29 Jan
2006
29 Jan
'06
11:48 p.m.
In article <200601300623.BAA03447 at Sparkle.Rodents.Montreal.QC.CA>,
der Mouse <mouse at rodents.montreal.qc.ca> writes:
The compression algorithm alone does not determine the overall
compression on the data. PNG rearranges the data for more efficient
application of the compression algorithm. GIF just lumps together
8-bit bytes for compression, avoiding any spatial coherence that could
be exploited in the images. In other words, GIF treats the image data
as unstructured data, but PNG exploits the spatial coherence of image
data to achieve better overall compression.
Probably not needed for manual scans, unless there are both color
content and continuous tone grayscale or color images in the pages.
For black text on white background, either works.
However, there is another advantage that PNG has over GIF that Jim
didn't mention: PNG is not burdened by any patented IP.
--
"The Direct3D Graphics Pipeline"-- code samples, sample chapter, FAQ:
<http://www.xmission.com/~legalize/book/>
Pilgrimage: Utah's annual demoparty
<http://pilgrimage.scene.org>
> I
didn't see anything in the original request that called for
> anything PNG can do but GIF can't,
Reasons to use PNG over GIF:
- Better compression with faster decompression
speeds (PNG uses LZ77,
GIF uses LZ78)
You need to go reread the spec. GIF uses LZW, Lempel-Ziv-Welch, a
significant extension to LZ. - Color depths
*above* 8-bit (ie. 24-bit color)
- Alpha channel (256 levels of transparancy)
Neither of these appeared to matter from what I could see. If either
is necessary, PNG may indeed be called for.