> When I restart an old project, it will start from
> the begining to re-download all the pages again,
> which wastes time, how can I only download the pages
> that not exist on the local folders instead of
> download the exist pages again and again everytime I
> restart the project.

There is no way to know about new pages, unless they are linked to the old
pages (which have changed and are no longer old.)

When you Update a project, it asks the server has this file changed and only
downloads those that have. Except for dynamic html pages which are always
reported as new, checking for updates is fast. I get about 3/sec (dependent on
the round trip time to server,) size of the files are irrelevant.