1*2f083884Ss.makeev_local1.10 2*2f083884Ss.makeev_local* Iterative cluster fit is now considered to be a new compression mode 3*2f083884Ss.makeev_local* The core cluster fit is now 4x faster using contributions by Ignacio 4*2f083884Ss.makeev_localCastano from NVIDIA 5*2f083884Ss.makeev_local* The single colour lookup table has been halved by exploiting symmetry 6*2f083884Ss.makeev_local 7*2f083884Ss.makeev_local1.9 8*2f083884Ss.makeev_local* Added contributed SSE1 truncate implementation 9*2f083884Ss.makeev_local* Changed use of SQUISH_USE_SSE to be 1 for SSE and 2 for SSE2 instructions 10*2f083884Ss.makeev_local* Cluster fit is now iterative to further reduce image error 11*2f083884Ss.makeev_local 12*2f083884Ss.makeev_local1.8 13*2f083884Ss.makeev_local* Switched from using floor to trunc for much better SSE performance (again) 14*2f083884Ss.makeev_local* Xcode build now expects libpng in /usr/local for extra/squishpng 15*2f083884Ss.makeev_local 16*2f083884Ss.makeev_local1.7 17*2f083884Ss.makeev_local* Fixed floating-point equality issue in clusterfit sort (x86 affected only) 18*2f083884Ss.makeev_local* Implemented proper SSE(2) floor function for 50% speedup on SSE builds 19*2f083884Ss.makeev_local* The range fit implementation now uses the correct colour metric 20*2f083884Ss.makeev_local 21*2f083884Ss.makeev_local1.6 22*2f083884Ss.makeev_local* Fixed bug in CompressImage where masked pixels were not skipped over 23*2f083884Ss.makeev_local* DXT3 and DXT5 alpha compression now properly use the mask to ignore pixels 24*2f083884Ss.makeev_local* Fixed major DXT1 bug that can generate unexpected transparent pixels 25*2f083884Ss.makeev_local 26*2f083884Ss.makeev_local1.5 27*2f083884Ss.makeev_local* Added CompressMasked function to handle incomplete DXT blocks more cleanly 28*2f083884Ss.makeev_local* Added kWeightColourByAlpha flag for better quality images when alpha blending 29*2f083884Ss.makeev_local 30*2f083884Ss.makeev_local1.4 31*2f083884Ss.makeev_local* Fixed stack overflow in rangefit 32*2f083884Ss.makeev_local 33*2f083884Ss.makeev_local1.3 34*2f083884Ss.makeev_local* Worked around SSE floor implementation bug, proper fix needed! 35*2f083884Ss.makeev_local* This release has visual studio and makefile builds that work 36*2f083884Ss.makeev_local 37*2f083884Ss.makeev_local1.2 38*2f083884Ss.makeev_local* Added provably optimal single colour compressor 39*2f083884Ss.makeev_local* Added extra/squishgen.cpp that generates single colour lookup tables 40*2f083884Ss.makeev_local 41*2f083884Ss.makeev_local1.1 42*2f083884Ss.makeev_local* Fixed a DXT1 colour output bug 43*2f083884Ss.makeev_local* Changed argument order for Decompress function to match Compress 44*2f083884Ss.makeev_local* Added GetStorageRequirements function 45*2f083884Ss.makeev_local* Added CompressImage function 46*2f083884Ss.makeev_local* Added DecompressImage function 47*2f083884Ss.makeev_local* Moved squishtool.cpp to extra/squishpng.cpp 48*2f083884Ss.makeev_local* Added extra/squishtest.cpp 49*2f083884Ss.makeev_local 50*2f083884Ss.makeev_local1.0 51*2f083884Ss.makeev_local* Initial release 52*2f083884Ss.makeev_local 53