AMD Radeon 890M vs NVIDIA GeForce GTX 780M Comparison

Where Each One Wins

The AMD Radeon 890M is the clear winner in every measured head-to-head benchmark, but the two GPUs serve very different use cases. The Radeon 890M dominates in compute-oriented workloads, while the NVIDIA GeForce GTX 780M holds its own in legacy API scenarios and remains relevant for older software ecosystems.

The Radeon 890M wins both shared benchmarks decisively. In Geekbench OpenCL, it scores 37,254 versus 12,769 for the GTX 780M, a 65.7% advantage. In Geekbench Vulkan, the margin is even larger: 40,808 versus 12,696, a 68.9% lead. These are not close contests; the AMD part is roughly three times faster in both tests.

The GTX 780M has no benchmark wins in the head-to-head data. Its only advantage lies in its compatibility profile. It supports Vulkan 1.2.175 and DirectX 12 (11_0), while the Radeon 890M supports Vulkan 1.4 and DirectX 12 Ultimate (12_2). The older NVIDIA part also carries a 50th percentile ranking across all GPUs, slightly ahead of the Radeon 890M's 45th percentile. That percentile gap reflects the GTX 780M's stronger standing relative to the entire GPU database, despite losing the direct comparisons.

For users running modern compute workloads, the Radeon 890M is the obvious choice. Its FP32 throughput of 5.939 TFLOPS more than doubles the GTX 780M's 2.448 TFLOPS. The AMD part also offers 16 ray tracing cores, a feature the Kepler-based NVIDIA GPU lacks entirely. Games and applications that leverage DirectX 12 Ultimate features will only run properly on the Radeon 890M.

The GTX 780M's use case is narrower: legacy DirectX 11 titles, OpenGL 4.6 applications, and scenarios where its 160.0 GB/s dedicated memory bandwidth matters. Its 4 GB of GDDR5 memory is fixed and dedicated, unlike the Radeon 890M's system-shared memory, which depends entirely on the host system's RAM configuration.

Architecture Differences

The architectural gap between these two GPUs spans over a decade of development. The GTX 780M uses NVIDIA's Kepler architecture on a 28 nm process from TSMC. The Radeon 890M uses AMD's RDNA 3.5 architecture on a 4 nm process, also from TSMC. That process shrink enables enormous transistor density differences: the Radeon 890M packs 34,000 million transistors into a 233 mm² die, while the GTX 780M contains 3,540 million transistors on a 294 mm² die. The density figures tell the story: 145.9 million transistors per mm² for AMD versus 12.0 million per mm² for NVIDIA.

The Radeon 890M's chip is codenamed Strix Point and belongs to the Navi III IGP generation for mobile Strix Point platforms. The GTX 780M uses the GK104 chip from the GeForce 700M generation. The Radeon 890M is an integrated graphics processor (IGP), while the GTX 780M is a discrete MXM module.

Clock speeds differ dramatically. The GTX 780M runs at a base clock of 771 MHz with a boost of 797 MHz. The Radeon 890M has a much lower base clock of 400 MHz but boosts to 2,900 MHz. The AMD part's higher boost clock, combined with its newer architecture, drives its substantial performance advantage.

Memory architecture is fundamentally different. The GTX 780M has 4 GB of dedicated GDDR5 memory on a 256-bit bus, delivering 160.0 GB/s of bandwidth. The Radeon 890M uses system-shared memory with system-dependent bandwidth. This means the AMD GPU's memory performance varies with the host system's RAM, while the NVIDIA part has fixed, predictable bandwidth.

Feature support diverges sharply. The Radeon 890M includes 16 ray tracing cores, supports DirectX 12 Ultimate (12_2), and offers Vulkan 1.4. The GTX 780M has no ray tracing hardware, supports DirectX 12 (11_0) only, and tops out at Vulkan 1.2.175. Both GPUs support OpenGL 4.6.

The Radeon 890M's FP16 throughput matches its FP32 at 5.939 TFLOPS (1:1 ratio), a feature the GTX 780M does not list. The AMD part also uses a PCIe 4.0 x8 bus interface, while the GTX 780M connects via MXM-B (3.0).

Head-to-Head Benchmarks

The recorded head-to-head data contains only two shared benchmarks, and the Radeon 890M wins both by overwhelming margins.

In Geekbench OpenCL, the Radeon 890M scores 37,254 against the GTX 780M's 12,769. The delta is 65.7% in favor of AMD. This is the smaller of the two gaps, yet it still represents roughly three times the performance. OpenCL workloads that scale with compute unit count and clock speed heavily favor the AMD part, which combines a 2,900 MHz boost clock with a modern RDNA 3.5 architecture.

Geekbench Vulkan shows an even wider split. The Radeon 890M posts 40,808, while the GTX 780M manages 12,696. The 68.9% delta puts the AMD GPU at over three times the NVIDIA part's score. Vulkan is a low-overhead API, and the Radeon 890M's newer architecture handles it far more efficiently. The GTX 780M's Vulkan support stops at version 1.2.175, which may also limit its performance in newer Vulkan workloads.

The average benchmark scores from the wider database paint a more nuanced picture. The GTX 780M averages 11,261 across all its recorded benchmarks, while the Radeon 890M averages 9,210. This seems contradictory given the head-to-head results, but it reflects the different test suites each GPU was subjected to. The GTX 780M's nearest rivals include the AMD Radeon Pro WX 3200 at 11,228 (0.3% delta) and the AMD FirePro W4300 at 11,225 (0.3% delta). The Radeon 890M's nearest rivals include the AMD Radeon Vega 8 at 9,221 (-0.1% delta) and the NVIDIA GeForce GTX 960 at 9,273 (-0.7% delta).

The Radeon 890M's individual benchmark results vary widely. It scores 8,076 in Passmark G3D, 4,157 in Passmark GPU Compute, and 979 in Passmark G2D. Its DirectX-specific Passmark scores are low: 97 in DirectX 9, 73 in DirectX 11, 37 in DirectX 12, and 33 in DirectX 10. These low scores contrast sharply with its strong Geekbench results, suggesting the Radeon 890M's performance is highly workload-dependent. Its 3DMark Steel Nomad DX12 score of 590 is modest, but that test targets modern gaming workloads where the IGP's system-shared memory may bottleneck.

Specification Differences

The two GPUs differ in nearly every measurable specification. Manufacturing process: 28 nm for the GTX 780M versus 4 nm for the Radeon 890M. Transistor count: 3,540 million versus 34,000 million. Die size: 294 mm² versus 233 mm². Transistor density: 12.0M per mm² versus 145.9M per mm².

Clock speeds: the GTX 780M runs at 771 MHz base and 797 MHz boost, while the Radeon 890M runs at 400 MHz base and 2,900 MHz boost. Memory: the GTX 780M has 4 GB of GDDR5 on a 256-bit bus with 160.0 GB/s bandwidth; the Radeon 890M has system-shared memory with system-dependent bandwidth.

Compute resources: the GTX 780M has 1,536 shading units, 128 TMUs, and 32 ROPs. The Radeon 890M has 1,024 shading units, 64 TMUs, and 32 ROPs. Despite fewer shading units and TMUs, the Radeon 890M achieves far higher pixel and texture rates: 92.80 GPixel/s and 185.6 GTexel/s versus 25.50 GPixel/s and 102.0 GTexel/s.

FP32 performance: 2.448 TFLOPS for NVIDIA versus 5.939 TFLOPS for AMD. FP16: the GTX 780M has no listed FP16 figure, while the Radeon 890M matches its FP32 at 5.939 TFLOPS (1:1). Ray tracing cores: none for NVIDIA, 16 for AMD. TDP: 122 W for the GTX 780M versus 15 W for the Radeon 890M.

Form factor: MXM module for the GTX 780M versus IGP for the Radeon 890M. Bus interface: MXM-B (3.0) versus PCIe 4.0 x8. Power connectors: none for either. Display outputs: portable device dependent for both.

API support: DirectX 12 (11_0), OpenGL 4.6, Vulkan 1.2.175 for NVIDIA; DirectX 12 Ultimate (12_2), OpenGL 4.6, Vulkan 1.4 for AMD. Production status: end-of-life for the GTX 780M, active for the Radeon 890M. Release dates: May 10, 2013 for NVIDIA, July 14, 2024 for AMD.

FAQ

Q: Which GPU is faster in Geekbench OpenCL?

A: The AMD Radeon 890M scores 37,254 versus 12,769 for the NVIDIA GeForce GTX 780M, a 65.7% advantage.

Q: Does the GTX 780M support ray tracing?

A: No. The GTX 780M has no ray tracing cores, while the Radeon 890M includes 16 ray tracing cores.

Q: How do their power requirements compare?

A: The GTX 780M has a TDP of 122 W, while the Radeon 890M has a TDP of 15 W. The Radeon 890M draws far less power.

Q: Which GPU has better API support?

A: The Radeon 890M supports DirectX 12 Ultimate (12_2) and Vulkan 1.4. The GTX 780M supports DirectX 12 (11_0) and Vulkan 1.2.175. Both support OpenGL 4.6.

Q: What memory configurations do these GPUs use?

A: The GTX 780M has 4 GB of dedicated GDDR5 memory on a 256-bit bus with 160.0 GB/s bandwidth. The Radeon 890M uses system-shared memory with system-dependent bandwidth.

Q: How do their average benchmark scores compare to nearby GPUs?

A: The GTX 780M averages 11,261, placing it near the AMD Radeon Pro WX 3200 (11,228) and AMD FirePro W4300 (11,225). The Radeon 890M averages 9,210, near the AMD Radeon Vega 8 (9,221) and NVIDIA GeForce GTX 960 (9,273).

The Verdict

The AMD Radeon 890M is the superior GPU for modern workloads. It wins both head-to-head benchmarks by margins of 65.7% and 68.9%, offers more than double the FP32 throughput, and adds ray tracing support and newer API versions. Its 15 W TDP makes it dramatically more power-efficient than the GTX 780M's 122 W. Users running current games, DirectX 12 Ultimate titles, or Vulkan applications should choose the Radeon 890M without hesitation.

The NVIDIA GeForce GTX 780M retains value only in specific legacy scenarios. Its 50th percentile ranking across all GPUs edges out the Radeon 890M's 45th percentile, and its dedicated 4 GB GDDR5 memory with 160.0 GB/s bandwidth provides fixed performance that does not depend on host system RAM. Users with older software that relies on DirectX 11 or earlier APIs may find the GTX 780M adequate, but its end-of-life production status and lack of ray tracing make it a poor choice for future-proofing.

The data is unambiguous: the Radeon 890M represents a generational leap over the GTX 780M in raw compute, API support, and efficiency. The GTX 780M's only remaining strengths are its dedicated memory and slightly higher percentile ranking, neither of which compensates for its massive performance deficit in the shared benchmarks. For any new purchase or upgrade, the AMD Radeon 890M is the only sensible option based on the recorded measurements.