NVIDIA GeForce GTX 480 vs NVIDIA GeForce GTX 590 Comparison

The NVIDIA GeForce GTX 480 and the NVIDIA GeForce GTX 590 represent two distinct approaches to high-end graphics within the Fermi architecture lineage. The GTX 480, built on the original GF100 chip, was the flagship of the GeForce 400 generation. Its successor, the GTX 590, utilizes the GF110 chip and belongs to the GeForce 500 series. While both cards are now end-of-life products, their benchmark data reveals a surprising outcome that defies the generational naming convention. The analysis below examines the specifications and the single benchmark score available to determine where each card excels and which one emerges as the statistical victor.

FAQ

Q: Which card has a higher raw compute performance (FP32)?

A: The NVIDIA GeForce GTX 480 leads in FP32 performance with 1,345.0 GFLOPS, while the NVIDIA GeForce GTX 590 trails slightly with 1,244.2 GFLOPS.

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

A: The GTX 480 has a memory bandwidth of 177.4 GB/s, which is higher than the GTX 590's 164.0 GB/s. Both cards use GDDR5 memory on a 384-bit bus, but the GTX 480 runs its memory at a faster effective speed of 3.7 Gbps compared to the GTX 590's 3.4 Gbps.

Q: How do the two cards compare in terms of shading units and texture mapping units (TMUs)?

A: The GTX 590 has a higher count of both: 512 shading units and 64 TMUs, compared to the GTX 480's 480 shading units and 60 TMUs. Despite this, the GTX 480 achieves higher pixel and texture rates, at 21.03 GPixel/s and 42.06 GTexel/s respectively, versus the GTX 590's 19.46 GPixel/s and 38.91 GTexel/s.

Q: Which card has a higher power consumption and what power connectors does it require?

A: The GTX 590 has a significantly higher thermal design power (TDP) of 365 W, requiring two 8-pin power connectors. The GTX 480 has a TDP of 250 W and uses a single 6-pin and a single 8-pin connector. Consequently, the GTX 590 has a higher recommended power supply rating of 750 W, while the GTX 480 suggests a 600 W unit.

Q: In the benchmark data, which card performs better, and by how much?

A: The GTX 480 wins the single Geekbench OpenCL benchmark, scoring 13,300 points against the GTX 590's 12,830 points. This represents a 3.7% performance advantage for the GTX 480.

Q: How does each card rank against all other GPUs in the database?

A: The GTX 480 holds a percentile rank of 53, placing it slightly above the GTX 590, which has a percentile rank of 52. This aligns with the benchmark scores, where the GTX 480 has an average score of 13,300 versus 12,830 for the GTX 590.

Where Each One Wins

Based on the available benchmark results, the NVIDIA GeForce GTX 480 is the clear winner in the compute arena. It secures the only head-to-head victory, posting a 3.7% higher score in the Geekbench OpenCL test. This indicates that for workloads that leverage OpenCL compute tasks, the older GTX 480 offers a tangible performance edge over its newer counterpart. The data shows the GTX 480 also has superior theoretical throughput rates, including higher pixel fill rate (21.03 GPixel/s vs 19.46 GPixel/s) and texture fill rate (42.06 GTexel/s vs 38.91 GTexel/s), suggesting it may be better suited for tasks that are fill-rate limited.

On the other hand, the NVIDIA GeForce GTX 590 does not win in any benchmark category from the data. Its wins are purely on the specification sheet. The GTX 590 boasts a higher number of shading units (512 vs 480) and TMUs (64 vs 60), which theoretically provides more parallel processing capacity. It also features a more robust display output configuration, with three DVI ports and a mini-DisplayPort, compared to the GTX 480's two DVI ports and a single mini-HDMI. For a user prioritizing multiple simultaneous display connections, the GTX 590's output options are a clear advantage.

Architecture Differences

The two cards are built on different iterations of the Fermi architecture. The GTX 480 is based on the original "Fermi" architecture using the GF100 chip, while the GTX 590 utilizes the refined "Fermi 2.0" architecture with the GF110 chip. This architectural revision is a key differentiator, even though both are manufactured on the same 40 nm process node at TSMC. The GTX 590's GF110 chip is slightly smaller, with a die size of 520 mm², and contains 3,000 million transistors, compared to the GTX 480's 529 mm² die with 3,100 million transistors. This results in a nearly identical transistor density of 5.8M / mm² for the GTX 590 and 5.9M / mm² for the GTX 480. The shift to Fermi 2.0 in the GTX 590 appears to have allowed for more shading units and TMUs within a slightly smaller die. However, the benchmark data suggests these architectural changes did not translate into higher compute performance, as the GTX 480's GF100 chip delivers a higher FP32 throughput. Both cards support DirectX 12 (11_0) and OpenGL 4.6, and neither features dedicated ray tracing or tensor cores.

Specification Differences

The specification sheets for the two cards reveal several divergences beyond the core architectural naming. The GTX 480 has a higher memory clock speed, running at 924 MHz (3.7 Gbps effective), compared to the GTX 590's 854 MHz (3.4 Gbps effective). This contributes to the GTX 480's superior memory bandwidth of 177.4 GB/s versus 164.0 GB/s. While both cards have the same 1536 MB of GDDR5 memory and a 384-bit bus width, the memory speed difference is a decisive factor in the bandwidth disparity.

In terms of compute units, the GTX 590 has more shading units (512) and TMUs (64) than the GTX 480 (480 and 60, respectively). However, the pixel rate and texture rate are higher on the GTX 480. The GTX 590 has a much higher TDP of 365 W, requiring a 750 W power supply, while the GTX 480 is rated at 250 W with a 600 W suggested PSU. The physical dimensions also differ, with the GTX 590 being longer at 279 mm (11 inches) compared to the GTX 480's 267 mm (10.5 inches). The GTX 590 also has specified height and width dimensions of 111 mm and 40 mm, respectively. Display outputs differ, with the GTX 480 offering 2x DVI and 1x mini-HDMI 1.3a, while the GTX 590 provides 3x DVI and 1x mini-DisplayPort. The GTX 480 was released on 2010-03-25 with a launch MSRP of 499 USD, and the GTX 590 followed on 2011-03-23 with a launch MSRP of 699 USD.

Head-to-Head Benchmarks

The head-to-head comparison is limited to a single test, but it is decisive. In the Geekbench OpenCL benchmark, the NVIDIA GeForce GTX 480 scores 13,300 points, while the NVIDIA GeForce GTX 590 scores 12,830 points. This gives the GTX 480 a 3.7% lead in performance. This result is significant because it shows the newer, higher-spec card underperforms its predecessor in this specific compute workload. The GTX 480's higher memory bandwidth and FP32 throughput appear to be more impactful than the GTX 590's additional shading units and TMUs in this test.

Examining the nearest rivals for each card provides further context. The GTX 480's closest competitor is the AMD Radeon Pro 555X, which scores 13,321, a mere 0.2% difference, effectively making them statistical peers. The GTX 590's closest rival is the AMD Radeon Pro 455, which scores 12,831, showing a 0% delta. This places both cards in a similar performance tier, but the GTX 480 sits at the higher end of that tier. The data shows that despite its higher percentile rank and average score, the GTX 480's victory margin over the GTX 590 is modest but clear.

The Verdict

From the data, the NVIDIA GeForce GTX 480 is the superior card for general compute performance. It wins the only available benchmark by a 3.7% margin, has a higher FP32 throughput, and offers greater memory bandwidth. Users focused on OpenCL-accelerated applications or tasks where raw compute and fill rates are paramount should favor the GTX 480. Its lower TDP of 250 W is also a practical advantage, simplifying power supply requirements.

The NVIDIA GeForce GTX 590, despite its newer Fermi 2.0 architecture, higher unit counts, and higher launch MSRP, does not deliver a performance win in the data. Its strengths lie in its feature set, specifically the triple DVI and mini-DisplayPort outputs, which offer more flexibility for multi-monitor setups. For a user needing to drive three or more displays directly from the card, the GTX 590's output configuration is a distinct benefit. However, in terms of raw computational power, the benchmark results indicate that the GTX 590 is slightly behind. The verdict is clear: the older GTX 480 is the performance pick, while the GTX 590 offers a richer display feature set.