Floating point can represent numbers smaller than 1e-16, but cannot represent relative differences of this size such as 1 - 1e-16. Such differences are likely to occur for 'skew' matrices, since vml is likely to internally compute values such as b * x + f * y where x and y may be 1 for example. That should just work though, as it should be possible to use a skew matrix with b=1 and f=0 or vice versa.
On 16 Jun 2010 17:35, "William" <[email protected]> wrote: On Jun 16, 4:57 pm, Ben Appleton <[email protected]> wrote: > > What was the value of e? > e = 0.317, so this makes (e*f) less than 1e-16 (and less than 2^-53 which is the limit for double precision mantissa) and maybe this is zero if just the mantissa is shown? I understand what you are saying now, underflow is when a small number gets rounded to zero. -- You received this message because you are subscribed to the Google Groups "Google Maps JavaScript AP... -- You received this message because you are subscribed to the Google Groups "Google Maps JavaScript API v3" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/google-maps-js-api-v3?hl=en.
