NVIDIA GeForce GTX 1070 vs NVIDIA GeForce MX350 Comparison
NVIDIA GeForce GTX 1070
GeForce MX350
PERFORMANCE BENCHMARKS
Analysis: NVIDIA GeForce GTX 1070 vs NVIDIA GeForce MX350
The NVIDIA GeForce MX350 is a low-power mobile GPU built for thin-and-light laptops, while the NVIDIA GeForce GTX 1070 is a desktop-class card from the same Pascal architecture generation. The benchmark data shows a decisive split, with the GTX 1070 dominating in every recorded test, but the MX350 holds its own as a surprisingly capable option in its narrow niche.
Head-to-Head Benchmarks
The two available head-to-head benchmark results paint a clear picture of the performance gap. In the Geekbench OpenCL test, the GTX 1070 scores 44,700 against the MX350's 8,689. That is a delta of -80.6% for the MX350, meaning the GTX 1070 delivers roughly 5.1 times the raw compute performance in this workload. This is not a close contest; the GTX 1070's massive advantage in shading units, memory bandwidth, and memory bus width translates directly into an overwhelming lead in this synthetic compute test.
The Geekbench Vulkan result is closer but still lopsided. The GTX 1070 scores 22,121, while the MX350 manages 13,077, a delta of -40.9% for the MX350. While the MX350 is clearly behind, the smaller gap in Vulkan suggests that the MX350's architecture handles this API relatively better than OpenCL, possibly due to driver optimizations or the workload's characteristics. Still, the GTX 1070 wins this test by a comfortable margin of 69%.
Looking at the average benchmark scores, the MX350 has an average of 10,883 across its two recorded tests, while the GTX 1070 averages 9,780 across a much larger suite of benchmarks. This apparent inversion is misleading: the MX350's average is pulled up by its Vulkan score, while the GTX 1070's average is dragged down by low scores in Passmark DirectX 9 (197), DirectX 10 (82), and DirectX 11 (100) tests. The GTX 1070's Passmark G3D score of 13,498 and GPU Compute score of 6,102 reflect its true gaming and compute strength. The MX350's percentile ranking of 49 versus the GTX 1070's 47 further highlights that these averages are not directly comparable due to different test suites.
The GTX 1070 also wins the 3DMark Steel Nomad DX12 test with a score of 1,082, a test the MX350 did not participate in. The GTX 1070's Geekbench Metal score of 18,801 is also notable, indicating strong performance in Apple's Metal API, though the MX350 has no comparable data point.
FAQ
Q: Is the GTX 1070 always faster than the MX350?
A: Yes, in every head-to-head benchmark recorded, the GTX 1070 wins. The Geekbench OpenCL delta is -80.6% for the MX350, and the Vulkan delta is -40.9%. The GTX 1070 also has a higher pixel rate (107.7 GPixel/s vs 23.49 GPixel/s) and texture rate (202.0 GTexel/s vs 46.98 GTexel/s).
Q: What is the biggest performance difference between the two?
A: The largest gap is in Geekbench OpenCL, where the GTX 1070's score of 44,700 is 414% higher than the MX350's 8,689. This reflects the GTX 1070's 6.463 TFLOPS FP32 performance versus the MX350's 1.879 TFLOPS.
Q: Does the MX350 have any advantages in memory technology?
A: Both use GDDR5 memory. The MX350 has 2 GB on a 64-bit bus with 56.06 GB/s bandwidth, while the GTX 1070 has 8 GB on a 256-bit bus with 256.3 GB/s bandwidth. The GTX 1070 has 4.6 times the bandwidth.
Q: Which card has a higher boost clock?
A: The GTX 1070 has a boost clock of 1683 MHz, which is higher than the MX350's 1468 MHz. The GTX 1070's base clock of 1506 MHz also exceeds the MX350's boost clock.
Q: Can the MX350 run modern games at high settings?
A: The data suggests limited capability. Its 2 GB memory capacity and 56.06 GB/s bandwidth are restrictive, and its FP32 performance of 1.879 TFLOPS is far below the GTX 1070's 6.463 TFLOPS. Its 49th percentile ranking indicates it is roughly mid-pack among all GPUs, but that is not a high bar for gaming.
Q: Which card has better Vulkan performance relative to its OpenCL performance?
A: The MX350 shows relatively stronger Vulkan performance. Its Vulkan score (13,077) is 50% higher than its OpenCL score (8,689). The GTX 1070's Vulkan score (22,121) is only about half of its OpenCL score (44,700), suggesting the MX350's architecture is more balanced across these APIs.
Architecture Differences
Both GPUs are built on NVIDIA's Pascal architecture, but they are fundamentally different chips. The MX350 uses the GP107S chip, fabricated on a 14 nm process at Samsung, while the GTX 1070 uses the GP104 chip, fabricated on a 16 nm process at TSMC. The MX350's smaller process node allows for a transistor density of 25.0M per mm², slightly higher than the GTX 1070's 22.9M per mm². However, the GTX 1070 has far more transistors overall: 7,200 million versus 3,300 million, and a larger die size of 314 mm² versus 132 mm².
The shading unit counts are drastically different. The MX350 has 640 shading units, 32 texture mapping units (TMUs), and 16 render output units (ROPs). The GTX 1070 has 1,920 shading units, 120 TMUs, and 64 ROPs — exactly three times the shading units, 3.75 times the TMUs, and four times the ROPs. This explains the GTX 1070's massive lead in pixel rate (107.7 vs 23.49 GPixel/s) and texture rate (202.0 vs 46.98 GTexel/s).
Memory architecture differs fundamentally. The MX350 uses a 64-bit memory bus with 2 GB of GDDR5, while the GTX 1070 uses a 256-bit bus with 8 GB. The GTX 1070's memory clock is 2002 MHz (8 Gbps effective) versus the MX350's 1752 MHz (7 Gbps effective), but the real difference is the bus width. The GTX 1070's 256.3 GB/s bandwidth is 4.6 times the MX350's 56.06 GB/s.
Neither card has ray tracing or tensor cores, as both predate NVIDIA's RTX line. The MX350 uses a PCIe 3.0 x4 interface, while the GTX 1070 uses PCIe 3.0 x16, which limits the MX350's ability to feed data from the CPU. Both support DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. FP16 performance is also identical in ratio at 1:64, with the GTX 1070 offering 101.0 GFLOPS versus the MX350's 29.36 GFLOPS.
Specification Differences
The GTX 1070 uses the larger GP104 chip, while the MX350 uses GP107S. Both are Pascal, but the GTX 1070 is fabricated on a 16 nm TSMC process, while the MX350 is on a 14 nm Samsung process. The GTX 1070 has 7,200 million transistors on a 314 mm² die; the MX350 has 3,300 million on 132 mm².
Clock speeds favor the GTX 1070: base 1506 MHz vs 1354 MHz, boost 1683 MHz vs 1468 MHz. Memory clocks are 2002 MHz (8 Gbps effective) vs 1752 MHz (7 Gbps effective). Memory capacity is 8 GB vs 2 GB, with bus widths of 256-bit vs 64-bit, yielding bandwidth of 256.3 GB/s vs 56.06 GB/s.
The GTX 1070 has 1,920 shading units, 120 TMUs, and 64 ROPs, versus 640, 32, and 16 on the MX350. Pixel rates are 107.7 GPixel/s vs 23.49 GPixel/s, and texture rates are 202.0 GTexel/s vs 46.98 GTexel/s. FP32 performance is 6.463 TFLOPS vs 1.879 TFLOPS. FP16 is 101.0 GFLOPS vs 29.36 GFLOPS.
Power consumption is a major divider: the GTX 1070 has a TDP of 150 W, requires a 1x 8-pin power connector, and needs a 450 W suggested PSU. The MX350 has a TDP of 20 W, uses no power connectors, and is listed as "Portable Device Dependent" for display outputs. The GTX 1070 is a dual-slot card measuring 267 mm long, 112 mm high, and 40 mm wide, with outputs for 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.4a. The MX350 has no fixed dimensions or display outputs, as it is designed for mobile integration.
Release dates differ: the MX350 launched on 2020-02-09, while the GTX 1070 launched on 2016-06-09. Both are end-of-life. The GTX 1070 has a launch MSRP of 379 USD; the MX350 has no launch MSRP. The GTX 1070's predecessor is GeForce 900 and successor is GeForce 20; the MX350 has no listed predecessor or successor.
The Verdict
The data is unambiguous: the GTX 1070 is the superior GPU in every measurable way. Its 6.463 TFLOPS FP32 performance is 3.4 times the MX350's 1.879 TFLOPS. Its 256.3 GB/s memory bandwidth is 4.6 times higher. Its 8 GB VRAM is quadruple the MX350's 2 GB. In the head-to-head OpenCL test, the GTX 1070 is 414% faster; in Vulkan, it is 69% faster. Any workload that stresses compute, memory, or pixel throughput will heavily favor the GTX 1070.
However, the MX350 is not without merit. Its 20 W TDP is just 13% of the GTX 1070's 150 W. It requires no power connectors and is designed for portable devices. Its 49th percentile ranking is actually higher than the GTX 1070's 47th percentile, though this is due to the different benchmark suites used. In the MX350's nearest rivals, it sits close to the GTX 1650 SUPER (-1.5%) and RX 550 (-1.7%), indicating it is a solid entry-level mobile GPU.
The verdict is context-dependent. For a desktop system where power and space are not constraints, the GTX 1070 is the obvious choice. For a thin-and-light laptop where the 20 W TDP is a hard requirement, the MX350 is the only viable option between these two. The GTX 1070 cannot be installed in such a device, and the MX350 cannot drive a high-refresh-rate 1440p monitor. Choose based on the physical form factor, not raw performance.
Where Each One Wins
The GTX 1070 wins in every computationally intensive scenario. Its 8 GB VRAM and 256-bit bus make it suitable for high-resolution textures and large datasets. Its 107.7 GPixel/s pixel rate handles 4K gaming or multiple monitors. Its 6.463 TFLOPS FP32 performance suits GPU compute tasks like rendering, simulation, or machine learning inference. Its Passmark G3D score of 13,498 and 3DMark Steel Nomad DX12 score of 1,082 confirm strong gaming credentials. The GTX 1070 also wins in every head-to-head benchmark and has a wider display output suite (1x DVI, 1x HDMI 2.0, 3x DisplayPort 1.4a).
The MX350's wins are more situational. Its 20 W TDP enables fanless or passively cooled designs in ultraportable laptops. Its 14 nm Samsung process is newer than the GTX 1070's 16 nm TSMC process, though this does not translate to a performance advantage. Its 49th percentile ranking is higher than the GTX 1070's 47th, but this is an artifact of the benchmark set. In absolute terms, the MX350's nearest rivals are the GTX 1650 SUPER and RX 550, both of which are close to its average score of 10,883. The MX350 wins on power efficiency and portability, not on output. For light productivity, web browsing, and esports titles at low settings, the MX350 is adequate. For anything demanding, the GTX 1070 is the only choice.