NVIDIA GeForce GTX 590 vs NVIDIA GeForce GTX 660 Ti Comparison

FAQ

Q: Which GPU has a higher average benchmark score in the database?

A: The NVIDIA GeForce GTX 660 Ti records an average benchmark score of 15063, while the NVIDIA GeForce GTX 590 records 12830. The GTX 660 Ti is ahead by 17.8% in the head-to-head Geekbench OpenCL test.

Q: How do the two cards compare in memory capacity and bus width?

A: The GTX 660 Ti has 2 GB of GDDR5 memory on a 192-bit bus, while the GTX 590 has 1536 MB of GDDR5 memory on a 384-bit bus. Despite the narrower bus, the GTX 660 Ti's memory runs at 1502 MHz (6 Gbps effective) versus 854 MHz (3.4 Gbps effective) for the GTX 590.

Q: What are the pixel and texture rate differences?

A: The GTX 660 Ti achieves a pixel rate of 27.44 GPixel/s and a texture rate of 109.8 GTexel/s. The GTX 590 delivers 19.46 GPixel/s and 38.91 GTexel/s. The GTX 660 Ti leads in both categories.

Q: Which card has a higher transistor density?

A: The GTX 660 Ti, built on a 28 nm process, has a transistor density of 12.0M per mm². The GTX 590, built on a 40 nm process, has a density of 5.8M per mm².

Q: What are the power connector requirements for each card?

A: The GTX 660 Ti requires 2x 6-pin power connectors and has a TDP of 150 W. The GTX 590 requires 2x 8-pin power connectors and has a TDP of 365 W.

Q: Which card supports Vulkan in the recorded API data?

A: The GTX 660 Ti lists Vulkan 1.2.175 support, while the GTX 590 lists no Vulkan support in the database. Both support DirectX 12 (11_0) and OpenGL 4.6.

The Verdict

The data points clearly toward the NVIDIA GeForce GTX 660 Ti for nearly every usage scenario. Its average benchmark score of 15063 places it in the 57th percentile among all GPUs, while the GTX 590 sits at the 52nd percentile with a score of 12830. The 17.8% advantage in the Geekbench OpenCL test is the only direct head-to-head measurement available, and it favors the GTX 660 Ti.

For users prioritizing compute workloads or modern API compatibility, the GTX 660 Ti is the obvious selection. It supports Vulkan 1.2.175, whereas the GTX 590 has no Vulkan entry in the database. The GTX 660 Ti also runs on the PCIe 3.0 x16 interface, compared to the GTX 590's PCIe 2.0 x16.

The GTX 590 does retain a few physical advantages. Its 384-bit memory bus provides 164.0 GB/s of bandwidth, slightly higher than the GTX 660 Ti's 144.2 GB/s. It also has double the ROP count at 48 versus 24. However, these advantages do not translate into a benchmark win in the recorded data.

Users with an older system constrained to PCIe 2.0 might still consider the GTX 590, but the power requirements are substantially higher at 365 W TDP and a suggested 750 W PSU, versus 150 W TDP and a suggested 450 W PSU for the GTX 660 Ti. The benchmark evidence favors the newer card.

Head-to-Head Benchmarks

The only direct comparison in the database is the Geekbench OpenCL test. The GTX 660 Ti scores 15113, while the GTX 590 scores 12830. This represents a 17.8% delta in favor of the GTX 660 Ti.

Context from the nearest rivals reinforces the gap. The GTX 660 Ti sits within 0.1% of the AMD Radeon Pro 560X, which scores 15082. It trails the AMD Radeon RX 7600 by only 0.7% (that card scores 15171). The GTX 660 Ti is 0.9% behind the NVIDIA GeForce RTX 3050 OEM, which scores 15199. These are modern cards, and the 2012-era GTX 660 Ti is essentially matching them in this workload.

The GTX 590's closest rival is the AMD Radeon Pro 455, which scores 12831, a 0% delta. The AMD FirePro W5100 scores 12847, putting it 0.1% ahead of the GTX 590. The NVIDIA GeForce GTX 670 scores 12773, which is 0.4% behind the GTX 590. The GTX 590's position among these rivals indicates it performs at a lower tier than the GTX 660 Ti in the recorded OpenCL measurement.

The 17.8% delta between the two cards is not a marginal difference. It is a decisive margin, placing the GTX 660 Ti in a different performance class. The GTX 590's texture rate of 38.91 GTexel/s and pixel rate of 19.46 GPixel/s are both well below the GTX 660 Ti's corresponding figures of 109.8 GTexel/s and 27.44 GPixel/s.

Specification Differences

The memory configurations differ significantly. The GTX 660 Ti has 2 GB of GDDR5 on a 192-bit bus with 144.2 GB/s bandwidth. The GTX 590 has 1536 MB on a 384-bit bus with 164.0 GB/s bandwidth. The GTX 590's wider bus gives it higher bandwidth, but the GTX 660 Ti's memory clock is much higher at 1502 MHz versus 854 MHz.

The compute unit counts are starkly different. The GTX 660 Ti has 1344 shading units, 112 texture mapping units, and 24 ROPs. The GTX 590 has 512 shading units, 64 TMUs, and 48 ROPs. The GTX 660 Ti has more than double the shading units and nearly double the TMUs, while the GTX 590 has double the ROPs.

Clock speeds are not directly comparable because the GTX 590 lists no base or boost clock in the database. The GTX 660 Ti has a base clock of 915 MHz and a boost clock of 980 MHz.

Process node and physical dimensions differ. The GTX 660 Ti is built on a 28 nm process with a die size of 294 mm² and 3,540 million transistors. The GTX 590 uses a 40 nm process with a die size of 520 mm² and 3,000 million transistors. The GTX 660 Ti is shorter at 241 mm (9.5 inches) versus 279 mm (11 inches) for the GTX 590. The GTX 590 is also thicker, with listed dimensions of 111 mm height and 40 mm width.

Power delivery is a major differentiator. The GTX 660 Ti has a 150 W TDP, uses 2x 6-pin connectors, and suggests a 450 W PSU. The GTX 590 has a 365 W TDP, uses 2x 8-pin connectors, and suggests a 750 W PSU.

Bus interface and display outputs also differ. The GTX 660 Ti uses PCIe 3.0 x16 and offers 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2. The GTX 590 uses PCIe 2.0 x16 and offers 3x DVI and 1x mini-DisplayPort.

Architecture Differences

The GTX 660 Ti is based on the GK104 chip with Kepler architecture, while the GTX 590 is based on the GF110 chip with Fermi 2.0 architecture. These are two distinct design generations from NVIDIA.

The manufacturing process is a critical difference. The GTX 660 Ti uses a 28 nm process at TSMC with a transistor density of 12.0M per mm². The GTX 590 uses a 40 nm process at TSMC with a transistor density of 5.8M per mm². The newer process allows the GTX 660 Ti to pack more transistors into a smaller die: 3,540 million in 294 mm² versus 3,000 million in 520 mm².

The GTX 660 Ti belongs to the GeForce 600 generation, with a release date of August 15, 2012. The GTX 590 belongs to the GeForce 500 generation, released March 23, 2011. The GTX 660 Ti's predecessor is the GeForce 500 series and its successor is the GeForce 700 series. The GTX 590's predecessor is the GeForce 400 series and its successor is the GeForce 600 series.

API support reflects the architectural differences. The GTX 660 Ti lists Vulkan 1.2.175 support, while the GTX 590 has no Vulkan entry. Both support DirectX 12 (11_0) and OpenGL 4.6. The GTX 660 Ti's FP32 compute is rated at 2.634 TFLOPS, while the GTX 590 is rated at 1,244.2 GFLOPS.

The GTX 590's dual-GPU design is not explicitly stated in the database, but its specification pattern, with a 520 mm² die and 365 W TDP, indicates a fundamentally different internal layout compared to the single-die GTX 660 Ti. The GTX 660 Ti achieves higher compute throughput with a lower power envelope.

Where Each One Wins

The GTX 660 Ti wins in the recorded benchmark data, taking the only head-to-head test with a 17.8% margin. It also wins on compute throughput, with 2.634 TFLOPS FP32 versus the GTX 590's 1,244.2 GFLOPS. Its pixel rate of 27.44 GPixel/s and texture rate of 109.8 GTexel/s are both substantially higher.

The GTX 660 Ti is the better choice for users running OpenCL workloads, given the 15113 score versus 12830. Its Vulkan support makes it more viable for modern applications that rely on that API. Its lower TDP of 150 W and suggested 450 W PSU also make it easier to integrate into existing systems, especially those with limited power headroom.

The GTX 660 Ti's higher memory clock of 1502 MHz and larger 2 GB capacity provide an edge in scenarios where memory speed matters. Its PCIe 3.0 x16 interface offers greater bandwidth potential for data transfer between the CPU and GPU.

The GTX 590 wins in memory bandwidth, with 164.0 GB/s versus 144.2 GB/s, and in ROP count, with 48 versus 24. These attributes historically benefit high-resolution rendering with heavy antialiasing, where fill-rate and memory bandwidth are stressed. However, the database does not include a benchmark that isolates these strengths, and its overall score of 12830 places it below the GTX 660 Ti.

The GTX 590 also offers more display outputs in terms of DVI ports, with 3x DVI compared to the GTX 660 Ti's 2x DVI. Users requiring multiple DVI connections for legacy monitors might prefer the GTX 590, though the GTX 660 Ti adds an HDMI 1.4a and a DisplayPort 1.2 for modern display connectivity.

In practical terms, the GTX 660 Ti is the pick for compute-heavy tasks, modern API support, and power efficiency. The GTX 590 retains niche advantages in raw memory bandwidth and ROP throughput, but the recorded data shows the GTX 660 Ti as the stronger overall performer.