Hi Folks,
In my humble opinion, as the number of extremely long levels
grows, it will take more and more time for the highscore
maintainer (=Donald) to handle split playbacks and follow debates
about the topic.
I looked into the source, and saw that there are
five or six places where code must be changed to get rid
of the 65500 limit. One needs to change WORD to DWORD at several
places. The only speciality is that we need to handle
current playbacks and extended playbacks.
When saving:
if the playback buffer is >= 65535, then write 65535 as size
in the header, and size on extra 4 bytes after the header.
When loading:
if the size in the header is 65535, then read size from
the extra 4 bytes.
In the high score file:
This is not so obvious, as we have 10 bytes pro level.
I'd say for initials we use 4 bytes, the trailing 2 zeros
could be used as high-order word for the number of moves.
Then for shots, we'll still have the limit of 65535.
Has anyone a better idea?
I don't think it takes much time to compile
a version 4.1 with these corrections.
regards.M.
In my humble opinion, as the number of extremely long levels
grows, it will take more and more time for the highscore
maintainer (=Donald) to handle split playbacks and follow debates
about the topic.
I looked into the source, and saw that there are
five or six places where code must be changed to get rid
of the 65500 limit. One needs to change WORD to DWORD at several
places. The only speciality is that we need to handle
current playbacks and extended playbacks.
When saving:
if the playback buffer is >= 65535, then write 65535 as size
in the header, and size on extra 4 bytes after the header.
When loading:
if the size in the header is 65535, then read size from
the extra 4 bytes.
In the high score file:
This is not so obvious, as we have 10 bytes pro level.
I'd say for initials we use 4 bytes, the trailing 2 zeros
could be used as high-order word for the number of moves.
Then for shots, we'll still have the limit of 65535.
Has anyone a better idea?
I don't think it takes much time to compile
a version 4.1 with these corrections.
regards.M.