On Mon, Mar 8, 2010 at 4:10 PM, bratliff <[email protected]> wrote: > A quick follow up. Two minutes is really not enough. Each zoom > change or map type change can trigger a couple dozen tile requests. > The browser will limit the number in flight at one time. The API > cannot control the time at which a request is dispatched once it is > submitted. A batch of twenty to thirty tiles could easily exceed two > minutes. It is OK to prioritize tiles if a preempted tile can be > fetched later.
We now queue image requests in JS, only fetching at most 12 images at a time. This allows to cancel requests if a tile falls out of view before the request has been forwarded to the browser, which reduces latency for a quick sequence of zooms. To recover from stalled tiles we abandon a tile fetch after 2 minutes. This can still fail in a few ways: - The bandwidth could be so low that loading 12 tiles x 25kB = 300kB takes more than 2 minutes. - The internet connection could drop out temporarily. - The page could be making many other HTTP requests, blocking our tile requests. I appreciate your feedback; we haven't finished optimizing tile loading yet. The API does not reload a discarded tile if it returns > to view. Once it is considered missing, it is lost forever. > > > -- > 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]<google-maps-js-api-v3%[email protected]> > . > For more options, visit this group at > http://groups.google.com/group/google-maps-js-api-v3?hl=en. > > -- 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.
