NVIDIA CMP 70HX vs NVIDIA GeForce GTX 980 Ti Comparison

The NVIDIA CMP 70HX and the NVIDIA GeForce GTX 980 Ti represent two distinct eras of GPU design, one aimed at the cryptomining market and the other at flagship gaming. The benchmark data shows a clear performance hierarchy between the two, with the older GTX 980 Ti holding a decisive lead in the available tests. This analysis will detail the specific performance gaps, architectural differences, and specification disparities that define this matchup.

Head-to-Head Benchmarks

The head-to-head benchmark results provide a stark illustration of the performance difference between these two cards. The data shows the GTX 980 Ti winning both available tests, securing a 2-0 victory over the CMP 70HX. This outcome is surprising given the generational gap and the newer process node of the CMP 70HX, but the numbers are unambiguous.

In the Geekbench OpenCL test, the GTX 980 Ti scores 43,513 points, while the CMP 70HX manages only 25,135 points. This translates to a delta of -42.2% for the CMP 70HX, meaning it performs roughly 42% worse than its rival. This is a massive margin that places the two cards in entirely different performance tiers. The GTX 980 Ti’s advantage is not marginal; it is a substantial lead that would be felt in any compute-heavy workload.

The Vulkan test paints a similar picture, though with a slightly narrower gap. Here, the GTX 980 Ti again takes the lead with a score of 47,724, against the CMP 70HX’s 35,817. The delta in this test is -24.9%, still a significant loss for the CMP 70HX, but less dramatic than the OpenCL result. This suggests that the GTX 980 Ti’s architecture scales better in this particular API, or that the CMP 70HX’s design is more severely hampered in OpenCL workloads.

Looking at the CMP 70HX’s average benchmark score of 30,476, it sits just 0.1% above the AMD Radeon RX 6700 and 1.3% above the AMD Radeon RX 6800. In contrast, the GTX 980 Ti’s average score of 28,020 places it in close competition with the AMD Radeon Pro W5500X, which it leads by 0.2%. The performance landscape is therefore quite different for each card; the CMP 70HX is competing with modern mid-range GPUs, while the GTX 980 Ti is holding its own against a set of professional and mobile workstation parts.

The CMP 70HX’s best individual performance is its Geekbench Vulkan score of 35,817, which is still significantly lower than the GTX 980 Ti’s Vulkan result. The data indicates that the GTX 980 Ti is simply the more powerful card in every measurable way. For a mining-focused product, this performance deficit might be acceptable, but in a head-to-head comparison, it is a clear and decisive loss.

The Verdict

From a pure performance standpoint, the data makes a compelling case for the NVIDIA GeForce GTX 980 Ti. It wins both benchmark tests by significant margins, making it the obvious choice for anyone prioritizing compute performance. The GTX 980 Ti’s scores of 43,513 and 47,724 in OpenCL and Vulkan, respectively, dwarf the CMP 70HX’s corresponding results. Benchmark results indicate that the GTX 980 Ti is the stronger card in this pairing.

The CMP 70HX, with its 75th percentile ranking among all GPUs, is a capable card in its own right. However, it is outclassed here. Its average benchmark score of 30,476 is lower than the GTX 980 Ti’s 28,020 when considering the head-to-head deltas, but the percentile ranking shows the GTX 980 Ti is in the 73rd percentile, a very close overall standing. The data suggests that while the CMP 70HX is a solid performer, the GTX 980 Ti remains a formidable piece of hardware even against a much newer card.

Who should pick which? For a user with a workload that leverages Vulkan or OpenCL, the GTX 980 Ti is the clear winner. Its performance advantage is too substantial to ignore. The CMP 70HX, by contrast, is a niche product designed for a specific purpose. It has no display outputs and is intended for a mining rig. In a general-purpose computing context, the GTX 980 Ti is the better choice. The data does not support any other conclusion; the GTX 980 Ti is the superior performer.

Given that the GTX 980 Ti has a launch MSRP of 649 USD, it was a high-end product at its release. The CMP 70HX, with no launch MSRP listed, was a product with a different market positioning. The choice here is straightforward: the GTX 980 Ti delivers the performance, while the CMP 70HX is a specialized tool with a clear performance handicap.

Architecture Differences

The two GPUs are built on fundamentally different architectures and manufacturing processes. The NVIDIA CMP 70HX is based on the Ampere architecture, utilizing the GA104 chip, while the NVIDIA GeForce GTX 980 Ti uses the older Maxwell 2.0 architecture with the GM200 chip. This single fact explains many of the performance and feature disparities between the two.

The manufacturing process highlights a significant generational leap. The CMP 70HX is fabricated on an 8 nm process at Samsung, whereas the GTX 980 Ti uses a 28 nm process from TSMC. This process shrink allows the CMP 70HX to pack 17,400 million transistors into a 392 mm² die, resulting in a transistor density of 44.4M per mm². In contrast, the GTX 980 Ti has 8,000 million transistors on a much larger 601 mm² die, giving it a density of just 13.3M per mm². This shows how much more efficient the newer manufacturing node is.

The CMP 70HX’s Ampere architecture brings modern features that the Maxwell 2.0 GTX 980 Ti lacks. Most notably, the CMP 70HX includes 30 RT Cores and 120 Tensor Cores, which are absent on the GTX 980 Ti. This means the CMP 70HX supports hardware-accelerated ray tracing and AI-based technologies, while the GTX 980 Ti does not. The API support reflects this, with the CMP 70HX boasting DirectX 12 Ultimate (12_2) support versus the GTX 980 Ti’s DirectX 12 (12_1).

Memory technology also differs significantly. The CMP 70HX uses 8 GB of GDDR6X memory on a 256-bit bus, achieving a bandwidth of 608.3 GB/s. The GTX 980 Ti uses 6 GB of GDDR5 memory on a wider 384-bit bus, but its bandwidth is lower at 336.6 GB/s. This gives the newer card a major bandwidth advantage, which is crucial for many compute tasks.

The core configurations are also divergent. The CMP 70HX has 3,840 shading units, 120 TMUs, and 64 ROPs. The GTX 980 Ti has fewer shading units at 2,816, but more TMUs (176) and ROPs (96). This results in different theoretical fill rates; the GTX 980 Ti has a higher pixel rate of 103.3 GPixel/s and texture rate of 189.4 GTexel/s, while the CMP 70HX trails with 89.28 GPixel/s and 167.4 GTexel/s. The FP32 performance tells a different story, with the CMP 70HX reaching 10.71 TFLOPS versus the GTX 980 Ti’s 6.060 TFLOPS.

Specification Differences

The specification sheets for these two cards reveal a host of differences beyond their core architectures. While both are dual-slot cards with a similar length of 267 mm, their other specifications diverge considerably.

• Process Node: The CMP 70HX is built on an 8 nm process, while the GTX 980 Ti uses a 28 nm process.
• Memory: The CMP 70HX has 8 GB of GDDR6X memory, whereas the GTX 980 Ti has 6 GB of GDDR5 memory.
• Memory Bus: The CMP 70HX uses a 256-bit bus, while the GTX 980 Ti uses a wider 384-bit bus.
• Memory Bandwidth: The CMP 70HX offers 608.3 GB/s of bandwidth, more than the GTX 980 Ti’s 336.6 GB/s.
• Base Clock: The CMP 70HX has a base clock of 1365 MHz, which is higher than the GTX 980 Ti’s 1000 MHz.
• Boost Clock: The CMP 70HX boosts to 1395 MHz, while the GTX 980 Ti boosts to 1076 MHz.
• Power Connectors: The CMP 70HX requires a single 12-pin connector, while the GTX 980 Ti uses one 6-pin and one 8-pin connector.
• Suggested PSU: The CMP 70HX recommends a 200 W power supply, whereas the GTX 980 Ti recommends a 600 W unit.
• Bus Interface: The CMP 70HX uses a PCIe 1.0 x4 interface, while the GTX 980 Ti uses PCIe 3.0 x16.
• Display Outputs: The CMP 70HX has no display outputs, while the GTX 980 Ti has 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.2.
• TDP: The GTX 980 Ti has a TDP of 250 W, while the CMP 70HX’s TDP is not listed.
• Dimensions: The GTX 980 Ti has a defined width of 40 mm, while the CMP 70HX’s width is not specified.

The CMP 70HX’s higher clocks and newer memory give it a theoretical compute advantage, but its limited PCIe interface and lack of display outputs are clear indicators of its mining-focused design. The GTX 980 Ti, with its standard display outputs and higher power requirement, is a conventional graphics card.

FAQ

Q: Which GPU is faster in the available benchmarks?

A: The NVIDIA GeForce GTX 980 Ti is faster in both head-to-head tests, winning Geekbench OpenCL and Vulkan with scores of 43,513 and 47,724, respectively, versus the CMP 70HX’s 25,135 and 35,817.

Q: What is the performance gap in the Geekbench OpenCL test?

A: The CMP 70HX trails the GTX 980 Ti by -42.2% in the Geekbench OpenCL test, indicating a significant performance deficit.

Q: Does the CMP 70HX support modern features like ray tracing?

A: Yes, the CMP 70HX includes 30 RT Cores and 120 Tensor Cores, and supports DirectX 12 Ultimate (12_2), while the GTX 980 Ti has no RT or Tensor Cores and supports DirectX 12 (12_1).

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

A: The CMP 70HX has a memory bandwidth of 608.3 GB/s, which is significantly higher than the GTX 980 Ti’s 336.6 GB/s.

Q: Which card has a higher transistor density?

A: The CMP 70HX, built on an 8 nm process, has a transistor density of 44.4M per mm², compared to the GTX 980 Ti’s 13.3M per mm² on a 28 nm process.

Q: Can the CMP 70HX be used for display output like the GTX 980 Ti?

A: No, the CMP 70HX has no display outputs, whereas the GTX 980 Ti has 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.2 outputs.