Monday, June 23, 2014

RAW or not RAW, that is the Question

If you would have asked me that question some months ago, the stereotypical answer would have been: RAW!
You may read a certain doubt between the lines, which however is only true under very specific circumstances.
I am a big fan of shooting photos in raw format and treating those in RawTherapee, DarkTable and Lightroom. Raw data, i.e. data as recorded by the sensor, gives a huge advantage in post processing, since nothing is compressed (read: thrown) away.

As a regular reader, you may recall an earlier post about the Canon IXUS 140 used in raw, to great success, while jpg images, processed in-camera, had the edge concerning distortion.
As a matter of fact, the in-camera correction could always be applied to the raw image by use of post processing software.

However, everything changed with the purchase of the Fujifilm X100S!
The X100S not only comes with an enhanced sensor, with the XTRANS filter array, it also improves the sharpness of the JPG-image by using information about the 23mm prime lens of the camera.
What does that mean? As soon as a focal condition is established, the entire optical system is known. That is exactly what Fuji uses in the JPG converter algorithms.

Presently, such algorithms are not implemented in any software able to deal with XTRANS raw-files.

And here is the dilemma:
  1. Shooting in RAW allows me for tweaking the files in terms of exposure.
  2. Shooting in JPG gives the full advantage of Fuji's JPG converter.
Of course, the primary answer is simple: Write both file to the memory card and use which ever one is better! That's what I am doing right now.

Fuji took it a step beyond: RAW-photos can be tweaked in-camera and exported to JPG using the lens data to increase resolution. I have not done this yet, folks on the interweb write that being a really fiddly procedure.

To answer the question I posed in the title: use RAW when using any camera other than the X100S.

When using the X100S, you want to record both, the raw and the jpg. Make sure to have all parameters for jpg set to your likings and exposed spot on. If exposure is spot on and all the settings are ideal, the jpg image might actually be much better than the raw with all tweaking attempts... but only then!

Have a look at a screen capture that shows both (click to see on full resolution),  the jpg on the left and the raw on the right:
JPG vs RAF (raw)

Just in this particular case, i.e. the Fuji X100S, I will primarily use the JPG-file, for its enhanced sharpness.

Up to now, with any other camera I own, I would prefer the raw file!

Sunday, June 15, 2014

Exposure w/o Metering - Part 6

Today was a mixed bag of light, i.e. very clear and also cloudy skies, in sometimes rapid succession.
Hence, I went out to photograph the shadows. I was talking about in the last post. What the day did not provide me with, at least not during the time I was hunting shadows, was a full (bright) overcast. Of the 3 promised conditions, we are therefore missing one.

Down to business. You will see that my subject will not deserve a place in a gallery, however, I believe several effects of bright sunlight versus light overcast can be easily demonstrated.

The first image refers to shadows cast by unobstructed sunlight. When this short of shadows can be seen, the sunny 16 rule applies, i.e. stop down to f/16, shutter speed = s/ISO.
hard light (direct sun) casting sharp shadows and a high contrast image

The second image show sunlight shining through a thin layer of clouds. This is the condition when you want to open up the aperture to f/11.
soft light (light clouds) washing out the shadows
You might notice that the upper image has a much warmer overall tone. This is actually created by the dominance of the direct sunlight over the blue-sky ambient light. In the second image, the ambient light starts to take over filling the scene with blue light, hence the colder color tone.

Another thing to learn here. the total amount of light present in the shadows (in theory) is identical and caused only by the ambient light, i.e. the blue sky. Why did a write in theory? Very simple, in real life reflections, e.g. from buildings, cars, etc., can occur. It is fair to say, I believe, that in the above images, real life meets the theory pretty well.

For techies under my readers, the following remark. Both shots are taken with the X100S in aperture priority, so that the images are technically correctly exposed. Also, I used RAW files, thereby preventing color temperature correction influencing the outcome.

Finally, for the f/8 condition, just imagine that there is no shadow cast at all, and all you see is pavement and dirt, even more blue-ish.

Concluding, to understand the illumination, watch the shadows!

Exposure w/o Metering - Part 5

Up to now, I was referring to a clear blue sky and direct sunlit scenes. Can we use the "sunny 16" rule under conditions other than sunny?
Yes, no problem at all!

Let's assume that for this discussion, it is mid-day and we are photography a frontlit scene. Also, we don't care about the D.o.F for the time being.

In full sunlight, as the rule suggest, we are now photographing with an aperture of f/16. This is very hard light, throwing very sharp shadows. Hence, whenever you see very clear and sharp shadows,  you should consider using f/16. BTW, this light would be referred to a hard light.

With a light cloudy sky, the direct sunlight is somewhat attenuated and the contribution of scattered light increases. Shadows are a bit washed out under such lighting conditions. When I say washed out, I refer to clearly visible, having blurry soft edges. To compensate for this drop in illumination, we need to open up to f/11.

When the sky is heavily overcast, but still bright, w/o any direct sunlight, the only natural source of illumination is scattered light. Scattered light comes from all direction, hence, no shadows will be present. This lighting condition is referred to as soft light and would require to open up one stop to f/8.

With a rainy sky, a lot of light is absorbed. I figure it is not necessary to detect this condition by looking at shadows... to me the best indicator of rain is getting wet. The mentioned loss of light due to absorption requires us to yet again open up one stop to arrive at f/5.6.

Somewhat obvious, in heavy rain, we would yet again open up one stop to f/4. However, I figure it is pretty rare to go out shooting under such weather conditions.

I hope to be able to shoot some frames showing the different shadows of the first 3 conditions mentioned above. If so, I will show those in a future post.

Exposure w/o Metering - Part 4

In the last part, I had a look at the influence of the incident illumination angle. To prevent the post from getting too long, I essentially skip the explanation why that is.
However, since this is one of the more important things to the sunny 16 rule, I would like to pick up the topic once again.

The basic principle behind the angle of incident light can be illustrated by the phases of the moon.
Please have a look, e.g. on this web-page.

0˚ between illumination and line of sight
If the sun is directly behind us, and the moon is directly in front of us, that would be full moon. In this phase we all the sunlight shining on the moon.

If the sun is on either side of us, this happens during the first and third quarters of the moon, very obviously, we see only half the light illuminating the moon.
In terms of photography, reducing the amount of illumination by half reflects 1 f-stop. To compensate for this lost stop, we need to open the aperture by one stop, or double the exposure time.

When we are actually looking into the light source, the analogy to the phases of moon starts to become a little weak. The reason for that is, that we can't see any of the sunlight illuminating the moon, still the moon is somewhat visible, due to earth-shine. Earth-shine is the sunlight that the earth reflects, which then illuminates the moon. As I said, this is when the analogy phases out.
However, we still can learn something about illumination of a scene in photography.
In earth, we are surrounded by an atmosphere, which itself scatters light. This scattered light is the illumination of a scene shot under backlit conditions. Of course, scattered light is less intense than direct light. To compensate for this, we need to further open the aperture.
This, of course, only holds true if details of a scene are to be photographed. For a backlit profile photo, the exposure has to be set to match the brightness of the light source, i.e. stopping down the exposure, rather than open it up.

Saturday, June 14, 2014

Exposure w/o Metering - Part 3

In the earlier parts of the series about determining exposure w/o metering, I kept a relatively theoretical approach to the effects of illumination, i.e. luminosity.
In this post, I will cover another aspect of luminosity, however, this time I will provide some real life examples.

Previously I wrote about the influence that time and latitude have on the rule "sunny 16". In this post, I will have to make use of the influence that local time.

For taking the following images, I went out later than 6pm local (MEST), therefore, I opened the aperture by 1 stop to f/11.
The 3 following photos were taken from the very same spot, with an essentially cloudless sky, and the camera set to s/250 and ISO250. Please observe the shadows to determine the direction of light.

0˚, f/11, s/250, ISO250, front lit

90˚, f/8, s/250, ISO250, side lit

180˚, f/5.6, s/250, ISO250, back lit

In the last photo, you may notice the lens flair, which was created by shooting into the glaring sun.

You will notice that the mid-ground (the trees) is perfectly exposed in all three photos. Of course the highlights will be blown out when shooting into the light... no surprise here.

Comparing the 0˚ and the 90˚ photos, you will notice the balance of the tree's highlights and shadows are the same, while contrast and highlights are more pronounced at the 90˚ shot. At 180˚, the contrast is somewhat "back to normal", when accepting the blown out highlights.
There is also a noticeable difference in the exposure of the sky. At 0˚, the sky is exposed the darkest. Since the brightness of the sky is independent from the scene, when opening up at 90˚, the sky portion of the image receives more light and is therefore brighter in the photograph. (At this angle, a blue sky can be recovered by use of a polarizing filter, which I will discuss in another post sometime later.)

Concluding: use the sunny 16 rule when shooting front-lit scenes. Open up 1 stop for side-lit scenes, and 2 stops for back-lit scenes.

I hope that this brings you a bit closer to understanding exposure and lighting under given circumstances.

In future parts of this series I will explain how to determine the illumination under circumstances other than an open blue sky.

Exposure w/o Metering - Part 2

In part 1, I discussed the rule "sunny 16" and the possibilities opened by variations of the rule according to the use of aperture stops combined with shutter speeds.

In this part, I would like to introduce the first constraints to the rule, at least what I believe to be the first constraint: time of day (actually, the interweb in pretty silent about this topic) and time of year.

As mentioned in part 1, sunny 16 applies to mid day, i.e. when the sun is the highest.
Since today, everything is available world-wide, I would like to add that this statement holds true only to a certain range of geographical latitudes! Actually, my understanding is, that the sunny 16 rule is a rule valid for latitudes between about 30˚ - 60˚ N/S.

To my understanding, the rule is all about taking pictures of your family during summer vacation. The rule reflects the amount of illumination available to the scene.

As a physicist, I would like to inform you about the hardness of sun light, i.e. light being collimated, in other words, the light rays are all parallel.
At a perpendicular angle of incidence, the light rays are closest to one another, meaning the illumination (light yield) is at a maximum.
At an angle parallel to the incidence of the light, the illumination is minimal.

Confused?! Take an orange and put it in sunlight. You will observe that the level of illumination drops, dependent on the angle of illumination.

The same happens on Earth. Dependent on the time of day, the sun will illuminate the earth's surface more or less, that's the influence of the longitude. The same holds for the latitude, although this changes with the time of year.

I figure, when traveling, one has to adapt the sunny 16 rule according to the latitude and the time of year.

I would do the following in general:
  • latitudes between 0˚ and 30˚: stop down by one stop
  • latitudes between 60˚ and 90˚: open up by one stop
  • in spring/autumn, between 6:00 and 9:00 local time: open up by one stop
  • in spring/autumn, between 15:00 and 18:00 local time: open up by one stop
 Of course, this needs adaption depending on the longitude.

Remember, I still talk write about a front-lit scene.

Exposure w/o Metering - Part 1

This will be a multi-part story about the technique of taking properly manually exposed photographs without the use of any light metering device.

30+ years ago, I heard about the technique called "sunny 16". However, being relatively young that time, with film being expensive and a heavy burden on my pocket money, I did not dare to use it, in particular since I was mainly shooting color slide film, which demands spot on exposure.
Consequently, I bough a relatively inexpensive light meter, a "Gossen Bissix 2" which did a great job, I still have it to the present day.
The light meter was used essentially only to measure the illumination, not the light reflected from the scene, thereby being independent from the color/brightness of the scene itself. Actually, I bought the light meter for exactly that, not sure if I used it ever for metering the scene itself.

Although the light meter did a great job, it costs additional time to take an accurate meter reading. Some shots were gone before the lighting was determined. To the time of expensive film, I thought "tough luck, no photo now, more luck next time"...

Today, I do not rely on pocket money any longer, I got my own income now ;-) So I decided to finally buy a camera which is able bring back the good old manual times, Fujifilm's X100S. The X100S is probably the best camera I ever owned, for sure it is the best compact point&shoot available on the market, which is definitely reflected by it's financial resistance, being somewhat a short circuit in the wallet.  Right, today, I could even afford to shoot more film, but where can you buy film today?!

Back to the topic: the "sunny 16" rule.

The rule is pretty simple:
  • On a sunny cloudless day,
  • mid day between 9am and 3pm local time,
  • the exposure time is the inverse of the ISO sensitivity
  • at an aperture of f/16
  • when the sun is directly behind the photographer.
What does that mean? In essence, this is what a light meter does, when used for measuring the illumination of a scene. A light meter, with the diffuser/filter closed, pointed at the sun a mid day on a cloudless day, should read f/16 for 1s/100 with an ISO of 100. In the film days, ISO100 was a pretty normal film to use. A shutter time of s/100, however, did not exist. The closest to s/100 is s/125, which consequently is the shutter speed in this example.

However, there might be adaptations to this rule, e.g. if you wanted a shallower depth of field. Using an aperture of f/16 creates a very wide D.o.F., which is good for landscape photos, but may ruin a portrait. With a camera of those days, the shortest shutter time was s/1000, which would reflect an aperture of f/5.6 for an exposure equivalent to the example above, creating a shallower depth of field however.

Depth of Field
In general, whenever you need to open the aperture by 1 stop, the exposure time needs to be divided by 2.
  • f/16 - s/125
  • f/11 - s/250
  • f/8 - s/500
  • f/5.6 - s/1000
The above aperture exposure time pair will result in the exact same exposure of the image, however, the depth of field will be very different.

So far, so good. I will end this post here, w/o any example pics...

Some of the constraints metioned, I will write about in further parts of this series. I also intend to show examples shot with the X100S with manual settings.

For now, I would like you to absorb the rule "sunny 16" and the equivalence in exposures.

Friday, June 6, 2014

Goin' old skool ... sort of

In earlier posts, you might have seen that I love the Panasonic Lumix LX7 a lot for its great low light performance, the stepped zoom (simulating various primes) and the low noise shutter. Although the LX7 is one of my all time favorite cameras, in terms of handling, it lets me down in street photography.

In other posts, you might have read about me praising the Olympus micro-4/3 system for having this super-fast AF and the cool street photos that could be shot at 9mm and 15mm focal length, using the body cap lenses. However, the shutter, although relatively quiet, compared to an SLR camera's shutter, is rather noisy, in particular when compared to the LX7.

Having hesitated for quite a while, I finally bought a Fuji X100S. And I love it!

Why hesitating in the first place? Well, for what it provides, the X100S seem well overpriced, not to call it expensive. One gets an APSC sized sensor, hooked up with a 23mm lens. That's it! No zoom, no bells, no whistles.
Well, really?! Concerning the picture taking part, that might be true... However, concerning the handling part of the camera, we are looking at a gem!

The X100S offers a hybrid view finder, i.e. an optical view finder which is enriched by an overlay of data and can be toggled to an electronic view finder by the twitch of a finger.
In my shooting, I regularly switch between the 2 modes of the view finder. Once you learned where the frame is, the optical VF will serve perfectly in street photography, since it does show a wider view than the frame captured on the sensor, giving the photographer a lot more time when persons are predictably moving into the frame.

The advantages of the sophisticated view finder set apart, the X100S offers one more invaluable feature which is often neglected: silence!
Just like the LX7, the X100S is nearly inaudible, thanks to the leaf-shutter built into the lens.
I was taking pics at an art's fair in my town and nobody noticed it.
Rijswijk art fair
Personally, I think that the above photo does belong to the group of weaker photos I took over the years. The interesting part about this shot is, the artists did not notice me when I was sticking my X100S (35mm full frame equivalent) right into his face.

The Fuji X100S gave me something back, which I was missing in my recent photography: the memories of my youth!
When I was young, my father shot with an Agfa Optima 500 sensor (42mm lens). At some stage, I was using this camera too. In the optical view finder, Agfa displayed "frame lines" to reflect the frame to be found on the film in the optical view finder of the camera.
Besides having those frame indications, the Agfa was next to inaudible, concerning shutter noise, just like the LX7 or the X100S.

And here is what I love about the pricey X100S, the optical view finder with the mirrored in frame lines and shooting information.
Thank you Fuji for bringing back the memories of my youth with a contemporary camera that performs beyond any doubts!