NVIDIA GeForce GTX 1080 vs NVIDIA GeForce GTX 590 Comparison

# NVIDIA GeForce GTX 590 vs NVIDIA GeForce GTX 1080

The NVIDIA GeForce GTX 590, a Fermi 2.0-era dual-GPU flagship from the GeForce 500 generation, and the NVIDIA GeForce GTX 1080, a Pascal-based single-chip powerhouse from the GeForce 10 series, represent two distinct eras of NVIDIA engineering. In the single available head-to-head benchmark, the GTX 1080 dominates decisively, but the data tells a more nuanced story about architectural evolution and workload suitability. The GTX 1080 delivers a Geekbench OpenCL score of 51,204 against the GTX 590's 12,830, a 74.9% margin that underscores the generational leap, yet the GTX 590 still holds relevance in legacy DirectX 11 scenarios and compute tasks where its dual-chip design excels.

Where Each One Wins

The benchmark data is unambiguous: the GTX 1080 wins the sole head-to-head test, Geekbench OpenCL, with a score of 51,204 versus the GTX 590's 12,830. That 74.9% delta is not a narrow edge but a categorical victory. The GTX 1080's OpenCL performance places it in a different performance tier entirely, and this single metric aligns with its broader benchmark suite. Across its full benchmark portfolio, the GTX 1080 shows strength in DirectX 12 (PassMark score of 55), DirectX 11 (124), DirectX 10 (93), DirectX 9 (211), and compute workloads (PassMark GPU Compute score of 7,614). Its Geekbench Metal score of 23,824 and Vulkan score of 30,398 further indicate versatility across modern APIs.

The GTX 590, by contrast, has no benchmark wins in the head-to-head comparison. Its only recorded benchmark is Geekbench OpenCL at 12,830, which places it in the 52nd percentile of all GPUs — just above the GTX 1080's 51st percentile, though that percentile ranking reflects the GTX 1080's average score across many tests, not its OpenCL peak. The GTX 590's nearest rivals include the AMD Radeon Pro 455 (12,831, delta 0%), the AMD FirePro W5100 (12,847, delta -0.1%), and the AMD Radeon 740M (12,870, delta -0.3%), showing it sits in a tightly clustered mid-range performance band. The GTX 590's strength lies not in modern API performance but in its raw dual-GPU architecture, which was designed for multi-monitor setups and high-resolution gaming in its era, though the data available here does not quantify those scenarios.

Architecture Differences

The architectural gap between these two GPUs is immense. The GTX 590 is built on TSMC's 40 nm process node with 3,000 million transistors packed into a 520 mm² die, yielding a transistor density of 5.8M per mm². The GTX 1080 uses TSMC's 16 nm node, housing 7,200 million transistors in a 314 mm² die — a 22.9M per mm² density that represents a 4x improvement in packing efficiency. This density advantage allows the GTX 1080 to deliver far more compute in a smaller physical footprint.

The GTX 590's Fermi 2.0 architecture features 512 shading units, 64 texture mapping units, and 48 ROPs. Its memory subsystem comprises 1,536 MB of GDDR5 on a 384-bit bus, delivering 164.0 GB/s of bandwidth. The GTX 1080's Pascal architecture scales this dramatically: 2,560 shading units (5x more), 160 TMUs (2.5x more), and 64 ROPs. Memory jumps to 8 GB of GDDR5X on a 256-bit bus, with bandwidth nearly doubling to 320.3 GB/s. The GTX 1080's base clock of 1,607 MHz and boost clock of 1,733 MHz stand in contrast to the GTX 590's unspecified base/boost clocks, with only its memory clock listed at 854 MHz (3.4 Gbps effective).

Compute throughput tells the story most clearly. The GTX 590 delivers 1,244.2 GFLOPS of FP32 performance, while the GTX 1080 reaches 8.873 TFLOPS — a 7x increase. Pixel rate jumps from 19.46 GPixel/s to 110.9 GPixel/s, and texture rate from 38.91 GTexel/s to 277.3 GTexel/s. The GTX 1080 also supports FP16 at 138.6 GFLOPS (1:64 ratio), a feature the GTX 590 lacks entirely. API support reflects the generational divide: the GTX 590 supports DirectX 12 (11_0) and OpenGL 4.6, but no Vulkan, while the GTX 1080 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The GTX 1080 also uses PCIe 3.0 x16 versus the GTX 590's PCIe 2.0 x16, doubling interface bandwidth.

Power and physical characteristics diverge sharply. The GTX 590 draws 365 W TDP requiring a 750 W suggested PSU and dual 8-pin connectors; the GTX 1080 draws 180 W TDP with a 450 W suggested PSU and a single 8-pin connector. Both are dual-slot cards, with the GTX 590 measuring 279 mm (11 inches) long, 111 mm (4.4 inches) tall, and 40 mm (1.6 inches) wide — nearly identical to the GTX 1080's 267 mm (10.5 inches) length, 112 mm (4.4 inches) height, and 40 mm (1.6 inches) width. Display outputs differ: the GTX 590 offers 3x DVI and 1x mini-DisplayPort, while the GTX 1080 provides 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.4a.

Head-to-Head Benchmarks

The only direct comparison available is Geekbench OpenCL, and it is a rout. The GTX 1080 scores 51,204 against the GTX 590's 12,830, a delta of -74.9% from the GTX 1080's perspective. This means the GTX 590 achieves only about a quarter of the GTX 1080's OpenCL performance. In practical terms, the GTX 1080 completes the same OpenCL compute workload in roughly a quarter of the time, or alternatively, it processes nearly four times as many elements in the same period. This aligns with the FP32 throughput figures: 8.873 TFLOPS versus 1,244.2 GFLOPS is a 7.13x raw compute advantage, though real-world OpenCL scaling is less than linear due to memory bandwidth and driver overhead.

The GTX 590's score of 12,830 places it in a tight cluster with its nearest rivals — the AMD Radeon Pro 455 at 12,831 (0% delta), the AMD FirePro W5100 at 12,847 (-0.1%), and the AMD Radeon 740M at 12,870 (-0.3%) — while the GTX 670 trails at 12,773 (0.4%). This indicates that in OpenCL, the GTX 590 performs at a level comparable to entry-level professional GPUs from several generations later. The GTX 1080's score of 51,204, however, sits well above its nearest rivals in the average benchmark list: the GTX 960A at 11,998 (-0.3%), the AMD Radeon RX 6500 XT at 11,842 (1%), the GTX 1660 at 11,680 (2.4%), and the AMD Radeon RX 7800 XT at 11,627 (2.9%). Note that the GTX 1080's average benchmark score across all tests is 11,960, which is significantly lower than its OpenCL peak — this reflects the wide variance in its performance across different API and workload types.

The GTX 1080's broader benchmark suite reveals its versatility. Its PassMark G3D score of 15,586 is strong, while its DirectX 9 score of 211 and DirectX 11 score of 124 show strong legacy API support. The DirectX 12 score of 55 is notably lower, which may reflect the test's sensitivity to driver optimization rather than raw hardware capability. Its PassMark G2D score of 888 indicates solid 2D performance, and the GPU Compute score of 7,614 confirms compute-heavy workloads are a strength. The Geekbench Metal score of 23,824 and Vulkan score of 30,398 round out a comprehensive modern API profile.

FAQ

Q: Which GPU wins the head-to-head benchmark?

A: The GTX 1080 wins the sole head-to-head test, Geekbench OpenCL, with a score of 51,204 versus the GTX 590's 12,830, a 74.9% margin.

Q: How does the GTX 590 compare to its nearest rivals in OpenCL?

A: The GTX 590's 12,830 OpenCL score sits within 0.4% of the AMD Radeon Pro 455 (12,831), AMD FirePro W5100 (12,847), AMD Radeon 740M (12,870), and NVIDIA GeForce GTX 670 (12,773).

Q: What is the memory bandwidth difference between the two cards?

A: The GTX 1080 provides 320.3 GB/s of bandwidth from 8 GB of GDDR5X on a 256-bit bus, while the GTX 590 offers 164.0 GB/s from 1,536 MB of GDDR5 on a 384-bit bus.

Q: What are the power requirements for each card?

A: The GTX 590 has a 365 W TDP with a 750 W suggested PSU and dual 8-pin power connectors; the GTX 1080 has a 180 W TDP with a 450 W suggested PSU and a single 8-pin connector.

Q: Which card supports Vulkan?

A: The GTX 1080 supports Vulkan 1.4, while the GTX 590 has no Vulkan support listed. Both support DirectX 12, but the GTX 1080 uses version 12_1 versus the GTX 590's 11_0.

Q: How do their transistor counts and process nodes compare?

A: The GTX 590 uses 3,000 million transistors on a 40 nm TSMC process with a 520 mm² die; the GTX 1080 uses 7,200 million transistors on a 16 nm TSMC process with a 314 mm² die.

The Verdict

The data is unambiguous: the GTX 1080 is the superior GPU for virtually all modern workloads. Its 74.9% OpenCL lead over the GTX 590 is backed by a 7x advantage in FP32 compute (8.873 TFLOPS versus 1,244.2 GFLOPS), twice the memory bandwidth (320.3 GB/s versus 164.0 GB/s), and over five times the memory capacity (8 GB versus 1,536 MB). The GTX 1080 also draws half the power (180 W versus 365 W TDP), requires a less demanding PSU (450 W versus 750 W), and supports modern APIs including Vulkan 1.4 and DirectX 12_1. Its 16 nm process node delivers 22.9M transistors per mm² versus the GTX 590's 5.8M, enabling the performance leap in a smaller die.

The GTX 590's case rests entirely on its legacy position. Its 52nd percentile ranking among all GPUs is marginally higher than the GTX 1080's 51st percentile, but this is a statistical artifact of the GTX 1080's broader benchmark suite pulling its average down — the GTX 1080's OpenCL score alone is four times higher. The GTX 590's dual-GPU design was built for a 2011-era landscape of multi-monitor gaming and SLI-optimized titles, and its 3x DVI output configuration reflects that. The GTX 1080, with its HDMI 2.0 and three DisplayPort 1.4a outputs, is built for modern displays and VR. For anyone choosing between these two today, the GTX 1080 is the only reasonable pick: it wins the only direct benchmark, offers dramatically better efficiency, and supports the APIs that modern software requires. The GTX 590 is a historical artifact — an impressive one for its time, but one whose performance ceiling, power draw, and API limitations relegate it to a footnote in NVIDIA's evolution from Fermi to Pascal.