NVIDIA GeForce GT 1010 vs NVIDIA GeForce GTX 670MX Comparison

The Verdict

The recorded data presents a straightforward outcome. The NVIDIA GeForce GT 1010 wins the only head-to-head benchmark available in the database, the Geekbench OpenCL test, by a margin of 9.4%. Its score of 6698 places it at the 38th percentile of all GPUs, while the GTX 670MX scores 6125 and sits at the 33rd percentile. For any user choosing strictly between these two based on the measured performance data, the GT 1010 is the faster part.

The GTX 670MX does hold advantages in other areas, but not in raw benchmark performance. It offers a larger memory pool of 3 GB versus 2 GB, a wider 192-bit memory bus versus 64-bit, and significantly higher bandwidth at 67.20 GB/s. However, none of these specifications translate into a benchmark win in the database. The GT 1010 is the pick for compute performance.

The GTX 670MX, being a mobile part with portable device dependent display outputs, is clearly designed for laptops. The GT 1010, with its single-slot profile, 147 mm length, and DVI plus mini-HDMI outputs, is a desktop card. Users with a desktop system should choose the GT 1010. Users constrained to a mobile platform have only the GTX 670MX as the realistic option, despite its lower benchmark score.

Where Each One Wins

The benchmark data gives the GT 1010 the sole victory. In the Geekbench OpenCL test, it surpasses the GTX 670MX by 9.4%. This is the only recorded head-to-head comparison, and the GT 1010 wins it. The GTX 670MX does not win any recorded benchmark.

Looking at the broader performance context, the GT 1010 sits at the 38th percentile, which places it above the GTX 670MX at the 33rd percentile. The GT 1010's closest rivals are the AMD Radeon R7 M370 at 6764 (1% higher), the AMD FirePro M5100 at 6830 (1.9% higher), the AMD Radeon R7 M460 at 6612 (1.3% lower), and the AMD Radeon HD 7730M at 6581 (1.8% lower). The GTX 670MX's rivals are different, including the Intel Iris Pro Graphics P6300 at 5712 (0.2% higher) and the NVIDIA GeForce GTX 550 Ti at 5731 (0.2% higher). This grouping suggests the GT 1010 competes in a higher performance tier than the GTX 670MX.

The GTX 670MX wins on specifications that matter for certain workloads, even if not reflected in the OpenCL score. Its memory bandwidth of 67.20 GB/s is 39.8% higher than the GT 1010's 48.06 GB/s. Its texture rate of 48.08 GTexel/s is more than double the GT 1010's 23.49 GTexel/s. Its pixel rate of 12.02 GPixel/s is slightly higher than the GT 1010's 11.74 GPixel/s. For texture-heavy workloads, the GTX 670MX has a theoretical edge, but the measured OpenCL performance does not confirm this advantage.

Architecture Differences

The two GPUs come from different architecture generations. The GT 1010 uses the Pascal architecture on a 14 nm process from Samsung. The GTX 670MX uses the older Kepler architecture on a 28 nm process from TSMC. This process difference is significant: the GT 1010 packs 1,800 million transistors into a 74 mm² die, giving a transistor density of 24.3M per mm². The GTX 670MX has 3,540 million transistors on a 294 mm² die, for a density of 12.0M per mm².

The chip designs differ substantially. The GT 1010 uses the GP108 chip with 256 shading units, 16 texture mapping units, and 8 ROPs. The GTX 670MX uses the GK104 chip with 960 shading units, 80 TMUs, and 24 ROPs. Despite having fewer shading units, the GT 1010 achieves a higher FP32 compute rating in terms of efficiency per unit, though the absolute numbers tell a different story. The GTX 670MX delivers 1,153.9 GFLOPS of FP32 performance versus 751.6 GFLOPS for the GT 1010. The GTX 670MX also has higher absolute rates: 48.08 GTexel/s versus 23.49 GTexel/s, and 12.02 GPixel/s versus 11.74 GPixel/s.

Clock speeds are very different. The GT 1010 runs at a base clock of 1228 MHz and boosts to 1468 MHz. The GTX 670MX runs at a fixed 601 MHz with no boost. This clock advantage helps the GT 1010 overcome the GTX 670MX's larger core configuration in the OpenCL benchmark.

Memory architecture also differs. The GT 1010 has 2 GB of GDDR5 on a 64-bit bus with 48.06 GB/s bandwidth. The GTX 670MX has 3 GB of GDDR5 on a 192-bit bus with 67.20 GB/s bandwidth. The GTX 670MX's memory clock is 700 MHz (2.8 Gbps effective) versus 1502 MHz (6 Gbps effective) for the GT 1010, but the wider bus gives the GTX 670MX the bandwidth advantage.

The power profiles diverge as well. The GT 1010 has a TDP of 30 W and a suggested PSU of 200 W. The GTX 670MX has a TDP of 75 W. Neither card requires external power connectors. The GT 1010 uses a PCIe 3.0 x4 interface, while the GTX 670MX uses PCIe 3.0 x16.

API support differs. The GT 1010 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The GTX 670MX supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The GT 1010 has a higher DirectX feature level and a newer Vulkan version.

FAQ

Q: Which GPU has the higher benchmark score?

A: The NVIDIA GeForce GT 1010 scores 6698 in Geekbench OpenCL, which is 9.4% higher than the GTX 670MX's 6125. The GT 1010 also has a higher percentile ranking at 38 versus 33.

Q: Does the GTX 670MX have any performance advantage?

A: The GTX 670MX has higher theoretical throughput in several areas: 1,153.9 GFLOPS FP32 versus 751.6 GFLOPS, 48.08 GTexel/s versus 23.49 GTexel/s, and 67.20 GB/s memory bandwidth versus 48.06 GB/s. However, it does not win any recorded benchmark test.

Q: Which card has more memory?

A: The GTX 670MX has 3 GB of GDDR5 memory on a 192-bit bus. The GT 1010 has 2 GB of GDDR5 on a 64-bit bus.

Q: What are the power requirements?

A: The GT 1010 has a TDP of 30 W and a suggested PSU of 200 W. The GTX 670MX has a TDP of 75 W and has no suggested PSU listed. Neither card requires external power connectors.

Q: Which GPU is more efficient per watt?

A: The GT 1010 achieves a higher OpenCL score of 6698 with a 30 W TDP, while the GTX 670MX scores 6125 with a 75 W TDP. The GT 1010 delivers more benchmark performance per watt.

Q: Are these cards from the same architecture generation?

A: No. The GT 1010 uses the Pascal architecture on a 14 nm Samsung process with the GP108 chip. The GTX 670MX uses the Kepler architecture on a 28 nm TSMC process with the GK104 chip.

Head-to-Head Benchmarks

The database records exactly one head-to-head benchmark between these two GPUs. In the Geekbench OpenCL test, the GT 1010 scores 6698 against the GTX 670MX's 6125. The GT 1010 wins by 9.4%. This is a decisive margin, though not an overwhelming one. The GTX 670MX has no recorded benchmark wins against the GT 1010.

The GT 1010's score of 6698 places it in a competitive range with its nearest rivals. The AMD Radeon R7 M370 scores 6764, just 1% higher. The AMD FirePro M5100 scores 6830, 1.9% higher. The AMD Radeon R7 M460 scores 6612, 1.3% lower. The AMD Radeon HD 7730M scores 6581, 1.8% lower. These are all close margins, indicating the GT 1010 sits in a tight performance cluster.

The GTX 670MX's average benchmark score of 5721 is lower than its OpenCL score of 6125 because it also has a Geekbench Vulkan score of 5316. The Vulkan result pulls its average down. The GTX 670MX's nearest rivals include the Intel Iris Pro Graphics P6300 at 5712 (0.2% higher), the NVIDIA GeForce GTX 550 Ti at 5731 (0.2% higher), the AMD Radeon HD 8790M at 5691 (0.5% higher), and the NVIDIA Quadro M500M at 5604 (2.1% lower). The GTX 670MX is essentially tied with these parts, all within a 2.1% range.

The biggest win in this comparison belongs to the GT 1010, with its 9.4% OpenCL advantage. The GTX 670MX's theoretical advantages in texture rate and memory bandwidth do not translate into a benchmark win. The GT 1010's higher clock speeds and newer architecture appear to compensate for its smaller core configuration. The data shows a clear, if modest, performance hierarchy: the GT 1010 is the faster GPU in this pairing.

Specification Differences

The two cards differ in nearly every major specification category.

Process and Die: The GT 1010 uses a 14 nm Samsung process with 1,800 million transistors on a 74 mm² die (24.3M per mm²). The GTX 670MX uses a 28 nm TSMC process with 3,540 million transistors on a 294 mm² die (12.0M per mm²).

Architecture and Chip: The GT 1010 is Pascal with the GP108 chip. The GTX 670MX is Kepler with the GK104 chip.

Clocks: The GT 1010 has a base clock of 1228 MHz and a boost clock of 1468 MHz. The GTX 670MX has a base clock of 601 MHz and no boost (601 MHz). Memory clocks are 1502 MHz (6 Gbps effective) for the GT 1010 and 700 MHz (2.8 Gbps effective) for the GTX 670MX.

Memory: The GT 1010 has 2 GB GDDR5 on a 64-bit bus with 48.06 GB/s bandwidth. The GTX 670MX has 3 GB GDDR5 on a 192-bit bus with 67.20 GB/s bandwidth.

Core Configuration: The GT 1010 has 256 shading units, 16 TMUs, and 8 ROPs. The GTX 670MX has 960 shading units, 80 TMUs, and 24 ROPs.

Performance Rates: The GT 1010 has a pixel rate of 11.74 GPixel/s and a texture rate of 23.49 GTexel/s. The GTX 670MX has a pixel rate of 12.02 GPixel/s and a texture rate of 48.08 GTexel/s. FP32 is 751.6 GFLOPS for the GT 1010 and 1,153.9 GFLOPS for the GTX 670MX.

Power: The GT 1010 has a TDP of 30 W and a suggested PSU of 200 W. The GTX 670MX has a TDP of 75 W and no suggested PSU. Neither needs external power connectors.

Interface and Form Factor: The GT 1010 uses PCIe 3.0 x4 and is a single-slot card at 147 mm (5.8 inches) long. The GTX 670MX uses PCIe 3.0 x16 and has no listed dimensions. Display outputs are 1x DVI and 1x mini-HDMI 2.0 for the GT 1010, while the GTX 670MX is portable device dependent.

API Support: The GT 1010 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The GTX 670MX supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175.

Release Timing: The GTX 670MX was released in 2012, while the GT 1010 came later in 2021. Both are end-of-life products. The GTX 670MX belongs to the GeForce 600M generation with a GeForce 500M predecessor and GeForce 700M successor. The GT 1010 belongs to the GeForce 10 generation with a GeForce 900 predecessor and GeForce 20 successor.