[ReneG]

Quote:

Originally Posted by wikianswerer Having a couple of hundred thousands rows is perfectly fine, as long as : they are indexed properly and your queries are done properly (i.e. using the right indexes, for instance)

and

Quote:

Originally Posted by wikianswerer I maintain a 90GB reporting database (for web server logs) which has several tables with 100s of millions of rows the biggest is 318m. I can get results from a standard select query with a join here and there (under moderate load) in 10 - 50ms.

Not sure about the comparison of BlueG's MySQL plugin to what webservers use, but it was made to handle tons of data.

Quote:

Originally Posted by YJIET IMO saving objects to a file is more suitable for this kind of use. In my object editor, I save objects to a file in this format: pawn Код: //model, pos x, pos y, pos z, rot x, rot y, rot z, ... 496C|391E27F8|42E5EBE1|4439C3D1|368483F1|43A8C000|42B3FFEE 496C|397FB81A|43197BA4|443C586D|36C3EF17|43B04000|42B3FFED 4964|39B4E559|434169F9|443CAC16|3705A89D|37000000|42B3FFE8 Integers and floats are stored as hex values, which can then be easily exploded with sscanf: pawn Код: new     val[7]; sscanf(str, "p<|>a<x>[7]", val); CreateObject(val[0], Float:val[1], Float:val[2], Float:val[3], Float:val[4], Float:val[5], Float:val[6]); And that took me less than one second to load 900 objects (1.8Ghz single core processor).

Props.