How would you be able to represent anything close to a JPG as a vector graphic? I can understand if it is an image with areas of solid color like a logo (in which case, why would it be a jpeg in the first place?) but how is it anywhere near possible to represent a truecolor, hi-res photo as a vector graphic that wouldn't be at least 100 times the size of a JPG? The only methods I've seen that do this are more lossy than the JPG compression itself *and* they are at least 10-20 times as large as a JPG and browsers can't even process them fast enough for them to be useful. Please help me understand what it is you're talking about.
On Fri, Nov 16, 2012 at 10:53 PM, Joe Doll <[email protected]>wrote: > ** > > > I don't see it that way. PNG only lags JPG compression by about 3 times > at moderate to good quality. SVG lags JPG by about 1.5 times if the > image has good color segmentation. When closely matching shapes are > group together (aka shape grouping), SVG can use the <def> tag to exceed > JPG compression. JPG would then lag SVG by about 2 times (but I haven't > fully tested it). When entropic shaping (e.g. making a shape that is > nearly a circle a circle) reduces noise even further, JPG lags SVG by 10 > times. After entropic shaping, almost all of the signal is extracted > from the noise (aka very little noise left), then the data is in > position for raster to vector conversion. Yes, SVG is vector, but the > last step provides for something more than a set of lines that are stuck > together. In other words, SVG has continuous mathematical functions. In > our tests, it seems that JPG lags this type of SVG by 100 times. > > I do agree that its not about trying to replace other technologies, but > some technologies are not positioned to have a future. The whole raw > signal can be stored in PNG, so I see PNG as a permanent technology. I > do not see what role either JPG or PDF will play when SVG is advanced to > noise-free continuous mathematical function. > > SVG has a permanent future on the Internet, but more than that, the > ability to provide high quality data to electronic circuits will improve > the performance of any electronic circuit that currently relies on > frequency based signal conditioning which is nearly all circuits. A > circuit that relies on noise free continuous mathematics for its data, > will be about 100 times more capable than the way we design electronics > today (frequency based signal conditioning). This is the inverse of the > old computer rule, "Garbage in, garbage out". > > [Non-text portions of this message have been removed] ------------------------------------ ----- To unsubscribe send a message to: [email protected] -or- visit http://groups.yahoo.com/group/svg-developers and click "edit my membership" ----Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/svg-developers/ <*> Your email settings: Individual Email | Traditional <*> To change settings online go to: http://groups.yahoo.com/group/svg-developers/join (Yahoo! ID required) <*> To change settings via email: [email protected] [email protected] <*> To unsubscribe from this group, send an email to: [email protected] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
