatically at F5.6, although we loose DOF in corners.
The good news is that Z7 sensor is capable of very high and fine resolution showing fine texture which used to be attainable by MF only.

[bjanes]

UPDATE - I updated original post with Z7 at apertures 5.6 and 8. Lighting & weather is different, but that's the best available now in NYC:)

Bernard, you are correct, sharpness on Z7 improves dramatically at F5.6, although we loose DOF in corners.
The good news is that Z7 sensor is capable of very high and fine resolution showing fine texture which used to be attainable by MF only.

Lloyd Chambers is not that impressed with the 24-70 f/4 S and has noted weak corners. This may be due to field curvature (could be verified by refocusing at the corners). He also notes that distortion correction in software stretches the peripheral pixels with resulting loss of microcontrast. To get a better assessment of the Z7 sensor capabilities, one should use a good prime with known capabilities, such as the Zeiss series.

Bill

[faberryman]

Lloyd Chambers is not that impressed with the 24-70 f/4 S and has noted weak corners. This may be due to field curvature (could be verified by refocusing at the corners). He also notes that distortion correction in software stretches the peripheral pixels with resulting loss of microcontrast. To get a better assessment of the Z7 sensor capabilities, one should use a good prime with known capabilities, such as the Zeiss series.

So much for the vaunted large throat diameter ushering in a new era in optics.

[MichaelEzra]

I agree, there is surprisingly *lots* of software correction on the 24-70 S corners. Even though they look terrible compared to the stellar center, they are pretty good compared to many other Nikon mount lenses.

The Z7 sensor capabilities are best illustrated in the center.

If there is a lens which can project as sharp of an image into the corners, this sensor would record that, almost like it does in the center:)

[alan_b]

So much for the vaunted large throat diameter ushering in a new era in optics.

It’s a compact zoom lens that costs only $600 new.  I’m pretty impressed, YMMV.