NVIDIA GeForce GTX 1660 SUPER vs NVIDIA Quadro K4200 Comparison

# NVIDIA GeForce GTX 1660 SUPER vs NVIDIA Quadro K4200

The NVIDIA GeForce GTX 1660 SUPER is a modern Turing-architecture card from 2019, while the NVIDIA Quadro K4200 is a Kepler-based professional card from 2014. Benchmark data shows the GTX 1660 SUPER dominates in every recorded test, with the Quadro K4200 offering only its professional pedigree as a differentiator.

FAQ

Q: Which card has the higher average benchmark score?

A: The GeForce GTX 1660 SUPER averages 12,986 across all benchmark tests, while the Quadro K4200 averages 12,398. The GTX 1660 SUPER sits at the 53rd percentile of all GPUs, compared to the K4200’s 52nd percentile.

Q: How large is the performance gap in compute workloads?

A: In Geekbench OpenCL, the GTX 1660 SUPER scores 52,490 versus the K4200’s 12,313, a 326.3% advantage. In Geekbench Vulkan, the gap widens to 357.5%, with scores of 57,102 and 12,482 respectively.

Q: What are the memory specifications of each card?

A: The GTX 1660 SUPER has 6 GB of GDDR6 memory on a 192-bit bus with 336.0 GB/s bandwidth. The Quadro K4200 has 4 GB of GDDR5 memory on a 256-bit bus with 172.8 GB/s bandwidth.

Q: Which card is more power-efficient?

A: The K4200 has a lower TDP at 108 W compared to the GTX 1660 SUPER’s 125 W. However, the GTX 1660 SUPER delivers roughly 2.4x the FP32 performance (5.027 TFLOPS vs 2.107 TFLOPS) while drawing only 17 W more.

Q: Do both cards support modern APIs?

A: The GTX 1660 SUPER supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The Quadro K4200 supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175 — so the GTX 1660 SUPER has a higher DirectX feature level and newer Vulkan version.

Q: What is the form factor difference?

A: The GTX 1660 SUPER is a dual-slot card measuring 229 mm (9 inches) long, 111 mm tall, and 35 mm wide, requiring a 1x 8-pin power connector. The K4200 is a single-slot card at 241 mm (9.5 inches) long and 111 mm tall, requiring only a 1x 6-pin connector.

The Verdict

The data is unambiguous: the GeForce GTX 1660 SUPER wins both head-to-head benchmark tests outright and holds a 588-point average benchmark advantage. For anyone prioritizing raw compute performance, gaming capability, or modern API support, the GTX 1660 SUPER is the only sensible choice. Its 326.3% OpenCL and 357.5% Vulkan leads over the K4200 make the older card effectively non-competitive in these workloads.

The Quadro K4200’s only practical advantages are physical: it is single-slot, slightly longer, and draws 17 W less power. Builders with strict slot constraints or low-power chassis might consider it, but the performance penalty is severe. The K4200 also uses a PCIe 2.0 x16 interface versus the GTX 1660 SUPER’s PCIe 3.0 x16, further limiting its data transfer capabilities.

Neither card is in production; both are end-of-life. The GTX 1660 SUPER launched with a 229 USD MSRP, while the K4200’s launch price is not recorded. For any modern workload, the GTX 1660 SUPER is the clear pick. The K4200 only makes sense in legacy professional systems where its slot profile and power draw are non-negotiable constraints.

Head-to-Head Benchmarks

Two benchmark tests were run on both cards, and the GTX 1660 SUPER won both decisively.

In Geekbench OpenCL, the GTX 1660 SUPER scored 52,490 against the K4200’s 12,313. That is a 326.3% delta — more than four times the performance. This test reflects general compute throughput, and the gap is consistent with the massive differences in FP32 throughput (5.027 TFLOPS vs 2.107 TFLOPS) and memory bandwidth (336.0 GB/s vs 172.8 GB/s).

In Geekbench Vulkan, the results are even more lopsided. The GTX 1660 SUPER scored 57,102, while the K4200 managed only 12,482 — a 357.5% delta. Vulkan is a modern low-overhead API, and the K4200’s older Kepler architecture with Vulkan 1.2.175 support (versus Vulkan 1.4 on the GTX 1660 SUPER) clearly struggles to keep pace.

The GTX 1660 SUPER also shows its strength in other benchmark categories where the K4200 has no recorded scores. In PassMark G3D, the GTX 1660 SUPER scores 12,699; in PassMark GPU Compute, it scores 5,076. DirectX tests show 189 in PassMark DirectX 9, 104 in DirectX 11, 65 in DirectX 10, and 50 in DirectX 12. The K4200 lacks any PassMark results in the data, so direct comparisons in those tests are unavailable — but the existing head-to-head results already establish a clear hierarchy.

Specification Differences

The two cards differ in nearly every measurable specification.

The GTX 1660 SUPER uses a TU116 chip on TSMC’s 12 nm process, packing 6,600 million transistors on a 284 mm² die with a transistor density of 23.2M per mm². The K4200 uses a GK104 chip on TSMC’s 28 nm process, with 3,540 million transistors on a 294 mm² die at 12.0M per mm² density. The GTX 1660 SUPER achieves higher density in a slightly smaller die.

Clock speeds tell a similar story. The GTX 1660 SUPER runs at 1530 MHz base and 1785 MHz boost. The K4200 idles at 771 MHz base and boosts to just 784 MHz — less than half the base clock of the newer card.

Memory configurations diverge completely. The GTX 1660 SUPER uses 6 GB of GDDR6 at 1750 MHz (14 Gbps effective) on a 192-bit bus, delivering 336.0 GB/s. The K4200 uses 4 GB of GDDR5 at 1350 MHz (5.4 Gbps effective) on a 256-bit bus, delivering 172.8 GB/s. The wider bus on the K4200 cannot compensate for the older memory technology and lower clock speed.

Compute resources: the GTX 1660 SUPER has 1408 shading units, 88 TMUs, and 48 ROPs. The K4200 has 1344 shading units, 112 TMUs, and 32 ROPs. Despite fewer TMUs, the GTX 1660 SUPER achieves much higher texture and pixel rates: 157.1 GTexel/s versus 87.81 GTexel/s, and 85.68 GPixel/s versus 21.95 GPixel/s.

FP32 performance is 5.027 TFLOPS for the GTX 1660 SUPER versus 2.107 TFLOPS for the K4200. The GTX 1660 SUPER also supports FP16 at 10.05 TFLOPS (2:1 ratio), while the K4200 has no FP16 capability listed.

The GTX 1660 SUPER is dual-slot with a 1x 8-pin connector and a 125 W TDP. The K4200 is single-slot with a 1x 6-pin connector and a 108 W TDP. Both recommend a 300 W PSU. Display outputs differ: the GTX 1660 SUPER offers 1x DVI, 1x HDMI 2.0, and 1x DisplayPort 1.4a; the K4200 offers 1x DVI and 2x DisplayPort 1.2.

Architecture Differences

The architectural gap spans two generations of NVIDIA design. The GTX 1660 SUPER is built on Turing (the GeForce 16 generation), while the K4200 uses Kepler (Quadro Kepler Kx200 generation).

The manufacturing process moved from 28 nm (K4200) to 12 nm (GTX 1660 SUPER). Transistor count nearly doubled, from 3,540 million to 6,600 million, even though die size shrank slightly from 294 mm² to 284 mm². This is reflected in transistor density: 23.2M per mm² versus 12.0M per mm².

Memory architecture changed from GDDR5 to GDDR6. The GTX 1660 SUPER’s memory clock is 1750 MHz versus the K4200’s 1350 MHz, and effective data rate is 14 Gbps versus 5.4 Gbps. Bus width actually decreased from 256-bit to 192-bit, but the newer memory technology more than compensates — bandwidth nearly doubles from 172.8 GB/s to 336.0 GB/s.

Compute capabilities diverge sharply. The GTX 1660 SUPER supports FP16 with a 2:1 ratio, reaching 10.05 TFLOPS; the K4200 has no FP16 support. DirectX support differs as well: the GTX 1660 SUPER reaches DirectX 12 (12_1), while the K4200 is limited to DirectX 12 (11_0). Vulkan versions also differ — 1.4 on the GTX 1660 SUPER versus 1.2.175 on the K4200. Both support OpenGL 4.6.

The bus interface advanced from PCIe 2.0 x16 (K4200) to PCIe 3.0 x16 (GTX 1660 SUPER). Neither card has ray tracing or tensor cores — both entries list null values for RT and tensor core counts. The GTX 1660 SUPER’s predecessor is GeForce 10, with successor GeForce 20; the K4200’s predecessor is Quadro Fermi, with successor Quadro Maxwell.

Where Each One Wins

The GeForce GTX 1660 SUPER wins in every benchmark category where both cards have recorded scores. In Geekbench OpenCL, it is 326.3% faster; in Geekbench Vulkan, 357.5% faster. Its average benchmark score of 12,986 beats the K4200’s 12,398 — a 588-point margin that places it at the 53rd percentile versus the 52nd percentile of all GPUs.

The GTX 1660 SUPER is the clear choice for compute-heavy workloads, modern API-based rendering, or any task that benefits from higher memory bandwidth (336.0 GB/s vs 172.8 GB/s). Its FP32 throughput of 5.027 TFLOPS is roughly 2.4x the K4200’s 2.107 TFLOPS, and its FP16 capability adds another dimension for compatible workloads. The 6 GB GDDR6 frame buffer is also 50% larger than the K4200’s 4 GB GDDR5.

The Quadro K4200 wins in a narrow set of physical attributes. It is single-slot versus dual-slot, making it viable in dense chassis configurations. It draws 108 W versus 125 W — a modest power savings. It is slightly longer at 241 mm versus 229 mm, but the same height at 111 mm. Its 1x 6-pin power connector is less demanding than the GTX 1660 SUPER’s 1x 8-pin, though both cards recommend a 300 W PSU.

For gaming or general desktop use, the K4200 has no meaningful advantage. For professional ISV applications that rely on the Quadro brand’s certified drivers, the K4200 might still function in legacy workflows, but the benchmark data shows it will do so at a fraction of the GTX 1660 SUPER’s speed. The GTX 1660 SUPER’s nearest rivals — the RTX 3050 Ti Mobile (0.4% faster), Radeon RX 580 (0.4% slower), and Tesla M2090 (0.7% slower) — all cluster within 1% of its average score, indicating it sits comfortably in a competitive mid-range tier. The K4200’s nearest rivals, including the Tesla K20Xm (1.8% faster) and GTX 670 (2.9% faster), show it similarly positioned among older hardware — but at a far lower absolute performance level.