by randall » Jul 12, 2001 @ 10:24pm
There isn't a mysterious "big map error" phenomenon. The error could have happened for a number of reasons, none of which I can point out at this moment.<br><br>Here is the way I would do it:<br><br>Create an area, fine tune it, optimise and then freeze it. (meaning you don't work on it anymore because it is PERFECT).<br><br>Create a NEW area, fine tune it, optimise it, then freeze it. Glue it to the first part and ensure it compiles without any errors. Test it, and FREEZE those two sections.<br><br>Move on to the next area, fine tune it, optimise it, then freeze it. Glue it to the first and second parts... I think I get the idea now.<br><br>This will ensure your map will be as large as possible, with all bugs worked out along the way. Also, the individual compile times of the small maps will be much more managable. If you do it right, the r_speeds will be much more managable too.<br><br>By using your method, a single large area, you may get halfway into the map and then discover the compilers can't handle any more. Then you'll have a single large map, with huge areas that are incomplete. And the compile time keeps increasing, which makes it difficult to make changes to decrease the r_speeds.