Cigarette cards revisited

No. 15: Officer, 1813, 5th (Brandenburg) Regiment of Dragoons.

Back in 2009, you may recall that I experimented with photographing a Prussian Landwehr drummer from a set of cigarette cards. At the time, I had it in mind to photograph the whole set and post them here on BfK but I never got round to it because I had white balance issues with the photographs. Recently, we upgraded our printer at BfK HQ to a swanky new wireless one that can also be used as a scanner. So, tonight, I dropped another one of the cards from the set under the scanner to see how it would turn out. I’ve got to say that the results are excellent and incredibly quick and easy to achieve.

So, for the record, this is card no. 18 of a set of 24 entitled “Napoleonic Uniforms” from Doncella (a division of John Player and Sons, Imperial Tobacco Limited). The set was issued in 1980 and it was probably actually issued with cigars rather than cigarettes. For your interest the blurb on the back of the card reads:

The painting from which this study is taken is entitled ‘Aftermath of the Peoples’ Battle of Leipzig, 16-19th October 1813′. It shows a mixed group of civilian and uniformed figures resting, tending the wounded and embracing each other after three days of intensive, close-quarter fighting in the streets of the city, at the end of which Napoleon was forced to withdraw across the Elbe to avoid encirclement. The officer polishing his sword wears a single-breasted ‘leibrock’ in the distinctive light blue of the Prussian dragoon regiments, with the standard grey overalls (trousers) worn by all ranks and a cloth cap, or feldmutze, often worn on campaign in place of the heavier shako. The cap band, collar and cuffs are in the regimental colour of black.

As I noted with the cigarette card of the drummer, there are some errors in the historical accuracy of the uniform information but the illustration is executed with some panache. I’d be interested to know more about the painting referred to in the blurb, if anybody has details.

Now that I’ve found a quick and way to get good quality images of these cigarette cards, you can expect the rest of the set to follow in due course.

Photography set-up

Recently I noted that I had reached the stage where I’m consistently happy with the miniature photography results I’m getting. I also promised to write it up – as much for my benefit (in case I forget how I achieved it) as for any interested BfKers.

Before I go any further, I make no claims to expertise in photography. What I describe here is the result of a lot of experimentation and exploration of blind alleys over the last couple of years. I reckon almost anybody could get good results by emulating and adapting my approach to suit their own circumstances. What I propose to do is split this description into four sections: the physical set-up, the photography equipment I use, taking the pictures themselves and an incy-wincy bit about digital post-processing. My aim throughout is to get crisp, evenly-lit pictures of miniatures against a white background that require the minimum of post-processing.

Let’s start with the physical set-up: the first point to note is that I set up figures for photography in natural ambient light. I’m fortunate that the dining room at the back of my house has big French windows that let in plenty of light during the day. So I tend to do photography in the early evening (during the Summer) or at the weekend but I avoid sunset because the light changes then.

I use plain old A4 white paper as the backdrop and I simpy prop it up against something like a box or pile of books to provide a curved background. I’ve learnt that one sheet of paper isn’t a good idea because it tends to be surprisingly translucent so I typically place two or three sheets on top of each other. The other thing that I got wrong for a long time was that I would put the miniature directly on this paper. Nope, that’s not the way to go because the figure casts shadows across the white paper that appear in the frame of the photo and are difficult to get rid of without lots of post-processing or having several lights positioned at different angles. I can’t be bothered with lighting – I never seem to get rid of the shadows and I get annoying reflections – even off figures that appear to be comlpetely matt varnished.

The figure is placed on a pedastal and its shadows fall in the valley behind.

The breakthrough came when I realised that I could place the figure on top of a small pedastal of some sort. Then the shadows fall in the valley behind the pedastal and in front of the white background. This area is outside the frame of the photo and so the shadows cease to be an issue. As an added bonus, when the camera is focussed on the figure in this set-up, the background is out of focus and thus its whiteness seems more even. The even better news is that the pedastal doesn’t have to be anything special. You can improvise a solution with all kinds of things like an upturned mug or eggcup – I happen to use my stacking white ceramic paint palettes. Ideally, you want something white and sufficiently tall to create a valley for the shadows to fall into (a couple of inches is enough). That’s a lot of words to explain something relatively simple but the picture on the right should make everything perfectly clear.

Secondly equipment: now this is, of course, an area where budget constraints come into play. We all have to cut our cloth according to our finances and what I use might seem expensive to some but pretty basic to others. Fundamentally, as long as your equipment lets you do certain things, then you should be able to get good results. The single most important factor here is a camera that has a fully manual mode that let’s you control both aperture (f-stop) and shutter speed (exposure). I rely on natural ambient lighting, so I need to be able to vary both these parameters to get enough light in through the lens and have a decent depth of field (more of that another time).

That means I often work with slow shutter speeds which in turn means that hand-holding the camera would result in blurred images. So I use a tripod to keep the camera steady. And, as a bit of overkill, I use an infra-red remote shutter release control but that’s not necessary if your camera has functionality for delayed shutter release.

Being specific then, because I know photographers like to know this sort of thing: my camera is a second-hand Nikon D70 DSLR. Second-hand cameras like this can represent really good value for money especially if you buy from a reputable source (I was lucky here) and it’s your first foray into DSLRs. The lens I usually use is the kit lens that came with the camera – an AF-S DX 18-70mm f/3.5-44.5G IF ED (phew, what a mouthful) zoom lens. The tripod is a lightweight Slik Spint Pro – it’s probably not a good choice for outdoor use in windy weather but it’s just perfect this job. And the last toy I use is a Nikon ML-L3 remote control which removes the risk of camera shake from my finger pressing the shutter release.

Automatic mode is used to capture starting values for aperture and exposure, then I switch to manual mode to take the actual pictures.

And so to taking the pictures themselves. Here’s the process I follow:

1. Set up the figure on its pedastal against the white background.
2. Position the camera on its tripod so that it’s as close to the figure as possible while still retaining the ability to autofocus with the lens at its highest zoom.
3. Set the camera to Automatic mode – that’s the mode in which the camera makes all the decisions about aperture, exposure etc for you. On most DSLRs this is labelled as “Auto” on the mode selection dial but check your own camera’s manual.
4. On my camera, you can partially depress the shutter release button and the camera will autofocus and record the settings its selected so that you can read them off the display. Do this and make a note of the aperture and exposure selected by the camera.
5. Now set the camera to Manual mode – usually denoted by a letter “M” on the mode selection dial but, again, check your own camera’s manual. Set the aperture to the one you recorded from the camera in Automatic mode.
6. Now we’re going to use a technique photographers called “exposure bracketing”. The idea is to take a series of photos, each one at a different exposure (shutter speed) – some of these will turn out over-exposed or under-exposed but one of the photos should turn out just right (well, hopefully!). I tend to take five pictures – one at the exposure I recorded from the camera in Automatic mode, two shorter exposures and two longer exposures. A lot of cameras provide specific functionality to do this for you (look up exposure bracketing in the manual). Mine does provide this option but only for a series of three photos so I normally take my five photos by manually adjusting the exposure between each one.
7. Before finishing, I preview the photos on the camera’s display just to check that it looks like got one or more candidates that appear to have the right level of exposure. If I suspect that I haven’t got one that’s right, I take a couple more pictures at different exposures.

I love digital cameras. The process I describe above takes only a few minutes with my DSLR and I can check the results almost immediately. Ironically, exposure bracketing is a technique that pre-dates digital photography and was (still is) used by film photographers to ensure that they capture at least one good shot of their chosen subject because the time and effort needed to go back and do the shoot again was significant (sometimes impossible) if they discovered, after developing the film, that none of their photos had the correct exposure.

After the photography session, I download the pictures on to the computer, select the best ones and do as little post-processing as possible. There’s all sorts of picture editing software you can use across a range of prices and complexity – anything from freeware like GIMP through to the professional kit like Adobe Photoshop, so I’m not going to go into the specifics of how to use a particular software package but rather make some general comments. The main four software functions I use are the ones that let me control level equalization (I normally let the software apply this automatically to be honest), crop the image tightly round the figure, resize the image to fit in a blog posting and save it as a high quality JPEG (I take pictures in RAW format, which isn’t web-compatibile).

On another occassion, I’ll write a little about the topic of depth of field. It isn’t much of an issue when photographing individual figures but it’s useful to understand if you want to take pictures of a whole base of figures and get as many of them in focus as possible.

Experimental photo shoot

Test shot of Perry plastic officer

A quiet Sunday afternoon provides the perfect opportunity to experiment with taking better photos of miniatures and put into practice some of the things I’ve learnt recently. On the one hand, I chose quite a challenging subject – a completely grey Perry plastic French officer. On the other, this was the first time I’ve been able to deploy the tripod and infra-red remote with the camera. I also assembled a couple of lamps from around the house to provide light sources from different angles.

For the technically-minded among you (and so that I’ve got a record of what I did) here are some of the more gory details…

Firstly, set-up: the camera was mounted on the tripod and I actually used my telephoto zoom lens at its fullest zoom (300mm). I set the camera to use aperture preference and started at f8 – the idea being to get a good depth of field. As it turned out, this figure holds a sword that (with the figure’s arm) sticks out about 30mm in front of the figure’s body, which was a bit of an issue.

I blu-tacked a couple of sheets of white A4 paper to the worksurface as a background (just one sheet seemed to be too transparent under the lights). The lamps were positioned to the front and side of the figure at 45 degree angles to catch it in the crossfire, if you see what I mean. This was intended to reduce shadows on the background.

Then I did something I’ve never tried before – I used the camera’s whitebalance functionality to set a custom whitebalance for this particular set-up. I’m sure the method for doing this varies from camera to camera so I won’t go into the specifics here. Look up “whitebalance” in your camera’s manual and follow the instructions. Essentially, I was able to get my camera to measure the lighting conditions on the white background and adjust its own whitebalance setting accordingly. This should remove those yellowish, redish or bluish tinges you sometimes see on pictures taken under artificial lights.

Once that was sorted, I set the camera to work with the infra-red remote shutter release (these tiny gizmos are really great fun), positioned the miniature in shot and started taking photos at various apertures.

Then I downloaded the photos on to the computer and did a little post-processing. I selected the image with the best depth of field and cropped it tightly. Then I tried various combinations of resizing, automatic equalization and lightening. The original images had still come out a little dark and the background was still a bit grey (but tonally neutral) so these tweaks fixed those problems to a reasonable extent.

The final image isn’t ideal but its certainly a promising start. On the plus side, there’s no nasty flash glare, shadows on the background aren’t an issue and I was able to get the background down to a nice unobstrusive and neutral almost-white with a combination of the custom whitebalance and a little post processing.

Painted Perry horse

There’s still room for improvement, though. The main thing is that I’d like to get brighter lighting on to the figure. That ought to reduce the need for post-processing and enable me to go up to higher apertures which will improve sharpness and depth of field. I think choosing a grey figure to photograph was asking for trouble in the first place but once we get to ones wearing paint, that ought to make life easier. At the end of the session, I took a couple of shots of the Perry cuirassier horse I painted at the end of last year to see how things would differ for a larger and painted figure. As you can see, this is easily a few steps on from the pictures I took back then. Given this, you may well be wondering why I bothered photographing the naked plastic figure at all.

Well, it’s all part of my preparation for painting Saxons. I’m anticipating an exciting package from Calpe Towers any day now and I’ve been giving a lot of consideration to conversions with Perry plastic heads and the unresolved issue of suitable paint choices for Saxon officers’ surtouts. So messing around with this Perry officer gives me a chance to kill both birds with one stone. The figure in the photo has actually been decapitated and had one of the spare Perry heads fitted (a natty number with a damaged shako). I did the conversion with some drilling and pinning and applied green stuff to hold the pin in place and fill in the gaps. And that part of the exercise went very smoothly.

The next stage is some painting. As mentioned above, the light blue for the surtout is one target. Peter F. has suggested Vallejo Sky Blue (VMC961) but I’ve got to work out the palette to go with it. The other target is a new white palette because I’m still not happy with the one I’m using at present and there’s an awful lot of white on those Saxons! So the shot of the naked plastic is a reference point for the rest of the test painting.

Cleaned up card

Cigarette card with levels adjusted

A couple of days ago I posted the first results of my experiments with photographing cigarette cards. At the time I mentioned that the outcome was sub-optimal because of how the white background on the cards was coming out grey.

Well, my thanks are due to Pete B, who posted a comment with a link to a tutorial on Cool Mini Or Not (CMON) that explains how to edit the Levels settings in photo editing software. So here’s my attempt to reproduce that piece of digital wizardry for you to compare and contrast with the first image.

Not bad at all – and only a few seconds work. I wonder if an even better result could be achieved when I have a better source image to manipulate.