AMD Radeon RX 7650 GRE vs NVIDIA GeForce RTX 5090 Mobile Comparison
AMD Radeon RX 7650 GRE
GeForce RTX 5090 Mobile
PERFORMANCE BENCHMARKS
Analysis: AMD Radeon RX 7650 GRE vs NVIDIA GeForce RTX 5090 Mobile
The NVIDIA GeForce RTX 5090 Mobile and AMD Radeon RX 7650 GRE represent two very different interpretations of a modern GPU, one aimed at flagship mobile performance and the other at a more traditional desktop slot. The benchmark data shows a clear overall winner in raw compute, but the architecture and positioning of each card reveal distinct use cases. The RTX 5090 Mobile dominates the two shared benchmark tests by a wide margin, while the RX 7650 GRE counters with a more accessible physical design and a newer process node for its class.
Where Each One Wins
The data is unambiguous: the NVIDIA GeForce RTX 5090 Mobile wins every single head-to-head benchmark recorded. In the 3DMark Steel Nomad DX12 test, the RTX 5090 Mobile scores 5871 against the RX 7650 GRE's 2336, a 151.3% advantage. In Geekbench OpenCL, the gap is similarly lopsided, with the RTX 5090 Mobile posting 201834 versus 83109, a 142.9% delta. There are no benchmark categories where the AMD card comes out ahead.
However, the win condition for each card depends entirely on context. The RTX 5090 Mobile is a laptop-class part with a 95 W TDP and an IGP slot width, meaning it is designed to be soldered onto a motherboard in a portable system. Its 24 GB of GDDR7 memory and 896.0 GB/s of bandwidth make it a compute-heavy monster for tasks like 3D rendering or AI workloads. The RX 7650 GRE, by contrast, is a dual-slot desktop card with a 170 W TDP, a 1x 8-pin power connector, and a suggested PSU of 450 W. It is a drop-in component for a standard desktop build, and its 8 GB of GDDR6 memory is more modest but still capable for mainstream gaming at its intended resolution.
The percentile rankings tell a similar story. The RTX 5090 Mobile sits at the 84th percentile of all GPUs, while the RX 7650 GRE is at the 83rd percentile. This is a razor-thin difference in overall standing, despite the massive gap in raw scores. The reason is that the RTX 5090 Mobile's average benchmark score of 45152 is dragged down by its low power limits and thermal constraints in mobile chassis, while the RX 7650 GRE's average score of 42723 is more representative of its desktop performance. In short, the RTX 5090 Mobile wins on absolute performance, but the RX 7650 GRE wins on consistency within its form factor.
Architecture Differences
The architectural gulf between these two GPUs is vast. The RTX 5090 Mobile uses the GB203 chip on NVIDIA's Blackwell 2.0 architecture, fabricated on a 5 nm process at TSMC. It packs 45,600 million transistors onto a 378 mm² die, yielding a transistor density of 120.6M per mm². The RX 7650 GRE, meanwhile, is built on the Navi 33 chip with AMD's RDNA 3.0 architecture, codenamed Hotpink Bonefish. It uses a 6 nm process, also from TSMC, but has just 13,300 million transistors on a 204 mm² die, with a density of 65.2M per mm².
The compute resources are equally divergent. The RTX 5090 Mobile has 10496 shading units, 328 TMUs, and 112 ROPs, along with 82 ray tracing cores and 328 tensor cores. The RX 7650 GRE offers 2048 shading units, 128 TMUs, and 64 ROPs, with 32 ray tracing cores and no tensor cores at all. The FP32 throughput tells the story: 31.80 TFLOPS for the NVIDIA card versus 22.08 TFLOPS for the AMD card. Both have 1:1 FP16 ratios, so those numbers double for half-precision work, but the NVIDIA card's tensor cores give it a hardware advantage for AI inferencing that AMD cannot match on this chip.
Memory is another clear differentiator. The RTX 5090 Mobile has 24 GB of GDDR7 on a 256-bit bus, delivering 896.0 GB/s of bandwidth. The RX 7650 GRE has 8 GB of GDDR6 on a 128-bit bus, with 288.0 GB/s. Clock speeds flip the expected narrative: the AMD card runs at a 1720 MHz base and 2695 MHz boost, with a 2350 MHz game clock, while the NVIDIA card is heavily power-limited with a 990 MHz base and 1515 MHz boost. Yet the NVIDIA card's massive shader count overwhelms the AMD card's higher clocks. The memory clocks are 1750 MHz (28 Gbps effective) for NVIDIA versus 2250 MHz (18 Gbps effective) for AMD.
Physical and interface differences further separate them. The RTX 5090 Mobile is an IGP (integrated graphics processor) with no power connectors and a PCIe 5.0 x16 interface, while the RX 7650 GRE is a dual-slot card measuring 204 mm in length and 115 mm in height, with a PCIe 4.0 x8 interface. The AMD card has a 1x 8-pin power connector and a suggested PSU of 450 W, plus display outputs of 1x HDMI 2.1a and 3x DisplayPort 2.1. The NVIDIA card's display outputs are listed as "Portable Device Dependent," meaning connectivity is determined by the laptop manufacturer. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.
Head-to-Head Benchmarks
The first shared test is 3DMark Steel Nomad DX12, a demanding DirectX 12 workload. The RTX 5090 Mobile scores 5871, which is 151.3% higher than the RX 7650 GRE's 2336. This is not a marginal win; it is a complete rout. The NVIDIA card's combination of 10496 shading units and 896.0 GB/s of bandwidth allows it to feed its compute units far more effectively than the AMD card's 2048 shading units and 288.0 GB/s can manage. Even though the AMD card boosts to 2695 MHz versus the NVIDIA card's 1515 MHz, the shader count difference of roughly 5x overwhelms the clock speed advantage.
The second shared test is Geekbench OpenCL, which measures general-purpose compute. The RTX 5090 Mobile scores 201834, a 142.9% improvement over the RX 7650 GRE's 83109. This benchmark is particularly revealing because it stresses raw FP32 throughput and memory bandwidth. The NVIDIA card's 31.80 TFLOPS of FP32 performance is 44% higher than the AMD card's 22.08 TFLOPS, and its memory bandwidth is more than 3x higher. The result is a score that is more than double the AMD card's output. The RTX 5090 Mobile also has a Geekbench Vulkan score of 198405 and a Passmark G3D score of 30034, though no equivalent data exists for the RX 7650 GRE in those tests.
Looking at the broader benchmark suite for the RTX 5090 Mobile, its Passmark scores show a mixed profile: a G2D score of 1057 is modest, but the GPU Compute score of 13401 is strong. The RX 7650 GRE has no additional benchmark data beyond the two shared tests, which limits direct comparison but does not change the outcome of the head-to-head. The deltaPct values in the nearestRivals data for the RTX 5090 Mobile show it trading places with cards like the AMD Radeon Pro 5500 XT (-0.5%) and NVIDIA GeForce RTX 4070 Ti (+0.8%), while the RX 7650 GRE is bracketed by the RTX 4070 SUPER (-1.2%) and RTX 5050 Mobile (-1.3%). This suggests the RTX 5090 Mobile competes with desktop-class GPUs, while the RX 7650 GRE sits in a lower performance tier.
FAQ
Q: Which GPU is faster in 3DMark Steel Nomad DX12?
A: The NVIDIA GeForce RTX 5090 Mobile scores 5871, which is 151.3% higher than the AMD Radeon RX 7650 GRE's 2336. This is the largest delta in the shared benchmarks.
Q: How much memory does each card have?
A: The RTX 5090 Mobile has 24 GB of GDDR7 on a 256-bit bus, while the RX 7650 GRE has 8 GB of GDDR6 on a 128-bit bus. Memory bandwidth is 896.0 GB/s for NVIDIA and 288.0 GB/s for AMD.
Q: What are the power requirements for the RX 7650 GRE?
A: The RX 7650 GRE has a 170 W TDP, requires a 1x 8-pin power connector, and has a suggested PSU of 450 W. The RTX 5090 Mobile has a 95 W TDP and uses no power connectors because it is an IGP.
Q: Do both cards support the same APIs?
A: Yes, both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. However, the RTX 5090 Mobile has 328 tensor cores for AI workloads, while the RX 7650 GRE has none.
Q: How do the cards compare in overall GPU rankings?
A: The RTX 5090 Mobile is at the 84th percentile of all GPUs, with an average benchmark score of 45152. The RX 7650 GRE is at the 83rd percentile, with an average score of 42723. They are nearly tied despite the huge performance gap in shared tests.
Q: What is the process node difference?
A: The RTX 5090 Mobile uses a 5 nm process at TSMC, while the RX 7650 GRE uses a 6 nm process. The NVIDIA chip has 45,600 million transistors, versus 13,300 million for AMD.
The Verdict
The data is one-sided. The NVIDIA GeForce RTX 5090 Mobile is the clear performance winner, taking both shared benchmarks with deltas above 140%. Its 24 GB of GDDR7 memory, 10496 shading units, and 31.80 TFLOPS of FP32 throughput make it a superior choice for any workload that demands raw compute, whether that is 3D rendering, machine learning inference, or high-resolution gaming on a laptop. The 82 ray tracing cores and 328 tensor cores give it hardware features that the RX 7650 GRE simply lacks.
The AMD Radeon RX 7650 GRE, however, is not without a case. Its 83rd percentile ranking is nearly identical to the RTX 5090 Mobile's 84th, and as a dual-slot desktop card with a 204 mm length and a 1x 8-pin power connector, it fits into a standard PC build without the constraints of a mobile chassis. Its 170 W TDP is higher than the RTX 5090 Mobile's 95 W, but that power budget is used to sustain a 2695 MHz boost clock, which helps it stay competitive in its own tier. The RX 7650 GRE also has a launch MSRP of 279 USD, which is a factual data point, though its performance relative to the RTX 5090 Mobile does not justify a head-to-head purchase decision.
For a laptop buyer, the RTX 5090 Mobile is the only option here, and the benchmark data shows it delivers desktop-class performance in a 95 W envelope. For a desktop builder who prioritizes ease of installation and a moderate power footprint, the RX 7650 GRE is a sensible choice, but the performance gap is enormous. The verdict is straightforward: choose the RTX 5090 Mobile if you need maximum compute, and choose the RX 7650 GRE if you need a conventional desktop GPU with a lower absolute score but a similar percentile ranking. The numbers do not lie — this is a 151.3% differential in one test and a 142.9% differential in the other.