“Field” test of Hasselblad H wide-angle lenses.

[jerome_m]

The object of the following tests is to determine the optimum apertures of different MF lenses for landscape pictures.

For landscape pictures, lenses should be tested on subject at large distances. Typical test charts are imaged relatively close and are therefore not optimal for that use. Therefore, I chose to take a photograph of the end of a field (hence the pun on “field test”…). To get some subject up to the angles of the picture, the camera was tilted sideways. Then, for presentation, the image was straightened in PS and a small strip from corner to centre was selected. The attached image should make the process clearer.

I tested the following lenses: HCD 24, HCD 28, HC 35, HC 50-II, HC 80, HC 100. This kind of test is less useful for longer lenses. All lenses were tested on an H4D-50, so 36,7 x 49,1 mm sensor. The individual images were treated by Phocus with default settings and therefore include digital corrections. In practice only lateral chromatic aberration makes a difference here (the lenses have a bit of it, which Phocus nicely removes).

Images and some comments will be posted in individual messages. A pdf full of technical info for all H lenses ca be downloaded here.

[jerome_m]

HCD 4.8 / 24

Introduced in 2012. Currently, the widest lens in H mount, almost too wide to be used alone. However, the lens can be used with the HTS system and is quite useful in that case (the HTS includes a x1.5 converter, so the lens will behave like a 35mm lens then). It is relatively small (smaller than the 28mm), but quite slow (f/4.8). Very rare used. Comes with an adapter for filters (105mm or 112mm, depending on sensor size, one adapter for both by turning it over, check wether you have it if you get that lens second-hand). The published MTF let us expect some degradation at the extreme corner. In the test image, fully open the lens is already excellent. Best aperture: f/8.
The “D” in the name of this lens stands for digital: the lens has a slightly reduced imaging circle and should not be used on film cameras. It will not work on a H1 or H2 camera, but will work on the H4x and H5x however.

[jerome_m]

HCD 4 / 28

Introduced in 2006. The standard ultra-wide for most H shooters and quite easy to find used. Becomes a 42mm on the HTS (and works very well). The published MTF let us expect some degradation at the extreme corner. In the test image, fully open the lens is already excellent. Best aperture: f/8.
The “D” in the name of this lens stands for digital: the lens has a slightly reduced imaging circle and should not be used on film cameras. It will not work on a H1 or H2 camera, but will work on the H4x and H5x however.

[jerome_m]

HC 3.5 / 35

Introduced with the H1 camera in 2002. The widest non “D” lens for H cameras. Easy to find second-hand and relatively cheap. The lens has poor reviews and indeed shows degraded performance in the 2/3 zone on the field test, full open. However, things clear up nicely at f/5.6. Best aperture: f/8.
This test does not show that, but this lens has nice bokeh and is quite nice on real-life subjects at shorter distances.

[jerome_m]

HC 3.5 / 50-II

Introduced in 2010 to replace the first 50mm model from 2002. I did not test the first version of the 50mm. The two versions have the same size and weight, the second can only be recognised by the discreet “II” after its name. Very rare used. The optics were vastly improved between the two versions. That version II can also be used with a special macro converter and then gives extraordinary results for product photography, also when using the HTS.
On the field test, it is simply already almost perfect full open. OK: if you really need to examine the extreme corners with a loupe, maybe use f/11 (look at the crane on the images).
That lens bokeh is a bit busy full open.

[jerome_m]

HC 2.8 / 80

Standard lens, sometimes sold with the camera as a kit. Introduced with the H1 camera in 2002. The cheapest lens in the H line, and easy to find used. Relatively compact, but the only H lens needing a 67mm filter amongst the ones tested here (the 120 macro also uses that size). Standard 6-elements double-gauss design, with the typical behaviour of this optical formula: on the field test corner fuzzyness full open, which clears nicely around f/5.6. Best aperture: f/8-f/11.
Nice bokeh, a bit of longitudinal chromatic aberration at f/2.8 (invisible on that test).

[jerome_m]

HC 2.2 / 100

The fastest lens in H mount (and one of the fastest MF lenses with central shutter), introduced in 2004. Modified 6-elements double-gauss design. More standard 77mm filter size than the 80mm. Sold as a portrait lens, not as a landscape lens and indeed f/2.2 shows the typical veiling of spherical aberration that is sought after for portrait lenses. Field test very similar to the one of the 80mm, best aperture if you want to use that lens for landscape: f/8-f/11.
Nice bokeh, but not that different than the 80mm. A bit of longitudinal chromatic aberration at f/2.2 and 2.8.

[Garry Sarre]

Interesting test Jerome, thanks for taking the time

[jerome_m]

Interesting test Jerome, thanks for taking the time

You are welcome. I was interested in the results myself and this is a test which is informative and relatively easy to set up.

Now, if someone would like to do the same kind of test for other camera systems...

[ErikKaffehr]

Hi Jerome,

Thanks for the images and the good info from Hasselblad.

Best regards
Erik

The object of the following tests is to determine the optimum apertures of different MF lenses for landscape pictures.

For landscape pictures, lenses should be tested on subject at large distances. Typical test charts are imaged relatively close and are therefore not optimal for that use. Therefore, I chose to take a photograph of the end of a field (hence the pun on “field test”…). To get some subject up to the angles of the picture, the camera was tilted sideways. Then, for presentation, the image was straightened in PS and a small strip from corner to centre was selected. The attached image should make the process clearer.

I tested the following lenses: HCD 24, HCD 28, HC 35, HC 50-II, HC 80, HC 100. This kind of test is less useful for longer lenses. All lenses were tested on an H4D-50, so 36,7 x 49,1 mm sensor. The individual images were treated by Phocus with default settings and therefore include digital corrections. In practice only lateral chromatic aberration makes a difference here (the lenses have a bit of it, which Phocus nicely removes).

Images and some comments will be posted in individual messages. A pdf full of technical info for all H lenses ca be downloaded here.

[Jim Kasson]

You are welcome. I was interested in the results myself and this is a test which is informative and relatively easy to set up.

Now, if someone would like to do the same kind of test for other camera systems...

Great work, and an inventive protocol. I think I'll pick up your challenge next time I test lenses.

Jim

[tjv]

Great examples, thanks. The 35mm really is the dog of the bunch, isn't it!

[jerome_m]

The 35mm really is the dog of the bunch, isn't it!

It is actually much better than its reputation. That lens illustrates a perverse effect of lens tests: because of relatively poor results on test charts, internet experts talked it down. In actual photography at medium distances it has a quite pleasing rendering. In landscape pictures, as shown here, one can simply close the aperture a bit for uniform sharpness. So: poor tests, poor reputation, but good pictures in actual use...

[UdoinDesert]

HC 2.8 / 80

....., but the only H lens needing a 67mm filter.

Hi Jerome,

Good job on that comparison test, this is quite a bit of work!
Just for information, the HC120 mk I and II lenses use 67mm filters as well.

Udo

[jerome_m]

Just for information, the HC120 mk I and II lenses use 67mm filters as well.

You are right, I forgot about the HC120. The HC 120 is huge, but its front lens is much smaller and it takes a 67mm filter. I corrected the review of the HC 80 accordingly.

[jerome_m]

I noticed today that the attachements are only visible for people who are logged in. Just in case someone wants to see the pictures, but is not a forum member on luminous landscape: I posted the same pictures to the Hasselblad digital forum (http://www.hasselbladdigitalforum.com). There they can be seen without registering.

[ondebanks]

Excellent work, Jerome. Although I am not and never will be a Hasselblad H user [can't live without a focal plane shutter!], I really commend your effort in taking and presenting these tests.

For me, as a disinterested observer, the lenses seem generally very impressive at the wider stops. The mark II 50mm does seem exceptionally good. But I expected better from the 100/2.2...it should be peaking at f/4, but doesn't really until f/8. 

I recall someone once writing [IIRC it was in the context of a shoot-off vs some RZ67 lenses] that the H lenses were deliberately not optimised for infinity/landscape use, but rather for near-field/portrait use. If that is indeed the case, it struck me as non-optimal (for my type of photography); but I wonder if it also means that the H lenses are capable of even better than your infinity tests suggest?

Ray

[ErikKaffehr]

Ray,

My reading is that the H-lenses are optimised for a wider range of distances. I would guess that some lenses have floating lens groups (variable air space may be a better word) to handle different focusing distances.

What I have seen from Hasselblads own MTF data the H-lenses generally perform better than the V-lenses.

Best regards
Erik

Excellent work, Jerome. Although I am not and never will be a Hasselblad H user [can't live without a focal plane shutter!], I really commend your effort in taking and presenting these tests.

For me, as a disinterested observer, the lenses seem generally very impressive at the wider stops. The mark II 50mm does seem exceptionally good. But I expected better from the 100/2.2...it should be peaking at f/4, but doesn't really until f/8. 

I recall someone once writing [IIRC it was in the context of a shoot-off vs some RZ67 lenses] that the H lenses were deliberately not optimised for infinity/landscape use, but rather for near-field/portrait use. If that is indeed the case, it struck me as non-optimal (for my type of photography); but I wonder if it also means that the H lenses are capable of even better than your infinity tests suggest?

Ray

[jerome_m]

There used to be an excellent article (PDF?) by Hasselblad available on the web that compared the H and V lenses including their respective optimisation.

You probably mean this article (pdf).