The Capturing Capital Of Culture

Paul McMullin gets to work in The Capital of Culture using 5x4, Nikon D3, Nikon D2x, and Digital Hasselblads along the way. McNamee acts as bag carrier!

This is a reduced scale, partial crop of an image from the Hasselblad H3D39.

Plunge in at the deep end – high end capture examined by Mike McNamee and Paul McMullin

High-end camera gear is something we look at infrequently in Professional Imagemaker. For many of our readership, the investment is out of reach, but it is something you might aspire to, should the lottery numbers come up! We thought it relevant to examine some of the older Hasselblad models, as the arrival of the latest H3D Mk II has caused a number of these older models to come available, as pre-owned gear.

There are three things to consider in regard to high-end digital. The first is how much extra quality do you get, the second is the value of that additional quality and finally how are you going to get your money back? A straight 'return on investment' calculation would be hard pressed to make a cogent case for the outlay around £25,000, bearing in mind that you need back-up gear as well, if you are a wedding shooter – a Hasselblad will fall from your hands as easily as a 35mm D-SLR (perish the thought!).

Faced with a choice between a 35mm D-SLR and a Hasselblad system we need to compute the overall package cost. To keep a common baseline, we have assumed that you need a body, a wide angle, a standard and a short telephoto lens to complete your kit. This is slightly unlikely in the ‘35mm’ kit – you would more likely go for a zoom, but we will let that pass. The figures are computed in the table below.

It is not a good way to go about valuing a camera but we have calculated the price per megapixel to provide a benchmark. In general terms the 35mmm D-SLRs are about half as much per pixel as the latest Hasselblads. If we compare the identical 22 megapixels of the Canon 1Ds MKIII and the Hasselblad H3D 22 then the leap in cost per pixel is more than two-fold.

Considering area though, the Hasselblad is providing 1.69 times the area of silicon to the scene compared to the D3 or 1Ds MkIII, but the price premium is near two-times. Or to put it yet another way, does the increase in cost of the Hasselblads, over the Nikon D3, of either 3.91 times or 5.60 times, justify the increase in image quality that you get?

Another issue is the relative ISO speeds. The Hasselblads are limited to 800 ISO, a less effective use of the silicon area available, which Nikon exploit all the way to 25,000 ISO on the D3. The compromise is ‘doubled up’ as well – the maximum apertures of the Hasselblad lenses are smaller than those available for 35mm D-SLRs.

No matter how you compute the RoI or how you consider usability, it is unlikely that you could convince a hard-nosed accountant that a Hasselblad investment was the best financial deal. However, this ignores one massive advantage that the Hasselblad has – status. If your wedding business revolves around making images of people who have arrived at the gig in Ferraris, Mercedes and Porsches you are still unlikely to be upstaged by the bride’s uncle Fred with his topend gear if you are wearing a Hasselblad. If that marks you out as ‘high end’, ‘top drawer’ or whatever, and subsequently gets you another wedding commission from your wealthy clientele, then the investment suddenly looks a lot better.

The fact that you cannot tell the difference between the technologies on the evidence of a ten-eight in the album, is as irrelevant as the fact that my £14 Asda watch keeps the time as accurately as a platinum-cased Rolex. The case for high-end commercial shooting is easier to make – if you are chasing quality for high-end clients they expect to pay for it. At 7200 pixels wide, the H3D can make a 24-inch spread at process colour resolution, let alone a double-page A4 spread for a glossy magazine. Most of this testing was carried out while Paul McMullin was shooting for The Architectural Review – they chose a Hasselblad shot for their front cover.

Technical Comparison

Our first comparison is from the top of the Liverpool Cathedral. It was not possible to arrange all the shots to cover an identical field area, we shot on 80mm with the Hasselblads and 105mm with the Nikon D3. The relevant data are as follows:

From each shot we took an 800 x 600 pixel clip for comparison and combined them into a single file. The curved roofs of corrugated cladding are a searching test for colour moiré fringing, which occurs at significant levels in both the Hasselblads but not in the D3. Overall the CVF performance was not up to the standard of the Nikon D3 in this particular shot although we found the result closer in others.

We compared the H3D and the D3 more closely. In the annotated composite the following remarks are pertinent:

A: The detail of the corrugated roof is obscured by the moiré fringing of the H3D.

B: There is spurious resolution of the railings, inferring that the gate has slanted rather than vertical rails.

C: The date ‘1855’ has more detail in the H3D shot.

D: There is significant colour fringing in the trees of the H3D image, similar noise is also present in the surface of the river.

Overall from this test it is obvious that pixel for pixel, the D3 is handling moiré better than the H3D. However, that is only part of the story. The H3D covers almost twice the field area of the D3 in this shot and given that the resolving power in the shots is similar, the H3D comes out well on top (see opposite). This is further illustrated by the detail from the signs on the hoarding around the construction site. The word KIRK is visibly poorer from the D3and the words ‘Banfield Construction’ can be read in the H3D image but not in that from the D3.

In these tests then the evidence says that there is a detectable improvement in the resolving power from the Hasselblad images even though the colour fringing could be a source of problems for say a fashion shoot involving checked or tartan materials.

The Software The Hasselblads come with FlexColor (we used version 4.7.1). This presented us with a proxy, lower-resolution image to review in under two seconds. The loupe facility allows the user to review four segments of the image at full resolution while adjusting critical parameters such as sharpness and noise control. Colour correction is made on the low resolution proxy image. The colour correction facilities are extensive and easy to use. Once the correction had been applied, we saved out a 112Mb TIFF file in 1m 38s to create a 7210x5400 pixel RGB, 8-bit file.

Join today and receive a fantastic free gift:

1. Monopod for Free - Valued at up to £44.99
2. Lastolite Tri-Grip for Free - Valued at up to £59.00
3. Lowepro Compu Daypack for FreeValued at up to £89