ソースフォージに CUDA-Z が GPU の性能テストを行うユーティリティであると見かけましたので、さっそく導入してみました。

[amazon_enhanced asin=”B004X31AW” container=”検証に使用した ELSA (NVIDIA) GTX 560 Ti OC ” container_class=”” price=”All” background_color=”FFFFFF” link_color=”000000″ text_color=”0000FF” /]

＝＝＝

■システム要件：

オペレーティングシステム: MacOSX, Linux, Windows 7, Windows Vista, Windows XP

—

■入手先：ソースフォージ

http://sourceforge.jp/projects/sfnet_cuda-z/releases/

—

■ Windows 7 x64 環境で実施例：テキスト出力

※ほかにもバッググラウンドで幾つかのアプリが動作しているので（MSSQL もｗ）、結果は芳しくありませんでした。

CUDA-Z Report
=============
Version: 0.6.133 SVN Built Jun 24 2010 10:38:20
http://cuda-z.sourceforge.net/
OS Version: Windows AMD64 6.1.7601 Service Pack 1
Driver Version: 275.33
Driver Dll Version: 4.0 (8.17.12.7533)
Runtime Dll Version: 3.0 (6,14,11,3000)
Core Information
----------------	Name: GeForce GTX 560 Ti	Compute Capability: 2.1	Clock Rate: 1800 MHz	Multiprocessors: 8	Warp Size: 32	Regs Per Block: 32768	Threads Per Block: 1024	Threads Dimensions: 1024 x 1024 x 64	Grid Dimensions: 65535 x 65535 x 65535	Watchdog Enabled: Yes	Integrated GPU: No	Concurrent Kernels: Yes	Compute Mode: Default
Memory Information
------------------	Total Global: 961.562 MiB	Shared Per Block: 48 KiB	Pitch: 2048 MiB	Total Constant: 64 KiB	Texture Alignment: 512 B	Texture 1D Size: 65536	Texture 2D Size: 65536 x 65535	Texture 3D Size: 2048 x 2048 x 2048	GPU Overlap: Yes	Map Host Memory: Yes	Error Correction: No
Performance Information
-----------------------
Memory Copy	Host Pinned to Device: 164.404 MiB/s	Host Pageable to Device: 156.472 MiB/s	Device to Host Pinned: 164.396 MiB/s	Device to Host Pageable: 156.318 MiB/s	Device to Device: 38.7264 GiB/s
GPU Core Performance	Single-precision Float: 914.648 Gflop/s	Double-precision Float: 115.108 Gflop/s	32-bit Integer: 459.414 Giop/s	24-bit Integer: 456.371 Giop/s
Generated: Sun Sep 11 05:14:29 2011

---

■CUDA-Z HTML出力

CUDA-Z Report

Version: 0.6.133 SVN Built Jun 24 2010 10:38:20 http://cuda-z.sourceforge.net/

OS Version: Windows AMD64 6.1.7601 Service Pack 1
Driver Version: 275.33

Driver Dll Version: 4.0 (8.17.12.7533)

Runtime Dll Version: 3.0 (6,14,11,3000)

Core Information

Name	GeForce GTX 560 Ti
Compute Capability	2.1
Clock Rate	1800 MHz
Multiprocessors	8
Warp Size	32
Regs Per Block	32768
Threads Per Block	1024
Threads Dimensions	1024 x 1024 x 64
Grid Dimensions	65535 x 65535 x 65535
Watchdog Enabled	Yes
Integrated GPU	No
Concurrent Kernels	Yes
Compute Mode	Default

Memory Information

Total Global	961.562 MiB
Shared Per Block	48 KiB
Pitch	2048 MiB
Total Constant	64 KiB
Texture Alignment	512 B
Texture 1D Size	65536
Texture 2D Size	65536 x 65535
Texture 3D Size	2048 x 2048 x 2048
GPU Overlap	Yes
Map Host Memory	Yes
Error Correction	No

Performance Information

Memory Copy
Host Pinned to Device	164.25 MiB/s
Host Pageable to Device	157.271 MiB/s
Device to Host Pinned	164.329 MiB/s
Device to Host Pageable	156.718 MiB/s
Device to Device	38.7502 GiB/s
GPU Core Performance
Single-precision Float	914.623 Gflop/s
Double-precision Float	115.113 Gflop/s
32-bit Integer	459.4 Giop/s
24-bit Integer	454.718 Giop/s