Preparing your panoramic material

SPi-V works with separate media files, referenced from a (set of) XML file(s). The SPi-V engine can use a range of input material. This article will tell you how to prepare your (panoramic) content for display in SPi-V.

Graphic elements

Most of the input material for SPi-V will be simple images.

File formats

SPi-V accepts images in the following formats:

• Jpeg
• PNG (can include 8 bit alpha-channel*)
• GIF (can include only 1 bit alpha-channel, PNG is recommended)
• BMP / PICT (uncompressed, not recommended)

The executable viewer can be made to accept more file types such as TIF, TGA and even PSD through the use of Macromedia Xtras.

*: Alpha-channels can also be specified in separate (greyscale) images. See the image node specs.

Dimensions

SPi-V can handle images in arbitrary dimensions, but in order to ensure your audience get a good experience, you must take the following in account:

Because of the way SPi-V uses hardware acceleration, all graphic material must fit into Video RAM simultaneously. If the amount of RAM available is not sufficient for the graphic material, the engine will fall back to CPU intensive software rendering. For computers with 32 Mb of Video RAM (eg nVidia TNT 2 or ATI Rage Pro and newer), this means you can realistically show 6 cubic faces of 1024*1024 pixels and some small branding elements.

Projections

SPi-V can show images in different panoramic projections:

• Flat images, also known as rectilinear

  The type of image you get from single photo through a 'normal' lens.

• Cubic images

  A shortcut for 6 flat images, each representing a face of a cube (ie: 6 square images with both horizontal and vertical field of view set to 90 degrees). These 6 images are compiled into a horizontal or vertical strip. The order of the faces in this strip should be images 'front', 'right', 'back', 'left', 'top', 'bottom' (either left to right or top to bottom).

• Cylindrical images

  The type of image you get from simple stitching applications such as Quicktime VR Authoring Studio or PhotoVista. Cylindrical images can not represent the full 180 degrees of vertical field of view. For cylindrical images with a high vertical field of view, it may be more efficient to convert the image to spherical format using eg PanoTools.

• Spherical images

  Also known as PSphere or Equirectangular.

Due to the nature of the SPi-V engine, cubic panorama format is recommended for full (360 * 180 degrees) panoramas as it gives the best performance. Spherical panoramas are displayed in a slightly lower quality to prevent graphical artifacts near the zenith and nadir; by default, filtering is turned off for spherical panoelements (see the 'antialias' attribute).

Sound

SPi-V offers support for (streaming) MP3, either mono or stereo. Playback can be looped, but looping may not be 100% 'gapless'. Due to limitations of the mp3 format, looping audio without a small/tiny period of silence is not possible.