2.Number of pixels and image quality

First of all, use software that displays a photo on a PC, such as Nikon Picture Project or Nikon View to open a photographic file you created. Next, enlarge the picture gradually. When the magnification reaches its maximum value (about 800%), closely observe a part of the screen. You can see a box with a single color and density. This square box is just a pixel.

<01>
Photo displayed by Nikon View (Browser). Selecting the command "Fill the Screen" displays the whole photo on the screen.

<02>
Photo enlarged at a magnification of 800%. Observe a part of the screen!

<03>
You can see that the photo consists of many boxes, each has a single color and density. This box is called a pixel.

Now you know that a pixel is a square box containing a single color and density, and that many pixels compose a digital image. Here, let's consider the number of pixels. Every digital camera has the menu "Recording Size." Nikon D50 allows you to select one of the following three sizes: L (3,008 × 2,000), M (2,256 × 1,496), and S (1,504 × 1,000).

The L size, for example, means that the long and short sides of an image consist of 3,008 and 2,000 pixels, respectively. The total number of pixels are about 6,000,000 (3,008 × 2,000), which is represented as 6M (M is an abbreviation for Mega = 10⁶). Find the number of pixels for the M and S sizes in the same fashion.

Millions of pixels or several mega-pixels can make a strong impression, but they just represent the number of square boxes composing an image.

<04>
Nikon D50 allows you to select one of the three recording sizes: L, M, and S. They are about 6, 3, and 1.5 million pixels respectively.

Note that the number of pixels is called the Recording Size. Assuming that each square box (pixel) is the same size, the greater the number of pixels, the larger the image. This is the reason for the first note in this paragraph.
It is often said that the greater the number of pixels, the higher the image quality. This means that if the image size is fixed, as the number of pixels increases, the size of the square box of a pixel decreases. That is to say, the denser image can reproduce a subject more precisely.

The User's Manual of Nikon D50 says that the L, M, and S sizes approximately correspond to the print size 38 × 25 cm, 29 × 19 cm, and 19 × 13 cm respectively, where the image resolution is 200 dpi.

Resolution is a measure of image clarity. The unit "dpi" is an abbreviation for dots per inch-the number of pixels per inch (approx. 2.5 cm).
The L size (6M) consists of 3,008 × 2,000 pixels. Dividing them by a resolution of 200 dpi gives you about 15 × 10 inches, which is almost equal to 38 × 25 cm.

A resolution of 200 dpi provides prints with sufficiently high quality and is realized by typical printers. In practical use, you can employ a lower resolution. If you want higher print quality, you need to use higher resolution.

I would like you to think about the size of the print you want to make. Most photos of family ceremonies or trips are printed out in the L (A6) size or the A5 size at maximum. This means that six million pixels corresponding to the B4 size are too many.

I recommend that you specify a small recording size where necessary. This provides many advantages: making photography speedier, increasing the number of pictures saved in your camera, shortening the data processing time, and comfortably copying photos to your PC.

<05>

<06>

<07>

<05>When viewing photos of different recording sizes from a distance, you can see no difference in quality. You can see a difference in clarity only when a part of them is enlarged.
<06>An A4 print sometimes can have adequate image quality even though it is made from an S-size photo (1.5 million pixels).
<07>You can identify each pixel when you make an A4 print from an image data with pixels not more than 0.75M-pixel.

There are many users who only display image data taken with their digital cameras on the screen of a PC, or just paste them on their websites or blogs. How many pixels do they need?
First of all, let's examine the number of pixels of the PC screen. If you have a loupe, view the screen of your PC through it. What a mystery! You see red, green, and blue spots only. These three colors and their intensity reproduce many colors. For the monitor screen, each pixel consists of a set of red, green, and blue spots.

<08>
When enlarging part of a PC monitor, you can see red, green, and blue spots. A set of them corresponds to a single pixel.

Well, how many pixels does the monitor have? There are several standards for PC monitors with different number of pixels. The following shows the typical ones. Note that the total number of pixels of each monitor is represented as a full screen.

SXGA (high-resolution PC monitor)1,280 × 1,024 ≈ 1.3 million pixels

XGA (normal-resolution PC monitor)1,024 × 768 ≈ 0.78 million pixels

VGA (general TV monitor)1,640 × 480 ≈ 0.38 million pixels

Generally, a PC monitor has only 780,000 pixels on its full screen. If 6M-pixel image data is correctly allotted to the monitor in a pixel-to-pixel correspondence, only a part (about 10 percent) of it appears.

Assuming that you simply display a photo on half the area of the monitor, the necessary number of pixels is 390,000 (780,000/2). Accordingly, hundreds of thousands of pixels are large enough to display photos on a PC monitor or to put them on your website or blog.

<09>
If 6M-pixel image data is displayed on the monitor in 100% scale (a factor of one), only a part of it appears.

<10>
Generally, a XGA monitor has only 780,000 pixels on its full screen. This number is reduced by the window frame of the viewer. Therefore, even a 300,000-pixel photo occupies about half of the screen.

<11>
These are options given by Nikon D50. Different models have different modes, but the basic ones are common to all our models.

The image quality mode gives you the following options:

• RAW: Presents raw data obtained by digitizing an electrical signal from the CCD. RAW means information that is not processed, so it is sometimes called "raw data." For more information, see the section "Using raw data."

• FINE/NORMAL/BASIC: Compressed image data is displayed in accordance with JPEG. The compression rates are 1/4, 1/8, and 1/16 respectively. The BASIC option has the highest compression rate and presents the smallest file size. For in-depth information, see the section "Using JPEG well."

• RAW+BASIC: Simultaneously records image data in two formats.

• In addition to the options above, TIFF is used by some models. Unlike JPEG, it saves image data without compression.

Unlike characters or numerical values, image data like a photo tends to have a remarkably large file size. Therefore, it is necessary to use data compression technology to decrease the file size and making the file easier to handle.

There are various data compression methods, but photographs require the reduction of the file size while maintaining the photographic image quality. For this purpose, JPEG, an image compression method, has been developed.

JPEG is an abbreviation for Joint Photographic Experts Group, which is the name of a working group who developed this data compression method for freeze-frame pictures. The group name has become established as the format name.

JPEG's key advantages are not only that it can greatly compact a file without degrading the photographic image quality, but it also allows changing the compression rate for a desired file size.

JPEG's disadvantages are that it gets rid of visually unnoticed parts from the image data. As a result, the higher the compression rate, the more the data is thinned out, that is, the worse the image quality.

Therefore, among the FINE, NORMAL, and BASIC options, the FINE option provides high-quality photos because of its low compression rate, while the BASIC option degrades the quality of pictures due to its high compression rate.

The NORMAL option is recommended from a general and practical point of view. However, the FINE option is essential to image processing. If sending a photo by e-mail, you need to use the BASIC option to reduce the file size.

Well, how different are these compressed photos? Perhaps you see almost no difference when you display them on a PC or print them out. Different users have different photographic purposes and vision. I recommend that you, yourself, compare photos compressed in such different formats.

<12>
Take a picture in each of the FINE, NORMAL, and BASIC options and compare the resulting photos.

<13>

<14>

<13>The FINE, NORMAL, and BASIC options show almost no difference when the photos appear on a monitor or are printed out.

<14>You can see a pattern like a growth ring in the background because the gradations of color and density are thinned out.

<15>

<16>

<15>These photos look identical even if you observe them very carefully. You have found a small difference! But it may be your imagination.

<16>The outline is not smooth and a mosaic pattern appears on the whole image.

Raw data is obtained only by digitizing an electrical signal from the CCD. We, at Nikon, call it NEF (Nikon Electricimage Format), which is not compatible with other makers' formats. The size of NEF data is very large because of the lack of compression. Nikon D50 needs more than 4M bytes per picture, which is constant and no recording size can be chosen.

The word "Raw" suggests that image data is created just after raw data is processed.

This means that raw data is a raw material for making image data. Accordingly, processing it differently gives you various image data. You can do it with the software "Nikon Capture."

Such data processing is normally carried out in your camera. However, by doing this time-consuming work by yourself, you can control the quality of a photo, such as color and gradation, to get high-quality image data.

I recommend that you use raw data for your hobby or business to get photos with the best quality rather than for normal or practical use.

<17>
This is a scene where Nikon Capture is opening raw data.

<18>
Nikon Capture allows you to adjust the image quality by using all parameters, such as exposure, color, and reproduction of gradation.

<19>
Nikon Capture allows you to save the processed image data in the desired format.

Nikon D50 and D70 have a command named "Finish Setup." This command is the optimization of several options, each producing a picture with the best edge enhancement, gradation compensation, and chroma depending to the purpose.

<20>
The Finish Setup command executed by Nikon D50. It has options that adjust the image quality based on purpose. Moreover, the Customize option allows you to separately adjust each setting as desired.

Edge Enhancement, as indicated by the term, makes the edge of a photo clearer. It is also known as sharpness.
Select a higher value for this option if you want to sharpen a building or landscape photo-to make the edge clearer.
Select a lower value if you want to soften the edge of a portrait.

<21>
The Edge Enhancement option offers five values: -2, -1, 0, +1, and +2. By default, A (automatic) is set up.

<22>

<23>

<22>Photo taken with an edge enhancement setting of -2 (weak). It looks soft, doesn't it?

<23>Photo taken with an edge enhancement setting of +2 (strong). The edge looks sharp, doesn't it?

Gradation means a change in shade or color. The higher the gradation, the smoother the change, which means low contrast. On the other hand, the lower the gradation, the higher the contrast, for example, white and black are clear or colors become vivid.
On a cloudy day, the contrast is low, so you can take a clear picture by setting the Gradation Compensation option at a higher value. If a subject is exposed to direct sunlight, the shade is clear and the contrast is high. In this case, you can soften highlights saturated with white and black shadows by selecting a lower value for this option.

<24>
The Gradation Compensation option offers five values: -2, -1, 0, +1, and +2. By default, A (automatic) is set up.

<25>

<26>

<25>Photo taken with a low contrast setting. The reproduced gradation between black and white becomes low.

<26>Photo taken with a high contrast setting. The image becomes sharp.

Chroma is the degree of saturation of a hue. As the chroma of a photo decreases, the saturation decreases - the color tone becomes less vivid. On the other hand, as the chroma increases, the photo becomes more vivid and the color tone is brighter.

Namely, the Chroma option allows you to finely adjust the saturation.

Select a lower value for this option if you want to take a photo with an impressive halftone, and select a higher value if you want to make the resulting photo's primary colors brighter.

<27>
The Chroma option allows you to select one of the following three values: standard, - (weak), and + (strong).

<28>

<29>

<28>Photo taken at the negative value. It looks somewhat monotone, doesn't it?

<29>Photo taken at the positive value. It looks a little more vivid, doesn't it?