I mentioned in my last blog that I had been using the Sony RX10 exclusively over the last week and in doing so I noticed a few things about how to get a good exposure. Here is what I learned:
When the highlights clip they literally fall of a cliff. This can make the areas around the blown highlights appear very ugly. The Olympus EM5 highlights by contrast seem to behave much more like film, which seem to be more gradual.
One of the features of the RX10 is that you can display “zebras” in the live view. These “zebras” show you where the scene is exceeding the dynamic range of the sensor and the highlights are blown. You can also set the level of this so that you see a warning before the damage is done. For my camera I have this set at 100%+ so that if I see zebras I know there is clipping which as mentioned above can look quite ugly. I do this because I shoot RAW and can usually recover some of the damage.
What I have found is that there just isn’t much headroom in the RAW files beyond the zebras so you need to take care. With most cameras I have found I can expose to the right (deliberately overexpose the image) and then correct this by careful processing of the RAW file. This typically results in a higher quality image with less shadow noise and more detail. With the Sony RX10 this doesn’t seem to be the case and leaving the camera to calculate the exposure without any compensation seems to render very good images.
So how much can I over expose the image by? Well it seems to be only 2/3 of a stop. BUT a nice feature I have noticed is that the histogram that you can display whilst taking the image seems to reflect what is being captured in the RAW file whilst the zebras seem to indicate where the JPEG file will blow the highlights. I have noticed that I can be showing the warning zebras (set at 100%) but the histogram shows no clipping. The JPEG will show clipping but when I get the RAW file into Lightroom I can fully recover the problem areas.
Here’s a query that I see with some regularity so I thought it would make a good blog topic. When photographing landscapes the sky is often lighter than the ground and this can cause the land to be either too dark in the final image or the sky is too light. Two common approaches to solve this problem are the Neutral Density graduated (ND Grad) filter and multiple exposures. Which do I recommend?
ND grads are filters that fit to the end of the camera lens and which are dark on one half and clear on the other. The dark part is placed on the lens to darken bright areas such as the sky, so balancing the exposure with the land. This results in a nicely exposed sky and ground, leading to a more pleasing image. I should say that if you want to know more about purchasing these filters, there is a full tutorial on my Lenscraft website at http://www.lenscraft.co.uk/training/160.html.
The approach with multiple exposures is to take as the name suggest multiple images, all identical except that the exposure changes. Typically this involves bracketing the exposure by say 1 stop above and below the correct exposure. The resulting images are then blended together in an image editor to achieve a final image with a balanced exposure. Alternatively you might choose to blend the images together using some form of HDR software.
Before saying which method I prefer I should make it clear that neither approach is perfect and both have advantages and disadvantages. Because of this I actually use both approaches from time to time although I do prefer one over the other.
The problems I see with the ND Grad filter are:
The graduate filters can be a little clumsy to use as you really need a holder and lens adapter to attach them to the camera. They actually add quite a bit of size to the camera, sometimes quite dramatically.
The filters are prone to scratching as they are usually made of optical plastic. If you buy the glass ones they are prone to cracking or breaking, which given their price makes you want to cry.
Although these filters should be neutral they often display a colour cast which is sometimes linked to the lighting and weather conditions. This can result in odd coloured skies even when the filters are neutral.
They can sometimes be difficult to line up so that their effect isn’t obvious. This isn’t so much of a problem with micro 43 cameras which have a small sensor.
The areas of light and dark don’t always line up in a straight line e.g. a tree cuts into the sky and ends up becoming darker because the filter cuts across it.
The Multiple Exposure method also has its difficulties:
The exposures really need to be identical for the best results. This often forces the use of a tripod.
You need to ensure you don’t change the lens focal length or point of focus between shots.
The longer exposure shot in the bracket sequence can sometimes be soft due to camera shake. Another reason to use a tripod.
When using a lightweight camera I don’t like to carry a tripod unless I have to.
Blending takes time and photo editing skills. Often I just can’t be bothered with all the extra work.
On balance and because I come from a film background where I used to shoot slide film, ND Grads are my preferred option. I will however shoot multiple exposures if the situation needs it and I feel the benefits of doing so will outweigh the additional processing time. Overall I would say both methods work and it’s a personal choice you feel happiest with.