AMD Radeon 840M vs NVIDIA GeForce RTX 5090 D Comparison

Head-to-Head Benchmarks

The recorded data contains no direct head-to-head benchmark comparisons between the AMD Radeon 840M and the NVIDIA GeForce RTX 5090 D. The database lists zero wins for the AMD part and zero wins for the NVIDIA part in direct matchup testing. This absence of comparative scores means the two GPUs cannot be ranked against each other through a single shared test run.

The NVIDIA GeForce RTX 5090 D does have a full benchmark profile in the database. Its 3DMark Steel Nomad DX12 score is 14,326 points. In Geekbench OpenCL, it records 310,674 points, and in Geekbench Vulkan, 376,915 points. Passmark results show a G3D score of 44,065 and a GPU Compute score of 28,396. The average benchmark score across all recorded tests is 77,712 points, placing the card in the 92nd percentile of all GPUs in the database.

The AMD Radeon 840M has no recorded benchmark scores and no average benchmark score. Its percentile ranking sits at 50, which reflects its position as a mid-tier integrated graphics solution, but the absence of any actual test data prevents a quantitative comparison with the RTX 5090 D.

For context, the RTX 5090 D's nearest rivals in the database include the AMD Radeon RX 6650M XT with an average score of 76,904 points, placing the RTX 5090 D 1.1% ahead of that card. The AMD Radeon RX 6850M XT averages 78,940 points, putting the RTX 5090 D 1.6% behind. The NVIDIA Tesla P100 PCIe 12 GB and 16 GB variants score 79,396 and 79,605 points respectively, with the RTX 5090 D trailing by 2.1% and 2.4% in those matchups. These figures indicate the RTX 5090 D sits in a competitive performance band among high-end GPUs, but no such comparative data exists for the Radeon 840M.

Architecture Differences

The two GPUs occupy opposite ends of the hardware spectrum. The AMD Radeon 840M is an integrated graphics processor built on the Krackan Point chip, using RDNA 3.5 architecture from the Navi III IGP generation. It is fabricated on a 4 nm process at TSMC. The NVIDIA GeForce RTX 5090 D uses the GB202 chip with Blackwell 2.0 architecture from the GeForce 50 generation, manufactured on a 5 nm process, also at TSMC.

Core configuration differences are substantial. The Radeon 840M contains 256 shading units, 16 texture mapping units, 8 render output units, and 4 ray tracing cores. It has no tensor cores listed. The RTX 5090 D contains 21,760 shading units, 680 texture mapping units, 176 render output units, 170 ray tracing cores, and 680 tensor cores. This represents a difference of 21,504 shading units, 664 TMUs, 168 ROPs, 166 RT cores, and 680 tensor cores in favor of the NVIDIA part.

Clock speeds also diverge. The Radeon 840M runs at a base clock of 400 MHz and a boost clock of 2,900 MHz. The RTX 5090 D operates at a base clock of 2,017 MHz and a boost clock of 2,407 MHz. While the AMD part has a higher boost clock, the NVIDIA card's base clock is substantially higher.

Memory architecture could not be more different. The Radeon 840M uses system shared memory, with system-dependent bandwidth and no dedicated VRAM. The RTX 5090 D has 32 GB of GDDR7 memory on a 512-bit bus, delivering 1.79 TB/s of bandwidth. Memory clock for the NVIDIA card is 1,750 MHz, listed as 28 Gbps effective. The Radeon 840M has no dedicated memory clock, relying on the system memory subsystem.

Compute throughput numbers show the scale of the gap. The Radeon 840M delivers 1,484.8 GFLOPS of FP32 performance and the same value for FP16, at a 1:1 ratio. The RTX 5090 D delivers 104.8 TFLOPS of FP32 and 104.8 TFLOPS of FP16, also at 1:1. Pixel rate for the AMD part is 23.20 GPixel/s, while the NVIDIA card reaches 423.6 GPixel/s. Texture rates are 46.40 GTexel/s versus 1,636.8 GTexel/s.

Transistor counts and die sizes are only listed for the RTX 5090 D: 92,200 million transistors on a 750 mm² die, with a transistor density of 122.9 million per square millimeter. The Radeon 840M's transistor count and die size are listed as unknown. Power consumption differs by a factor of over 38: the Radeon 840M has a TDP of 15 W, while the RTX 5090 D has a TDP of 575 W. The NVIDIA card requires a single 16-pin power connector and a suggested power supply of 950 W, while the AMD part uses no power connectors and is an IGP.

The bus interface differs as well: PCIe 4.0 x8 for the AMD part versus PCIe 5.0 x16 for the NVIDIA card. Display outputs on the RTX 5090 D include one HDMI 2.1b and three DisplayPort 2.1b connectors, while the Radeon 840M's display outputs are listed as portable device dependent. Physical dimensions are only recorded for the NVIDIA card: 304 mm in length, 137 mm in height, and 48 mm in width, occupying a dual-slot form factor. The AMD part is an IGP with no listed dimensions.

Where Each One Wins

Based on the recorded data, the NVIDIA GeForce RTX 5090 D wins in every measurable performance category. Its FP32 and FP16 throughput of 104.8 TFLOPS dwarfs the Radeon 840M's 1,484.8 GFLOPS. The pixel rate advantage is 423.6 GPixel/s versus 23.20 GPixel/s, and the texture rate advantage is 1,636.8 GTexel/s versus 46.40 GTexel/s. The RTX 5090 D has dedicated GDDR7 memory with 1.79 TB/s bandwidth, while the Radeon 840M relies on system shared memory with system-dependent bandwidth.

The RTX 5090 D also wins on feature set. It has 680 tensor cores for AI acceleration, while the Radeon 840M has none listed. The NVIDIA card has 170 ray tracing cores versus 4 on the AMD part. The 32 GB VRAM capacity enables workloads that the system-shared memory of the Radeon 840M cannot support. The RTX 5090 D's PCIe 5.0 x16 interface provides more bandwidth than the Radeon 840M's PCIe 4.0 x8 connection.

The Radeon 840M does hold advantages in specific areas. Its 15 W TDP makes it suitable for portable devices, and its IGP form factor requires no separate power connectors or cooling solution beyond what the system provides. The AMD part's 4 nm process node is smaller than the NVIDIA card's 5 nm node, which contributes to its lower power draw. The Radeon 840M's boost clock of 2,900 MHz exceeds the RTX 5090 D's boost clock of 2,407 MHz, though this does not translate into a performance advantage given the massive core count difference.

The RTX 5090 D's benchmark scores confirm its dominance in synthetic workloads. Its Passmark G3D score of 44,065 and GPU Compute score of 28,396, combined with the Geekbench Vulkan score of 376,915, place it in the 92nd percentile of all GPUs in the database. The Radeon 840M has no benchmark scores recorded, but its 50th percentile ranking suggests it sits at the median of all GPUs, which is consistent with an integrated graphics solution.

The Verdict

The data presents an unambiguous performance hierarchy. The NVIDIA GeForce RTX 5090 D is a dedicated high-end discrete GPU with 21,760 shading units, 32 GB of GDDR7 memory, and 104.8 TFLOPS of FP32 compute. The AMD Radeon 840M is an integrated GPU with 256 shading units, no dedicated memory, and 1,484.8 GFLOPS of compute. These are not competing products in the same class.

The RTX 5090 D's benchmark results place it in the 92nd percentile of all GPUs, with an average benchmark score of 77,712 points. Its nearest rivals in the database, the AMD Radeon RX 6650M XT and RX 6850M XT, sit within 1.6% of its average score, indicating the RTX 5090 D performs at a similar level to those mobile discrete GPUs. The Radeon 840M has no comparable data.

For workloads requiring high compute throughput, large memory capacity, or dedicated ray tracing and tensor cores, the RTX 5090 D is the only choice between these two. For low-power portable systems where energy efficiency and integration are priorities, the Radeon 840M serves a different purpose entirely. The 575 W TDP of the RTX 5090 D versus the 15 W TDP of the Radeon 840M illustrates that these products target different use cases.

The RTX 5090 D launched on January 29, 2025, with a launch MSRP of 2,299 USD. The Radeon 840M released on February 28, 2025, with no launch MSRP recorded. Both are listed as active in production. The NVIDIA card has a successor listed as GeForce 60, while the Radeon 840M's predecessor is Navi II IGP.

FAQ

Q: Which GPU has higher FP32 compute performance?

A: The NVIDIA GeForce RTX 5090 D delivers 104.8 TFLOPS of FP32 performance, while the AMD Radeon 840M delivers 1,484.8 GFLOPS, a difference of over 70 times in favor of the NVIDIA card.

Q: What memory configurations do these GPUs use?

A: The RTX 5090 D has 32 GB of GDDR7 memory on a 512-bit bus with 1.79 TB/s bandwidth. The Radeon 840M uses system shared memory with system-dependent bandwidth and no dedicated VRAM.

Q: How do their power requirements compare?

A: The Radeon 840M has a TDP of 15 W and requires no power connectors. The RTX 5090 D has a TDP of 575 W, requires one 16-pin power connector, and has a suggested power supply of 950 W.

Q: What are their ray tracing capabilities?

A: The RTX 5090 D has 170 ray tracing cores, while the Radeon 840M has 4 ray tracing cores.

Q: Do either of these GPUs have tensor cores for AI workloads?

A: The RTX 5090 D has 680 tensor cores. The Radeon 840M has no tensor cores listed in the database.

Q: What process nodes are used for each GPU?

A: The Radeon 840M is fabricated on a 4 nm process at TSMC, while the RTX 5090 D is fabricated on a 5 nm process at TSMC.