Hi all,

   I also meant to mention that moving the primary (large) offset into a 
transform will
also help a lot (not so many of your bits will be 'wasted' on essentially 
unchanging
data):

<path d="M489409.9,-4379209.97 [...] 

becomes:

<path transform="translate(489409.9,-4379209.97)" d="M 0,0 [...]



svg-developers@yahoogroups.com wrote on 05/10/2006 08:54:13 PM:

> Hi Bruce, Andreas,
> 
> svg-developers@yahoogroups.com wrote on 05/10/2006 05:57:28 PM:
> 
> > I agree the issue is coordinate accuracy but I am curious why the bug
> > only shows up in the relative format.  The starting coordinate is the
> > same.  Maybe the viewers parse relative coordinates differently.
> 
>    The problem is most likely that at the limit of precision you
> can get things like:
> 
>         x == (x+1)
> 
>    Thus if you add 1 ten times in a row you end up with the same x:
>         x' =x;
>       for (i=0; i<10; i++)
>            x' = x'+1;
>         x == x'
> 
>    However it is very likely that if you were to
> actually add 10 to x you would get a different number:
>         x != (x+10)
> 
>    So you would sort of see something like:
>         x == (x+1)
>         x == (x+2)
>         x == (x+3)
>         x != (x+4);  x' = x+4;  // this is a new 'representable value'
>         x' == (x+5)
>         x' == (x+6)
>         x' != (x+7); x'' == x+7  // another new 'representable value'
>         x'' == (x+8)
>         x'' == (x+9)
>         x'' == (x+10) 
> 
>    This is exactly the difference between absolute and relative
> coords.  With relative coords you start with a _really_large_ number
> and then add lots of relatively small numbers - meaning that the
> 'round off' is large in percentage terms of the small numbers.  So
> meaningful error can easily accumulate over many additions.
> 
>    However with absolute coordinates you are always giving it the
> whole (essentially infinite precision) number, which it then rounds
> to the nearest representable value - there opportunity for error to
> accumulate.
> 
>    You could compensate for this by occasionally inserting an
> absolute coordinate set (essentially resetting the error accumulation).
> 
> > --- In svg-developers@yahoogroups.com, "Andreas Neumann" <[EMAIL PROTECTED]>
> > wrote:
> > >
> > > hm, I'd be interested to know whats the problem here, as well.
> > > 
> > > I tested in other SVG viewers. Batik has the offset like ASV and
> > Firefox, Apples Safari 
> > > renders it correctly like Opera. Firefox 3 renders it the same way
> > than Firefox 1.5.
> > > 
> > > My guess is that some viewers have problems with coordinate accuracy
> > at certain viewBox 
> > > / coordinate range combinations, but I can't tell where the limits
> > are. I know that Firefox 
> > > usually has more problems with coordinate ranges than other SVG 
> viewers.
> > > 
> > > Any one knows more about valid coordinate ranges in SVG viewers?
> > > 
> > > Andreas
> > > 
> > > --- In svg-developers@yahoogroups.com, "brucerindahl" <rindahl@> 
> wrote:
> > > >
> > > > I have the following small portion of a GIS SVG mapping 
application. 
> 
> > > > The file shows a portion of a floodplain area in cyan using 
absolute
> > > > coordinates.  The red outline should be the same shape using 
> relative
> > > > coordinates.  The file displays perfectly in Opera 9 but is wrong 
in
> > > > IE/ASV and Firefox.  Any ideas on why this is happening??
> > > > 
> > > > http://www.lrcwe-data.com/tmp/relative.svg
> > > > 
> > > > Thanks
> > > > Bruce Rindahl
> > > >
> > >
> > 
> > 
> > 
> > 
> > 
> > 
> > 
> > -----
> > 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 unsubscribe send a message to: 
[EMAIL PROTECTED]
> -or-
> visit http://groups.yahoo.com/group/svg-developers and click "edit my 
membership"
> ---- 
> Yahoo! Groups Links
> 
> 
> 
> 
> 
> 
> 



------------------------ Yahoo! Groups Sponsor --------------------~--> 
Get to your groups with one click. Know instantly when new email arrives
http://us.click.yahoo.com/.7bhrC/MGxNAA/yQLSAA/1U_rlB/TM
--------------------------------------------------------------------~-> 

-----
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/

<*> 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/
 



Reply via email to