Test Methodology
Nvidia Titan X (Pascal) Specifications |
|---|
 |
GPU Comparisons |
|||||||||
|---|---|---|---|---|---|---|---|---|---|
GPU |
Memory |
Miscellaneous |
|||||||
Cores |
Base (MHz) |
Boost (MHz) |
Size (MB) |
Clock (MHz) |
Bus (bits) |
Power Config |
Driver |
||
| Nvidia Titan X (Pascal) | 3,584 |
1,417 |
1,531 |
12,228 |
10,008 |
384 |
8+6 |
368.98 |
|
| EVGA GeForce GTX 1080 FTW | 2,560 |
1,721 |
1,860 |
8,192 |
10,008 |
256 |
8+6 |
368.98 |
|
| Nvidia GeForce GTX 1080 Founders Edition | 2,560 |
1,607 |
1,734 |
8,192 |
10,008 |
256 |
8 |
368.39 |
|
| Nvidia GeForce GTX 1070 Founders Edition | 1,920 |
1,506 |
1,683 |
8,192 |
8,008 |
256 |
8 |
368.39 |
|
| MSI GeForce GTX 1060 Gaming X | 1,280 |
1,569 |
1,785 |
6,144 |
8,008 |
192 |
8 |
368.64 |
|
| Nvidia GeForce GTX Titan X (Maxwell) | 3,072 |
1.000 |
1,089 |
12,228 |
7,012 |
384 |
8+6 |
368.98 |
|
| Gigabyte GeForce GTX 970 Mini ITX OC | 1,664 |
1,076 |
1,216 |
4,096 |
7,012 |
256 |
8 |
368.81 |
|
| Asus GeForce GTX 960 Direct CU II | 1,024 |
1,253 |
1,317 |
2,048 |
7,200 |
128 |
6 |
368.39 |
|
| Sapphire Radeon RX 480 Nitro OC 8GB | 2,304 |
1,208 |
1,342 |
8,192 |
8,000 |
256 |
8 |
16.7.3 |
|
| Sapphire Radeon R9 Fury X | 4,096 |
- |
1,050 |
4,096 |
1,000 |
4,096 |
8+8 |
16.7.3 |
|
| Sapphire Radeon R9 390 Nitro | 2,560 |
- |
1,010 |
8,192 |
6,000 |
512 |
8+8 |
16.7.3 |
|
HEXUS Graphics Test Bench |
||
|---|---|---|
| Hardware Components | Product Page | |
| Processor | Intel Core i7-6700K (overclocked to 4.4GHz) | intel.com |
| CPU Cooler | Noctua NH-D15S | noctua.at |
| Motherboard | Asus Z170 Pro | asus.com |
| Memory | Crucial Ballistix Sport LT 32GB (2x16GB) DDR4-2400 | crucial.com |
| Power Supply | be quiet! Dark Power Pro 11 850W | bequiet.com |
| Primary Storage | 256GB Samsung 950 Pro | samsung.com |
| Secondary Storage | 512GB SK hynix Canvas SC300 | skhynix.com |
| Chassis | Fractal Design Define R5 Windowed | fractal-design.com |
| Monitor | Philips Brilliance 4K Ultra HD LED (288P6LJEB/00) | philips.co.uk |
| Operating system | Windows 10 (64-bit) | microsoft.com |
Benchmark Suite |
|
|---|---|
| Synthetic Benchmarks | |
| 3DMark | Fire Strike, Fire Strike Extreme, Fire Strike Ultra, Time Spy and Time Spy Stability Tests |
| Gaming Benchmarks and Settings | |
| Dirt Rally | DX11, 4xMSAA, Ultra Quality |
| Doom | OpenGL and Vulkan, TSSAA (8TX), Ultra Quality |
| Fallout 4 | DX11, TAA, Ultra Quality |
| Hitman | DX12, SMAA, Maximum Quality |
| Rise of the Tomb Raider | DX12, FXAA, Very High Quality |
| Tom Clancy's The Division | DX11, FXAA, Ultra Quality |
| Total War: Warhammer | DX12, MLAA, Ultra Quality |
| General Benchmarks | |
| Power Consumption | To emulate real-world usage scenarios, we record system-wide mains power draw both when idle and while playing Dirt Rally. |
| Temperature | To emulate real-world usage scenarios, we record GPU core temperature both when idle and while playing Dirt Rally. |
| Noise | A PCE-318 meter is used to record noise levels when idle and while playing Dirt Rally. |
| Overclocking | Maintaining out-the-box voltage, we increase the card's power target and aim for an optimal balance of core and memory overclock. With the heightened frequencies in place, the 3DMark, The Division and Rise of the Tomb Raider tests are re-run at 4K UHD settings. |
Notes
To minimise the impact of any CPU bottlenecks, our test platform is configured with a quad-core Intel Core i7-6700K processor overclocked to 4.4GHz across all cores and 32GB of Crucial Ballistix Sport LT DDR4 memory set to run at 2,400MHz using the built-in XMP profile.
For our high-end benchmark suite, we have an array of games tested at FHD (1,920x1,080), QHD (2,560x1,440) and 4K UHD (3,840x2,160) resolutions. Where applicable, any vendor-specific enhancements such as PhysX and TressFX are disabled to help even the playing field. In-game benchmark results are recorded as minimum and average frames per second.
We've done a fair bit of legwork for the review. An EVGA GTX 1080 FTW has been rebenchmarked on the same drivers as the Titan X for some honest-to-goodness comparisons between GPUs. The older, Maxwell-powered GTX Titan X has been added, on the same driver, while on the AMD side, we've upgraded the driver to 16.7.3 - the latest available at the time of testing - and rebenchmarked the trio of cards. This is important due to the recent improvements in Doom (Vulkan) and Tomb Raider performance.
Our review focusses on gaming performance alone, though we do plan to see how Titan X performs in a broader range of professional applications in the near future.
A peek at performance against GTX 1070 SLI
We haven't included formal numbers for the Titan X against an SLI or CrossFire setup because this is a single-GPU evaluation alone. However, those who need an inkling into how one ultra-high-end card performs against two premium graphics cards, represented in this instance by a couple of MSI GeForce GTX 1070 Gaming X in SLI, can reference the following table.
Titan X vs. GTX 1070 SLI |
||||
|---|---|---|---|---|
| Benchmark at 4K | Titan X | GTX 1070 SLI | ||
| 3DMark | 6,602 | 8,141 | ||
| Dirt Rally | 78.7 | 88.6 | ||
| Doom | 82.7 | 74.7 | ||
| Fallout 4 | 62.6 | 72.3 | ||
| Hitman | 67.3 | 32.1 | ||
| Tomb Raider | 67.4 | 85.2 | ||
| The Division | 55.1 | 59.8 | ||
The drivers are different, the games have since been updated, but it provides a simple reckoner of performance at 4K. Tuned perfectly, the GTX 1070 SLI is faster. Where games aren't/weren't tuned (Hitman), the Titan X's single-GPU approach always comes out on top. Oops, we got ahead of ourselves.