NVIDIA GeForce GTX 1660 Ti vs NVIDIA GeForce GTX 675M Comparison

Head-to-Head Benchmarks

The recorded data contains a single common benchmark between the NVIDIA GeForce GTX 1660 Ti and the NVIDIA GeForce GTX 675M: Geekbench OpenCL. In this test, the GTX 1660 Ti scores 51,029 points, while the GTX 675M scores 6,946 points. This yields a delta of 634.7% in favor of the GTX 1660 Ti, an extremely wide margin that reflects the generational gap between the two parts.

To put the GTX 1660 Ti's score into perspective, its average benchmark score across all recorded tests is 7,723. The nearest rivals in the database for the GTX 1660 Ti are the AMD Radeon 540 (average score 7,673, a delta of 0.7%), the AMD Radeon Pro WX 3100 (7,580, delta 1.9%), the AMD Radeon R7 250 (7,557, delta 2.2%), and the Intel Arc A310 (7,550, delta 2.3%). This means the GTX 1660 Ti sits within a tight cluster of similarly performing cards in aggregate, but its OpenCL result of 51,029 is a massive outlier relative to those averages, indicating that this specific workload heavily favors its architecture.

For the GTX 675M, the average benchmark score is 6,946, which is identical to its single recorded OpenCL result. Its nearest rivals are the NVIDIA GeForce GTX 680M (average 7,023, delta -1.1%), the NVIDIA T600 (7,035, delta -1.3%), the AMD Radeon R5 M240 (6,975, delta -0.4%), and the AMD FirePro M5100 (6,830, delta 1.7%). The GTX 675M is therefore positioned within a small performance band, trailing the GTX 680M by 1.1% and the T600 by 1.3%, while leading the R5 M240 by 0.4% and the FirePro M5100 by 1.7%. Against the GTX 1660 Ti, however, the delta is so large that the GTX 675M's position among its peers is almost irrelevant.

The head-to-head data shows one win for the GTX 1660 Ti and zero for the GTX 675M. The 634.7% delta is not a marginal improvement; it is a multi-generational leap. In practical terms, the GTX 1660 Ti delivers over seven times the OpenCL compute throughput of the GTX 675M. This single data point dominates the comparison, but it must be interpreted alongside the broader architectural and specification differences detailed below.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA GeForce GTX 1660 Ti has an average benchmark score of 7,723, compared to the NVIDIA GeForce GTX 675M's 6,946. The GTX 1660 Ti leads by 777 points in this aggregate metric.

Q: How do the two cards compare in the Geekbench OpenCL test?

A: The GTX 1660 Ti scores 51,029, while the GTX 675M scores 6,946. The GTX 1660 Ti wins with a delta of 634.7%.

Q: What is the transistor density difference between the two chips?

A: The GTX 1660 Ti, built on a 12 nm process, has a transistor density of 23.2 million transistors per square millimeter. The GTX 675M, built on a 40 nm process, has a density of 5.9 million transistors per square millimeter.

Q: Which GPU supports Vulkan?

A: The GTX 1660 Ti lists Vulkan 1.4 in its API support. The GTX 675M has no Vulkan entry in its recorded API data.

Q: What is the memory bandwidth of each card?

A: The GTX 1660 Ti has a memory bandwidth of 288.0 GB/s, while the GTX 675M has a memory bandwidth of 96.00 GB/s.

Q: Are both GPUs still in production?

A: No, both are marked as end-of-life in the database. The GTX 1660 Ti was released on 2019-02-21, and the GTX 675M was released on 2012-03-21.

The Verdict

The benchmark data indicates a decisive victory for the NVIDIA GeForce GTX 1660 Ti. In the only shared test, Geekbench OpenCL, the GTX 1660 Ti outperforms the GTX 675M by 634.7%. The GTX 1660 Ti also holds a higher average benchmark score (7,723 versus 6,946) and a higher percentile rank among all GPUs (41st percentile versus 39th percentile). For any workload represented by these metrics, the GTX 1660 Ti is the clear choice.

The GTX 675M, while competitive within its own era and among its nearest rivals (within 1.7% of the AMD FirePro M5100 and within 1.3% of the NVIDIA T600), cannot match the GTX 1660 Ti's compute performance. The data shows no test in which the GTX 675M wins, so from a pure performance standpoint, there is no scenario where it is preferable. The GTX 1660 Ti should be selected by anyone prioritizing raw OpenCL throughput, higher memory bandwidth, or modern API support. The GTX 675M remains relevant only as a historical comparison point for its generation.

Specification Differences

The two GPUs differ across nearly every recorded specification. The GTX 1660 Ti uses 6 GB of GDDR6 memory on a 192-bit bus, while the GTX 675M uses 2 GB of GDDR5 memory on a 256-bit bus. Despite the narrower bus, the GTX 1660 Ti achieves a bandwidth of 288.0 GB/s, three times the GTX 675M's 96.00 GB/s. The GTX 1660 Ti has 1,536 shading units, 96 texture mapping units, and 48 raster output units. The GTX 675M has 384 shading units, 64 TMUs, and 32 ROPs. The pixel rate of the GTX 1660 Ti is 84.96 GPixel/s versus 9.920 GPixel/s for the GTX 675M, and the texture rate is 169.9 GTexel/s versus 39.68 GTexel/s. Floating-point performance is 5.437 TFLOPS for the GTX 1660 Ti and 952.3 GFLOPS for the GTX 675M. The GTX 1660 Ti also supports FP16 at 10.87 TFLOPS (2:1), while the GTX 675M has no recorded FP16 capability.

Power and physical specifications also differ. The GTX 1660 Ti has a TDP of 120 W and uses a dual-slot form factor with a single 8-pin power connector, requiring a 300 W power supply. The GTX 675M has a TDP of 100 W and is an MXM module with no power connectors; it is dependent on the host device for power delivery. The bus interface is PCIe 3.0 x16 for the GTX 1660 Ti, while the GTX 675M uses MXM-B (3.0). Display outputs for the GTX 1660 Ti include 1x DVI, 1x HDMI 2.0, and 1x DisplayPort 1.4a, whereas the GTX 675M's outputs are listed as portable device dependent. The GTX 1660 Ti has dimensions of 229 mm by 111 mm by 35 mm, while the GTX 675M has no recorded dimensions.

Architecture Differences

The GTX 1660 Ti is built on the Turing architecture, specifically the TU116 chip, fabricated on a 12 nm process at TSMC. It contains 6,600 million transistors on a die size of 284 mm², yielding a transistor density of 23.2 million per square millimeter. The GTX 675M uses the Fermi 2.0 architecture with the GF114 chip, fabricated on a 40 nm process, also at TSMC. It contains 1,950 million transistors on a larger die of 332 mm², yielding a lower density of 5.9 million per square millimeter. The GTX 1660 Ti belongs to the GeForce 16 generation, while the GTX 675M belongs to the GeForce 600M generation. The GTX 1660 Ti's predecessor is GeForce 10, and its successor is GeForce 20. The GTX 675M's predecessor is GeForce 500M, and its successor is GeForce 700M.

In terms of API support, the GTX 1660 Ti supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The GTX 675M supports DirectX 12 (11_0) and OpenGL 4.6, but has no recorded Vulkan support. Neither GPU has ray tracing cores or tensor cores. The GTX 1660 Ti's clock speeds are a base of 1500 MHz and a boost of 1770 MHz, with memory running at 1500 MHz (12 Gbps effective). The GTX 675M has no recorded base or boost clock, but its memory runs at 750 MHz (3 Gbps effective).

Where Each One Wins

The GTX 1660 Ti wins in every recorded benchmark category. In Geekbench OpenCL, it is 634.7% ahead of the GTX 675M. Its average benchmark score is 11.2% higher (7,723 versus 6,946), and its percentile rank is 41 versus 39. The GTX 1660 Ti also has superior memory bandwidth (288.0 GB/s versus 96.00 GB/s), higher pixel and texture rates, and more shading units, TMUs, and ROPs. It supports Vulkan, which the GTX 675M does not, and has a higher FP32 throughput (5.437 TFLOPS versus 952.3 GFLOPS). For any compute-heavy task, the GTX 1660 Ti is the clear winner.

The GTX 675M's only recorded advantage is its lower TDP (100 W versus 120 W) and its MXM form factor, which is designed for portable devices. It also has a wider memory bus (256-bit versus 192-bit), but this does not translate into higher bandwidth due to the slower GDDR5 memory. The GTX 675M is competitive with its nearest rivals (within 1.7% of the AMD FirePro M5100 and within 1.3% of the NVIDIA T600), but against the GTX 1660 Ti, it has no measurable wins. Users seeking OpenCL compute, modern API support, or high-bandwidth memory should choose the GTX 1660 Ti. The GTX 675M is only relevant for legacy systems requiring an MXM module with lower power draw.