NVIDIA GeForce GTX 670MX vs NVIDIA GeForce GTX 770M Comparison

The NVIDIA GeForce GTX 770M is the clear performance winner over the GTX 670MX, delivering a 39.6% higher score in the shared Geekbench OpenCL benchmark. This advantage stems from a newer generation architecture, higher clock speeds, and faster memory, despite both cards sharing identical core counts and memory capacities. The GTX 770M sits in the 35th percentile of all GPUs, while the GTX 670MX ranks at the 33rd, confirming a modest but measurable performance gap that favors the newer mobile part.

FAQ

Q: Which GPU is faster in the shared benchmark?

A: The NVIDIA GeForce GTX 770M wins the only head-to-head benchmark (Geekbench OpenCL) with a score of 8552 versus the GTX 670MX’s 6125, a delta of 39.6%.

Q: Do both cards have the same memory configuration?

A: Yes, both the GTX 770M and GTX 670MX feature 3 GB of GDDR5 memory on a 192-bit bus. However, the GTX 770M’s memory runs at 1002 MHz (4 Gbps effective) versus 700 MHz (2.8 Gbps effective) on the GTX 670MX, yielding 96.19 GB/s versus 67.20 GB/s bandwidth.

Q: Are the core counts identical?

A: Yes, both the GTX 770M and GTX 670MX have 960 shading units, 80 texture mapping units, and 24 ROPs. The performance difference comes from clock speeds, not core counts.

Q: What is the transistor and die size difference?

A: The GTX 770M uses a GK106 chip with 2,540 million transistors on a 221 mm² die. The GTX 670MX uses a GK104 chip with 3,540 million transistors on a 294 mm² die. The GTX 670MX has a larger and more complex physical chip, yet it performs worse.

Q: Which GPU has a higher boost clock?

A: The GTX 770M has a base clock of 706 MHz and a boost clock of 797 MHz. The GTX 670MX operates at a fixed 601 MHz with no boost capability.

Q: Are both cards the same generation?

A: No. The GTX 770M belongs to the GeForce 700M generation (released May 2013), while the GTX 670MX belongs to the GeForce 600M generation (released September 2012). Both use the Kepler architecture and a 28 nm TSMC process node.

Architecture Differences

Both GPUs are built on NVIDIA’s Kepler architecture, fabricated by TSMC on a 28 nm process node. The fundamental difference lies in the underlying chip. The GTX 770M uses the GK106 die, which contains 2,540 million transistors spread across 221 mm², giving a transistor density of 11.5M per mm². The GTX 670MX uses the GK104 die, which is physically larger at 294 mm² and packs 3,540 million transistors, resulting in a slightly higher density of 12.0M per mm².

Despite the GTX 670MX having a larger and more transistor-rich chip, the GTX 770M achieves superior performance. This is primarily due to clock speeds. The GTX 770M runs at a base clock of 706 MHz with a boost of 797 MHz, whereas the GTX 670MX is locked at 601 MHz with no boost. This clock advantage translates directly into higher throughput rates: the GTX 770M delivers 15.94 GPixel/s pixel fill rate and 63.76 GTexel/s texture rate, compared to 12.02 GPixel/s and 48.08 GTexel/s on the GTX 670MX.

Memory architecture is another differentiator. The GTX 770M’s memory clock of 1002 MHz (4 Gbps effective) yields 96.19 GB/s bandwidth, while the GTX 670MX’s 700 MHz (2.8 Gbps effective) memory produces only 67.20 GB/s. Both use GDDR5 with a 192-bit bus and 3 GB capacity, but the GTX 770M’s faster memory clock gives it a 43% bandwidth advantage. The GTX 770M also uses an MXM-B (3.0) bus interface, while the GTX 670MX uses PCIe 3.0 x16, though both are portable-device dependent for display outputs. API support is identical: DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175.

Where Each One Wins

The GTX 770M wins the only direct comparison available, so its advantages are clear. In the Geekbench OpenCL test, it scores 8552 versus 6125 for the GTX 670MX — a 39.6% margin. This suggests the GTX 770M is better suited for compute-heavy workloads, such as OpenCL-accelerated applications, image processing, and general-purpose GPU tasks. Its higher memory bandwidth (96.19 GB/s vs 67.20 GB/s) and faster texture rate (63.76 GTexel/s vs 48.08 GTexel/s) further reinforce its strength in bandwidth-sensitive scenarios.

The GTX 670MX, while losing the head-to-head, still holds its own in specific contexts. Its larger GK104 die with 3,540 million transistors provides a more robust physical foundation, and its 33rd percentile ranking places it close to the GTX 770M’s 35th percentile. The GTX 670MX also has a separate Geekbench Vulkan score of 5316, which the GTX 770M lacks in the data, indicating it can handle Vulkan workloads in the absence of a competing score. For users prioritizing a fixed clock speed (601 MHz, no boost variance), the GTX 670MX offers predictable performance without thermal or power fluctuations.

In practical terms, the GTX 770M is the choice for raw speed and modern application demands. The GTX 670MX remains viable for older titles or workloads that do not stress memory bandwidth, but the data shows no benchmark where it outperforms the GTX 770M.

Specification Differences

The two GPUs differ across several key specification fields, despite sharing the same shading unit count (960), TMU count (80), ROP count (24), memory size (3 GB), memory type (GDDR5), memory bus width (192-bit), TDP (75 W), and power connector requirement (None).

| Specification | GTX 770M | GTX 670MX |

|---|---|---|

| Chip | GK106 | GK104 |

| Generation | GeForce 700M | GeForce 600M |

| Transistors | 2,540 million | 3,540 million |

| Die Size | 221 mm² | 294 mm² |

| Transistor Density | 11.5M / mm² | 12.0M / mm² |

| Base Clock | 706 MHz | 601 MHz |

| Boost Clock | 797 MHz | 601 MHz |

| Memory Clock | 1002 MHz (4 Gbps) | 700 MHz (2.8 Gbps) |

| Memory Bandwidth | 96.19 GB/s | 67.20 GB/s |

| Pixel Rate | 15.94 GPixel/s | 12.02 GPixel/s |

| Texture Rate | 63.76 GTexel/s | 48.08 GTexel/s |

| FP32 | 1.530 TFLOPS | 1,153.9 GFLOPS |

| Bus Interface | MXM-B (3.0) | PCIe 3.0 x16 |

| Slot Width | MXM Module | Not specified |

| Release Date | May 29, 2013 | September 30, 2012 |

| Predecessor | GeForce 600M | GeForce 500M |

| Successor | GeForce 800M | GeForce 700M |

The GTX 770M also has a Geekbench Metal score of 3448, which the GTX 670MX lacks. The GTX 670MX has a Geekbench Vulkan score of 5316, absent for the GTX 770M. Both support identical API versions.

Head-to-Head Benchmarks

The single available head-to-head benchmark is Geekbench OpenCL, and the results are decisive. The GTX 770M scores 8552, while the GTX 670MX scores 6125. This represents a 39.6% delta in favor of the GTX 770M — a substantial margin that indicates a generational leap rather than a minor incremental improvement.

This score gap is explained by the specification differences. The GTX 770M’s boost clock of 797 MHz is 32.6% higher than the GTX 670MX’s fixed 601 MHz. Its memory bandwidth of 96.19 GB/s is 43.1% higher. Its FP32 throughput of 1.530 TFLOPS exceeds the GTX 670MX’s 1,153.9 GFLOPS by roughly 32.6%. These numbers align closely with the 39.6% benchmark delta, confirming that clock speed and memory bandwidth are the primary drivers of the performance advantage.

The GTX 670MX’s only other benchmark scores are Geekbench Vulkan (5316) and Geekbench OpenCL (6125). The GTX 770M has no Vulkan score, so no comparison is possible there. For OpenCL, the GTX 770M’s 8552 is also higher than its own Geekbench Metal score of 3448, showing the card performs differently across APIs.

In the broader context, the GTX 770M’s average benchmark score of 6000 places it nearly even with the AMD Radeon RX 6400 (6001, 0% delta) and slightly ahead of the NVIDIA RTX PRO 6000 Blackwell Server (5996, 0.1% delta). The GTX 670MX’s average of 5721 sits just above the Intel Iris Pro Graphics P6300 (5712, 0.2% delta) and just below the NVIDIA GeForce GTX 550 Ti (5731, -0.2% delta). These rival comparisons show that both cards occupy similar tiers in the overall GPU landscape, but the GTX 770M sits at the higher end of that tier.

The Verdict

The data dictates a clear choice: the NVIDIA GeForce GTX 770M outperforms the GTX 670MX in every measurable shared metric. Its 39.6% lead in Geekbench OpenCL, combined with higher clocks, faster memory, and better fill rates, makes it the superior GPU for compute tasks and modern workloads. The GTX 770M’s 35th percentile ranking versus the GTX 670MX’s 33rd percentile confirms this hierarchy.

Buyers should select the GTX 770M if they prioritize raw performance, memory bandwidth, or OpenCL acceleration. Its boost clock of 797 MHz provides adaptive performance, and its 96.19 GB/s bandwidth handles data-intensive applications more effectively. The GTX 770M is also one generation newer (GeForce 700M vs 600M), offering a more modern feature set.

The GTX 670MX is only preferable in niche scenarios. Its larger GK104 die with 3,540 million transistors might suggest higher durability, but benchmark data does not support a performance benefit. Its fixed 601 MHz clock could appeal to users wanting consistent, non-boosting behavior. It also has a Vulkan score (5316) that the GTX 770M lacks, but no direct comparison exists to judge which is better. For any user choosing between these two, the GTX 770M is the verdict-driven pick based on all available evidence.