NVIDIA GeForce GTX 1070 vs NVIDIA GRID K2 Comparison

Head-to-Head Benchmarks

The recorded data contains two direct benchmark comparisons between the NVIDIA GeForce GTX 1070 and the NVIDIA GRID K2, both in Geekbench workloads. The results are decisive in favor of the GTX 1070, with the GRID K2 trailing by a substantial margin in each test.

In Geekbench Metal, the GTX 1070 scores 18,801 points against the GRID K2's 5,557 points. This represents a 238.3% advantage for the Pascal-based card, meaning the GTX 1070 delivers more than triple the performance of the Kepler board in this API. The gap is even larger in Geekbench OpenCL, where the GTX 1070 records 44,700 points versus 10,602 points for the GRID K2, a 321.6% delta. In practical terms, the GTX 1070 completes roughly four and a half times the work of the GRID K2 in this compute workload.

Context from the database's nearest-rival data reinforces the magnitude of this separation. The GTX 1070's average benchmark score sits at 9,780, with its closest competitors being the AMD FirePro W5000 at 9,803 (0.2% higher), the NVIDIA Quadro M2000M at 9,832 (0.5% higher), the NVIDIA Tesla M10 at 9,724 (0.6% lower), and the NVIDIA Tesla C2070 at 9,716 (0.7% lower). The GRID K2's average score of 8,080 places it in a different performance tier entirely, with its nearest rivals being the NVIDIA GeForce GTX 650 Ti Boost at 8,067 (0.2% higher), the NVIDIA GeForce 945M at 8,099 (0.2% lower), the NVIDIA GeForce GTX 650 Ti at 8,053 (0.3% higher), and the NVIDIA GeForce GTX 880M at 8,040 (0.5% higher). The performance gap between the two cards in the head-to-head tests is far larger than any delta observed within their respective rival clusters.

The GTX 1070 also holds a clear advantage in percentile ranking. It sits at the 47th percentile among all GPUs in the database, while the GRID K2 sits at the 42nd percentile. While both cards fall in the middle of the overall distribution, the GTX 1070's position is meaningfully higher, consistent with its benchmark scores.

The wins tally reflects this dominance: the GTX 1070 wins 2 head-to-head benchmarks, while the GRID K2 wins none. No test in the shared benchmark set favors the older card.

Where Each One Wins

The data shows a straightforward split: the GTX 1070 wins in every measured category, but the magnitude of its advantage varies by workload type, which suggests different use-case suitability.

For compute-heavy tasks exposed through OpenCL, the GTX 1070's lead is enormous. Its 44,700 OpenCL score, which is 321.6% ahead of the GRID K2, indicates that workloads relying on general-purpose GPU compute will see dramatic improvements when using the newer card. This includes tasks such as data processing, scientific simulations, and other parallel compute workloads that can leverage the card's 6.463 TFLOPS of FP32 throughput. The GRID K2's 2.289 TFLOPS FP32 figure is less than half of the GTX 1070's, which explains the lopsided OpenCL result.

For graphics workloads exposed through Metal, the GTX 1070's 18,801 score versus 5,557 for the GRID K2, a 238.3% lead, still represents a massive gap, though slightly less extreme than the OpenCL difference. This suggests the GTX 1070 is the stronger choice for any rendering or graphics-acceleration task, but the relative gap narrows somewhat compared to pure compute.

The GRID K2's strengths, such as they are, lie in its original design context rather than in the benchmark data. It is a dual-slot card with no display outputs, which indicates it was built for virtualized or server-side workloads rather than direct rendering to a monitor. Its 4 GB of GDDR5 memory and 160.0 GB/s of bandwidth are adequate for certain virtual desktop infrastructure scenarios, but the benchmark evidence does not show any workload where it outperforms the GTX 1070. The GTX 1070's 8 GB of GDDR5 memory and 256.3 GB/s bandwidth more than double the GRID K2's memory capacity and bandwidth, respectively.

The GTX 1070 also wins on efficiency metrics. Its 150 W TDP is lower than the GRID K2's 225 W, while delivering significantly higher performance. The suggested power supply rating for the GTX 1070 is 450 W, compared to 550 W for the GRID K2. This means the faster card demands less system power headroom, making it easier to integrate into a wider range of systems.

Architecture Differences

The two cards represent different generations of NVIDIA GPU architecture, and the database records substantial differences in their underlying designs.

The GTX 1070 uses the GP104 chip built on the Pascal architecture, fabricated on a 16 nm process at TSMC. It contains 7,200 million transistors on a 314 mm² die, yielding a transistor density of 22.9 million transistors per mm². The GRID K2, by contrast, uses the GK104 chip based on the older Kepler architecture, also fabricated at TSMC but on a 28 nm process. It contains 3,540 million transistors on a 294 mm² die, for a density of 12.0 million transistors per mm². The Pascal chip packs more than double the transistors into a slightly larger die, and the density difference reflects the newer manufacturing process.

Clock speeds tell a similar story. The GTX 1070 has a base clock of 1506 MHz and a boost clock of 1683 MHz. The GRID K2's base and boost clocks are not recorded in the database, but its memory runs at 1250 MHz (5 Gbps effective), compared to the GTX 1070's memory clock of 2002 MHz (8 Gbps effective). The faster memory clock contributes to the GTX 1070's 256.3 GB/s bandwidth versus 160.0 GB/s for the GRID K2.

The compute configurations differ in interesting ways. The GTX 1070 has 1,920 shading units, 120 texture mapping units, and 64 raster operation pipelines. The GRID K2 has 1,536 shading units, 128 TMUs, and only 32 ROPs. While the GRID K2 has slightly more TMUs, the GTX 1070 more than doubles its ROP count, which affects pixel throughput. The pixel rate for the GTX 1070 is 107.7 GPixel/s, while the GRID K2 manages 23.84 GPixel/s. Texture rates are 202.0 GTexel/s for the GTX 1070 and 95.36 GTexel/s for the GRID K2.

FP32 performance follows the same pattern. The GTX 1070 delivers 6.463 TFLOPS, while the GRID K2 delivers 2.289 TFLOPS. The GTX 1070 also records a FP16 figure of 101.0 GFLOPS (at a 1:64 ratio), while the GRID K2 has no recorded FP16 capability.

Memory configurations differ in capacity. The GTX 1070 ships with 8 GB of GDDR5 on a 256-bit bus. The GRID K2 ships with 4 GB of GDDR5, also on a 256-bit bus, but with lower effective memory speed. Both cards use a PCIe 3.0 x16 bus interface and are dual-slot designs with a length of 267 mm (10.5 inches).

API support shows the generational divide. The GTX 1070 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The GRID K2 supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The GTX 1070 has a higher DirectX feature level and a newer Vulkan version.

Physical and power characteristics also differ. The GTX 1070 requires a single 8-pin power connector and carries a 150 W TDP with a 450 W suggested PSU. The GRID K2 requires a 6-pin and an 8-pin connector, carries a 225 W TDP, and has a 550 W suggested PSU. The GTX 1070 offers display outputs (1x DVI, 1x HDMI 2.0, 3x DisplayPort 1.4a), while the GRID K2 has no display outputs at all, reinforcing its server-oriented design.

The GTX 1070 was released on 2016-06-09, while the GRID K2 was released on 2013-05-10. Both are marked as end-of-life production status. The GTX 1070 has a recorded predecessor in the GeForce 900 series and a successor in the GeForce 20 series; the GRID K2 has neither recorded.

The Verdict

The benchmark data points to a clear winner for nearly any use case. The GTX 1070 outperforms the GRID K2 by 238.3% in Geekbench Metal and by 321.6% in Geekbench OpenCL, while also ranking higher in the overall GPU percentile distribution (47th versus 42nd). It does so while consuming 75 W less power, requiring only a single 8-pin connector instead of dual connectors, and offering 8 GB of memory versus 4 GB.

For users selecting a card for graphics rendering, compute workloads, or general GPU acceleration, the GTX 1070 is the superior choice according to every recorded measurement. Its higher FP32 throughput, more than double the pixel rate, and significantly higher memory bandwidth make it better suited to demanding tasks. The GTX 1070's display outputs also make it usable in a standard desktop configuration, which the GRID K2 cannot support.

The GRID K2's only conceivable advantage is its original design purpose as a virtualization and cloud-gaming card, which is not reflected in the benchmark data. The database records no test where the GRID K2 wins. Its lower average benchmark score of 8,080 versus 9,780 for the GTX 1070 places it in a performance class closer to mobile and entry-level desktop parts. The GTX 1070's nearest rivals sit within 0.7% of its average score, while the GRID K2's nearest rivals also cluster tightly around its score, indicating both cards are well-positioned within their respective tiers, but those tiers are far apart.

The GTX 1070 also offers better API support, with DirectX 12 (12_1) versus the GRID K2's DirectX 12 (11_0), and Vulkan 1.4 versus Vulkan 1.2.175. This gives the newer card better forward compatibility with modern software.

The verdict from the data is unambiguous: the GeForce GTX 1070 is the faster, more efficient, and more capable card. The GRID K2 should only be considered for legacy virtualization deployments where its specific form factor and lack of display outputs are required, and even then, its performance disadvantages are substantial.

FAQ

Q: How much faster is the GeForce GTX 1070 than the GRID K2 in Metal benchmarks?

A: The GTX 1070 scores 18,801 in Geekbench Metal, while the GRID K2 scores 5,557, giving the GTX 1070 a 238.3% advantage.

Q: What is the OpenCL performance difference between these two cards?

A: In Geekbench OpenCL, the GTX 1070 scores 44,700 versus 10,602 for the GRID K2, a 321.6% lead for the GTX 1070.

Q: Which card has more memory and memory bandwidth?

A: The GTX 1070 has 8 GB of GDDR5 memory with 256.3 GB/s bandwidth. The GRID K2 has 4 GB of GDDR5 memory with 160.0 GB/s bandwidth.

Q: How does the power consumption compare?

A: The GTX 1070 has a TDP of 150 W and a suggested PSU of 450 W, while the GRID K2 has a TDP of 225 W and a suggested PSU of 550 W. The GTX 1070 requires one 8-pin connector, while the GRID K2 requires one 6-pin and one 8-pin connector.

Q: Do both cards support the same DirectX version?

A: No. The GTX 1070 supports DirectX 12 (12_1), while the GRID K2 supports DirectX 12 (11_0), which is a lower feature level.

Q: Which card has display outputs?

A: The GTX 1070 has 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.4a outputs. The GRID K2 has no display outputs.