NVIDIA Tesla K20Xm versus NVIDIA GeForce GTX 690
Comparaison des cartes vidéo NVIDIA Tesla K20Xm and NVIDIA GeForce GTX 690 pour tous les caractéristiques connus dans les catégories suivants: Essentiel, Infos techniques, Sorties et ports de vidéo, Compatibilité, dimensions et exigences, Soutien API, Mémoire, Technologies. Analyse du performance de référence des cartes vidéo: PassMark - G3D Mark, PassMark - G2D Mark, Geekbench - OpenCL, CompuBench 1.5 Desktop - Face Detection (mPixels/s), CompuBench 1.5 Desktop - Ocean Surface Simulation (Frames/s), CompuBench 1.5 Desktop - T-Rex (Frames/s), CompuBench 1.5 Desktop - Video Composition (Frames/s), CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s), GFXBench 4.0 - Manhattan (Frames), GFXBench 4.0 - T-Rex (Frames), GFXBench 4.0 - Manhattan (Fps), GFXBench 4.0 - T-Rex (Fps), GFXBench 4.0 - Car Chase Offscreen (Frames), GFXBench 4.0 - Car Chase Offscreen (Fps), 3DMark Fire Strike - Graphics Score.
Différences
Raisons pour considerer le NVIDIA Tesla K20Xm
- La carte vidéo est plus nouvelle: date de sortie 6 mois plus tard
- Environ 28% consummation d’énergie moyen plus bas: 235 Watt versus 300 Watt
- Environ 50% plus de taille maximale de mémoire: 6 GB versus 4 GB (4 GB per GPU) GDDR5
- Environ 1% meilleur performance en Geekbench - OpenCL: 16664 versus 16577
- Environ 7% meilleur performance en CompuBench 1.5 Desktop - Face Detection (mPixels/s): 49.455 versus 46.352
- Environ 79% meilleur performance en CompuBench 1.5 Desktop - Ocean Surface Simulation (Frames/s): 1666.354 versus 930.114
- Environ 60% meilleur performance en CompuBench 1.5 Desktop - T-Rex (Frames/s): 5.836 versus 3.656
- 3.1x meilleur performance en CompuBench 1.5 Desktop - Video Composition (Frames/s): 92.472 versus 29.836
- 2.8x meilleur performance en CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s): 275.229 versus 97.861
- 4.2x meilleur performance en GFXBench 4.0 - T-Rex (Frames): 7038 versus 1673
- 4.2x meilleur performance en GFXBench 4.0 - T-Rex (Fps): 7038 versus 1673
Caractéristiques | |
Date de sortie | 12 November 2012 versus 3 May 2012 |
Thermal Design Power (TDP) | 235 Watt versus 300 Watt |
Taille de mémore maximale | 6 GB versus 4 GB (4 GB per GPU) GDDR5 |
Référence | |
Geekbench - OpenCL | 16664 versus 16577 |
CompuBench 1.5 Desktop - Face Detection (mPixels/s) | 49.455 versus 46.352 |
CompuBench 1.5 Desktop - Ocean Surface Simulation (Frames/s) | 1666.354 versus 930.114 |
CompuBench 1.5 Desktop - T-Rex (Frames/s) | 5.836 versus 3.656 |
CompuBench 1.5 Desktop - Video Composition (Frames/s) | 92.472 versus 29.836 |
CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s) | 275.229 versus 97.861 |
GFXBench 4.0 - T-Rex (Frames) | 7038 versus 1673 |
GFXBench 4.0 - T-Rex (Fps) | 7038 versus 1673 |
Raisons pour considerer le NVIDIA GeForce GTX 690
- Environ 25% plus haut vitesse du noyau: 915 MHz versus 732 MHz
- Environ 43% taux plus haut de remplissage de la texture: 234 billion / sec versus 164.0 GTexel / s
- Environ 14% de pipelines plus haut: 2x 1536 versus 2688
- Environ 59% de meilleur performance á point flottant: 2x 3,130 gflops versus 3,935 gflops
- Environ 16% plus haut de vitesse de mémoire: 6008 MHz versus 5200 MHz
- Environ 25% meilleur performance en PassMark - G3D Mark: 5489 versus 4403
- 2.3x meilleur performance en PassMark - G2D Mark: 463 versus 198
- Environ 79% meilleur performance en GFXBench 4.0 - Manhattan (Frames): 1849 versus 1032
- Environ 79% meilleur performance en GFXBench 4.0 - Manhattan (Fps): 1849 versus 1032
Caractéristiques | |
Vitesse du noyau | 915 MHz versus 732 MHz |
Taux de remplissage de la texture | 234 billion / sec versus 164.0 GTexel / s |
Pipelines | 2x 1536 versus 2688 |
Performance á point flottant | 2x 3,130 gflops versus 3,935 gflops |
Vitesse de mémoire | 6008 MHz versus 5200 MHz |
Référence | |
PassMark - G3D Mark | 5489 versus 4403 |
PassMark - G2D Mark | 463 versus 198 |
GFXBench 4.0 - Manhattan (Frames) | 1849 versus 1032 |
GFXBench 4.0 - Manhattan (Fps) | 1849 versus 1032 |
Comparer les références
GPU 1: NVIDIA Tesla K20Xm
GPU 2: NVIDIA GeForce GTX 690
PassMark - G3D Mark |
|
|
||||
PassMark - G2D Mark |
|
|
||||
Geekbench - OpenCL |
|
|
||||
CompuBench 1.5 Desktop - Face Detection (mPixels/s) |
|
|
||||
CompuBench 1.5 Desktop - Ocean Surface Simulation (Frames/s) |
|
|
||||
CompuBench 1.5 Desktop - T-Rex (Frames/s) |
|
|
||||
CompuBench 1.5 Desktop - Video Composition (Frames/s) |
|
|
||||
CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s) |
|
|
||||
GFXBench 4.0 - Manhattan (Frames) |
|
|
||||
GFXBench 4.0 - T-Rex (Frames) |
|
|
||||
GFXBench 4.0 - Manhattan (Fps) |
|
|
||||
GFXBench 4.0 - T-Rex (Fps) |
|
|
Nom | NVIDIA Tesla K20Xm | NVIDIA GeForce GTX 690 |
---|---|---|
PassMark - G3D Mark | 4403 | 5489 |
PassMark - G2D Mark | 198 | 463 |
Geekbench - OpenCL | 16664 | 16577 |
CompuBench 1.5 Desktop - Face Detection (mPixels/s) | 49.455 | 46.352 |
CompuBench 1.5 Desktop - Ocean Surface Simulation (Frames/s) | 1666.354 | 930.114 |
CompuBench 1.5 Desktop - T-Rex (Frames/s) | 5.836 | 3.656 |
CompuBench 1.5 Desktop - Video Composition (Frames/s) | 92.472 | 29.836 |
CompuBench 1.5 Desktop - Bitcoin Mining (mHash/s) | 275.229 | 97.861 |
GFXBench 4.0 - Manhattan (Frames) | 1032 | 1849 |
GFXBench 4.0 - T-Rex (Frames) | 7038 | 1673 |
GFXBench 4.0 - Manhattan (Fps) | 1032 | 1849 |
GFXBench 4.0 - T-Rex (Fps) | 7038 | 1673 |
GFXBench 4.0 - Car Chase Offscreen (Frames) | 6875 | |
GFXBench 4.0 - Car Chase Offscreen (Fps) | 6875 | |
3DMark Fire Strike - Graphics Score | 0 |
Comparer les caractéristiques
NVIDIA Tesla K20Xm | NVIDIA GeForce GTX 690 | |
---|---|---|
Essentiel |
||
Architecture | Kepler | Kepler |
Nom de code | GK110 | GK104 |
Date de sortie | 12 November 2012 | 3 May 2012 |
Prix de sortie (MSRP) | $7,699 | $999 |
Position dans l’évaluation de la performance | 746 | 748 |
Genre | Workstation | Desktop |
Prix maintenant | $999 | |
Valeur pour le prix (0-100) | 7.01 | |
Infos techniques |
||
Vitesse du noyau | 732 MHz | 915 MHz |
Performance á point flottant | 3,935 gflops | 2x 3,130 gflops |
Processus de fabrication | 28 nm | 28 nm |
Pipelines | 2688 | 2x 1536 |
Taux de remplissage de la texture | 164.0 GTexel / s | 234 billion / sec |
Thermal Design Power (TDP) | 235 Watt | 300 Watt |
Compte de transistor | 7,080 million | 3,540 million |
Vitesse augmenté | 1019 MHz | |
Noyaux CUDA | 3072 | |
Sorties et ports de vidéo |
||
Connecteurs d’écran | No outputs | 3x DVI, 1x mini-DisplayPort, Two Dual Link DVI-I. One Dual link DVI-D. One Mini... |
Contribution d’audio pour HDMI | Internal | |
Soutien de G-SYNC | ||
HDCP | ||
HDMI | ||
Résolution VGA maximale | 2048x1536 | |
Soutien de plusiers moniteurs | ||
Compatibilité, dimensions et exigences |
||
Interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
Longeur | 267 mm | 11.0" (27.9 cm) |
Soutien de bus | PCI Express 3.0 | |
Hauteur | 4.376" (11.1 cm) | |
Options SLI | Quad | |
Connecteurs d’énergie supplementaires | Two 8-pin | |
Soutien API |
||
DirectX | 12.0 (11_0) | 12.0 (11_0) |
OpenGL | 4.6 | 4.2 |
Mémoire |
||
RAM maximale | 6 GB | 4 GB (4 GB per GPU) GDDR5 |
Bande passante de la mémoire | 249.6 GB / s | 384 GB / s |
Largeur du bus mémoire | 384 Bit | 512-bit (256-bit per GPU) |
Vitesse de mémoire | 5200 MHz | 6008 MHz |
Genre de mémoire | GDDR5 | GDDR5 |
Technologies |
||
3D Blu-Ray | ||
3D Gaming | ||
3D Vision | ||
3D Vision Live | ||
Adaptive VSync | ||
CUDA | ||
DirectX 11 | DirectX 11 | |
FXAA | ||
GPU Boost | ||
SLI | ||
TXAA |