AMD Radeon 890M vs NVIDIA GeForce GTX 460 Comparison

Head-to-Head Benchmarks

The only direct benchmark overlap in the database is Geekbench OpenCL, and the result is decisive. The AMD Radeon 890M scores 37,254, while the NVIDIA GeForce GTX 460 scores 7,925. That is a 370.1% advantage for the AMD part, meaning the 890M delivers nearly five times the OpenCL compute throughput of the older Fermi card. The delta is so large that it places the two GPUs in entirely different performance tiers, despite both being nominally classified as graphics processors.

The Radeon 890M also holds a substantial edge in the broader benchmark suite recorded for each product. Its average benchmark score is 9,210, a figure that sits 16.2% above the GTX 460's 7,925 average. In percentile terms, the 890M ranks at the 45th percentile among all GPUs in the database, while the GTX 460 sits at the 41st percentile. That gap, while modest in percentile terms, masks the fact that the 890M's nearest rivals include the NVIDIA GeForce GTX 960 (average score 9,273, delta -0.7%), the GeForce GTX 465 (9,294, -0.9%), and the GeForce GTX 850M (9,302, -1%). In other words, the integrated Radeon part is competing with midrange discrete GPUs from a few generations back, and it is doing so while drawing far less power.

The GTX 460, by contrast, has nearest rivals that cluster around its own score: the NVIDIA Quadro K4100M (7,906, delta +0.2%), the Quadro P5000 (8,039, -1.4%), the GeForce GTX 880M (8,040, -1.4%), and the GeForce GTX 650 Ti (8,053, -1.6%). These are all older or lower-tier parts, and the 460 is essentially at parity with them. The data shows a generational chasm: the 890M outscores the GTX 460 by a margin that no single benchmark metric can fully capture, because the 890M also supports newer API features, higher pixel and texture rates, and a much more advanced architecture.

Looking at the raw throughput numbers, the 890M's pixel rate is 92.80 GPixel/s versus 9.450 GPixel/s for the GTX 460, a 9.8x difference. Texture rate is 185.6 GTexel/s versus 37.80 GTexel/s, a 4.9x difference. FP32 compute is 5.939 TFLOPS versus 907.2 GFLOPS, a 6.5x difference. These are not incremental improvements; they are orders of magnitude. The GTX 460 was competitive in its era, but the recorded data shows it cannot keep pace with even an integrated solution from 2024.

FAQ

Q: How much faster is the AMD Radeon 890M than the GTX 460 in Geekbench OpenCL?

A: The 890M scores 37,254 versus 7,925 for the GTX 460, a 370.1% advantage. This is the only benchmark where both products have recorded scores.

Q: What is the average benchmark score for each GPU?

A: The Radeon 890M averages 9,210 across all recorded tests, while the GTX 460 averages 7,925. The 890M's average sits 16.2% higher.

Q: Which GPU has the higher percentile ranking?

A: The Radeon 890M is at the 45th percentile among all GPUs in the database, while the GTX 460 is at the 41st percentile. Both are mid-pack parts, but the 890M sits above the GTX 460.

Q: Does the GTX 460 have any benchmark where it wins?

A: No. The head-to-head data shows one win for the AMD Radeon 890M and zero wins for the NVIDIA GeForce GTX 460.

Q: What are the closest rivals to each GPU?

A: For the 890M, the nearest rivals are the AMD Radeon Vega 8 (9,221, -0.1%), NVIDIA GeForce GTX 960 (9,273, -0.7%), NVIDIA GeForce GTX 465 (9,294, -0.9%), and NVIDIA GeForce GTX 850M (9,302, -1%). For the GTX 460, they are the NVIDIA Quadro K4100M (7,906, +0.2%), Quadro P5000 (8,039, -1.4%), GeForce GTX 880M (8,040, -1.4%), and GeForce GTX 650 Ti (8,053, -1.6%).

Q: What does the 890M's position among rivals suggest about its performance class?

A: The 890M's nearest rivals are all discrete GPUs from NVIDIA's 900 series or AMD's Vega family, which places it in the range of midrange discrete graphics from roughly 2014-2017. The GTX 460, by contrast, sits alongside mobile and older desktop parts like the Quadro K4100M and GTX 650 Ti.

Architecture Differences

The architectural gap between these two GPUs is enormous, spanning 14 years of processor design, fabrication, and feature evolution. The AMD Radeon 890M is built on the RDNA 3.5 architecture and uses the Strix Point chip, which places it in the Navi III IGP generation for Strix Point Mobile. It is fabricated on a 4 nm process at TSMC, with 34,000 million transistors packed into a 233 mm² die, giving a transistor density of 145.9 million per mm². The GTX 460, in contrast, uses the Fermi architecture with the GF104 chip, part of the GeForce 400 generation. It is built on a 40 nm process, also at TSMC, but with only 1,950 million transistors on a much larger 332 mm² die, yielding a density of just 5.9 million per mm². The density difference is a 24.7x gap, a direct consequence of the shift from 40 nm to 4 nm lithography.

The 890M integrates 1,024 shading units, 64 texture mapping units, and 32 ROPs, along with 16 dedicated ray tracing cores. The GTX 460 has 336 shading units, 56 TMUs, and 24 ROPs, with no ray tracing hardware at all. The 890M also supports DirectX 12 Ultimate (feature level 12_2), Vulkan 1.4, and OpenGL 4.6. The GTX 460 only reaches DirectX 12 (feature level 11_0) and OpenGL 4.6, with no Vulkan support listed. That DirectX feature level difference is critical: the 890M can run modern titles with ray tracing and mesh shaders, while the GTX 460 is limited to legacy rendering paths.

Memory architecture is another fundamental split. The 890M uses system shared memory, with bandwidth described as system dependent. The GTX 460 uses 768 MB of dedicated GDDR5 on a 192-bit bus, delivering 86.40 GB/s. The 890M's memory clock is listed as system shared, while the GTX 460 runs its GDDR5 at 900 MHz (3.6 Gbps effective). The shared memory approach of the 890M means its performance depends heavily on the host system's RAM configuration, an important caveat when interpreting its benchmark scores.

Power and physical design also diverge sharply. The 890M is an IGP with a 15 W TDP, no power connectors, and a slot width of IGP. The GTX 460 is a dual-slot card rated at 160 W, requiring two 6-pin power connectors and a 450 W suggested power supply. The 890M's display outputs are portable device dependent, while the GTX 460 offers 2x DVI and 1x mini-HDMI 1.3a. The 890M uses a PCIe 4.0 x8 interface, while the GTX 460 uses PCIe 2.0 x16. The 890M is a 4 nm integrated solution for mobile devices; the GTX 460 is a 40 nm discrete card for desktop towers.

Specification Differences

The two GPUs differ across nearly every recorded specification. Process node: 4 nm for the 890M versus 40 nm for the GTX 460. Transistor count: 34,000 million versus 1,950 million. Die size: 233 mm² versus 332 mm². Transistor density: 145.9M per mm² versus 5.9M per mm². Shading units: 1,024 versus 336. TMUs: 64 versus 56. ROPs: 32 versus 24. Ray tracing cores: 16 on the 890M, none on the GTX 460. Pixel rate: 92.80 GPixel/s versus 9.450 GPixel/s. Texture rate: 185.6 GTexel/s versus 37.80 GTexel/s. FP32 compute: 5.939 TFLOPS versus 907.2 GFLOPS. FP16: the 890M supports 5.939 TFLOPS (1:1), while the GTX 460 has no recorded FP16 capability.

Clock behavior differs as well. The 890M has a base clock of 400 MHz and a boost clock of 2900 MHz. The GTX 460 has no base or boost clocks listed, only a memory clock of 900 MHz (3.6 Gbps effective). The 890M's memory is system shared in size, type, bus width, and bandwidth. The GTX 460 uses 768 MB of GDDR5 on a 192-bit bus with 86.40 GB/s bandwidth. TDP: 15 W versus 160 W. Slot width: IGP versus dual-slot. Power connectors: none versus 2x 6-pin. Suggested PSU: not listed for the 890M, 450 W for the GTX 460. Bus interface: PCIe 4.0 x8 versus PCIe 2.0 x16. Display outputs: portable device dependent versus 2x DVI and 1x mini-HDMI 1.3a. DirectX support: 12 Ultimate (12_2) versus 12 (11_0). Vulkan: 1.4 on the 890M, none on the GTX 460. Production status: active versus end-of-life. Release date: 2024-07-14 versus 2010-07-11. Predecessor: Navi II IGP versus GeForce 200. Successor: none for the 890M, GeForce 500 for the GTX 460. The GTX 460 had a launch MSRP of 199 USD.

Dimensions also differ: the GTX 460 is 210 mm (8.3 inches) long, while the 890M has no recorded physical dimensions because it is an integrated part with no board of its own.

Where Each One Wins

The AMD Radeon 890M wins in every measurable category in this comparison. It has higher compute throughput, higher pixel and texture rates, a much newer architecture, support for modern API features like ray tracing and DirectX 12 Ultimate, and it does all of this at a 15 W TDP. The 890M is the clear choice for anyone building a thin-and-light laptop or a compact mobile system where integrated graphics must handle modern workloads, including gaming at reasonable settings and compute tasks like OpenCL acceleration. Its 45th percentile ranking among all GPUs means it sits in the middle of the entire database, but that is remarkable for an IGP. Its nearest rivals are discrete GPUs like the GTX 960 and GTX 465, which draw far more power and require dedicated cooling and power connectors.

The GTX 460 retains one niche advantage: it is a discrete card with dedicated GDDR5 memory, so its performance does not depend on system RAM configuration. For a legacy system where the user wants a drop-in card for older games or basic desktop acceleration, the GTX 460 could still function, but its 160 W TDP, dual-slot cooler, two 6-pin connectors, and 450 W suggested PSU make it a poor fit for modern builds. Its percentile ranking of 41 places it just below the 890M, and its nearest rivals are all parts that are themselves outdated.

The data indicates a single clear winner. The 890M is faster, more efficient, more feature-rich, and still in active production. The GTX 460 is end-of-life, lacks modern API support, and offers a fraction of the compute performance. For any new system, the 890M is the logical choice. For a retro build or a legacy upgrade, the GTX 460 could still serve, but it cannot compete on any benchmark recorded in the database. The one benchmark where both have scores, Geekbench OpenCL, shows a 370.1% gap. No other metric in the dataset suggests the GTX 460 is competitive. The architecture difference alone, 4 nm RDNA 3.5 versus 40 nm Fermi, explains most of the performance chasm, and the rest is accounted for by the 890M's higher shading unit count, ray tracing cores, and vastly superior fill rates.