GPU Comparison

The NVIDIA GeForce 920MX and the NVIDIA GeForce RTX 5080 SUPER are separated by nearly a decade of mobile and desktop GPU evolution, representing the extremes of the current benchmark database. The 920MX is an end-of-life entry-level mobile chip from the GeForce 900M generation, while the RTX 5080 SUPER is an active, top-tier desktop card from the GeForce 50-series. The data shows they share almost no common ground in performance, architecture, or target use case, making this comparison a study in technological contrast rather than a competitive matchup.

Where Each One Wins

Based strictly on the benchmark results and specifications in the data pack, the GeForce 920MX wins in exactly zero benchmark categories, while the RTX 5080 SUPER also wins zero head-to-head benchmarks because no direct comparison tests exist between the two. The 920MX's benchmark results come from Geekbench OpenCL and Vulkan tests, scoring 4,068 and 2,987 respectively, giving it an average benchmark score of 3,528. The RTX 5080 SUPER's sole benchmark is 3DMark Steel Nomad DX12, where it scores 3,075, placing its average benchmark score at 3,075.

The performance gap is not merely numerical but generational. The 920MX delivers 508.4 GFLOPS of FP32 compute, while the RTX 5080 SUPER delivers 56.28 TFLOPS, a difference of over 110 times. The 920MX's texture rate is 23.83 GTexel/s, compared to the RTX 5080 SUPER's 879.3 GTexel/s, and the pixel rate difference is similarly stark: 7.944 GPixel/s versus 293.1 GPixel/s. The 920MX has 256 shading units, 24 TMUs, and 8 ROPs; the RTX 5080 SUPER has 10,752 shading units, 336 TMUs, and 112 ROPs. The RTX 5080 SUPER also carries 84 RT cores and 336 tensor cores, features entirely absent from the Maxwell-based 920MX.

Where each one "wins" is therefore defined by use case rather than benchmark score. The 920MX wins in power efficiency and portability: its 16 W TDP requires no power connectors and fits an MXM module slot, making it suitable for thin, low-power laptops. The RTX 5080 SUPER wins in every raw performance metric, but requires a 415 W TDP, a 1x 16-pin power connector, and a dual-slot footprint measuring 304 mm in length, 137 mm in height, and 40 mm in width. The 920MX's display outputs are listed as "Portable Device Dependent," while the RTX 5080 SUPER offers 1x HDMI 2.1b and 3x DisplayPort 2.1b, indicating desktop multi-monitor capability.

The Verdict

The data dictates a clear verdict: the RTX 5080 SUPER is the only choice for any compute-intensive or modern gaming workload, while the 920MX is confined to legacy, low-power mobile applications. The RTX 5080 SUPER's 24 GB of GDDR7 memory on a 256-bit bus provides 1.02 TB/s of bandwidth, versus the 920MX's 2 GB of DDR3 on a 64-bit bus delivering 14.40 GB/s. That is a 70-fold bandwidth advantage, which directly impacts texture streaming, high-resolution rendering, and any memory-bound workload.

The RTX 5080 SUPER supports DirectX 12 Ultimate (12_2), while the 920MX is limited to DirectX 12 (11_0). Both support OpenGL 4.6 and Vulkan 1.4, but the RTX 5080 SUPER's Blackwell 2.0 architecture adds hardware ray tracing and tensor cores that the Maxwell architecture lacks entirely. For anyone selecting a GPU for current-generation gaming, content creation, or AI-adjacent workloads, the RTX 5080 SUPER is the only viable option from this data. The 920MX should be chosen only for scenarios where the 16 W TDP and MXM form factor are absolute requirements, and where performance expectations are minimal.

The percentile data reinforces this split. The 920MX sits at the 21st percentile of all GPUs, while the RTX 5080 SUPER sits at the 19th percentile. These similar percentile ranks are misleading, they reflect different benchmark suites with different scoring scales rather than comparable performance. The 920MX's nearest rivals include the GeForce GT 545 (avg score 3,594, deltaPct -1.8) and the RTX 5000 Mobile Ada Generation (avg score 3,596, deltaPct -1.9), while the RTX 5080 SUPER's nearest rivals include the Quadro P1000 (avg score 3,163, deltaPct -2.8) and the GeForce 820A (avg score 2,983, deltaPct 3.1). The rival lists confirm that each card competes in entirely different performance strata.

Head-to-Head Benchmarks

No direct head-to-head benchmark results exist in the data pack, so a side-by-side walkthrough of matched tests is impossible. However, the individual benchmark scores and compute specifications provide a basis for comparison. The 920MX achieves 4,068 in Geekbench OpenCL and 2,987 in Geekbench Vulkan. The RTX 5080 SUPER achieves 3,075 in 3DMark Steel Nomad DX12. These tests measure different workloads, Geekbench targets general-purpose compute and Vulkan API performance, while 3DMark Steel Nomad is a modern DX12 gaming benchmark, so direct score comparison is not meaningful.

What is meaningful is the compute throughput delta. The RTX 5080 SUPER's FP32 performance of 56.28 TFLOPS is 110.7 times the 920MX's 508.4 GFLOPS. The texture rate advantage is 36.9 times (879.3 vs 23.83 GTexel/s), and the pixel rate advantage is 36.9 times (293.1 vs 7.944 GPixel/s). The memory bandwidth advantage is 70.8 times (1.02 TB/s vs 14.40 GB/s). These ratios indicate that no workload tested in the 920MX's benchmarks could approach the RTX 5080 SUPER's floor. The 920MX's average benchmark score of 3,528 is actually higher than the RTX 5080 SUPER's average of 3,075, but this is an artifact of the different test suites, Geekbench OpenCL tends to produce higher numerical scores than 3DMark Steel Nomad on any hardware.

The RTX 5080 SUPER's 3DMark Steel Nomad score of 3,075 places it 2.8% behind the Quadro P1000 (3,163) and 3.4% behind the Intel Arc Pro B60 (3,182), while running 3.1% ahead of the GeForce 820A (2,983) and 3.6% ahead of the GeForce GTX 860M (2,967). The 920MX's OpenCL/Vulkan scores place it 1.8% behind the GeForce GT 545 (3,594), 1.9% behind the RTX 5000 Mobile Ada Generation (3,596), and 2.4% behind the GeForce GT 735M (3,616), while running 2.7% ahead of the Quadro 2000M (3,434). These deltaPct values show that each card sits within a tight cluster of similar-performing parts, confirming that the 920MX and RTX 5080 SUPER occupy separate performance brackets.

FAQ

Q: Which GPU has higher FP32 compute performance?

A: The RTX 5080 SUPER delivers 56.28 TFLOPS of FP32 compute, which is over 110 times the 920MX's 508.4 GFLOPS.

Q: How do the memory configurations compare?

A: The RTX 5080 SUPER has 24 GB of GDDR7 on a 256-bit bus with 1.02 TB/s bandwidth, while the 920MX has 2 GB of DDR3 on a 64-bit bus with 14.40 GB/s bandwidth.

Q: What are the architectural differences in ray tracing support?

A: The RTX 5080 SUPER's Blackwell 2.0 architecture includes 84 RT cores and 336 tensor cores, while the 920MX's Maxwell architecture has no RT or tensor cores listed.

Q: Which card supports newer DirectX features?

A: The RTX 5080 SUPER supports DirectX 12 Ultimate (12_2), while the 920MX supports DirectX 12 (11_0). Both support OpenGL 4.6 and Vulkan 1.4.

Q: What is the power requirement difference?

A: The 920MX has a 16 W TDP with no power connectors, while the RTX 5080 SUPER has a 415 W TDP and requires a 1x 16-pin power connector.

Q: How do their physical sizes differ?

A: The 920MX uses an MXM module slot with no listed dimensions, while the RTX 5080 SUPER is a dual-slot card measuring 304 mm in length, 137 mm in height, and 40 mm in width.

Architecture Differences

The two GPUs are built on fundamentally different architectures and process nodes. The 920MX uses the GM108S chip based on Maxwell architecture, fabricated on TSMC's 28 nm process. The RTX 5080 SUPER uses the GB203 chip based on Blackwell 2.0 architecture, fabricated on TSMC's 5 nm process. The transistor counts reflect the generational leap: the 920MX packs 1,020 million transistors on a 77 mm² die, yielding a density of 13.2M / mm², while the RTX 5080 SUPER packs 45,600 million transistors on a 378 mm² die, yielding a density of 120.6M / mm². That is a 44.7-fold increase in transistor count and a 9.1-fold increase in density.

The compute architectures differ in core composition. The 920MX has 256 shading units, 24 TMUs, and 8 ROPs, with no RT cores or tensor cores. The RTX 5080 SUPER has 10,752 shading units, 336 TMUs, and 112 ROPs, plus 84 RT cores and 336 tensor cores. The FP16 support also differs: the 920MX lists no FP16 capability, while the RTX 5080 SUPER provides 56.28 TFLOPS of FP16 with a 1:1 ratio to FP32. The memory subsystem is entirely different as well, the 920MX uses DDR3 with 900 MHz base memory clock (1,800 Mbps effective), while the RTX 5080 SUPER uses GDDR7 with 2,000 MHz base memory clock (32 Gbps effective).

The bus interfaces reflect their target platforms. The 920MX uses PCIe 3.0 x8, appropriate for mobile integration, while the RTX 5080 SUPER uses PCIe 5.0 x16 for maximum desktop bandwidth. The API support shows the 920MX is capped at DirectX 12 (11_0), whereas the RTX 5080 SUPER reaches DirectX 12 Ultimate (12_2). Both support OpenGL 4.6 and Vulkan 1.4, but the RTX 5080 SUPER's hardware features enable capabilities the Maxwell architecture cannot expose through those APIs.

Specification Differences

The specification table highlights every differing field between the two GPUs. The 920MX belongs to the GeForce 900M generation, while the RTX 5080 SUPER belongs to the GeForce 50-series. The process nodes are 28 nm versus 5 nm. Transistor counts are 1,020 million versus 45,600 million. Die sizes are 77 mm² versus 378 mm². Transistor densities are 13.2M / mm² versus 120.6M / mm².

Clock speeds differ substantially: the 920MX has a base clock of 965 MHz and a boost clock of 993 MHz, while the RTX 5080 SUPER has a base clock of 2,295 MHz and a boost clock of 2,617 MHz. Memory clocks are 900 MHz (1,800 Mbps effective) for the 920MX versus 2,000 MHz (32 Gbps effective) for the RTX 5080 SUPER. Memory size is 2 GB versus 24 GB, memory type is DDR3 versus GDDR7, bus width is 64 bit versus 256 bit, and bandwidth is 14.40 GB/s versus 1.02 TB/s.

Shading units, TMUs, and ROPs are 256, 24, and 8 versus 10,752, 336, and 112. The RT cores and tensor cores exist only on the RTX 5080 SUPER (84 and 336, respectively). Pixel rate is 7.944 GPixel/s versus 293.1 GPixel/s, texture rate is 23.83 GTexel/s versus 879.3 GTexel/s, and FP32 is 508.4 GFLOPS versus 56.28 TFLOPS. TDP is 16 W versus 415 W, slot width is MXM Module versus Dual-slot, and power connectors are None versus 1x 16-pin. The bus interface is PCIe 3.0 x8 versus PCIe 5.0 x16, and display outputs are Portable Device Dependent versus 1x HDMI 2.1b and 3x DisplayPort 2.1b. DirectX support is 12 (11_0) versus 12 Ultimate (12_2), and FP16 is null versus 56.28 TFLOPS (1:1). Release dates are 2016-03-24 versus 2025-12-31, and production status is End-of-life versus Active. The launch MSRP for the RTX 5080 SUPER is 999 USD; the 920MX has no listed launch MSRP.