Here you can download SiSoftware Sandra 2014 which includes remote analysis, benchmarking and diagnostic features for PCs, servers and networks.
SiSoft Sandra (the System ANalyser, Diagnostic and Reporting Assistant) is an information & diagnostic utility.
SANDRA will provide most of the information (including undocumented) you need to know about your hardware, software and other devices whether hardware or software.
It works along the lines of other Windows utilities, however it tries to go beyond them and show you more of what's really going on. Giving the user the ability to draw comparisons at both a high and low-level. You can get information about the CPU, chipset, video adapter, ports, printers, sound card, memory, network, Windows internals, AGP, ODBC Connections, USB2, 1394/Firewire, etc.
SiSoftware Sandra, the latest version of the award-winning utility, includes remote analysis, benchmarking and diagnostic features for PCs, servers, mobile devices and networks.
At SiSoftware from version to version we focus four key areas of software development to make sure we give our users the best software experience. Quite simply our four drivers are:
Inclusion of new technologies that ensure our tools remain current.
Increased hardware and software support for existing modules.
Improving the user interface to make it a richer easier user experience.
Filling gaps in the product line in response to our customers' needs.
* New CPU benchmark: Scientific Analysis
- 3 common algorithms: GEMM (matrix multiply), FFT (fast Fourier transform), N-Body simulation - 2 precision: single/float, double/FP64 - SSE2, SSE3, AVX, FMA support where possible - same workload as GP scientific analysis benchmark - to directly compare CPU vs. GPGPU performance.
* New GP benchmark: Scientific Analysis
- 3 common algorithms: GEMM (matrix multiply), FFT (fast Fourier transform), N-Body simulation - 2 precision: single/float, double/FP64 - 2 interfaces: CUDA, OpenCL (with DirectX ComputeShader to come) - not using any vendor specific library like cuBLAS, cuFFT, MKL, etc. - same workload as CPU scientific analysis benchmark - to directly compare CPU vs. GPGPU performance.
* nV "Maxwell" GPU preliminary support *
- Mainly detection for SMM 5.0 GPUs.
- Platform: Still based on CUDA SDK 5.5 RTM; cannot move to SDK 6.0 RC as Sandra would then require CUDA 6.0 video drivers for all cards - and those are thin on the ground.
- Issues: some OpenCL benchmarks may still not work; since we don't have a card ourselves, we've no idea what the problem is. Could be due to driver returning too high work-group sizes that fail.
- SHA hashing: check out the scores in "GPGPU Crypto" benchmarks, this thing does far better than "Kepler" (perhaps due to far more registers per SMM/SMX). Enough to beat AMD? Only you know )
- Memory latencies: "in-page random" in "GPGPU Memory Latency" benchmarks for both Global & Texture memories, TLBs seem to have improved a lot; also huge L2 reduces memory pressure.