Sandra 2012 s’offre une nouvelle mise à jour (18.47)
Une quinzaine de jours à peine après la version 18.45, l’éditeur SiSoftware annonce la sortie de la version 18.47 de SANDRA 2012. Par l’intermédiaire de ce nouvel opus, le programme s’offre une mise à jour mineure qui vise à corriger quelques bugs et à apporter plusieurs modifications principalement au niveau des benchs. Vous trouverez le changelog complet ci-dessous.
– Télécharger Sandra 2012.06.18.47
Voilà le changelog complet :
– Thread Scheduler: You probably know that Sandra has always used its own scheduler. This has allowed Sandra to properly schedule threads before Windows’ kernel caught-up, e.g. Windows 7/2008 R2 with Hyper-Threading and Windows 8/2012 with Bulldozer.
– Benchmark: Multi-Media: enabled FMA4 Multi-Media code-path, thus improving Bulldozer performance by over 20% (versus AVX). FMA3 is also supported for Piledriver (and Haswell) which may be a bit faster still.
– Benchmark: Crypto: improved SHA256/SHA1 Bulldozer bandwidth by 50% by rolling back one SNB AVX optimisation. SNB/IVB scores are not significantly affected by this change. Similarly for AVX2, though nothing supports it yet.
– Benchmarks: Memory/Cache Bandwidth, Memory/Cache Latency: enabled large-pages (2MB) on Bulldozer by reading 2MB/TLB correctly. Assuming you granted yourself « lock pages in memory », using 2MB pages improves both performance and reliability by minimising TLB misses when using large memory blocks.
– Benchmarks: Memory Bandwidth, Cache Bandwidth: improved both FMA4 and FMA3 code-paths for better STREAM/Triad performance (versus AVX).
– Benchmark: Memory/Cache Latency: 32-bit/x86: Rolled back assembler > intrinsic change as the x86 compiler decided to generate non-optimal loop for the latency test. x64 compiler worked just fine.
– Fixes: Turbo multiplier detection on Bulldozer where P-States are not as expected (e.g. when overclocked manually or BIOS setting them up incorrectly).
Désolé, les commentaires sont fermés.