SheerVideo consists of a generic codec as well as a number specific codecs.

By choosing the generic Sheer codec, you let QuickTime automatically choose the (hopefully) appropriate Sheer pixel type.

By choosing a specific Sheer codecs, you choose the Sheer pixel type yourself. The choice should be based on the source pixel format and the formats accepted by and used by the applications in your SheerVideo workflow. Certain pixel-format conversions are inherently destructive, so choosing an inappropriate Sheer codec will almost certainly degrade the image &mdash even with Synchromy. Inherently destructive conversions include

• reducing the sample precision, such as from 10-bit to 8-bit
• reducing the chroma sampling, such as from 4:4:4 to 4:2:2
• converting between RGB and Y'CbCr without adequately increasing the sample precision by at least 2 bits

The Sheer codecs are:

• Sheer lets QuickTime automatically choose the appropriate Sheer pixel type.
• Sheer RGB[A] 8bf for broadcast-quality scanned film and computer-generated imagery, of pixel format ARGB, or, on PC+Windows, ABGR, RGBA, BGRA, RGB, or BGR
• Sheer RGB[A] 10bf for studio-quality scanned film and computer-generated imagery, of pixel format 10-bit 'b64a' and 'L64A', 10-bit 'b48r', 'R10k', 'r210', 'F10k', 'F210', 'S10k', or 'S210'
• Sheer Y'CbCr[A] 8bv 4:4:4[:4] for compositing broadcast-quality digitized video, of pixel format 'v408', 'VUYA', or 'r408'
• Sheer Y'CbCr[A] 10bv 4:4:4[:4] for compositing studio-quality digitized video, of pixel format 'v410', 10-bit 'V416', or 10-bit 'r4fl'
• Sheer Y'CbCr[A] 8bv 4:2:2[:4] for cutting and archiving broadcast-quality digitized video, of pixel format '2vuy', '2Vuy', or 'yuvs'
• Sheer Y'CbCr[A] 10bv 4:2:2[:4] for cutting and archiving studio-quality digitized video, of pixel format 'v210', 10-bit 'v216', or 10-bit 'Y216'
• Sheer Y'CbCr 8bw 4:2:2 for cutting and archiving digitized video of obsolete pixel format 'yuvu' or 'yuv2'

RGB stands for {Red, Green, Blue}, the color representation used by the human eye, scanners, and displays.

[A] stands for an optional Alpha channel specifying opacity or coverage, used for compositing.

Y'CbCr stands for {luma (Y), Chroma blue-yellow, Chroma red-cyan}, the color representation used by television.

8b stands for 8 bits of information per color channel, the standard precision of digital displays.

10b stands for 10 bits of information per color channel, the standard precision of digitized film and computer-generated imagery.

f stands for full-range, with no headroom and no footroom.

u stands for unbiassed video-range, with headroom for filter overshoot but no footroom for undershoot, otherwise per ITU-R BT.601-4.

v stands for video-range, with footroom and headroom for filter under- and overshoot, per ITU-R BT.601-4.

w stands for wide-range, with no footroom or headroom but with one value (0x80 = 128) unused; obsolete.

4:4:4 means that for every 4 luma pixels, there are 4 chroma pixels (on both even and odd lines).

4:4:4:4 means that for every 4 luma and chroma pixels, there are 4 alpha pixels.

4:2:2 means that for every 4 neighboring luma pixels, there are only 2 chroma pixels (on both even and odd lines), as in television.

4:2:2:4 means that for every 4 luma pixels (and for every 2 chroma pixels), there are 4 alpha pixels.

Note that in some cases the application or QuickTime itself may degrade the image by converting the source to a different pixel format before passing it to an encoder. Unfortunately, QuickTime does not provide any mechanism for the encoder to determine the source format.