home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!rosie!aozer
- From: aozer@next.com (Ali Ozer)
- Newsgroups: comp.sys.next.programmer
- Subject: Re: NeXT TIFF question
- Message-ID: <4481@rosie.NeXT.COM>
- Date: 26 Jul 92 01:46:23 GMT
- References: <17124@acorn.co.uk>
- Sender: news@NeXT.COM
- Organization: Next Computer, Inc.
- Lines: 19
-
- In article <17124@acorn.co.uk> David De Vorchik writes:
- >We are currently trying to decode some TIFF files that originated on a NeXT
- >box. These icons have a strange samples per pixel value (0x20002), we
- >assume this is something to do with masking of the image (as some of them
- >are icons).
-
- Bizarre. NeXTSTEP never generates TIFFs with that spp; spp is always 1, 2, 3,
- or 4. In fact, given that the SamplesPerPixel tag is a short, the value
- you specify doesn't look right. If it were to be 2 (which seems likely), then
- you probably have a grayscale image with alpha.
-
- Images with alpha aren't too complicated; they just have an extra channel
- holding the transparency information, and the actual image itself is
- premultipled with alpha. The alpha is stored as the last channel of data.
-
- NeXTSTEP 2.x denotes alpha with the private MATTE tag; NeXTSTEP 3.0 uses
- the newly created standard ExtraSamples tag.
-
- Ali, Ali_Ozer@NeXT.com
-