AMD Radeon 890M vs NVIDIA GeForce GTX 760 Comparison

The NVIDIA GeForce GTX 760 and AMD Radeon 890M represent two vastly different eras of GPU design. The GTX 760 is a 2013 discrete desktop card built on the 28 nm Kepler architecture, while the 890M is a 2024 integrated graphics processor (IGP) on the 4 nm RDNA 3.5 architecture, designed for mobile Strix Point processors. Despite the generational and form-factor gap, their average benchmark scores are remarkably close: the GTX 760 averages 9458 points, while the 890M averages 9210 points, a difference of roughly 2.7%. This places both in similar percentiles (46th and 45th respectively), but their performance characteristics diverge sharply in specific workloads, particularly compute-heavy APIs.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce GTX 760 has a slightly higher average benchmark score of 9458, compared to the AMD Radeon 890M's 9210. This puts the GTX 760 at the 46th percentile of all GPUs, while the 890M sits at the 45th percentile.

Q: How do the two GPUs compare in the Geekbench OpenCL test?

A: The AMD Radeon 890M wins decisively in Geekbench OpenCL, scoring 37254 against the GTX 760's 11299. This represents a 69.7% advantage for the 890M, showcasing its superiority in general-purpose compute workloads.

Q: What is the performance difference in the Geekbench Vulkan test?

A: The AMD Radeon 890M again dominates, scoring 40808 in Geekbench Vulkan compared to the GTX 760's 12551. The delta is 69.2%, indicating a massive lead for the integrated GPU in modern graphics API performance.

Q: Which GPU is a closer rival to the GTX 760 based on average score?

A: The NVIDIA GeForce GTX TITAN BLACK is the nearest rival to the GTX 760, with an average score of 9474, which is 0.2% lower. Other close rivals include the AMD Radeon R7 M380 (1.6% lower) and the NVIDIA GeForce GTX 850M (1.7% lower).

Q: What is the direct competitor to the AMD Radeon 890M in the benchmark database?

A: The AMD Radeon Vega 8 is the closest rival, with an average score of 9221, which is 0.1% higher than the 890M. The NVIDIA GeForce GTX 960 is also close, scoring 9273, which is 0.7% higher.

Q: What are the manufacturing process differences between the two?

A: The GTX 760 uses a 28 nm process at TSMC, while the 890M uses a much more advanced 4 nm process, also at TSMC. This results in a significantly higher transistor density for the 890M: 145.9M / mm² versus 12.0M / mm² for the GTX 760.

Architecture Differences

The architectural chasm between these two GPUs is profound. The GTX 760 is built on the Kepler architecture, using the GK104 chip, while the 890M employs the RDNA 3.5 architecture on the Strix Point chip. The GTX 760 is a discrete, dual-slot card from the GeForce 700 generation, whereas the 890M is an integrated graphics processor from the Navi III IGP (Strix Point Mobile) generation.

The most striking difference is in the transistor count and density. The GTX 760 houses 3,540 million transistors on a 294 mm² die, while the 890M packs 34,000 million transistors onto a smaller 233 mm² die. This is made possible by the 4 nm process node, a massive leap from the 28 nm node used by the GTX 760. Consequently, the 890M's transistor density is over 12 times higher, enabling far more complex compute capabilities in a smaller footprint.

The 890M also brings modern features that the GTX 760 lacks entirely. It includes 16 ray tracing cores, a feature absent from the older Kepler design. Additionally, the 890M supports DirectX 12 Ultimate (12_2) and Vulkan 1.4, while the GTX 760 is limited to DirectX 12 (11_0) and Vulkan 1.2.175. The 890M's FP16 performance is 5.939 TFLOPS (1:1), whereas the GTX 760 has no listed FP16 capability, only FP32 at 2.378 TFLOPS.

Memory architecture is another fundamental divide. The GTX 760 has dedicated 2 GB of GDDR5 on a 256-bit bus, delivering 192.3 GB/s of bandwidth. The 890M uses System Shared memory, meaning its bandwidth is "System Dependent" and its bus width is "System Shared." This shared memory design is typical for IGPs, trading dedicated bandwidth for flexibility and lower power consumption. The GTX 760's TDP is 170 W, while the 890M sips power at just 15 W, a difference that reflects their intended use cases.

Where Each One Wins

The GTX 760 wins in scenarios that favor its dedicated memory and older, but still relevant, compute strengths. Its 192.3 GB/s of fixed bandwidth and 256-bit bus give it a deterministic advantage in bandwidth-sensitive tasks where the 890M's shared memory might become a bottleneck. The benchmark data shows the GTX 760's average score of 9458 edges out the 890M's 9210, indicating that in aggregate, the discrete card remains slightly ahead in overall throughput as measured by the database's average. This suggests the GTX 760 may hold up better in legacy DirectX 9/10/11 workloads, though specific tests for those are not in the head-to-head data.

The AMD Radeon 890M wins decisively in modern, compute-heavy and API-advanced workloads. Its performance in Geekbench OpenCL (37254 vs 11299) and Geekbench Vulkan (40808 vs 12551) is nothing short of dominant, with margins of 69.7% and 69.2% respectively. This indicates the 890M is far superior for general-purpose GPU computing, modern game engines that leverage Vulkan, and any workload that can utilize its ray tracing cores. Its 5.939 TFLOPS FP32 and FP16 performance, combined with a 2900 MHz boost clock, make it a powerhouse for its 15 W power envelope.

The 890M also wins in efficiency and portability. Its 4 nm process and 15 W TDP mean it can operate in thin-and-light laptops with no power connectors, whereas the GTX 760 requires a 450 W power supply and dual 6-pin connectors. For users needing a capable GPU in a compact, power-efficient system, the 890M is the clear choice. The GTX 760, being end-of-life and dual-slot, is relegated to older desktop builds.

Specification Differences

The specification tables reveal stark contrasts in nearly every field. The process node differs from 28 nm (GTX 760) to 4 nm (890M). The GTX 760 uses the GK104 chip from the Kepler architecture, while the 890M is based on the Strix Point chip from RDNA 3.5. Transistor counts are 3,540 million versus 34,000 million, and die sizes are 294 mm² versus 233 mm².

Clock speeds are radically different. The GTX 760 has a base clock of 980 MHz and a boost clock of 1032 MHz, with memory clocked at 1502 MHz (6 Gbps effective). The 890M has a much lower base clock of 400 MHz but a significantly higher boost clock of 2900 MHz, and its memory clock is "System Shared." The GTX 760's memory configuration is fixed at 2 GB GDDR5 with a 256-bit bus and 192.3 GB/s bandwidth, while the 890M's memory is entirely system-dependent.

Compute unit counts also diverge. The GTX 760 has 1152 shading units, 96 TMUs, and 32 ROPs. The 890M has 1024 shading units, 64 TMUs, and 32 ROPs, but adds 16 RT cores. Pixel and texture rates reflect the newer architecture's efficiency: the 890M delivers 92.80 GPixel/s and 185.6 GTexel/s, compared to the GTX 760's 24.77 GPixel/s and 99.07 GTexel/s. FP32 performance is 2.378 TFLOPS for the GTX 760 versus 5.939 TFLOPS for the 890M.

Power and physical specs are opposites. The GTX 760 has a 170 W TDP, is dual-slot, and requires 2x 6-pin power connectors with a 450 W suggested PSU. The 890M has a 15 W TDP, is an IGP, and requires no power connectors. The GTX 760 measures 241 mm (9.5 inches) in length, while the 890M has no listed dimensions. Bus interfaces differ: PCIe 3.0 x16 for the GTX 760 versus PCIe 4.0 x8 for the 890M. The GTX 760 has fixed display outputs (2x DVI, 1x HDMI 1.4a, 1x DisplayPort 1.2), while the 890M's outputs are "Portable Device Dependent."

Head-to-Head Benchmarks

The head-to-head data is limited to two tests, but both are decisive wins for the AMD Radeon 890M. In Geekbench OpenCL, the 890M scores 37254 against the GTX 760's 11299, a delta of -69.7% from the perspective of the GTX 760. This means the 890M is roughly 3.3 times faster in this compute test. The result highlights the massive leap in general-purpose compute efficiency enabled by the RDNA 3.5 architecture and its high boost clock.

In Geekbench Vulkan, the 890M again wins with a score of 40808 versus the GTX 760's 12551, a delta of -69.2%. This test leverages modern graphics APIs, where the 890M's support for Vulkan 1.4 and ray tracing cores gives it a substantial advantage. The GTX 760's older Vulkan 1.2.175 support and lack of RT cores likely contribute to its lower score.

These results are consistent with the average benchmark scores, though the averages tell a different story. The GTX 760's average of 9458 is slightly higher than the 890M's 9210, suggesting that in other, unspecified benchmark tests (likely older DirectX or legacy workloads), the GTX 760 performs better. However, the two available head-to-head tests show the 890M's dominance in modern compute and graphics APIs. The 890M's win count in head-to-head benchmarks is 2, while the GTX 760 has 0, but the overall average scores remain close due to the GTX 760's strength in other areas not captured in this direct comparison.