AMD Radeon 610M vs NVIDIA GeForce GTX 460 SE Comparison

The NVIDIA GeForce GTX 460 SE and the AMD Radeon 610M are two graphics solutions separated by nearly twelve years and fundamentally different design philosophies. The GTX 460 SE is a discrete card built on the 40 nm Fermi architecture, while the Radeon 610M is an integrated graphics unit fabricated on TSMC's 6 nm process under the RDNA 2.0 architecture. The recorded benchmark data shows the older discrete card holding a clear performance lead in the one test where both were measured, but the integrated solution counters with dramatically lower power draw and a more modern feature set. This analysis walks through the measured numbers and the specifications that explain them.

Head-to-Head Benchmarks

The database contains a single common test between these two parts, and it produces an unambiguous result. In Geekbench OpenCL, the GTX 460 SE scored 6389 against 4535 for the Radeon 610M, a margin of 40.9 percent in favor of the NVIDIA card. That is a substantial gap for any pairing, and it exists despite the GTX 460 SE having launched in 2010 while the Radeon 610M arrived in 2022.

The raw compute figures help explain the result. The GTX 460 SE delivers 748.8 GFLOPS of FP32 throughput compared to 486.4 GFLOPS for the Radeon 610M. That works out to roughly a 54 percent raw compute advantage for the discrete card, which more than covers the 40.9 percent benchmark lead once memory behavior and driver differences are considered. The GTX 460 SE also carries its own 1024 MB of GDDR5 on a 256-bit bus with 108.8 GB/s of bandwidth, while the Radeon 610M shares system memory with a bandwidth figure the database records simply as system dependent.

There is a second recorded result for the Radeon 610M that is worth noting carefully. Its Geekbench Vulkan score of 6353 is nearly identical to the GTX 460 SE's OpenCL score of 6389, but these are different APIs and cannot be compared as a head-to-head result. The OpenCL pairing is the only apples-to-apples measurement available, and it belongs to the NVIDIA card.

Context from each part's nearest rivals reinforces the picture. The GTX 460 SE's average benchmark score of 6389 places it dead even with the NVIDIA GeForce GTX 580M, within 0.3 percent of the NVIDIA RTX PRO 5000 72 GB Blackwell, 0.9 percent ahead of the AMD Radeon Pro WX 4100, and 1 percent ahead of the AMD Radeon R7 M350 in that cluster. The Radeon 610M's average of 5444 sits 0.4 percent behind the NVIDIA Quadro M4000, 0.5 percent ahead of the AMD Radeon R7 M365X, 0.7 percent behind the AMD Radeon R7 M440, and 1 percent behind the NVIDIA GeForce GTX 765M. Against the full GPU population in the database, the GTX 460 SE lands in the 37th percentile while the Radeon 610M sits at the 32nd percentile, a modest but real gap consistent with the head-to-head outcome.

The Verdict

The data supports a straightforward recommendation: the GTX 460 SE wins on measured graphics performance, and the Radeon 610M wins on efficiency and modern platform features. If the decision is purely about benchmark scores, the 40.9 percent OpenCL advantage, the higher percentile ranking, and the stronger compute, texture, and memory subsystem all point to the discrete NVIDIA card.

If the decision is about what each part physically demands, the situation reverses completely. The GTX 460 SE has a 150 W TDP, occupies a dual-slot footprint, requires two 6-pin power connectors, and carries a suggested power supply of 450 W. The Radeon 610M has a 15 W TDP, requires no power connectors at all, and is an integrated solution that draws no slot. That is a tenfold difference in thermal design power between the two, and it defines where each part is viable. A laptop-class platform cannot host the Fermi card; a desktop build with a capable power supply and a free PCIe slot can.

Both parts are recorded as end-of-life in the database, so neither represents a current purchase option in the usual sense. The GTX 460 SE carried a launch MSRP of 160 USD; the Radeon 610M, as an integrated GPU, has no standalone launch MSRP recorded.

Where Each One Wins

The GTX 460 SE wins wherever raw GPU throughput and dedicated memory matter. Its 748.8 GFLOPS of FP32 compute, its 108.8 GB/s of dedicated GDDR5 bandwidth, its 31.20 GTexel/s texture rate, and its 32 render output units make it the stronger part for compute-bound graphics workloads and any task that benefits from a wide, fast memory bus. The OpenCL score confirms this in practice: 6389 versus 4535. Its rival cluster, which includes workstation-class parts like the Radeon Pro WX 4100, also suggests it holds its own in general GPU compute contexts.

The Radeon 610M wins in efficiency and feature support. At 15 W it delivers 486.4 GFLOPS FP32 and 972.8 GFLOPS FP16 thanks to a 2:1 half-precision rate, a capability the Fermi card's record does not include. Its Vulkan score of 6353 shows it performing well in a modern API that the GTX 460 SE does not support at all. With 2 ray tracing cores, DirectX 12 Ultimate at feature level 12_2, and Vulkan 1.4 support, it covers rendering features the Fermi card simply cannot expose. Its pixel rate of 7.600 GPixel/s is nearly equal to the GTX 460 SE's 7.800 GPixel/s, an interesting result given the power difference, though its texture rate of 15.20 GTexel/s is less than half the discrete card's.

In short: fixed-function modern rendering features, half-precision throughput, and power efficiency go to the AMD part. Dedicated memory bandwidth, texture throughput, and measured OpenCL compute go to the NVIDIA part.

FAQ

Q: Which GPU is faster in the shared benchmark? A: The GTX 460 SE scored 6389 in Geekbench OpenCL versus 4535 for the Radeon 610M, a 40.9 percent advantage for the NVIDIA card.

Q: How much more power does the GTX 460 SE need? A: Its TDP is 150 W with two 6-pin connectors and a suggested 450 W power supply. The Radeon 610M has a 15 W TDP and needs no power connectors.

Q: Does the Radeon 610M support ray tracing? A: Yes, it has 2 ray tracing cores, which the GTX 460 SE lacks entirely.

Q: How do they compare in memory? A: The GTX 460 SE has 1024 MB of GDDR5 on a 256-bit bus with 108.8 GB/s of bandwidth. The Radeon 610M uses shared system memory with system-dependent bandwidth.

Q: Where does each rank against all GPUs in the database? A: The GTX 460 SE sits at the 37th percentile; the Radeon 610M sits at the 32nd percentile.

Q: Which API support differs most? A: The Radeon 610M supports DirectX 12 Ultimate and Vulkan 1.4, while the GTX 460 SE supports DirectX 12 at feature level 11_0 with no Vulkan support listed.

Architecture Differences

The architectural gap between these two parts spans more than a decade of GPU development. The GTX 460 SE uses NVIDIA's Fermi architecture, implemented on the GF104 chip, part of the GeForce 400 generation, fabricated at TSMC on a 40 nm process. That die measures 332 mm² and packs 1,950 million transistors, for a density of 5.9 million transistors per square millimeter.

The Radeon 610M uses AMD's RDNA 2.0 architecture on the Mendocino chip, part of the Navi II integrated graphics line for mobile, fabricated at TSMC on a 6 nm process. Its die measures just 100 mm², less than a third of the Fermi die area, and transistor count is not recorded. The process node difference is the single largest architectural story here: a decade-plus of lithography advancement allows the integrated part to deliver comparable pixel throughput and meaningful modern features within a 15 W envelope.

Feature-level support also diverges sharply. The Fermi card offers no ray tracing or tensor cores, DirectX 12 at feature level 11_0, and OpenGL 4.6. The RDNA 2.0 part includes 2 ray tracing cores, DirectX 12 Ultimate at feature level 12_2, OpenGL 4.6, and Vulkan 1.4. Both use TSMC as foundry, one of the few points of commonality.

Specification Differences

The specification table highlights differences across nearly every field. Process node: 40 nm for the GTX 460 SE versus 6 nm for the Radeon 610M. Shading units: 288 versus 128. Texture mapping units: 48 versus 8. Render output units: 32 versus 4. Ray tracing cores: none versus 2.

Clocks tell the opposite story. The GTX 460 SE has no base or boost clock recorded, with memory running at 850 MHz (3.4 Gbps effective). The Radeon 610M lists a 1500 MHz base clock and a 1900 MHz boost clock, with system-shared memory.

Compute and throughput split the difference. FP32 goes to the NVIDIA card, 748.8 GFLOPS versus 486.4 GFLOPS, but FP16 exists only on the AMD side at 972.8 GFLOPS with a 2:1 ratio. Pixel rate is nearly tied at 7.800 versus 7.600 GPixel/s, while texture rate favors NVIDIA heavily at 31.20 versus 15.20 GTexel/s.

Memory is a categorical difference: 1024 MB of dedicated GDDR5 on a 256-bit bus with 108.8 GB/s of bandwidth, versus fully shared system memory with no fixed size, type, or bandwidth. Power delivery differs equally: 150 W, dual-slot, two 6-pin connectors and a 450 W suggested power supply, versus 15 W, integrated, no connectors, no suggested supply. The bus interfaces are PCIe 2.0 x16 against PCIe 4.0 x8. Display outputs are fixed on the NVIDIA side at two DVI ports and one mini-HDMI 1.3a, and portable-device-dependent on the AMD side. Physical length is recorded only for the GTX 460 SE at 210 mm, since the Radeon 610M has no standalone dimensions. Release dates are 2010 for the Fermi card and 2022 for the Mendocino integrated GPU, and both are now end-of-life products in the database.