Friday, June 24, 2011

Pseudo HDR

Digital Cameras are no doubt brilliant devices. Their tiny CCD or CMOS chips can capture an enormous amount of detail. However, their dynamic range- the range of brightness values they can capture in one picture, is limited, a lot more than the human eye.

You may have noticed a lot of pictures with "washed out"(completely white, lacking any detail) skies or too dark subjects. In some cases, this problem can be fixed by using a flash to balance the lighting. However, you cannot expect to light up a huge building using a flash, to balance it against the bright sky!

HDR or High Dynamic Range technique to the rescue! This technique involves taking the same picture at multiple exposure settings and then merging and tone-mapping them using a computer. Usually 3 photographs are taken, exposures separated by 1 to 2 stops. Most cameras have an "Auto-Bracketing" setting, which takes these 3 pictures automatically when you hold the shutter release down.

There are 2 problems with this technique:
1) you need a tripod or a steady surface to keep the camera from moving between the 3 shots.
2) you cannot have a moving object in the frame, or you will get multiple "ghosts" of the object.

Since I don't like using a tripod and searching for a steady surface can limit compositions, I prefer using another technique called pseudo HDR when needed. This technique doesn't extract as much dynamic range as normal HDR, but it is much simpler and can be used for handheld shots or moving objects.

So basically you shoot in RAW format-> this stores a lot more detail than jpeg(by detail here I mean variation of tones, not more pixels!). You have to underexpose the snap by 1 to 2 stops to preserve some detail in the sky while retaining detail in the shadows.

After this, we have to post process to extract all the detail we can from the shadows and highlights:
Open it in Adobe Camera RAW/ Photoshop/ Lightroom/ Aperture
Change the following settings roughly to these values. Tweak to your taste.
Contrast between 90 to 110
Brightness between 70 to 90
Blacks between 5 to 7
Recovery to 100%
now raise the brightness to about 100-120
use the adjustment tools:
Burn(darken) the sky and dodge(brighten) the rest
use a soft large brush and ~exposure -1.8 for burning and +1.0 for dodging

Use noise reduction, sharpening and a slight vignette for a finishing touch.

What I got was this:

Saturday, June 4, 2011

How to choose a camera

Long time no post, and that is because I'm in Chennai, about 500 miles away from home and I don't have my DSLR with me. So I'll try and post on topics which do not require it while I'm here.

Many people ask me which camera should they buy, so I thought of writing an article which could help them make informed decisions on this matter without having to consult me :)

The first thing to see is your budget. How much are you willing to spend? There will always be a better more expensive camera, so you need to draw the line on the price before doing anything else.

The next question to ask is, what will you be shooting most of the while? If you do most of your shooting in daytime and you are on a limited budget, a megazoom camera may be more appropriate than a DSLR, as you will need to spend money on lenses for the latter.
A DSLR will however give you full control over practically everything, besides giving you pixel level sharpness and low noise at high ISO(compared to a compact digicam), besides the flexibility of being able to use various lenses.

For DSLRs:
Check the make. Prefer Canon or Nikon to Sony. I'm not anti Sony or anything, but Canon and Nikon have a much better variety of lenses and their noise profiles are also better than Sony as of date.

Check for lenses and other accessories that you would buy over time and see how affordable they are to you. Check the burst rate of the camera in case you like shooting sports. Try and get the maximum viewfinder coverage and magnification that you can. Check a couple of reviews on www.dpreview.com or www.dcresource.com Finally, go to a shop and see what feels best in your hands.

For Non-DSLRs:
Decide whether you want a megazoom or a good quality compact or a normal digicam. Try and get one with Manual mode, as you would like to have control over your images. Make sure you have manual white balance on the camera. See if the camera if capable of saving RAW files. Forget megapixels, anything above 3 is good, as long as you dont have to print billboard sized images! Try to get as large a sensor as possible. Usually CCD sizes are 1/2.5". Bigger than that is good. 1/1.8" is used in a few good cameras.

Since you have only one lens on this camera, you want to make sure its the best you are getting. Besides Zoom range, you want the lens to be fast, as in have a large maximum aperture. The max aperture range will be quoted like f/2.8-f/5.6
This means that the max aperture when zoomed out completely is f/2.8 and when zoomed in completely is f/5.6. you want these numbers to be as small as possible.

Recommendations:
As of now
Entry level DSLR: Nikon D3100
Megazoom: Panasonic FZ 35
Good semi-pro Compact: Canon S95 or the latest G series

Thursday, May 19, 2011

About Flash: Part Two


Now what you know about the power& distance limitations of a flash, lets see some stuff you can do with a flash.
There are two ways to use a flash:
1. As the primary source of light
2. As a secondary source (fill flash)

As a secondary source, you would expose the image for ambient light and let the camera adjust the flash strength via TTL(through the lens) metering, while giving a bias if needed(+1 flash EV will double the power, -1 will half the power that the camera has otherwise calculated using TTL metering). Av(/A) and Tv(/S) modes expose for ambient light and use flash as a 'filler'. Also called slow sync when long shutter speeds are used (slower than about 1/30s)

As a primary source, u underexpose the background by a lot and use the flash's power to completely light your subject(or whatever part of it you want to light up). In a reasonably dark place, P(rogram) mode does this... it will not expose for ambient light, only for the flash via TTL.  Will usually use a shutter speed of 1/60-1/200s. When used in this manner, the flash freezes the subjects movement and leaves no noticeable blur, because the flash strobe has a very short time duration (~1/10,000s) and the camera is NOT exposing for ambient light.

When to use the flash?
Indoors: When there is insufficient light or to balance a subjects exposure with backlighting if a window is open in the background
Outdoors: for portraits/ group snaps, to lessen harsh shadows and to create catchlights (the twinkle in the eye that you usually see)

In this example, I used flash to balance the indoor and window exposure. It was about 3pm and had I not used flash, there would be two possible outcomes depending on how you expose the snap:
i) The subjects would be properly exposed, but the window in the background would be completely blown out(highlights clipped), ruining the snap
ii) The window would be exposed correctly but the subjects would be silhouettes!
The only way out was to correctly expose ambient light of the window with the lightmeter while letting the camera use TTL-metered flash to expose the subjects.

Flashes used directly can produce harsh shadows and ruin texture and it is advisable to diffuse them i.e. make the source seem bigger. There are many ways to do this, the most effective is bouncing it off the ceiling/wall. This changes the light source from being hard to soft.

I may have mentioned in the past about the X-sync speed of a flash, which is ~1/200s on most cams, and it wont allow you to shoot faster than this with the flash on. However you can use an external flash which supports high-speed sync to use flash at higher shutter speeds. This is especially useful for shooting portraits outdoors when you want to use a large aperture to blur the background as well as use the flash to prevent harsh shadows.

Monday, May 16, 2011

About Flash: Part One

A fundamental requirement for a photograph is LIGHT. This light can be of various kinds. Some important ones are:
Hard light- Perceived as a point source, gives well defined, contrasty shadows. e.g. the sun on a clear day
Soft light- A large source, gives soft shadows. e.g. An overcast sky(the sun is hidden behind clouds)
Transmitted- When the source of light is in the frame. e.g. A sunset
Each type of light has its uses (for portraits, soft light is generally preferred), and you cannot say one type of light is 'better' than another(in the absolute sense).

There are many instances when we need to artificially add light into the frame, for various reasons(some of which I will discuss in later parts of the article). One way of adding light is using a flash.

Flashes may be built into the camera (e.g. a pop-up flash) or may be an external unit. The power of a flash is expressed as a "guide number" in meters or feet(@ISO 100).

This formula will tell u how far a flash can expose a subject.

f-number x distance x sqrt(100/ISO)= Guide number

*sqrt= square root for those who dont know.

My canon 400D flash has a guide number of 12m... lets see what dat looks like in a practical situation:
Say I use a reasonable aperture of f/4 and ISO 400... then
4 x D x sqrt(100/400) = 12m
4 x D x 1/2= 12m
D=6m

So the flash can effectively expose a subject upto a distance of 6 meters, with the above settings.

If I keep the aperture constant and change the ISO to 100, the max distance the flash will expose is 3m.
for ISO 800, it will be 6 x 1.414= 8.5m

The reason for the square-root is because of the way light intensity varies with distance (inverse-square rule).
Say I have a constant source of light. Suppose, when I stand at a distance of 10 m from it, I receive 100 lux of light intensity, if I go another 10 meters away from the source (total of 20m from the light source now), I will receive 25 lux.(1/4th the original intensity when at double the distance)

#Remember that doubling the f-number (say from f/4 to f/8) reduces the light by a factor of the square of the ratio(in this case (8/4)^2 or by a factor of 4), that is why there is no square root function required for the f-number.

For 'ordinary lenses', the maximum aperture available varies with zoom. Take an 18-55 f/3.5-5.6 lens. When zooming in therefore, the maximum range of the flash is reduced. (With a constant aperture lens, this does not happen)

------------ More coming later in Part Two---------------------------

Sunday, May 8, 2011

About Histograms

One major advantage of Digital over Film is that you can see the results instantaneously, and if needed change the settings and re-take the shot. We normally use the camera's LCD screen for that. However, LCD screens can be deceptive. A more accurate way of determining whether the exposure was correct or not is using the histogram. Almost all digital cameras have them, so lets learn how to read and interpret them.

First of all, a histogram is nothing but a bar chart. Lets take a look at a simple example of a bar chart:
This bar chart is of a hypothetical mark distribution in a class. On the Y(vertical) axis, we have number of students. On the X(horizontal) axis, we have categorized marks into 4 groups, 0-25, 25-50, 50-75 and 75-100. You can see that the height of each bar is proportional to the number of students getting marks that fall into that group.

A histogram is basically the same thing. Instead of number of students on the Y axis, we have number of pixels. On the X-axis, instead of categorizing students by their marks, we categorize pixels by their brightness values, into 256 categories, ranging from 0-255 (for jpeg, 8-bits=>2^8=256 values). 0 represents pure black, while 255 represents pure white. A typical histogram looks like this:

You will not see all this, of-course. The camera will only display the distribution, to save space:
So now, the taller the bars in a specific region, the more the number of pixels having those brightness values. So how do you use this for exposure?
A normal image will have a full spectrum of tones ranging from black to white, well distributed. If the histogram has too many pixels on the left side, it means your image is underexposed, while if there are too many on the right side, the image is over exposed. If you have a very tall bar just near 0 or 255, it means the shadows or highlights respectively are clipped(lost all detail). You may have many times experienced highlight clipping... the sky becomes completely white, losing all detail of clouds, etc.

See this example of histograms of differently exposed images:
Courtesy: http://www.thecolorblindphotographer.com/

So next time you take a picture, don't rely on the LCD screen. Take a look at the histogram too!(and underexpose/overexpose if necessary)

PS: Sometimes the histogram SHOULD have more values on one side, for example snow. It should appear white-ish in the image, so you want most of the pixels to be on the right side. However using the histogram, make sure the highlights DO NOT get clipped.

Wednesday, May 4, 2011

The Zoom Perspective


You may have heard people say that a Wide Angle lens expands space while a Tele Photo lens compresses it... What exactly does that mean? 

Have a look at these 2 pictures:

focal length=17mm


focal length= 55mm

I've tried to keep the shoe(say subject) size the same in both photos by moving away when shooting on the tele end. But what is the big difference between these two?
Its the BACKGROUND, or rather the size of it.

When shooting with a wide angle lens, the background becomes a lot smaller than the foreground(as if when viewed with the naked eyes, they were separated by a large distance- hence the term expansion of space).

On the other hand, when using a tele lens, the relative difference between the size of the background and foreground is much less( as if when viewed with the naked eye, they are close by-> hence compression of space).

This also means that a wide angle lens gives more background than a tele lens.

I think the following diagram is self-explainatory as to why it happens:


So when composing an image, if you want to show a subject in its environment, use a wide angle lens. However, if your purpose is only the subject, a telephoto is preferable, so as to reduce the amount of clutter in the background.

Note: the 'normal' perspective of human vision is said to be equivalent to a  50mm lens on a 35mm film SLR/ full frame DSLR (sensor size same as 35mm film), which is ~30mm on a Canon crop body and ~35mm on a Nikon crop body DSLR.

Monday, May 2, 2011

Why I still love my nifty-fifty


left: Canon 17-55mm f/2.8 USM IS (f/2.8 is the fastest they make zoom lenses till date)
right: Canon 50mm f/1.8 (popularly called the nifty-fifty)

Although I recently got an AWESOME Canon 17-55 f/2.8 constant aperture zoom lens (which allows f/2.8 regardless of how much I zoom in, unlike other typical zooms [eg canon 18-55 f/3.5-5.6]), I still love my 50mm lens. Here's why:
1. Size: See the comparison picture above!
2. Weight: the nifty is just 130g in weight, while the 17-55 weights 645g, which is heavier than my 400D body!(and can get quite tiring if used over long periods of time)
3. Speed(aperture): The nifty can take in more than double the amount of light than the 17-55, when both apertures are set wide open( f/1.8 vs f/2.8), effectively allowing higher shutter speeds to be used with the same ISO thereby reducing motion blur, or alternatively lower ISO to be used with the same shutter speed thereby reducing noise.
4. Price: Primes are cheaper than zooms(especially large aperture zooms). Period!
5. No Zoom!- Zooms tend to make me lazy. The nifty makes me walk around searching for the best vantage point to shoot from.
6. DoF: The shallowness of the depth of field at f/1.8 is enchanting!
The nifty is also more flair resistant and suffers from less chromatic aberrations (like purple fringing at high contrast edges)

I've tested the sharpness of both lenses @ f/2.8 and found them very similar. I surprisingly found the 17-55 slightly sharper, but just by a hair, probably because of the USM accuracy. Both these lenses blow away normal zoom lenses in terms of sharpness.


Depth of field example @ f/1.8

I highly recommend buying the 50mm f/1.8 (or the f/1.4 if you can afford it- 3 times the price of the f/1.8). It should be the 2nd lens you own (the first being the kit lens that comes with your DSLR). I guess the 35mm [f/2(canon) or f/1.8(nikon)] is also great for non full frame DSLRs.