AMD Radeon 760M vs NVIDIA GeForce GTX 670M Comparison

The NVIDIA GeForce GTX 670M and AMD Radeon 760M represent two very different eras of mobile graphics, separated by over a decade of architectural evolution. The GTX 670M is an end-of-life discrete mobile GPU from 2012, built on a 40 nm process, while the 760M is an active integrated graphics processor from 2024, built on a current-generation 4 nm node. The benchmark data tells a story of generational leaps, but also reveals that raw compute performance does not always translate directly to a higher overall percentile ranking. In the single available head-to-head test, the AMD part delivers a decisive victory, yet the two GPUs sit within a few percentile points of each other in the global database, highlighting the complex nature of comparing legacy and modern hardware.

Where Each One Wins

The AMD Radeon 760M is the clear winner in the available head-to-head benchmark, taking the sole win in Geekbench OpenCL with a score of 20,255 against the GTX 670M's 6,513. This represents a massive 67.8% performance advantage for the AMD part in this compute-focused test. The 760M also shows its breadth in other benchmark suites, posting a strong Geekbench Vulkan score of 30,336 and a Passmark G3D score of 5,310, indicating solid general-purpose graphics capability. Its Passmark G2D score of 890 further suggests it handles 2D desktop workloads and basic display tasks with ease, a critical factor for an integrated part that must manage the entire system's visual output.

The NVIDIA GeForce GTX 670M, despite losing the head-to-head, still has a niche. Its single Geekbench OpenCL score of 6,513 places it in the 38th percentile of all GPUs, which is actually higher than the 760M's 35th percentile ranking. This suggests that in the broader context of the benchmark database, the older card's performance profile is slightly more competitive relative to its contemporaries than the newer AMD part is relative to its own peers. However, the data does not provide any specific test where the GTX 670M wins outright. Its advantage is purely contextual, resting on its percentile ranking rather than a direct benchmark victory. For a user looking at legacy applications or drivers that are optimized for older NVIDIA architectures, the GTX 670M may still hold its own, but the data shows no scenario where it outpaces the 760M.

Architecture Differences

The architectural gap between these two GPUs is vast, reflecting their different release eras. The NVIDIA GeForce GTX 670M is built on the Fermi 2.0 architecture, using the GF114 chip fabricated on a 40 nm process at TSMC. This chip contains 1,950 million transistors on a die size of 332 mm², resulting in a transistor density of 5.9M per mm². In contrast, the AMD Radeon 760M uses the RDNA 3.0 architecture with the Phoenix chip, also fabricated by TSMC but on a much more advanced 4 nm process. The Phoenix chip packs 25,390 million transistors into a smaller 178 mm² die, achieving a staggering density of 142.6M per mm². This density advantage is a primary driver of the 760M's performance capabilities.

The memory configurations could not be more different. The GTX 670M uses 1536 MB of dedicated GDDR5 memory on a 192-bit bus, delivering 72.00 GB/s of bandwidth. The 760M, as an integrated graphics processor, relies on System Shared memory, meaning its bandwidth is System Dependent and its memory type and bus width are also listed as System Shared. This is a fundamental design difference: one is a discrete solution with its own fast memory, while the other is an IGP that must share system resources. The 760M compensates with a much higher clock speed, boasting a boost clock of 2599 MHz compared to the GTX 670M's memory clock of 750 MHz (3 Gbps effective), though the GTX 670M's base and boost clocks are not listed.

In terms of compute units, the 760M has 512 shading units, 32 TMUs, and 16 ROPs, while the GTX 670M has 336 shading units, 56 TMUs, and 24 ROPs. Interestingly, the older NVIDIA part has more TMUs and ROPs, but the AMD part has more shading units and features 8 dedicated ray tracing cores. The AMD part also supports DirectX 12 Ultimate (12_2) and Vulkan 1.4, whereas the NVIDIA part is limited to DirectX 12 (11_0) and has no Vulkan support listed. The 760M's pixel rate is 41.58 GPixel/s and its texture rate is 83.17 GTexel/s, far exceeding the GTX 670M's 8.372 GPixel/s and 33.49 GTexel/s. The 760M also leads in raw FP32 compute with 5.323 TFLOPS versus the GTX 670M's 803.7 GFLOPS, and it offers FP16 performance at a 1:1 ratio.

Head-to-Head Benchmarks

The only direct head-to-head benchmark available is Geekbench OpenCL, and it is a decisive victory for the AMD Radeon 760M. The 760M scores 20,255, while the NVIDIA GeForce GTX 670M scores 6,513. This yields a deltaPct of -67.8%, meaning the 760M is 67.8% faster than the GTX 670M in this test. This is not a marginal win; it is a generational wipeout. The 760M's score is more than three times higher, indicating that its architecture, higher clock speeds, and superior compute throughput are overwhelming the older Fermi-based design.

Looking at the broader benchmark suites for the AMD part, the results reinforce its strength. In Geekbench Vulkan, the 760M scores 30,336, which is even higher than its OpenCL score, suggesting that its RDNA 3.0 architecture is particularly well-optimized for modern graphics APIs. Its Passmark G3D score of 5,310 and Passmark GPU Compute score of 2,840 further demonstrate its capability in both rasterization and compute workloads. The GTX 670M has no comparable benchmark results in the data, so the OpenCL test is the only direct comparison point. However, for context, the GTX 670M's nearest rivals include the NVIDIA GeForce GT 555M with an average score of 6,493 (0.3% difference) and the Intel UHD Graphics P750 with 6,554 (-0.6% difference). The 760M's nearest rivals include the AMD Radeon RX 6400 with 6,001 (0.3% difference) and the NVIDIA Quadro P2000 with 6,049 (-0.5% difference). This shows that while the 760M dominates the GTX 670M, its average benchmark score is actually lower due to the inclusion of other tests, placing it in a similar performance bracket to the GTX 670M's rivals.

The Verdict

Based strictly on the data, the AMD Radeon 760M is the superior performer for any modern workload. It wins the only direct head-to-head benchmark by a massive 67.8% margin, and its additional benchmark scores in Vulkan, Passmark G3D, and Passmark GPU Compute demonstrate a well-rounded capability that the GTX 670M simply cannot match. The 760M's 512 shading units, 8 ray tracing cores, and support for DirectX 12 Ultimate make it a far more future-proof option for gaming and compute tasks. Its active production status and 2024 release date mean it is a current product, whereas the GTX 670M is end-of-life with a 2012 release date. Anyone building a new system or looking for a capable integrated GPU should choose the Radeon 760M without hesitation.

The NVIDIA GeForce GTX 670M should only be considered by someone with a legacy system that requires a specific MXM-B (3.0) module and cannot accommodate a newer part. Its 75 W TDP is higher than the 760M's 15 W, making it less power-efficient, and its lack of Vulkan support and older DirectX 12 (11_0) feature level limit its software compatibility. Its only statistical advantage is its 38th percentile ranking versus the 760M's 35th, but this is a reflection of the broader database and not a performance win. In a direct comparison, there is no scenario in the data where the GTX 670M comes out ahead. It is a relic of its era, and the data unequivocally shows the 760M as the superior choice for any current application.

FAQ

Q: Which GPU is faster in the only direct benchmark test?

A: The AMD Radeon 760M is significantly faster, scoring 20,255 in Geekbench OpenCL compared to the NVIDIA GeForce GTX 670M's 6,513, a 67.8% advantage.

Q: What are the architectural differences between the two chips?

A: The GTX 670M uses NVIDIA's Fermi 2.0 architecture on a 40 nm process, while the 760M uses AMD's RDNA 3.0 architecture on a 4 nm process. The 760M has 512 shading units and 8 ray tracing cores, while the GTX 670M has 336 shading units but more TMUs (56 vs 32) and ROPs (24 vs 16).

Q: How does the memory configuration differ?

A: The GTX 670M has 1536 MB of dedicated GDDR5 memory on a 192-bit bus with 72.00 GB/s bandwidth. The 760M relies on System Shared memory, with its bandwidth listed as System Dependent.

Q: What is the TDP of each GPU?

A: The NVIDIA GeForce GTX 670M has a TDP of 75 W, while the AMD Radeon 760M has a significantly lower TDP of 15 W.

Q: Which GPU has better API support?

A: The AMD Radeon 760M supports DirectX 12 Ultimate (12_2) and Vulkan 1.4, while the NVIDIA GeForce GTX 670M only supports DirectX 12 (11_0) and OpenGL 4.6, with no Vulkan support listed.

Q: What are the production statuses of these GPUs?

A: The NVIDIA GeForce GTX 670M is end-of-life, released in March 2012, while the AMD Radeon 760M is active, released in January 2024.