Small Beginnings

by Mark Graham


Starting out...

When you open a canvas in Satori you create a space that will contain your work. This space may have nominal dimensions already - as when you import a bitmap at a fixed pixel resolution – but, when starting from scratch, it is a comfort to have some idea of the final size and shape of your image (landscape, portrait, square etc) before you start. As Confucius says “It furthers one to have somewhere to go”. This is why you begin a Satori session by setting the dimensions of your new canvas.

"The Picture Properties dialog is displayed whenever you use the New Canvas tool."

These dimensions may be changed at any time and your work resized accordingly. This is because Satori is Resolution Independent. Unlike traditional bitmap editors, enlarging your work in Satori will cause no degradation or loss of picture quality. In the example above, a canvas with dimensions of 320 pixels by 240 pixels has been specified - a size suitable for an image for a website. This is not an obstacle to the later production of a high resolution image suitable for outputting as a poster.

"Your work may be re-sized at any time."

This is because your work is stored internally as vector instructions. The results of these instructions are displayed on your monitor, and in any output, as an array of pixels (a bitmap) of the requested dimensions. Any size of bitmap may be requested at any time and either rendered as output to a standard printer, a bitmap file on disk or displayed in a View window on your monitor. Multiple, concurrent views are permissible and Zoom tools are used to select and display details of any sub-section of the whole canvas.   Note: Data that is already of a fixed resolution I.E imported pixel data will behave as in traditional bitmap editors and degrade on enlargement (due to the usual practice of applying a Bi-Cubic Convolution filter to try to ameliorate the “pixellation” effect).

"The original image of Tracy (by John Gordon) - 720 x 480 pixels."


Size isn't everything...

A small area of this image of Tracy (originally 720 x 486) has been enlarged (32 times magnification) and clearly shows the pixels that compose this bitmap. After filtering, the detail is softened but the effect is blurry and out-of-focus.

"After enlargement but before filtering."

"After filtering, the image is blurry."

Satori brush strokes and shapes, however, are always redrawn at the best possible quality for the view (or output) requested.

In the example below, as you zoom into the image, you can see that the imported bitmap shows heavy filtering with loss of detail but that the brushing simply reveals more detail.

"Satori brush strokes remain sharp at all times."

It is important, in order to preserve quality, that your specified output is sufficient for your needs. In practice, this is always a compromise between what you want to do, what you must do and what you can do. In other words you will be limited by external factors such as available resources (RAM, virtual memory, disk space) and the performance and capabilities of the various available output media (printer, graphics card, Internet etc).

As a courtesy to the user, presets are offered - such as standard paper sizes -and many film, video and digital imaging formats are selectable. Once selected, these settings and dimensions are retained until you choose to alter them. The assumption being that if you begin work on a graphic for a website it is unlikely to be required as a poster or a movie! But it can happen. It surely has happened. Using many traditional bitmap editors this kind of unexpected change to a project, particularly late in the day, can render work-in-progress effectively useless due to the prior commitment to a given resolution or set of dimensions.

The freedom to re-size and re-scale at will, without the slightest concern that your actions will have any harmful effect on your work, can be liberating.

Free yourself from the tyranny of pixels and transcend their constraints!


How many colors?!?

We began, at the start of this article, with a canvas of 320 x 240 pixels. This defines the physical dimensions - now we can select the number of colors we want to use to display our image.

Once again, the proper solution will be defined by what is appropriate for the media (output) and necessary for the task in hand.

When producing graphics for the web, for example, it is common practice to design images using 16.7 million colors (32 bit) and then optimize special versions to display on the net (GIF files etc) that employ only 256 colors (8 bit).

Satori offers the option of using 281 trillion colors (64 bit mode) to display images that may be saved at any color depth at output. Since the human eye can only detect a few million colors and since your monitor/printer probably cannot display or reproduce so many colors, it is reasonable to ask why one would ever need 281 trillion. The reason, immediately clear to anyone who has ever tried to apply color corrections to high-contrast images or to computer shaded objects, is visible in the undesirable banding, moiré pattern and dithering artifacts that can be produced as the result of bashing images around on a computer. 281 trillion possible colors gives a lot of "headroom" with which to work and with this comes the peace of mind to know that, no matter what kind of extreme manipulation you make to an image, it will never show degradation in color content. Remember - you can always produce a 16 million (or 256) color image at will. And this can always optionally be calculated either from an operation calculated in 64 bit mode (16 bits per pixel RGBA) or by that calculation being made entirely using only 16.7 million colors (32 bit mode - 8bits per pixel RGBA). For more information on the benefits of working at high color depths see the article in the installed Satori Reference Guide, "64 bit and the Quality of Color" by James Hughes. You can also output a 64 bit image using only 256 colors using Satori's Web Optimizer tools.

Click the 64 bits Per Pixel checkbox in the Color Depth panel.

For now, we have chosen the physical dimensions and number of colors to use in our canvas - 320 x 240 pixels in 64 bit mode. The Picture Properties dialog contains other tools for setting the Pixel Aspect Ratio (shape of pixel - square/non-square), for specifying 1:1 Mode (meaning the allocation of one pixel on the image window per pixel of the input/output bitmap) and for setting the nature of the "background". For this exercise I recommend leaving the Pixel Aspect at the default setting (1), putting a tick in the 1:1 Picture to Pixels checkbox (not strictly necessary), and selecting Black (check radio button) as the option under Background.


Now it's your turn...

If we now click OK on the Picture Properties dialog, an, as yet empty (except for black), canvas of theses dimensions and color depth is opened and displayed in a new window on the Satori desktop.

By default, when starting Satori, the last-used brush style is recalled ready for use. Click to open the Brushes tab on the Paint Actions palette and make a brush selection from the library of preset brush types displayed on the tab. Also by default, the color red and a size of 30 pixels will be applied and these settings are suitable for this example. You can make color selections by clicking inside the Color Cubes that are displayed on the Color palette. You make brush size selections using the tools on the Paint Actions palette, Size tab. Make some scribbles near the center of the canvas/window. They might look like this...

"My scribble on a 320 x 240 pixel canvas."

Click the In button on the Zoom Controls palette 5 times to make a 32 times magnification view of a tiny detail from the exact center of your canvas. At first it looks blocky and pixellated. Click the Hi-Rez button on the Zoom Controls palette to cause the view to redrawn with the amount detail appropriate to the level of zoom.


"At 32 times magnification each pixel is greatly enlarged."

"After Hi-Rez, the true detail contained in the image is revealed."

This tiny detail can be used to generate a high resolution output image - such as for a poster - by the simple expedient of rendering at the required output size/resolution/format. From the Canvas Actions palette, Crop Region tools click Make From View to make a crop view of the canvas using the current Zoom View. Finally select Save Bitmap As.. and set the dimensions for the output file along with the destination filename and format. By default the Use Crop Area option is checked - set any value in pixels or inches/cms - the image will be redrawn at the new dimensions (rendered) and the result saved to disk.

It may, of course, also be printed.

"The detail from a nominally "small" canvas can be used to generate large, high resolution output."


In conclusion...

The knowledge that you can, at any time, elect to alter any detail or any action without compromise and without the requirement for inelegant “history” or other processor or resource intensive procedures extends this “liberation” by suggesting completely new ways of working. Because a Satori canvas (CVS file) stores instructions, not pixels, much smaller files are needed to reproduce your work – these files contain references to find (and use) other files and do not contain those files themselves. This is called “non-destructive” editing because you never alter the source files directly in any way. In effect, when working with imported files, you only save the changes that you make. The original files are left untouched and thereby, the disk space requirement for storing changes drops exponentially – typically to about 50k - 100k per image - notwithstanding the fact the image is, say, A2 at 720 dpi (or, indeed, a sequence of images such as from a movie or video clip).

Saving only instructions (instead of pixels) and using file referencing (instead of file copying) offers real production advantages by reducing overheads such as operator and system time, system resources, disk space etc and thereby reduces production costs that, in turn, lead to savings and profits.