AMD Radeon 820M vs NVIDIA GeForce RTX 5090 Comparison
AMD Radeon 820M
GeForce RTX 5090
PERFORMANCE BENCHMARKS
Analysis: AMD Radeon 820M vs NVIDIA GeForce RTX 5090
The Verdict
The AMD Radeon 820M and NVIDIA GeForce RTX 5090 occupy opposite ends of the GPU spectrum. The Radeon 820M is an integrated graphics processor with a 15 W TDP, designed for portable devices where power draw and space are constrained. The RTX 5090 is a dual-slot discrete flagship with a 575 W TDP and a 950 W suggested PSU, intended for maximum performance in desktop systems. The database records no head-to-head benchmark wins for the Radeon 820M, while the RTX 5090 holds all recorded wins. The RTX 5090 sits at the 92nd percentile against all GPUs in the database, while the Radeon 820M sits at the 50th percentile. Any comparison between these two is a comparison of capability tiers, not direct competition. The Radeon 820M serves systems that cannot accommodate a discrete card, while the RTX 5090 serves workloads that demand the highest recorded scores. The choice is dictated by the host platform, not by benchmark preference.
Architecture Differences
The Radeon 820M uses the Krackan Point 2 chip, built on RDNA 3.5 architecture, and belongs to the Navi III IGP generation for Strix Point Mobile. It is fabricated on a 4 nm process at TSMC. The RTX 5090 uses the GB202 chip, built on Blackwell 2.0 architecture, and belongs to the GeForce 50 generation. It is fabricated on a 5 nm process at TSMC. The Radeon 820M's transistor count and die size are recorded as unknown, while the RTX 5090 has 92,200 million transistors on a 750 mm² die, giving a transistor density of 122.9M per mm².
The Radeon 820M has 128 shading units, 8 texture mapping units, and 4 ROPs. It includes 2 RT cores and no tensor cores. The RTX 5090 has 21,760 shading units, 680 texture mapping units, and 176 ROPs. It includes 170 RT cores and 680 tensor cores. The Radeon 820M's base clock is 400 MHz with a boost clock of 2800 MHz. The RTX 5090's base clock is 2017 MHz with a boost clock of 2407 MHz, and its memory runs at 1750 MHz with 28 Gbps effective speed.
Memory architecture is fundamentally different. The Radeon 820M uses system shared memory with system dependent bandwidth, meaning its performance scales with the host system's RAM. The RTX 5090 has 32 GB of dedicated GDDR7 memory on a 512 bit bus, delivering 1.79 TB/s of bandwidth. The Radeon 820M has no power connectors and uses the PCIe 4.0 x8 interface. The RTX 5090 uses a single 16-pin power connector and the PCIe 5.0 x16 interface. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The Radeon 820M's display outputs are portable device dependent, while the RTX 5090 provides 1x HDMI 2.1b and 3x DisplayPort 2.1b.
Head-to-Head Benchmarks
The database records no head-to-head benchmark entries between these two parts, and the win counts reflect that: the Radeon 820M has 0 wins and the RTX 5090 has 0 wins in direct comparison. However, the RTX 5090 has a substantial set of individual benchmark scores that illustrate its performance tier. In 3DMark Steel Nomad DX12, the RTX 5090 scores 18,355. In Geekbench OpenCL, it scores 334,370, and in Geekbench Vulkan, it scores 376,728. PassMark results show a G3D score of 39,650, a GPU compute score of 26,756, and a G2D score of 1,413. DirectX-specific PassMark tests show scores of 226 for DX10, 341 for DX11, 185 for DX12, and 395 for DX9.
The Radeon 820M has no recorded benchmark scores in the database, so its measured performance cannot be compared numerically. The RTX 5090's average benchmark score is 79,842, which places it among the top tier of GPUs. Its nearest rivals in the database include the NVIDIA Tesla P100 PCIe 16 GB at 79,605 average score, a 0.3% delta, and the NVIDIA Tesla P100 PCIe 12 GB at 79,396, a 0.6% delta. The AMD Radeon RX 6850M XT scores 78,940, a 1.1% delta below the RTX 5090. The AMD Radeon Pro Vega 64X scores 80,959, a 1.4% delta above the RTX 5090. These deltas indicate the RTX 5090 is closely matched with these workstation and mobile flagships, with differences under 1.5% in either direction.
Specification Differences
The two parts differ in nearly every recorded specification field. The Radeon 820M is an IGP with a 15 W TDP, while the RTX 5090 is a dual-slot card with a 575 W TDP. The Radeon 820M has no power connectors; the RTX 5090 uses 1x 16-pin. The Radeon 820M's bus interface is PCIe 4.0 x8; the RTX 5090 uses PCIe 5.0 x16. The Radeon 820M's memory is system shared; the RTX 5090 has 32 GB of GDDR7. The Radeon 820M's bus width is system shared; the RTX 5090 has a 512 bit bus. Bandwidth is system dependent for the Radeon 820M and 1.79 TB/s for the RTX 5090.
Compute resources differ by orders of magnitude. The Radeon 820M has 128 shading units, 8 TMUs, and 4 ROPs. The RTX 5090 has 21,760 shading units, 680 TMUs, and 176 ROPs. The Radeon 820M has 2 RT cores and no tensor cores; the RTX 5090 has 170 RT cores and 680 tensor cores. Pixel rate is 11.20 GPixel/s for the Radeon 820M versus 423.6 GPixel/s for the RTX 5090. Texture rate is 22.40 GTexel/s versus 1,636.8 GTexel/s. FP32 performance is 716.8 GFLOPS versus 104.8 TFLOPS, and FP16 is 716.8 GFLOPS (1:1) versus 104.8 TFLOPS (1:1).
The process nodes differ: 4 nm for the Radeon 820M and 5 nm for the RTX 5090. Transistor count and die size are unknown for the Radeon 820M but are 92,200 million and 750 mm² for the RTX 5090. The Radeon 820M has no recorded dimensions, while the RTX 5090 measures 304 mm in length, 137 mm in height, and 40 mm in width. The Radeon 820M has no launch MSRP; the RTX 5090 has a launch MSRP of 1,999 USD. Release dates differ: the Radeon 820M released on 2025-02-28, and the RTX 5090 released on 2025-01-29. The Radeon 820M's predecessor is Navi II IGP, while the RTX 5090's predecessor is GeForce 40 and its successor is GeForce 60.
FAQ
Q: Which GPU has more shading units?
A: The RTX 5090 has 21,760 shading units, while the Radeon 820M has 128 shading units.
Q: What memory does each GPU use?
A: The Radeon 820M uses system shared memory with system dependent bandwidth. The RTX 5090 uses 32 GB of GDDR7 on a 512 bit bus with 1.79 TB/s bandwidth.
Q: How do the power requirements compare?
A: The Radeon 820M has a 15 W TDP and no power connectors. The RTX 5090 has a 575 W TDP, uses a 1x 16-pin connector, and requires a 950 W suggested PSU.
Q: What are the process nodes for each chip?
A: The Radeon 820M uses a 4 nm process at TSMC. The RTX 5090 uses a 5 nm process at TSMC.
Q: Does the Radeon 820M support ray tracing?
A: Yes, it has 2 RT cores and supports DirectX 12 Ultimate (12_2). The RTX 5090 has 170 RT cores with the same DirectX feature level.
Q: What is the RTX 5090's average benchmark score relative to its nearest rivals?
A: The RTX 5090 scores 79,842 on average. It is 0.3% above the Tesla P100 PCIe 16 GB, 0.6% above the Tesla P100 PCIe 12 GB, 1.1% above the RX 6850M XT, and 1.4% below the Radeon Pro Vega 64X.
Where Each One Wins
The Radeon 820M wins in power efficiency and physical footprint. Its 15 W TDP means it can operate in portable devices without discrete power connectors, and its IGP slot width means it occupies no expansion slot. It uses the PCIe 4.0 x8 interface, which is sufficient for an integrated part, and its display outputs are portable device dependent, reflecting its intended use in laptops and compact systems. The 4 nm process node is smaller than the RTX 5090's 5 nm node, which contributes to its low power profile. The Radeon 820M's system shared memory approach eliminates the need for dedicated VRAM, simplifying the host design.
The RTX 5090 wins in every performance category recorded in the database. Its 104.8 TFLOPS FP32 performance is orders of magnitude above the Radeon 820M's 716.8 GFLOPS. Its 423.6 GPixel/s pixel rate and 1,636.8 GTexel/s texture rate dwarf the Radeon 820M's 11.20 GPixel/s and 22.40 GTexel/s. The RTX 5090's 32 GB GDDR7 memory with 1.79 TB/s bandwidth provides a dedicated memory pool that the Radeon 820M cannot match with system shared memory. The RTX 5090's 680 tensor cores enable AI workloads that the Radeon 820M cannot handle, as it has no tensor cores. The RTX 5090 also has 170 RT cores versus the Radeon 820M's 2.
The RTX 5090's benchmark scores demonstrate its performance class. Its 3DMark Steel Nomad score of 18,355, Geekbench OpenCL score of 334,370, and Geekbench Vulkan score of 376,728 represent the upper tier of recorded results. Its PassMark G3D score of 39,650 and GPU compute score of 26,756 confirm its compute capability. The 92nd percentile ranking against all GPUs places it among the fastest parts in the database. The Radeon 820M sits at the 50th percentile with no recorded benchmark scores, which reflects its position as an entry-level integrated solution rather than a performance part.
For a portable device that needs basic graphics output with minimal power draw, the Radeon 820M is the appropriate choice. For a desktop system that needs maximum compute, ray tracing, and memory bandwidth, the RTX 5090 is the appropriate choice. The data does not support any other conclusion.