![\"kega\"](\"https:\/\/www.emu-france.com\/wp-content\/uploads\/2015\/10\/kega.png\")
<\/a><\/p>\n<\/p>\n
Les changements:<\/p>\n
<\/p>\n
*maj2*<\/p>\n
Les *.rpi RLUT ont \u00e9t\u00e9 modifi\u00e9s de fa\u00e7on \u00e0 voir le nom R-LUT (reverse color – LUT) apparaitre dans Kega pour plus de clart\u00e9, aucun autre changement.<\/p>\n
<\/p>\n
*maj*:<\/p>\n
Ajouts des versions 1.1 + RLUT:<\/p>\n
– further speedups due to more code optimizations (see the source for details)<\/em> <\/p>\n The dir \u00ab\u00a0RLUT\u00a0\u00bb contains the same six versions, but compiled with a reverse color-LUT.<\/em> <\/p>\n Ajouts des filtres v1.0 de Zenju (modifi\u00e9s par milo1012): <\/p>\n <\/p>\n T\u00e9l\u00e9charger Kega Fusion Filters Plugins (09\/10\/2015) (1,1 Mo)<\/a><\/p>\n
\n\u00a0 -> around 15 % faster on most machines<\/em>
\n– removal of conversion buffers (were ~10 MiB for the 4x version)<\/em>
\n\u00a0 and now using a color lookup-table instead (< 1 MiB memory)<\/em>
\n– output is now copied line-wise directly after computation<\/em>
\n\u00a0 -> seems to fix threaded version for VBA-M<\/em>
\n– added reverse color-LUT variant for an additional ~5 % speed increase,<\/em>
\n\u00a0 but requires 64 MiB memory for each plugin, therefore optional<\/em>
\n– compiled with TDM-GCC (slight speed increase and now requires i686 compatible CPU)<\/em><\/p>\n
\nThis provides an additional ~5 % speed increase.<\/em>
\n(at least on the \u00ab\u00a0modern\u00a0\u00bb systems I tested, may be less for older systems)<\/em>
\nThe drawback is that this requires 64 MiB memory for each plugin version loaded,<\/em>
\nwhich would be 6*64 = 384 MiB when all six plugin versions are loaded by Kega!<\/em>
\n(not sure about VBA-M)<\/em>
\nI recommend to use just your \u00ab\u00a0main\u00a0\u00bb plugin version in the RLUT-variant,<\/em>
\nfor a reasonable memory footprint or for low-mem systems, and the remaining<\/em>
\n(if you still want or need them in that case) in the normal variant.<\/em><\/p>\n
\n– 4xBRZ.rpi\u00a0\u00a0\u00a0\u00a0\u00a0 4xBRZ – normal version – scales to e.g. 1280×960 (Mega Drive – PAL)<\/em>
\n– 4xBRZ-MT.rpi\u00a0\u00a0 4xBRZ – (multi-)threaded version<\/em>
\n– 3xBRZ.rpi\u00a0\u00a0\u00a0\u00a0\u00a0 3xBRZ – normal version – scales to e.g. 960×720<\/em>
\n– 3xBRZ-MT.rpi\u00a0\u00a0 3xBRZ – (multi-)threaded version<\/em>
\n– 2xBRZ.rpi\u00a0\u00a0\u00a0\u00a0\u00a0 2xBRZ – normal version – scales to e.g. 640×480<\/em>
\n– 2xBRZ-MT.rpi\u00a0\u00a0 2xBRZ – (multi-)threaded version<\/em><\/p>\nI’ve tested the threaded version quite thorough,<\/em> but I can’t guarantee that it’s completely bug-free.<\/em><\/del>
\n Using the plugin for VBA-M shows the threaded picture set-up, but that’s the fault of<\/em> VBA-M since it doesn’t wait until the function returns before displaying the frame.<\/em><\/del>\u00a0 ok avec <\/em>1.1<\/em> (\u00e0 priori !<\/em>)<\/em>
\n<\/em><\/del><\/span><\/p>\n