NVIDIA GeForce GT 745M vs NVIDIA GeForce RTX 5080 SUPER Comparison

Head-to-Head Benchmarks

The benchmark data in this comparison is unusual: the two GPUs were not tested on a common set of workloads. The NVIDIA GeForce GT 745M has recorded scores in three compute-oriented tests, while the NVIDIA GeForce RTX 5080 SUPER has a single DirectX 12 gaming benchmark result. Direct cross-test comparisons are therefore not possible. Instead, the database shows each GPU against its own nearest rivals, which reveals how far apart these two products really are in performance class.

The GT 745M posts scores of 2777 in Geekbench Metal, 3580 in Geekbench OpenCL, and 5502 in Geekbench Vulkan. Its average benchmark score is 3953, placing it at the 23rd percentile of all GPUs in the database. That percentile ranking means roughly three-quarters of all recorded desktop and mobile graphics processors outperform it. The nearest rivals confirm this position: the AMD Radeon R5 M420 scores 3956, a delta of -0.1% relative to the GT 745M, and the NVIDIA GeForce 830M scores 3957, also -0.1%. The NVIDIA Quadro K2000 is 0.3% ahead at 3964, while the NVIDIA Quadro 2000D trails by 0.6% at 3930. These deltas are all within a single percentage point, meaning the GT 745M sits in a very tight performance cluster of entry-level mobile GPUs from the same era.

The RTX 5080 SUPER, by contrast, records a 3DMark Steel Nomad DX12 score of 3075. Its average benchmark score is identical at 3075, and it sits at the 19th percentile of all GPUs. This percentile figure is counterintuitive at first glance, but the data is clear: the nearest rivals are not modern flagship parts. The NVIDIA Quadro P1000 scores 3163, which is 2.8% higher. The Intel Arc Pro B60 scores 3182, 3.4% higher. The NVIDIA GeForce 820A scores 2983, 3.1% lower, and the NVIDIA GeForce GTX 860M scores 2967, 3.6% lower. The RTX 5080 SUPER is therefore bracketed by a professional workstation card, an Intel professional GPU, and two older mobile gaming chips, all within a 6.4% spread. This indicates that in the database's current benchmark set, the RTX 5080 SUPER's recorded result does not reflect its architectural position; rather, it appears in a sparse data region where few comparable high-end results exist.

The biggest win for the GT 745M is its Vulkan score of 5502, which is substantially higher than its own OpenCL score of 3580 and its Metal score of 2777. That Vulkan figure suggests the Kepler architecture, despite its age, handles compute workloads in that API with relative efficiency. The biggest win for the RTX 5080 SUPER is less about its absolute score and more about the context: it is only 2.8% behind the Quadro P1000 and 3.4% behind the Arc Pro B60 in a modern DX12 test, while the GT 745M's nearest rivals are all within 0.6% of its score. The RTX 5080 SUPER is competing in a different league entirely, even if the percentile data does not currently show it.

Where Each One Wins

The GT 745M wins in scenarios where its specific API strengths matter. Its Geekbench Vulkan score of 5502 is its best result, and it exceeds its OpenCL score by 53.7% (5502 versus 3580). For users running Vulkan-based compute or legacy applications that leverage that API, the GT 745M is clearly the stronger of its own results. Its Metal score of 2777 is the weakest of the three, indicating that macOS or Metal-based workloads would not extract the same performance. The OpenCL score of 3580 sits in between, making it a moderate compute performer for its class. All three scores place the GPU in the 23rd percentile, so the practical takeaway is that this is a low-end mobile part best suited to lightweight, older, or API-specific workloads.

The RTX 5080 SUPER wins in all modern gaming and compute scenarios by virtue of its architecture and specifications, though the single recorded benchmark does not capture that breadth. Its 3DMark Steel Nomad DX12 score of 3075 is the only data point, but it is a modern DirectX 12 Ultimate test, which is more demanding than anything the GT 745M could run. The RTX 5080 SUPER also has dedicated ray tracing cores (84) and tensor cores (336), features the GT 745M lacks entirely. In any workload that uses those hardware units, the RTX 5080 SUPER has an unassailable advantage. The GT 745M cannot execute ray-traced or tensor-accelerated tasks at all, so the RTX 5080 SUPER wins by default in those categories.

For use-case splitting, the GT 745M is a fit for legacy 32-bit or early DirectX 11-era titles, given its DirectX 12 (11_0) API support. The RTX 5080 SUPER, with DirectX 12 Ultimate (12_2), Vulkan 1.4, and OpenGL 4.6, is built for current-generation games and future titles. The RTX 5080 SUPER also wins decisively in memory bandwidth: 1.02 TB/s versus 64.00 GB/s, a 16-fold difference. That bandwidth advantage matters for high-resolution textures and data-intensive compute. The GT 745M's 2 GB GDDR5 memory is a hard ceiling for modern game assets, while the RTX 5080 SUPER's 24 GB GDDR7 provides ample headroom.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The GT 745M has an average benchmark score of 3953 across three tests, while the RTX 5080 SUPER has an average score of 3075 from a single test. However, these scores come from different benchmark suites and cannot be directly compared as a measure of overall performance.

Q: How does each GPU compare to its nearest rivals?

A: The GT 745M is effectively tied with its nearest rivals: the AMD Radeon R5 M420 is 0.1% lower, the NVIDIA GeForce 830M is 0.1% lower, the NVIDIA Quadro K2000 is 0.3% higher, and the NVIDIA Quadro 2000D is 0.6% lower. The RTX 5080 SUPER has a wider spread: the Quadro P1000 is 2.8% higher, the Intel Arc Pro B60 is 3.4% higher, the GeForce 820A is 3.1% lower, and the GTX 860M is 3.6% lower.

Q: What is the percentile ranking of each GPU?

A: The GT 745M sits at the 23rd percentile of all GPUs in the database. The RTX 5080 SUPER sits at the 19th percentile. Both are in the lower quartile, but the RTX 5080 SUPER's ranking is based on only one benchmark result, which may not represent its true standing.

Q: Does the RTX 5080 SUPER support ray tracing?

A: Yes, the RTX 5080 SUPER has 84 dedicated ray tracing cores. The GT 745M has no ray tracing cores, so it cannot perform hardware-accelerated ray tracing.

Q: Which GPU has more shading units?

A: The RTX 5080 SUPER has 10,752 shading units, compared to the GT 745M's 384 shading units. This is a 28-fold difference in raw shader count.

Q: What memory configurations do the two GPUs use?

A: The GT 745M uses 2 GB of GDDR5 memory on a 128-bit bus, yielding 64.00 GB/s bandwidth. The RTX 5080 SUPER uses 24 GB of GDDR7 memory on a 256-bit bus, yielding 1.02 TB/s bandwidth.

Specification Differences

The two GPUs differ in nearly every measurable specification. The GT 745M has 384 shading units, 32 texture mapping units, and 16 raster output units. The RTX 5080 SUPER has 10,752 shading units, 336 TMUs, and 112 ROPs. The RTX 5080 SUPER also adds 84 ray tracing cores and 336 tensor cores, neither of which exists on the GT 745M.

Memory configuration is another major split. The GT 745M has 2 GB of GDDR5 on a 128-bit bus with 64.00 GB/s bandwidth. The RTX 5080 SUPER has 24 GB of GDDR7 on a 256-bit bus with 1.02 TB/s bandwidth. Memory clock differs as well: the GT 745M runs at 1000 MHz (4 Gbps effective), while the RTX 5080 SUPER runs at 2000 MHz (32 Gbps effective).

Clock speeds are only specified for the RTX 5080 SUPER, with a base of 2295 MHz and a boost of 2617 MHz. The GT 745M has no base or boost clock recorded in the database. Pixel and texture rates follow the hardware counts: the GT 745M achieves 4.392 GPixel/s and 17.57 GTexel/s, while the RTX 5080 SUPER achieves 293.1 GPixel/s and 879.3 GTexel/s. FP32 compute is 421.6 GFLOPS for the GT 745M versus 56.28 TFLOPS for the RTX 5080 SUPER. The RTX 5080 SUPER also has FP16 performance at 56.28 TFLOPS (1:1), while the GT 745M has no recorded FP16 figure.

Power and physical specifications diverge sharply. The GT 745M has a 45 W TDP and is an integrated graphics processor (IGP) with no power connectors. The RTX 5080 SUPER has a 415 W TDP, is dual-slot, and requires a single 16-pin power connector. Bus interface differs: PCIe 3.0 x16 for the GT 745M versus PCIe 5.0 x16 for the RTX 5080 SUPER. Display outputs are portable-device-dependent for the GT 745M, while the RTX 5080 SUPER offers 1x HDMI 2.1b and 3x DisplayPort 2.1b. The RTX 5080 SUPER has physical dimensions of 304 mm length, 137 mm height, and 40 mm width; the GT 745M has no recorded dimensions.

Architecture Differences

The GT 745M is built on the Kepler architecture using the GK107 chip, manufactured on a 28 nm process at TSMC. It contains 1,270 million transistors on a die size of 118 mm², giving a transistor density of 10.8 million per mm². The RTX 5080 SUPER uses the Blackwell 2.0 architecture with the GB203 chip, manufactured on a 5 nm process at TSMC. It contains 45,600 million transistors on a die size of 378 mm², yielding a transistor density of 120.6 million per mm². The density difference is 11.2-fold, reflecting the much newer process technology.

The GT 745M belongs to the GeForce 700M generation and supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. Its production status is end-of-life, and it was released in 2013, succeeding the GeForce 600M and preceding the GeForce 800M. The RTX 5080 SUPER belongs to the GeForce 50-series with Blackwell 2.0, supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. It is active in production, released in 2025, and has no recorded predecessor or successor.

The architecture gap is generational. Kepler (2013) predates ray tracing and tensor core hardware entirely. Blackwell 2.0 (2025) includes dedicated RT cores (84) and tensor cores (336), and its FP16 throughput matches FP32 at a 1:1 ratio, a capability the Kepler chip does not expose. The GT 745M's 28 nm process is five generations behind the 5 nm node used for the RTX 5080 SUPER. The transistor count difference is 36-fold (1,270 million versus 45,600 million), and the die size difference is 3.2-fold (118 mm² versus 378 mm²), yet the RTX 5080 SUPER packs over 11 times more transistors per square millimeter.

The Verdict

The data in the database makes the performance hierarchy unambiguous. The GT 745M is an end-of-life, 2013-era entry-level mobile GPU with 384 shading units, 2 GB GDDR5, and a 45 W TDP. Its benchmark scores place it in a tight cluster with other low-end parts like the Radeon R5 M420 and GeForce 830M, all within 0.6% of each other. The RTX 5080 SUPER is an active, 2025-era high-end desktop GPU with 10,752 shading units, 24 GB GDDR7, 84 RT cores, and a 415 W TDP. Even though its single recorded benchmark score of 3075 places it at the 19th percentile, that result is bracketed by professional and older mobile parts, not by modern high-end GPUs. The specification gap is so large that no workload in the database would favor the GT 745M over the RTX 5080 SUPER, except for the narrow case of a benchmark suite that only the older GPU was tested on.

Who should pick the GT 745M? Anyone maintaining a legacy 2013-era laptop or needing a GPU that runs DirectX 11-level workloads with minimal power draw. Its 45 W TDP and IGP form factor make it suitable for thin, low-power portable devices. Its Vulkan score of 5502 is respectable for its class, so users stuck with this hardware can still run Vulkan-based applications at modest settings. It is not a gaming GPU by modern standards, and its 23rd percentile ranking reflects that.

Who should pick the RTX 5080 SUPER? Anyone building a current-generation desktop for gaming, ray tracing, or tensor-accelerated compute. The hardware is unambiguous: 84 RT cores, 336 tensor cores, 56.28 TFLOPS FP32, and 1.02 TB/s memory bandwidth. The 3DMark Steel Nomad DX12 score of 3075 is the only recorded result, but the architectural capabilities far exceed anything the GT 745M can offer. The RTX 5080 SUPER is also dual-slot with a 16-pin connector, meaning it requires a full-size desktop chassis and adequate power delivery. The launch MSRP is 999 USD, and the production status is active, so it is a current purchase option. The GT 745M has no launch MSRP recorded, as it was an OEM mobile part.

In short, the GT 745M is a historical artifact of the Kepler era, while the RTX 5080 SUPER is a modern Blackwell part with a 36-fold transistor advantage and hardware features that did not exist in 2013. The database shows them at different percentiles (23 versus 19), but those rankings are artifacts of sparse benchmark coverage. The specification and architecture data tells the real story: there is no overlap in capability, target platform, or intended workload. Choose the GT 745M only for legacy compatibility; choose the RTX 5080 SUPER for anything current or future.