Archive for the ‘Reviews etc’ Category

Not Quite Déjà-vu

Thursday, July 19th, 2018

This morning I took a look at the front page of Café Royal Books, a small independent publishing house based in Southport, England originally set up in 2005 by Craig Atkinson as a “way to disseminate drawings and photographs, in multiple, affordably, quickly, and internationally without relying on ‘the gallery’“.

Since 2012, Café Royal Books has published at least weekly an ongoing series of publications presenting mainly ‘British Documentary Photography since 1960’. As he says on the site:

“This type of work has historically been neglected, in the UK and overseas by major institutions. It is often neglected by the photographer too, possibly because there has been no outlet, as such, for it.”

The publications usually present a series of images by a single photographer on a single project. It may be the work from a single event or representing a much longer project.  CRB has produced some larger works, but these weekly publications are generally between 24 and 40 pages, more a zine than a book, with the aim of building up a comprehensive survey of the area of work. Some photographers are represented by quite a few such volumes, in some cases more than 20, while others have preferred to stop at a single issue.

Atkinson keeps down costs, wanting to keep the issues affordable – currently £6 each for most.  You can get every title (except the special editions etc) with a 60 issue subscription – roughly the annual output – and there are also limited editions in a boxed set of 100 books every 100th title aimed “at public collections, so the books remain accessible.”

Among the photographers who have already had issues published are some very well-known names – including Martin Parr, Jo Spence, Daniel Meadows, Brian Griffin, David Hurn, Victor Sloan, Chris Killp, Paul Trevor and others, but some of the best books are by people you may well never have heard of.

The three most recent titles are Diane Bush — The Brits, England in the 1970s,
Ian MacDonald — Greatham Creek 1969–1974 and Janine Wiedel — Chainmaking: The Black Country West Midlands 1977, each worth a look, and you can page through them on the web site. Another recent title is John Benton-Harris — The English, where I have to declare an interest, as I helped John translate his ideas into digital form. It’s a great introduction to the work of this photographer who came to London to photograph Churchill’s funeral and stayed here as one of our most perceptive observers – and was also largely responsible for the seminal 1985 Barbican show ‘American Images 1945-80‘, providing most of the ideas and contacts and doing much of the legwork for which others were rather better at taking most of the credit.

But the déjà-vu? It came on the back cover of a book by another US visitor to this country, Diane Bush, who was here from 1969 for ten years, becoming a part of the Exit Photography Group with Paul Trevor and Nicholas Battye which produced ‘Down Wapping‘. On the back cover of her ‘The Brits, England in the 1970s’ was a picture of a car parked in front of a fence, using the reflections of that fence. It isn’t the same car nor I think the same fence, nor quite the same treatment, but I immediately thought of my picture when I saw hers.

Parked car, Vauxhall, Lambeth, 1978 – Peter Marshall

I don’t think there is much possibility that I had seen her picture when I took mine, but have a nagging suspicion that somewhere, by some photographer, is a similar image that we both had seen before making our pictures.

There are no adverts on this site and it receives no sponsorship, and I like to keep it that way. But it does take a considerable amount of my time and thought, and if you enjoy reading it, a small donation – perhaps the cost of a beer – would be appreciated.

My London Diary : London Photos : Hull : River Lea/Lee Valley : London’s Industrial Heritage

All photographs on this and my other sites, unless otherwise stated, are taken by and copyright of Peter Marshall, and are available for reproduction or can be bought as prints.

To order prints or reproduce images


Chiswick House and Gardens

Saturday, June 30th, 2018

I’m not sure that the London Borough of Hounslow’s tourist industry needs encouragement, though despite having one of the world’s busiest airports (and if our current government gets its way having managed to pass a vote on its expansion, one that will certainly be, despite all promises, busier, noisier and more polluting if an extra runway actually gets build) on its doorstep it doesn’t seem to be a major tourist attraction. Most are only too keen to get away from the LB Hounslow as fast as they can, by tube, taxi or expensive train.

But there certainly are parts worth a visit. Some houses – Chiswick House, Syon and Osterley and areas of housing; a splendid piece of riverside, a few interesting small museums including Hogarth’s House. And quite a few parks with points of interest, the finest of which is perhaps the grounds of Chiswick House. And these gardens are free to visit, open all the year round.

I grew up a few miles away and visited the gardens here a few time when young, brought by my father, a keen gardener, who always carried a pair of scissors in his pocket to take the occasional cutting when no-one was looking, and when near home, some seeds of the decorative giant thistle echinops to scatter in any likely looking bare patch of ground because they were so good for the bees he kept.

Then I think the gardens were overgrown, and the giant conservatory dilapidated, but both have now been much restores. The huge greenhouse – which people always said was built by Joseph Paxton of Chatworth and Crystal Palace fame (he was a gardener and not a footballer) but wasn’t – was really built by mistake. It was built in 1812-13 to the designs of Samuel Ware and was then the longest ever, at 2 feet over a 100 yards. The 6th Duke of Devonshire was then owner of the house and ordered it to put his camellia collection in. These plants, which grow wild in much of south-east Asia, became a great craze at the end of the 18th and early 19th centuries, selling for high prices and making fortunes for some nurseries.  Some have survived from back then and are now very rare.

I suspect my father took a few cuttings as usual, and there may still today be some from here surviving in some of the gardens in Hounslow he looked after, though the better of these are now long covered with flats.

It was probably the high prices that made people want to molly-coddle them, or perhaps just ignorance. They grow in their native soils in greater extremes of weather than we see in England, and just a few days earlier I had seen one in a front garden not far from our home with its flowers covered by snow. Though conservatories like this one would have been pleasant places to sit and walk and view your collection of plants, so perhaps the roof and heating were more for the benefit of people rather than plants.

And if you are not a gardener and wouldn’t know a camellia from a chrysanthemum, you will almost certainly have some familiarity with the drink made from the dried leaves of one variety of camellia. We call it ‘tea’.

The rest of the gardens were really the start of a new movement in garden design, the first “natural” garden in what became known as the ‘English Landscape Movement’, designed by William Kent for Lord Burlington, who also had the house built to his own designed, aided by Kent and copying the work from various Italian buildings, particularly those of Palladio.  The house was built in 1726-9 but work went on with gardens for some years. Although the gardens may be natural they contain many unnatural features of classical origin, the house and gardens representing Burlington’s attempt to create a Roman villa situated in a symbolic Roman garden.

Photographers will of course recognise them as they featured in several fine pictures by Bill Brandt, and of course many other photographers including Edwin Smith.  I’ve made a few pictures there in the past, but make no claims for these hurried snaps taken on a dull and wet day on an outing with others who were keen to get back to the car.

Both house and gardens are worth a visit, and the gardens and conservatory have been well restored (doubtless thanks to idiots buying lottery tickets.) There is a charge of entry to the house which is run by English Heritage, but the gardens are free. There is also a newly opened restaurant in which we had lunch, but I’d recommend a riverside walk from here to one of the Hammersmith pubs for a proper meal. You could also visit (on a Thursday or Saturday aftenoon) the small William Morris Society museum on Hammersmith Mall, and if sufficiently organised to book in advance, take the fascinating guided tour of Emery Walker’s house in Hammersmith Terrace.

Chiswick House Gardens


There are no adverts on this site and it receives no sponsorship, and I like to keep it that way. But it does take a considerable amount of my time and thought, and if you enjoy reading it, a small donation – perhaps the cost of a beer – would be appreciated.

My London Diary : London Photos : Hull : River Lea/Lee Valley : London’s Industrial Heritage

All photographs on this and my other sites, unless otherwise stated, are taken by and copyright of Peter Marshall, and are available for reproduction or can be bought as prints.

To order prints or reproduce images


Lange & Winship at the Barbican

Monday, June 18th, 2018

Opening shortly at the Barbican is ‘Dorothea Lange / Vanessa Winship – A photography double bill‘, with Dorothea Lange: Politics of Seeing showing together with Vanessa Winship: And Time Folds in the Art Gallery there from 22 June —2 September 2018, presenting the work of two photographers I greatly admire.

I’ve several times printed a copy of Lange’s best-known picture, ‘Migrant Mother‘ from the high-quality large Tiff file that I years ago downloaded from the Library of Congress, and have written on several occasions about this and other works such as her ‘White Angel Breadline‘ from 1933 which prompted her career as a documentary photographer.

The show apparently has a large section on this work, and you can read more about it and see the some variants on a page at the Library of Congress, where you can see all her work for the FSA (a search using the term ‘Lange, Dorothea’ yields over 4000 items, though not all are photographs), and find more about various shows of her work. On the Library of Congress they reproduce Lange’s own story about how she made the picture, written for Popular Photography in 1960:

“I saw and approached the hungry and desperate mother, as if drawn by a magnet. I do not remember how I explained my presence or my camera to her, but I do remember she asked me no questions. I made five exposures, working closer and closer from the same direction. I did not ask her name or her history. She told me her age, that she was thirty-two. She said that they had been living on frozen vegetables from the surrounding fields, and birds that the children killed. She had just sold the tires from her car to buy food. There she sat in that lean- to tent with her children huddled around her, and seemed to know that my pictures might help her, and so she helped me. There was a sort of equality about it.”

But apparently Florence (Owens) Thompson, the woman in the picture saw it differently, according to her grandson’s recollection (I think recorded in Anne Whiston Spirn’s book on Lange Daring to Look, and mentioned in my 2008 post on that)

 “a well-dressed woman jumped out of a smart newish car and started taking pictures, getting closer with each shot. Florence decide to ignore her.

After taking the pictures, Lange is said to have told Florence who she was and that she was working for the Farm Security Administration and to have promised that the pictures would not be published. Next day they made the front page of all the newspapers.”

Lange gave a long interview to Richard Doud in 1964, a year before her death. You can hear 12 seconds of her voice and read the lengthy transcript  in the Smithsonian Oral History Collection.

Some brief biographical details I wrote almost 20 years ago about Lange may be of interest:

Dorothea Lange was born in New Jersey. She gave up training to be a teacher to become a photographer, working part-time in the portrait studio of Arnold Genthe before studying with Clarence White.

She moved to California, meeting Imogen Cunningham and opening her own portrait studio. In the early 1930s she began to take pictures of people suffering from the effects of the Depression, such as the ‘White Angel Breadline‘ in San Francisco in 1933.

The following year she met sociologist Paul Taylor who she was to marry (after divorcing her first husband, painter Maynard Dixon) and began to work for various Government projects, most notably the Farm Security Adminstration.

Her career was interupted by illness for almost ten years from 1945, following this she travelled extensively around the world with her husband before settling down to photograph things ‘close at hand‘ around her home and family.

One single picture she took for the FSA stands as an icon of the depression. ‘Migrant Mother‘ shows a mother looking worried into the distance, as if wondering what future there is for her. One child lies sleeping on her lap, two older children frame her, turned away from the photographer with their heads bowed. Lange recorded that the mother was aged 32 with 7 children; they were migrant pea-pickers but the harvest had been ruined by frost so there was no work. They had already sold the tyres from their car for food and were now living in it, keeping alive on wild birds the children caught.

Surprisingly the article in yesterdays Observer, Dorothea Lange: Politics of Seeing review – a visionary whose camera never lied by Laura Cumming fails to even mention that the show is on together with Winship’s, though possibly this was made clear elsewhere in the print edition. I’ve long been a fan of Vanessa Winship, and have several times mentioned her work here (I think this is the 15th.)

The best of these posts is I think  Sweet Nothings – Vanessa Winship written in 2009 which included a couple of her portraits from Turkey.  In a more recent post, I quoted from Sean O’Hagan’s blog in The Guardian:

“From Mississippi to the Black Sea, Winship’s poetic, masterful photographs show how hard it is for people to belong … so why don’t British galleries acknowledge her as this large Madrid retrospective does? She deserves it”

At the time I commented: “Though I’m afraid the explanation is unfortunately rather simple. She is a real photographer, and there is no major British gallery with a real interest in photography.” It is great to see her work acknowledged at last in the Barbican show.


Reviews and Real Life

Thursday, May 24th, 2018

Before I buy a new camera, I always spend a long time reading through the various reviews available on the web, both the highly detailed ones such as those by Digital Photography Review and those by photographers who have spent some weeks actually working with the cameras and give a more personal evaluation of the strengths and weaknesses.

Of course they have there uses, and are certainly a lot more reliable than the opinions posted as reviews by some popular Internet figures or those user comments collected on some sites. Most internet reviews are not worth the bits they are written with – which come more or less free.

While it’s relatively easy to get the technical details right – it usually just means transferring them from the manufacturer’s data sheet – what is usually more important in practice is how the camera and user interact. Harder to quantify or even describe, and something that in the end is very individual.

I currently use two Nikon bodies, a D810 and a D750, and the differences in specifications are not great. The D810 is heavier and has better weather protection, and the shutter sounds smoother – and of course it costs rather more. The placement of buttons on the backs are different and the D750 lacks a separate ‘OK’ button, using instead the centre of the four-way controller. It’s a minor nuisance.

The D810 produces larger files, but normally those from the D750 are perfectly adequate and the D810 merely serve to fill up your backups faster. I do prefer the larger files for my panoramic work, where the D810 also scores with tilt indicators for up and down as well as for left and right. But apart from the relatively small increase in image size, the image quality is more or less the same.

The two most significant differences for me are actually things I’ve never seen pointed out in reviews. On the D810 it is possible to lock the aperture (and I think shutter speed) when using the camera on a manual setting. This means you can decide you want to work at – for example – 1/500th f8 and you can do so, allowing the camera to alter the ISO to give correct exposure – if possible on its auto-ISO setting. Unless you choose an aperture and shutter combination which is outside the range you have set for auto-ISO but with a little photographic nous this is unlikely.

You can use the same technique with the D750, but there appears to be no way to lock the aperture or shutter speed in manual mode. Whichever you leave on the primary command dial (in my case aperture) gets readily knocked around as you (or at least I) use the camera. The command dial doesn’t have sufficient resistance to movement for a clumsy fumble-finger like me, and I can suddenly find that I’ve taken a few frames at f45 rather than the f8 I’d intended, which can easily result in considerable underexposure even if I’ve set a maximum ISO of 12,800.

I’ve tried various ways to counter this. The most successful seems to be to tape over the command dial, but then sometimes you do want to change aperture, and have to peel back the tape which then often falls off and gets lost. The black masking tape I used also gets uncomfortably sticky.

My latest attempt to ameliorate the problem has been to reverse the direction of travel of the aperture control, which hopefully will mean that I get f4 instead of f45 when things go pear-shaped. Since I normally work with a wide angle at f5.6 or f8 this will only mean a two stop difference so perhaps less of a problem, and the depth of field with a wide angle will usually still be sufficeint. I’ll also reduce the maximum ISO when using the technique, perhaps to ISO3200 so the risk of overexposure will be less. It’s only very special conditions that need a higher value.

The second annoying difference is when using an FX lens in 1.2x or DX mode, which is a cheap way of effectively getting a longer telephoto. On the D810 there is a well-hidden way to get the viewfinder to display a greyed out area around the actual image, so that details outside the frame are visible but it is very clear they will not be recorded.

Quite why this should be obtained by setting Custom Setting a6 AF Point Illumination to OFF is a mystery only known to an inner circle of Nikon Illuminati, who may also be able to tell you (though they may have to kill you afterwards) why the equivalent CS a5 does not give the same effect on the D750.

Of course you do get a viewfinder frame to show the reduced image area. But as I can tell you from too frequent experience, it isn’t always easily noticed. It wouldn’t matter if it were not for the D750 to occasionally decide to spontaneously change the image area. It is possible to map the change to one of the function buttons or pressing the command dials, which could easily lead to such a mysterious event, but I’ve certainly never done so.

Modern cameras are simply too complicated, giving too many options. Both these cameras have instruction manuals with over 500 pages and there are just too many interdependencies, with one setting affecting another. Back in the old days you could just pick up almost any camera and just use it. Perhaps it’s time for a campaign for real cameras that real people can use. Or are they called phones?

American History of American Photography

Wednesday, January 31st, 2018

Only Part 1 of Marc Falzon‘s short survey of the history of American photography is currently available (and featured in PDN Pulse), but it’s 8 minutes offers an interesting introduction to the subject, dealing with it through 3 key curatorial figures, Alfred Stieglitz, Minor White and John Szarkowski (part one only deals with Stieglitz) and a few key photographers.

Among those photographers, so far only Timothy O’Sullivan and Walker Evans have been the only to have been looked at in any depth, with a passing mention of a few others.

It’s an introduction that tries to demonstrate what is peculiarly American about American photography and makes some interesting points, while of course minimising or neglecting much European work of the era.

Where it descends to the ridiculous is in suggesting that photographers such as Atget and Henri Cartier-Bresson were somehow the product of American advances in the medium and the influence of Stieglitz. What influence there was clearly flowed the other way.

It wasn’t the school around Stieglitz that motivated Walker Evans – except in a reaction which dismissed it as well as the newer modernism of photographers such as Paul Strand, but the vernacular photography of the many unknown American photographers – and the work of Atget which was brought to New York by Berenice Abbott and Julien Levy.

What I find most annoying about this video is however the way that the photographs are shown, generally starting zoomed into a detail and wandering around the image, with only at best a fleeting glimpse of the image as a whole. Framing is so intrinsic to our medium and we need to see and study pictures in their entirety – perhaps occasionally zooming in to view significant details.

Peak Design Fault?

Friday, January 19th, 2018

Last year I needed a new camera strap. Well no, really I didn’t but one of my straps had broken and I found myself walking around holding one of my two cameras in my hand all afternoon. I couldn’t really complain about the strap breaking as it was probably over 20 years old having seen out the life of at least half a dozen cameras. Each of which had come in a box with a strap inside, most of which are probably somewhere in my loft.

But I seldom use the straps that come ‘free’ with the camera, and certainly not for the cameras I work with. Most of them must have added at least 50p to the overall cost of the package, and tend to feel like that. My favourite neck strap for a heavy DSLR (though I started using them with SLRs) is still the Optech Pro Strap with its nice wide neoprene cushioning which means you hardly feel the weight – and once I started using this I no longer had an aching neck at the end of the day’s work.

But good though it is, you can’t really use two of them at the same time – though they do sell special rigs for the purpose, a double sling and a dual harness – I really didn’t like the look of these. Functional they doubtless are, but there was something about having to be strapped into a rig that didn’t appeal to me.

So I looked around for alternatives and read the reviews. Some straps were clearly for the fashion conscious, to go with expensive leather cases (and I never before realised how expensive these could be) and it was easy to dismiss these. Others had the kind of narrow belting that slowly beheads you from behind as you work, but there were a few that seemed worth a try, and in particular the Peak Design Slide. This would keep my second camera at my side, well out of the way of the one on my Optech strap carried high on my chest, but enable me to quickly bring it up to my eye. So I ordered one, and have been using it for quite a while now, and am generally fairly pleased with it.

The strap attaches a little differently to a normal neck strap with special short quick-release anchors which are left in place on the camera. The web site and packaging shows these fitting directly onto the lug on the camera body, and while this was never going to work with the anchors that came with my slide, the latest design has thinner and stronger cords which can be pulled through the hole with a strong thread, It is a neater solution than using the normal split rings supplied with the camera, and avoids any possible damage to the camera body. You can also buy extra anchors making it a ten-second job to switch the strap from one camera to another, and it is equally fast to change from the Slide to a wrist strap from the Peak Design range.

There are two ‘quick adjustment’ sliders on the slide, to allow rapid adjustment of the length of the strap, and it is about these – or at least one of these – that I find Peak Design’s design at fault. They show a diagram with the rear adjust being used for “long term adjustment based on body type and preference” while the front adjuster is designed for “quick adjust”. But the two adjusters are identical and neither has a lock.

The consequence of this is that whenever I use the strap for any length of time, both adjusters end up at the bottom and the camera gets to hang rather closer to my knees than my waist, which is something of a nuisance. Yes, it is ease and quick to adjust, but when I’m busy working the last thing I should need to do is to fiddle with my camera strap. I want to firmly set the maximum length of the strap around 6 inches shorter but Peak provide no way to do this.

They rightly say that the strap resists a sharp pull without changing length, but somehow – perhaps by being knocked by my camera bag or just by the normal movements of my legs and body and the sliding up to my eye to take pictures which cause them to slowly creep, the adjusters inevitably work their way down.

My part solution to this problem involves using a safety pin which prevents movement of the rear adjuster, but a fudge like that shouldn’t be necessary with a high price product like this, and safety pins do sometimes come undone. When I’m sure I know exactly the length I want I’ll perhaps put a few stitches in.

It wouldn’t be difficult to design a rear adjuster than locked more firmly in place, and keeping the current front adjust would retain the quick adjustment that is a part of the design philosophy. A quick Google shows me I’m not the only one with this problem, and people have written to the support page asking how they can prevent it happening. They get told it is a part of the design, but like me I’m sure the questioners regard it as a design fault.

So while I’m generally happy with this, and have just bought another Peak Design product, a wrist strap (Cuff) for my newly acquired Fuji X-E3 (a detailed review here), it does seem to me that the company should perhaps stop spitting in the face of customers and rethink on that rear adjuster.

More about the X-T3 later. I’ve been quite excited about it, as it seems to have fixed most of the niggles I’ve had with previous Fuji-X cameras. It seems far more responsive than the X-T1, focusing faster with all my lenses and not yet leaving me waiting when I want to take a picture. I’ve just sent the X-T1 in for repair – again – so I can’t directly compare the two, and I’m still working my way through the manual – some settings are rather hard to find and understand – but the only thing I’ve found less impressive so far is the slightly smaller viewfinder. I suppose the big difference is the fixed back, but if you can live without that then I think the X-E3 is the best in the current Fuji range. Of course an upgrade to the X-T2 is rumoured for later in the year, so things may change.

So far as straps are concerned, if you only carry one camera and are still using the neck strap that came in the box, I’d highly recommend the relatively small cost of an Optech Pro, which will almost certainly get rid of those neck pains and headaches at the end of a long day at work. But I’ve grown to like using a sling strap with the camera resting on my right hip as an even more functional and relaxing way to carry a camera. If you are above average height (or girth?) the Peak Design Slide may be the right length for you, but until they fix the creep problem for the rear adjuster I’d hesitate to recommend it for those of average and below size. There are many other sling straps available at a wide range of cost – from around £2 to £150 – including one that looks very similar to the Peak at around a tenth of the cost though clearly without the same quality of finish and probably of materials. I’d hesitate before entrusting a camera and lens costing £2000 or more to some of them.

Something completely different?

Tuesday, April 18th, 2017

Most new cameras that are announced are pretty much the same as our old ones, though the marketing guys would like us all to get all excited about a few more mega-pixels or slightly faster autofocus and other minor improvements, many of which we will never use. And often we fall for it and spend our cash on the latest model, hoping it will in some way revolutionize our picture-making.

Of course all these improvements do add up over time. The Nikon D810 I now use is significantly better in so many ways than my first digital SLR, the D100 in 2002, with its small, dim, viewfinder and its 6Mp sensor. But by the time they brought out the D200 or D300, Nikon had more or less sorted out the major issues, though they tempted me with full frame in 2008 and I upgraded to the D700 the following year. Nikon had actually said for some years that we didn’t need to have full-frame for digital cameras, and essentially they were correct for pretty much all I do – the main advantage is a little more light into the viewfinder, though often I work in DX format as I like being able to view outside the frame – an advantage the Nikon marketing department keep quiet about.

Incidentally, this is a feature that works better on the D810 (and 800) than on the D750. With the D750 you see the viewfinder frame, but it is easier to ignore, especially in the heat of the moment when what you see gets exciting, and I’ve several times framed and taken what I think would have been a great image, only to find later that I’d actually lost half of it and only taken the central part. With the D810 there is some esoteric combination of custom settings (don’t ask me – search in your manual) which allows you to grey out the non-image area of the viewfinder. You can still see the area outside the picture, but it’s very obvious it isn’t a part of the picture.

But there haven’t really been any great advances. I’d still be using the D700 if it hadn’t gone legs up after a little over 520,000 exposures (not bad as it was only rated for 150,000.) It does have a few other minor faults, but it’s still sitting on my desk waiting for me to bother to get an estimate for repair, though I’m convinced it will be beyond economic – along with my 16-35mm f4 lens.

Over the years I’ve read about several new cameras that claimed to be truly revolutionary, and the latest is the Light L16. When I first read about it around a year ago I was frankly a disbeliever, and certainly not convinced enough to risk the thousand dollars or so to back the campaign and get an early camera. The video on its web site explains the concept, with 16 lenses with focal length equivalents of 28 to 150mm, using 10 of them each time you take a picture and digitally combining their images. They promise this will give ‘DLSR quality’ equivalent to using 28 to 150mm lenses from a package the size of a rather fat smartphone.

The camera will produce 52Mp images and will enable you to choose focus and depth of field after taking the picture. I remain more than a little sceptical, but cameras should soon be going out to those who pre-ordered and are expected be on sale to the rest of us towards the end of the year. If it lives up to their expectations, they may be correct in their claim it will be the most significant advance in camera technology since the first Leica. Just a shame they don’t have a wide-angle version!

Fuji X-Trans blues

Saturday, March 11th, 2017

I’ve been testing Fuji-X cameras for some years now, hoping that I would find a lighter alternative to my Nikons, but never quite feeling I could make the break.

It started with the X100. A lovely camera in many ways, but one that had me tearing out my hair when so many times I pressed the shutter release – and nothing happened. It was (and is) a camera with a habit of going into a deep sleep, and the fastest way to revive it is to switch off and start again. By which time the picture had often vanished. And then there was the battery life…

Had it been small enough to fit into a pocket I might have continued to use it, but it was too big to be a compact camera, and not really flexible enough for many uses. The X-Pro1 was bigger and heavier, but I quite liked it, at least with a limited range of fixed lenses. The optical viewfinder was fine, but the electronic one not great. And I still felt I was carrying a lot of camera and lenses for a fairly limited system. I still get it out and use it occasionally, but not often, and not for long.

The X-E2 was a little smaller and lighter – and let me use the zoom lenses with an almost OK electronic viewfinder. But autofocus was often too slow. But its a camera I still use occasionally when I want to carry two cameras.

The best of those I own is the XT1. A better viewfinder and faster focus – though it is a bit larger than the X-E2. But it has too many quirks. Last time I used it, although I’d set the camera to save RAW files, a number of them came out as jpegs, and with some lenses it adds around 35 magenta to the images, while other lenses give correct colour. And like the others, it eats up batteries.

By now I’ve a considerable investment in Fuji-X lenses, but largely use the Fuji system as a camera for when I’m not working. When things get serious, the Nikons come out again. The weight difference is still there, though the Nikon always seems to last a full day on  a single battery, whereas I carry a bag with 4 spares for the Fuji, and sometimes need them all.

I’m wondering now whether the XT20 will be the solution to most of the problems I have with Fuji, but perhaps it’s a false hope and I should cut my losses and sell off all the Fuji gear I have.

Fuj-X has a strong following on the web, often showing what seems to me missionary zeal. I’ve never felt that the Fuji Image quality was inherently better than that I get from Nikon, and if I do get equivalent quality, then its down to lenses. When quality and size matters the Nikon D750 and D810 are still ahead with their full-frame sensors.

Fuji makes great play of the advantages of their X-Trans sensors, but while this may have some advantages, it also has its problems. The study in PetaPixel, X-Trans vs Bayer Sensors: Fantastic Claims and How to Test Them debunks those claims and makes clear that the publicity about the X-Trans sensor is just that, publicity, and that the different filter pattern has no real advantages – and indeed slightly lets the cameras down. Though of course the the differences are small and insignificant for all normal purposes.

Fuji’s latest camera announcements are, as usual, tempting. Reviewers and Fuji evangelists are generally waxing lyrical over the X-T20 and also the X100F as well as the rather too large and considerably too expensive (for me) GFX. My son, still a loyal user of the original X100, is thinking about updating, but for the moment I’m keeping my hand in my pocket.

I still remember with some bitterness the Leica M8, given glowing write-ups by some of the world’s greatest photographers as well as reviews that made it seem a near perfect machine, the digital camera that all of us who had once been bitten by the Leica bug would die for. Enough to persuade me to invest the ridiculous amount that it cost, only to find it was virtually unusable, at least for colour – and needed a special filter on every lens. Of course, Leicas have improved since then, and with every new model people have said that they’ve got it right now. And the M10 does look tempting…

Back in the real world (a decent Leica outfit with a few new lenses would cost what I live on for a year) those new Fujis do look tempting, but I’ll wait and see how others get on with them before making a decision. Or perhaps get out that old Leica M2 and go back to using film. It’s still a fine machine, working smoothly despite being 60 years old last time I tried it out. Later models never quite matched the feel of using it, though they did add a few useful things. And photographers made and sold the only thing it really lacked, a rewind lever.

Upsizing & AI

Monday, February 13th, 2017

Back in January 2016, when the Chinese bought up Corbis and Demotix where I was placing many of my pictures and handed their entrails to Getty, I began to put many of the pictures I was taking into Facebook albums as well as sending them to agencies.

I hadn’t been happy with Demotix for some years – particularly since they sold out to Corbis, but they had the advantage for me of presenting my work as coherent stories with text and captions that were easy to link to – so I could post links to these stories for my friends on Facebook and also others, particularly those who had taken part in the events I had photographed.

I’ve lost count of how many albums I put on Facebook in 2016, though at a rough guess they would contain around 5000 pictures. So when I read an advert on Facebook offering to make an album of my year I was intrigued to see what they would make of them, and clicked the links, though I had at the time no intention of buying the book from ‘Re-Snap‘.

It took several minutes before the book appeared as a 100 page A4 hardback and I was able to page through it’s roughly 100 pages, each with 5 images in a variety of layouts, mostly more or less uncropped. The selection they made wasn’t entirely random – they state:

“The images are selected by analyzing all the images step by step. For instance, we filter out photos by looking at several quality aspects (like blurriness). Our system also automatically looks at the amount of faces and the micro expressions of the faces. Of course we filter out similar photos by looking at the pixels. After this part our deep learning network will look at correlations the pictures in your uploaded photo set. In this way we analyse what kind of picture you would like to have in your photo book.”

All of my images on-line at Facebook are uploaded at 600×400 pixels and with my watermark – exactly the same images as I post on My London Diary (though sometimes these are a little tidied up as I generally post them there rather later.) Printing at the normal standard of 300dpi would result in images only 2 inches wide, and while the smallest images in the book are not much larger than this (I think around 3.5 inches) others are significantly larger – around 8 inches) and on the covers and a few inner images cropped to an 8.5 inch square they are even further stretched – to 12×8.5 inches on the cover (which also requires a slight crop. A little simple maths gives a figure of around 47 dpi.

The book actually looked pretty good on screen, and I couldn’t resist going ahead with the order. Of course I could have done a better job myself, more selective and with better quality originals, and the selection algorithm hadn’t included a number of my favourite pictures from the year.  There were just too many pictures from some events that were just a little too similar for me to have included them. But as a fairly random selection of typical images from the year it seemed useful, the kind of thing I could possibly had to people who ask me ‘So what do you photograph?’ as a view of my current work.

Possibly one day I’ll make a better book of 2016, but it would probably take me a week or so to put together, and still cost almost as much as the roughly £50 for this volume (allegedly on special offer), so I clicked and went ahead. It’s possible to edit the book, take out pictures and choose replacement images, but I decided to make only a single change, removing an awards certificate for this blog. Essentially I wanted to retain the automatic selection which was reflected in the title I added to the cover, ‘Peter Marshall 2016 – A random selection‘. I missed one picture that I should have removed, by my friend Townly Cooke who died last year – though I was happy with it appearing once, I should have removed a second, smaller version of the same image, but I must have turned over the page too quickly to notice it.

Obviously the quality of the highly upsized images is noticeably poorer, but the pictures still work, and however Re-Snap upsizes them, it does a pretty good job. Obviously the larger images don’t stand up to detailed scrutiny, but if you saw them in – for example – the pages of a newspaper, they would be acceptable. My main complaint is that the printing inside the book lacks vibrancy, almost as if it had been printed with watered-down inks, rather than any lack of detail or sharpness. It’s acceptable but lacks the punch of the original sRGB files, though the colour balance seems more or less spot on.

Seeing the result, I’m rather pleased that I now watermark all the new images I put on Facebook – and the watermarks are present throughout this book. It doesn’t stop people using my images without permission, sometimes complete with watermark, but more often cropping it or even removing it in Photoshop, both of which seem a clear admission of guilt. If only I could find a few abusers worth suing in the UK courts. I wrote a post here some while ago about an images that I had found used without permission on over 80 web sites, none of which on investigation I felt were worth pursing.

And recently, Google have published a paper on Pixel Recursive Super Resolution which doesn’t make for easy reading, but essentially shows how a believable image (rather than the actual original) can be created using neural networks from even a very small 8×8 pixel pixillated image. Unsurprisingly it works better on reasonably predictable subject matter, such as faces (which after all generally have the same number of eyes, a nose and a mouth) than other subjects.

And the most predictable images of all must be the photographs of ‘celebrities’ that so obsess our popular media. There can’t really be any need for yet another photograph of any of them, yet outside the Albert Hall yesterday was a pen crammed full of photographers with hardly room to swing a lens, when inputting a little blur and typing in a name – for example ‘Ken Loach‘ – could surely have generated an equally newsworthy image.

It was of course not his image, but his film, I Daniel Blake, and his speech that were newsworthy, telling to the nation the truth about the terrible treatment of benefit claimants by the DWP, something that all those who visit our job centres or talk to those who do already know, but which the complacent well off like to assure us doesn’t happen. And while some news outlets reported it, the BBC did their best to play it down, ignoring it in their early bulletins, though rather grudgingly reporting it later.

I mention him mainly because I’ve photographed him on a number of occasions in the past, but there was no way I’d ever cover an event like that. I was having much more fun a few miles to the west at the Willesden Green Wassail.

Fuji blues – and greens

Monday, January 30th, 2017

I’ve been using Fuji-X cameras for some years when I want something a little lighter and more compact than the Nikon DSLRs. They are usually the cameras I turn to when I’m not photographing events and don’t need the file size of the D810. The cameras I take on holiday.

But though I like some things about them, and have got some decent results, I also have some reservations. They are just too complicated and the controls and menu system lacks the simple and logical pattern of the Nikons. And there are too many ways in which they are just not so responsive and so reliable.

I can leave a Nikon switched on all day, safe knowing that the battery won’t run down and the camera will respond at the slightest touch of the shutter release. With all the Fujis, even if you switch off the rear screen and digital viewfinder, and don’t make any exposures, the battery still runs down. It’s kind of the worst of both worlds in that it Fuji cameras have an auto switch off that you can set, which does switch the battery off after the set time, but fails to stop the battery running down. When you want to take the next picture, even if the battery is still in juice, you have to wait a second or two while the camera starts up – and while you miss the pictures.

With the first Fuji I bought, the fixed lens X100,  things sometimes got even worse and the camera refused to switch on, either until you took the battery out and back again or pressed the shutter for ages, turned around three times widdershins and said the magic word. At least later models more or less fixed that, but still too often meant that when you pressed the release nothing at all happened.

Then there is the colour. Most digital cameras I think have problems with intense reds, losing the highlights, but Fuji has its problems with greens as well – unless you like your grass super-emerald rather than au naturel. And my XT1 has another problem – which needs extra work on the raw files – in that some Fuji lenses give results that are far too magenta, typically needed a correction of perhaps -35M in Lightroom. It doesn’t seem to be something that every XT1 suffers from, though I have found fellow sufferers, and possibly it could be solved by sending the camera back to Fuji for repair, but it only came to my notice after the guarantee period (when the camera went back to Fuji twice for other issues) was up when I bought another lens.

Then there are the mickey-mouse modes. I’d like to ignore them, but the combined ISO and mode knobs on the XT-1 are tricky to use, and changing ISO all to often puts you into what Fuji laughably call the Advanced Filter mode. The two dials are supposed to move independently – and sometimes they do, but other times both turn together. The resulting images are not nice. Jpegs rather than RAW and with impossible to correct contrast or colour or both. You can’t convert from ‘Dynamic Color’ or any of the others back to sensible colour, though you can just about get a half decent black and white.

It’s a shame because all of the Fuji cameras I’ve bought – X-Pro1, X-E1 as well as those already mentioned  have been almost there. Delightful in many ways, which is perhaps why I’ve several time bought another, but…. I’m even hoping that Fuji have at last got it right in the X-T20, and mug that I am, I’ll probably buy it.

And then there are those X-Trans sensors. It seemed a good idea to get away from the Bayer pattern, but I’ve never been convinced that they really improve things, though Lightroom at least seems now to have learnt how to get usable results after a really shaky start. But if  you still feel they are definitely an improvement (and Fuji cameras seem to have a unique facility to produce fanboys) you should read an article by Jonathan Moore Liles, which I first saw in PetaPixel, but is a little better read on medium com, where the pictures are larger.

Its an article which I think more or less destroys the claims of superiority of the X-trans sensor, which can perhaps at worst be seen as a marketing gimmick and at best simply a different balance between colour and luminance, and one which has some unfortunate side effects. In practical terms these are seldom if ever particularly important, and are often mitigated or eliminated by suitable corrective processing which I tend to apply fairly routinely in Lightroom. There is a tendency in portraits – whether on Nikon or Fuji – for faces to look a little flat that a little brushing with a positive value of ‘Clarity’ can improve, and the whites of the eyes (sclera to use the technical term preferred by Liles) usually benefit from a little more brightness and contrast which I think reduces the colour bleed into them.

Then there is the question of the Raw processing software preferred by Liles:

The RAW file was processed using Darktable’s Markesteijn demosaicking algorithm (3-pass mode) with a single iteration 9×9 chroma median filter followed by application of a bilateral filter on the chroma channel and light sharpening. The color profile is my own, generated from shots of a Wolf Faust IT8 chart and should accurately represent the colors in front of the camera.

Most of us just rely on Lightroom, though Fuji purists stick with the free Silkypix converter that Fuji provides, insisting on its superiority. Like me, unless you are a Linux user you have probably never heard of darktable, but Googling takes me to the site which tells me “darktable is an open source photography workflow application and raw developer. “  There is a MacOS version but not one for Windows, though I wouldn’t be rushing to try it out if there was, and there is a page about its X-Trans support which gives you some idea about that Markesteijn thingy, but includes the statement  “Though darktable now can read and process X-Trans files, there are plenty of opportunities to improve camera support. In particular, as mentioned in “What’s involved with adding support for new cameras“, each camera model could benefit from its own basecurve, white balance presets, lens correction, and noise profile.”

Although I have nothing against open-source software – and there are several programs I use or have used (including before Lightroom got better another RAW converter) – I think the use of it here is a serious weakness in the argument. First it entirely locates the article in the high geek rather than photographic sphere, and secondly it raises doubts about whether this particular software is as effective as that recommended by the manufacturer (and privy to their unreleased data), or to the commercial software such as Lightroom (and Adobe have enjoyed some cooperation from Fuji) and Phase One.

Of course, the debates about X-Trans and Bayer are marginal. Both can produce decent images and the differences between the two will seldom be apparent or important. Photography really isn’t about the minutiae of technical differences, but about what your images speak.