NVIDIA GeForce GTX 1060 6 GB vs NVIDIA GeForce GTX 780 Comparison

Head-to-Head Benchmarks

The benchmark data presents a fascinating generational clash. The GTX 1060 6 GB and GTX 780 are separated by three years of architecture, and the results show that progress is not always linear. In the database’s head-to-head tests, the GTX 1060 6 GB claims two decisive victories, while the GTX 780 secures one significant win.

The most lopsided result comes in Geekbench Metal, where the GTX 1060 6 GB scores 14,587 against the GTX 780’s 10,114. That is a 44.2% advantage for the Pascal card. In this test, the GTX 1060 6 GB is the definitive winner. The OpenCL benchmark tells a similar story. The GTX 1060 6 GB posts 32,322, while the GTX 780 manages 22,863. The delta is 41.4% in favor of the newer GPU. In both Compute API tests, the GTX 1060 6 GB demonstrates a substantial lead.

However, the Vulkan benchmark flips the script. Here, the GTX 780 outperforms the GTX 1060 6 GB by a significant margin. The GTX 780 scores 24,514, while the GTX 1060 6 GB scores 18,659. This represents a 23.9% advantage for the older Kepler card. This is a notable result, as it suggests that in certain low-level API workloads, the GTX 780’s larger memory interface and raw compute resources can still be leveraged effectively.

Looking at the average benchmark scores, the GTX 1060 6 GB holds the overall edge. Its average score is 21,856, which places it in the 67th percentile of all GPUs. The GTX 780’s average is 19,164, sitting in the 64th percentile. The difference in average score is 2,692 points, which is about 14% higher for the GTX 1060 6 GB. This means that across a broad spectrum of tasks, the newer card is consistently faster, despite its narrower memory bus and fewer shading units.

The head-to-head data also positions each card against its own rivals. The GTX 1060 6 GB’s nearest rival is the NVIDIA Quadro RTX 5000, which averages 21,629, a mere 1% difference. It also trades blows with the AMD Radeon RX 6700S, which is 1.3% behind, and the AMD Radeon RX 5700, which is 1.4% behind. This shows the GTX 1060 6 GB is clustered with much newer, higher-end parts. The GTX 780, on the other hand, sits right next to the NVIDIA TITAN Xp, with a -0.1% delta, and the NVIDIA Tesla K20m, with a 0.4% delta. Its performance is aligned with a professional-grade compute card.

In summary, the head-to-head shows that the GTX 1060 6 GB is the stronger card by a wide margin in Metal and OpenCL, but the GTX 780 is the clear winner in Vulkan.

The Verdict

The choice between these two cards is straightforward based on the recorded data. For any workload that relies on Metal or OpenCL, the NVIDIA GeForce GTX 1060 6 GB is the definitive pick. It is 44.2% faster in Metal and 41.4% faster in OpenCL. The data indicates that this is a massive generational leap in compute efficiency and raw throughput.

The GTX 780’s single victory in Vulkan is significant but does not compensate for its losses elsewhere. In the average benchmark score, the GTX 1060 6 GB is also ahead, and it achieves a higher percentile ranking. While the GTX 780 has a higher memory bandwidth and more TMUs, that does not translate to an overall win.

Who should pick the GTX 780? Only a user whose primary workload is exclusively Vulkan-based applications. In that specific scenario, it provides a 23.9% performance benefit. For everything else, the GTX 1060 6 GB is the better choice. The GTX 1060 6 GB also consumes less power, but that is a secondary consideration for pure performance analysis.

The verdict is that the GTX 1060 6 GB is the stronger, more versatile card. It wins more tests, has a higher average score, and occupies a higher percentile. The GTX 780 is a niche option for a specific API.

FAQ

Q: In which benchmark test does the GTX 1060 6 GB show its largest advantage over the GTX 780?

A: The GTX 1060 6 GB shows its largest advantage in the Geekbench Metal test, where it scores 14,587 versus the GTX 780’s 10,114, a 44.2% difference.

Q: Does the GTX 780 win any benchmark against the GTX 1060 6 GB?

A: Yes, the GTX 780 wins the Geekbench Vulkan test. It scores 24,514 compared to the GTX 1060 6 GB’s 18,659, which is a 23.9% advantage.

Q: What is the average benchmark score for each card?

A: The GTX 1060 6 GB has an average benchmark score of 21,856. The GTX 780 has an average benchmark score of 19,164.

Q: Which card has a higher percentile ranking among all GPUs?

A: The GTX 1060 6 GB ranks in the 67th percentile, while the GTX 780 ranks in the 64th percentile.

Q: How many head-to-head tests does each card win?

A: The GTX 1060 6 GB wins two of the three head-to-head tests (Metal and OpenCL), while the GTX 780 wins one (Vulkan).

Q: What is the closest rival to the GTX 1060 6 GB in the database?

A: The closest rival is the NVIDIA Quadro RTX 5000, with an average score of 21,629, which is a 1% difference from the GTX 1060 6 GB’s score.

Specification Differences

The two NVIDIA cards diverge significantly in their core specifications. The GTX 1060 6 GB utilizes a 16 nm process node, while the GTX 780 is built on a 28 nm process. The GTX 1060 6 GB has a smaller die size of 200 mm², compared to the 561 mm² die of the GTX 780. This results in a transistor density of 22.0M per mm² for the GTX 1060 6 GB versus 12.6M per mm² for the GTX 780, despite the GTX 780 having more total transistors (7,080 million vs 4,400 million).

Clock speeds are also quite different. The GTX 1060 6 GB has a base clock of 1506 MHz and a boost clock of 1709 MHz. The GTX 780 is lower in both respects, with a base clock of 863 MHz and a boost clock of 902 MHz. Memory configuration is a major difference: the GTX 1060 6 GB has 6 GB of GDDR5 on a 192-bit bus, yielding a bandwidth of 192.2 GB/s. The GTX 780 has 3 GB of GDDR5 on a 384-bit bus, which provides a higher bandwidth of 288.4 GB/s.

The compute unit counts differ. The GTX 1060 6 GB has 1280 shading units, 80 texture mapping units (TMUs), and 48 ROPs. The GTX 780 has 2304 shading units, 192 TMUs, and 48 ROPs. The GTX 780 has more shading units, but the GTX 1060 6 GB has higher pixel and texture rates in some respects. The pixel rate is 82.03 GPixel/s for the GTX 1060, but only 43.30 GPixel/s for the GTX 780. The texture rate is 136.7 GTexel/s for the 1060 and 173.2 GTexel/s for the 780.

Power consumption is a stark difference. The GTX 1060 6 GB has a TDP of 120 W, while the GTX 780 has a TDP of 250 W. The power connector requirements also differ: the GTX 1060 6 GB uses a single 6-pin connector, while the GTX 780 needs a 6-pin and an 8-pin connector. The suggested PSU is 300 W for the GTX 1060 and 600 W for the GTX 780.

The display outputs are different. The GTX 1060 6 GB offers 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.4a. The GTX 780 offers 2x DVI, 1x HDMI 1.4a, and 1x DisplayPort 1.2. The cards also have different dimensions: the GTX 1060 6 GB is 250 mm long, while the GTX 780 is 267 mm long.

Architecture Differences

The architectural gap is the most important factor in the performance differences. The GTX 1060 6 GB is based on the Pascal architecture, using the GP106 chip. The GTX 780 is based on the older Kepler architecture, using the GK110 chip. This is the fundamental reason for the performance differences.

The process node is a major architectural shift. The GTX 1060 6 GB is fabricated on TSMC’s 16 nm process, while the GTX 780 uses TSMC’s 28 nm process. This is a direct reason why the GTX 1060 6 GB can achieve higher clock speeds and better power efficiency. The GTX 1060 6 GB’s transistors are smaller and more densely packed, allowing for more compute units in a smaller area.

The GTX 1060 6 GB features the Pascal architecture, which introduces a different memory hierarchy and compute scheduling compared to Kepler. While the GTX 780 has more raw shading units (2300 vs 1280), the GTX 1060 6 GB achieves higher throughput in most tests. This is due to the improved instruction efficiency and higher clock speeds of Pascal. The GTX 1060 6 GB also has a significantly higher pixel rate (82.03 vs 43.30 GPixel/s), indicating a more efficient rasterizer.

The two also support different API versions. The GTX 1060 6 GB supports DirectX 12 (12_1) while the GTX 780 supports DirectX 12 (11_0). The GTX 1060 6 GB also has a newer Vulkan implementation, supporting version 1.4, while the GTX 780 supports version 1.2.175. The GTX 1060 6 GB has a more modern display engine, with support for HDMI 2.0 and DisplayPort 1.4a, while the GTX 780 has HDMI 1.4a and DisplayPort 1.2.

The FP16 performance is also a differentiator. The GTX 1060 6 GB has an FP16 rate of 68.36 GFLOPS with a 1:64 ratio, while the GTX 780 has no listed FP16 performance. This is a feature of the newer Pascal architecture. The GTX 780 is a product of the Kepler era, which was designed before these modern API features were widely adopted. The GTX 1060 6 GB is the newer, more refined architecture.