I had the oppurtunity to test a GeForce GTX TITAN Black (an ASUS model) for few hours and I compared it with the basic GTX TITAN card.
Quick recap of GTX Titan features:
GeForce GTX TITAN
GPU: Kepler GK110 @ 836MHz (base) / 876MHz (boost)
– CUDA cores: 2688 (14 SMX)
– 224 TMUs, 48 ROPs
GeForce GTX TITAN Black
– GPU: Kepler GK110 @ 889MHz (base) / 980MHz (boost)
– CUDA cores: 2880 (15 SMX)
– 240 TMUs, 48 ROPs
Let’s see if the additional SMX and the higher clock speed will make a significant difference.
Testbed
– CPU: Intel Core i5 4670K @ 3.4GHz
– Motherboard: GIGABYTE G1.SNIPER M5
– Memory: 16GB DDR3 @ 1600MHz
– OS: Win 8.1 PRO 64-bit
– Graphics drivers: R344.11
– PSU: Corsair AX860i
Unigine Valley 1.0
Settings: Extreme HD (1920×1080 fullscreen, DX11, …).
+0.2% – 64.4 FPS, Score: 2695 – GeForce GTX TITAN |
| 64.3 FPS, Score: 2690 – GeForce GTX TITAN Black |
Catzilla 1.3
Settings: 720p
| 22432 – GeForce GTX TITAN |
| -2.2% – 21950 – GeForce GTX TITAN Black |
Aliens vs Predators DX11
Settings: 2560×1440 fullscreen
| +3.1% – 93.5 FPS – GeForce GTX TITAN Black |
| 90.7 FPS – GeForce GTX TITAN |
FurMark 1.14.1
Settings: Preset:1080 (1920×1080 fullscreen)
| +3.8% – 4245 points (70 FPS) – GeForce GTX TITAN Black |
| 4088 points (68 FPS) – GeForce GTX TITAN |
GpuTest 0.7.0
Settings: 1920×1080 fullscreen, no AA
Julia FP32
| +10.3% – 41625 points (692 FPS) – GeForce GTX TITAN Black |
| 37737 points (628 FPS) – GeForce GTX TITAN |
Julia FP64
| +11.2% – 16753 points (278 FPS) – GeForce GTX TITAN Black |
| 15054 points (250 FPS) – GeForce GTX TITAN |
Volplosion
| +4.2% – 3276 points (54 FPS) – GeForce GTX TITAN Black |
| 3144 points (52 FPS) – GeForce GTX TITAN |
MSI Kombustor 3.5.0
Settings: Preset:1080 (1920×1080 fullscreen, 4X MSAA)
OpenCL Julia4D (GPU)
| +6% – 6921 points (115 FPS) – GeForce GTX TITAN Black |
| 6525 points (108 FPS) – GeForce GTX TITAN |
Lakes of Titan X32 (Tessellation)
| +11.6% – 9966 points (166 FPS) – GeForce GTX TITAN Black |
| 8926 points (148 FPS) – GeForce GTX TITAN |
3DMark 2013
Ice Storm
| 160580 – GeForce GTX TITAN |
| -1.7% – 157957 – GeForce GTX TITAN Black |
Cloud Gate
| +0.5% – 18346 – GeForce GTX TITAN Black |
| 18240 – GeForce GTX TITAN |
Sky Diver
| +1% – 21360 – GeForce GTX TITAN Black |
| 21128 – GeForce GTX TITAN |
Fire Strike
| +3.8% – 9212 – GeForce GTX TITAN Black |
| 8867 – GeForce GTX TITAN |
Quick conclusion
The GTX TITAN Black brings a rough +5% up to 11% of processing power for applications that requires pure GPU compute power. In other applications, the difference is insignificant.
Some pictures of the beast:
In other words a waste of money
I agree.
What was the point of the Black?
Other than getting fools like me to buy a pair, when I already own a pair of the originals.
Nvidia’s marketing strategy has been quite disappointing lately.
I don’t know why. They have completely dominated the gpu market for years, with no let up in sight.