summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)Author
2014-01-13(svn r26255) -Codechange: improve performance of brightness adjustment (MJP)rubidium
2014-01-13(svn r26254) -Codechange: MOVD is better in some cases than PINSRD (MJP)rubidium
2014-01-13(svn r26253) -Codechange: remove margin limitation to preserve block type (MJP)rubidium
2014-01-13(svn r26252) -Codechange: deduplicate code for margin case (MJP)rubidium
2014-01-13(svn r26251) -Codechange: prepare Draw code for code deduplication (MJP)rubidium
2014-01-13(svn r26250) -Codechange: deduplicate darkening (e.g. shadow) code (MJP)rubidium
2014-01-13(svn r26249) -Codechange: simplify/cleanup ALPHA_BLEND macro (MJP)rubidium
2014-01-13(svn r26248) -Fix (r26223): missed compiler warning (MJP)rubidium
2014-01-13(svn r26247) -Fix [FS#5854, FS#5855]: Possible out of bounds reads with the ↵rubidium
sse blitters (MJP)
2014-01-12(svn r26246) -Update from WebTranslator v3.0:translators
danish - 4 changes by Hatsen
2014-01-12(svn r26245) -Codechange: No need for a map, if a vector suffices.frosch
2014-01-12(svn r26244) -Feature: [NewGRF] Add StringCodes 9A 1B, 9A 1C and 9A 1D to ↵frosch
display amounts of cargo.
2014-01-12(svn r26243) -Cleanup: Move TTDPStringIDToOTTDStringIDMapping to newgrf.cpp, ↵frosch
it's only used there.
2014-01-12(svn r26242) -Codechange: Use GRFFile that filled the TextRefStack to ↵frosch
translate StringIDs from the TextRefStack, instead of passing stuff via global vars.
2014-01-12(svn r26241) -Codechange: Remember the GRFFile which filled the TextRefStack ↵frosch
in the TextRefStack.
2014-01-12(svn r26240) -Codechange: Pass the GRFFile to ↵frosch
GetErrorMessageFromLocationCallbackResult instead of the GRFID.
2014-01-12(svn r26239) -Fix: Check that there is space left in the string parameter ↵frosch
array, before pushing NewGRF parameters to it.
2014-01-12(svn r26238) -Codechange: Use StringParameters::GetDataLeft to check for ↵frosch
left space in the param array.
2014-01-12(svn r26237) -Cleanup: Remove unused SCC_STRING_ID.frosch
2014-01-12(svn r26236) -Fix: Don't overreserve after autorefit, but do reserve mail ↵fonsinchen
for aircraft
2014-01-11(svn r26235) -Fix: Decimal and digit separators were swapped for Korean languageplanetmaker
2014-01-10(svn r26234) -Update from WebTranslator v3.0:translators
bulgarian - 69 changes by kokobongo
2014-01-09(svn r26233) -Update from WebTranslator v3.0:translators
swedish - 2 changes by spacejens
2014-01-08(svn r26232) -Update from WebTranslator v3.0:translators
korean - 6 changes by telk5093 portuguese - 1 changes by frosch
2014-01-05(svn r26228) -Fix: missing variable initializationfonsinchen
2014-01-05(svn r26226) -Fix (r26212): Compilation in visual studio 2008 failedzuu
2014-01-04(svn r26225) -Document: more tidbits about cpuid and PIC (Eagle_Rainbow)rubidium
2014-01-03(svn r26224) -Update from WebTranslator v3.0:translators
brazilian_portuguese - 4 changes by Tucalipe
2014-01-03(svn r26223) -Codechange: Silence some compile warnings about unititialized ↵planetmaker
variables in the SSE blitters (MJP)
2014-01-03(svn r26222) -Codechange: use SSE instructions for transparent colour ↵rubidium
"remap" in the SSE2/SSSE3 blitters as well
2014-01-03(svn r26221) -Codechange: normalize constant names in anim-sse4 with the ↵rubidium
others (MJP)
2014-01-03(svn r26220) -Fix?: another attempt at getting cpuid to finally work ↵rubidium
"everywhere"
2014-01-03(svn r26219) -Fix: cpuid code still didn't work right in some situationsrubidium
2014-01-03(svn r26217) -Change: blitter autoselection based on full animation state, ↵rubidium
so the non-animated optimizations can actually be used in an easy manner (based on patch by MJP)
2014-01-03(svn r26216) -Change: the blitter selection to also consider the ↵rubidium
32bpp-sse4-anim blitter (MJP)
2014-01-03(svn r26215) -Codechange: rework code so one can test if a blitter factory ↵rubidium
exists before attempting trying to instantiate an instance
2014-01-02(svn r26214) -Add: specialised animated SSE4 blitter (MJP)rubidium
With 32bpp base set about 15-20% faster in the Draw function (slower with 8bpp base set). Overall, with 32bpp base set, about 5% faster.
2014-01-02(svn r26213) -Add: specialised non-animated SSE4 blitter (MJP)rubidium
With 32bpp base set about 40% faster than 32bpp-optimized, or about 10% for 8bpp base sets in the Draw function. Respectively about 8 and 1% of total run time
2014-01-02(svn r26212) -Add: specialised non-animated SSSE3 blitter (MJP)rubidium
With 32bpp base set about 35% faster than 32bpp-optimized, or about 10% for 8bpp base sets in the Draw function. Respectively about 6 and 1% of total run time
2014-01-02(svn r26211) -Add: specialised non-animated SS2 blitter (MJP)rubidium
With 32bpp base set about 30% faster than 32bpp-optimized, or about 10% for 8bpp base sets in the Draw function. Respectively about 5 and 1% of total run time
2014-01-02(svn r26210) -Codechange: add infrastructure for not registering a blitterrubidium
2014-01-02(svn r26209) -Codechange: remove some template magic and simplify some coderubidium
2014-01-02(svn r26208) -Fix (r26197): the cpuid assembly didn't work when PIC was enabledrubidium
2014-01-02(svn r26207) -Codechange: move the CPUID flag detection into cpu.cpprubidium
2014-01-02(svn r26206) -Fix [FS#5829]: Run everything from ini, obg, obs, obs, ... ↵frosch
files through str_validate.
2014-01-02(svn r26205) -Feature: SSE 4.1 sprite sorter, improving the sorting ↵rubidium
performance significantly (MJP) For example with GCC 4.8, x86_64 Linux, Intel i5-3337U this patch improves the performance of Pile, Treham and Hamac test save games by about 10% in over-all run time at fast forward at 1920x1080 when zoomed out and when trees are not disabled.
2014-01-02(svn r26204) -Codechange: rework GfxBlitter to remove some unneeded ↵rubidium
operations when there is no sub sprite (based on patch by MJP)
2014-01-02(svn r26203) -Codechange: unduplicate GfxMainBlitter and GfxMainBlitterViewportrubidium
2014-01-02(svn r26202) -Cleanup: since ZOOM_LEVEL_NORMAL 0 and there aren't any ↵rubidium
negative zoom levels, there is no reason to account for negative zoom levels in the zoom functions (based on patch by MJP)
2014-01-02(svn r26201) -Fix: unscaling by a negative value is the same as scaling by a ↵rubidium
positive number