Comparing images from Hasselblad H4D50, Sony A7Rii and Leica SL

[Peter Walker]

Advertisement (gone after registration)

I’m not one for spending a lot of time pixel peeping but, last Sunday, I noticed that, over the last couple of years, by coincidence, I had taken a photo of the same subject (a ruined temple with lots of fine details), at the same place, from roughly the same distance, at roughly the same time of day, with three cameras:

1. Hasselblad H4D50 - prime lens 100mm (~60mm) (4.5x6cm sensor @ 50 Mpxl)
2. Sony A7Rii - prime lens 55mm (35mm sensor @ 42 Mpxl)
3. Leica SL - zoom lens at about 50mm (35mm sensor @ 24 Mpxl)

 

Coincidentally, I had used roughly the same ISO, shutter speed and aperture all 3 times.

 

So I started doing some comparisons in LR, zoomed in on the TIF files at 100%.  Nothing scientific at all, just looking around the images: centre, corners, details, colours, sky, etc.

 

From a pure resolution point of view, I could not see any difference, either in the centre or in the corners.  This means that: 

• there’s no penalty from the lesser number of pixels;
• that the Leica SL zoom performed as well as the two prime lenses, and;
• in the 3 years since the H4D-50 sensor was built, sensor technology has advanced such that a CMOS 35mm sensor is equivalent to the older 4.5x6cm CCD sensor.

From a colour point of view, the SL was discernibly better - against the 'blad, not by much, but better.  Much more realistic than the Sony colours.

 

Other noticeable differences: skies are definitely better rendered by the SL sensor, better gradation, smoother, less colour grains.

 

From a usability point of view, the SL is way ahead of the other two.  The hit ratio of quality images (focus, exposure) has been much higher than the other two cameras.

 

After 40 years of knowing that "zoom = quality loss", with this SL, I need have no more concerns about using a zoom.  With an additional bonus from this weather-sealed camera: one lens on the camera all day long, never have to expose the sensor to dust. No more juggling lens swaps.

 

But kudos to the Sony too for keeping up with the two more expensive systems.  But, it suffers greatly in usability and a lot of the lenses, by comparison, are poor, especially the zooms.  The Zeiss 55mm that I mentioned above stands out in the Sony range as an exceptional lens.  In my experience, none of the Sony zoom lens come close to the image quality of the Leica SL 24-90mm. And I wouldn't have the strength to lug around the Hasselblad zoom...

 

Regards

Peter

Edited January 29, 2016 by peterbkk

[ramarren]

+1

 

Without intending any disparagement to the other cameras, all I can say is that with the SL and its zoom, the R lenses, the M-P, and the M lenses, I'm done. I no longer think about image quality and what equipment might be better suited to my photography. I think about what to make photographs of and how to improve my technique. ... This is a good place to be. 

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[Herr Barnack]

@Peter,

Not to question your findings, but it does seem curious that three different cameras with sensors of 24MP, 42MP and 50MP can offer essentially the same image quality.  Add to that the fact that Nikon's newly announced flagship DSLR (the D5) comes with a 20.8MP sensor and it seems that sensors have been improved to the point that 20ish megapixels are all that is needed for the vast majority of applications.

[Ingo]

The 24-90 is the reason why I bought the SL. I've never seen a zoom lens that performed so well.

[tom0511]

I cant believe your findings. Something must have gone wrong when you took the medium format image.

[jmahto]

To notice resolution difference between 50 and 24 megapixel, one needs to have the appropriate test subject (assuming shooting technique is optimized) since there are only 1.5 pixels in 50mega pixels for every pixel in 24 mega pixel sensor. A lens test chart (or a real world picture with very fine lettering) will be needed to show the difference.

 

In a general real world picture the technique will make a bigger difference than extra 0.5 pixels (per pixel). I am only talking about resolution and not about the "look" of bigger sensor.

Edited January 29, 2016 by jmahto

[Peter Walker]

Advertisement (gone after registration)

I cant believe your findings. Something must have gone wrong when you took the medium format image.

 

 

I assume that by "can't believe" you mean that my "findings" are unexpected, not that you are questioning my veracity.

 

Of course, my findings are not from a technical analysis.  For that, you'd need to control a lot more factors.  But, I'll leave that for someone else and get out to take some real photos.

 

Don't misunderstand my point: all 3 are excellent cameras.  All three of the images that I compared are excellent images.  Sharp as a tack.  Lovely exposure.  Great colours.   All three images made it through my very-stringent review process into my permanent library of "keeper" images.

 

My point is that, even at 100% side-by-side viewing, resolution difference between 3 excellent images is not perceivable (in a non-technical comparison).  And that, for my kind of "walk-about" photography, the 24-90mm zoom on the SL can produce images as good as prime lenses on the other two systems.  Plus the colour and handling benefits...

 

Why wasn't the Hasselblad H4D-50 with the 100mm prime lens perceivably better than the Leica SL with the 24-90mm zoom lens?  I don't know for sure, but maybe:

• the newer sensor matches the performance of the older style large sensors?
• the Fujinon lenses on the 'blad are not that great (certainly not as good as the Zeiss lenses on my V series 'blad)?
• the heavier camera is more difficult to hand-hold?
• the OIS in the Sony and the Leica make a difference?
• mirror movement reduces the advantage of larger sensor?

I repeat: the Hasselblad images are great - just not significantly better.

 

Regards

Peter

Edited January 30, 2016 by peterbkk

[FlashGordonPhotography]

I don't have the 'blad but I do have the other two you mention.In my experience the Sony most certainly out resolves the Leica and the 55mm is a tad better than the 24-90. This implies that you have a subject with that much detail in it.

 

For me the thing is that there aren't as many subjects that have that level of detail, to see a significant difference, in real world shooting. Sure for detailed landscape shots where I might use good technique and a tripod to ensure maximum file quality and IF I make a big print, there's a difference. But 90% of what I shoot just don't have enough in the subject for the resolution difference to be useful. If I shoot a wedding the bride's skin doesn't have enough detail to see any real world difference between 24 and 42MP. Even if it did the makeup would fill in the gaps. When i shoot homes I could get away with 16MP. A painted wall looks the same, regardless of resolution.

 

The second thing I see is that it's actually not that easy to get the maximum out of even a 24MP sensor. There is so much going on that can take the edge off maximum resolution. It could be technique. Handholding 42MP, even with Sony's brilliant IBIS isn't that easy. Even on a tripod a breath of wind or something like a bit of environmental haze can make all those extra pixels hard to appreciate. Heck, even the amount of coffee I had today makes a bigger difference than extra resolution most of the time.

 

So I agree with you. In the real world, most of the time, there's little appreciable difference in files over 20+ MP. The main reason the Sony's appeal to me isn't the resolution. It's the dynamic range, which is noticeable. I do dislike Sonys skin tones though.

 

Gordon