AMD Radeon 860M vs NVIDIA GeForce GTX 980 Ti Comparison

# NVIDIA GeForce GTX 980 Ti vs AMD Radeon 860M

The NVIDIA GeForce GTX 980 Ti and AMD Radeon 860M occupy entirely different corners of the GPU landscape: one is a 2015 flagship desktop card built on a 28 nm process with a 250 W TDP, while the other is a 2025 integrated graphics processor (IGP) designed for mobile systems with a 15 W TDP. Benchmark data shows the GTX 980 Ti holding a decisive performance lead in both shared tests, but the Radeon 860M counters with a much more modern feature set and dramatically lower power consumption. In the two head-to-head benchmarks, the GTX 980 Ti wins both — by 91.2% in Geekbench OpenCL and 58.9% in Geekbench Vulkan. However, the Radeon 860M sits at the 72nd percentile of all GPUs versus the GTX 980 Ti's 73rd, indicating that despite the older card's raw score advantage, the integrated part is surprisingly competitive in overall standing. The GTX 980 Ti averages 28,020 across all benchmarks, while the Radeon 860M averages 26,401 — a gap of just 6.1% in aggregate, far smaller than the head-to-head deltas suggest.

The Verdict

The data paints a clear picture: the NVIDIA GeForce GTX 980 Ti is the performance winner for anyone who needs maximum compute throughput in desktop workloads. Its Geekbench OpenCL score of 43,513 crushes the Radeon 860M's 22,759, and its Vulkan result of 47,724 beats the 860M's 30,043 by a wide margin. In direct comparison, the GTX 980 Ti delivers 91.2% higher OpenCL performance and 58.9% higher Vulkan performance. This card is for users running demanding applications that leverage raw GPU compute — the sort of workloads where its 6,060 TFLOPS FP32 throughput and 336.6 GB/s memory bandwidth matter. It also holds a 73rd percentile ranking among all GPUs, edging out the Radeon 860M's 72nd.

However, the Radeon 860M is the rational choice for portable systems where power efficiency and modern API support take priority. Its 15 W TDP is a fraction of the GTX 980 Ti's 250 W, and it requires no power connectors — it draws entirely from the host system. The 860M supports DirectX 12 Ultimate (12_2) and features 8 ray tracing cores, capabilities the GTX 980 Ti lacks entirely, as it tops out at DirectX 12 (12_1). For users building or buying a thin-and-light laptop, the 860M's 72nd percentile ranking — just one point below the GTX 980 Ti — is remarkable for an integrated part. The GTX 980 Ti, by contrast, is end-of-life, requires a 600 W power supply, and needs a dual-slot chassis with a 267 mm length. The verdict: choose the GTX 980 Ti for raw performance in a desktop, the Radeon 860M for a modern, efficient mobile system with ray tracing support.

FAQ

Q: Which GPU has higher raw performance in shared benchmarks?

A: The NVIDIA GeForce GTX 980 Ti wins both head-to-head tests: 43,513 vs 22,759 in Geekbench OpenCL (a 91.2% advantage) and 47,724 vs 30,043 in Geekbench Vulkan (a 58.9% advantage).

Q: How do their overall benchmark averages compare?

A: The GTX 980 Ti averages 28,020 across all benchmark results, while the Radeon 860M averages 26,401. The GTX 980 Ti is ahead by approximately 6.1%, and both cards sit within one percentile point of each other (73rd vs 72nd).

Q: What are the key architectural differences?

A: The GTX 980 Ti uses NVIDIA's Maxwell 2.0 architecture on a 28 nm TSMC process with 8,000 million transistors on a 601 mm² die. The Radeon 860M uses AMD's RDNA 3.5 architecture on a 4 nm TSMC process, with transistor count and die size listed as unknown.

Q: Does the Radeon 860M support ray tracing?

A: Yes. The Radeon 860M includes 8 ray tracing cores and supports DirectX 12 Ultimate (12_2). The GTX 980 Ti has no ray tracing cores and only supports DirectX 12 (12_1).

Q: What about memory configurations?

A: The GTX 980 Ti has 6 GB of GDDR5 memory on a 384-bit bus with 336.6 GB/s bandwidth. The Radeon 860M uses system-shared memory with system-dependent bandwidth, meaning it relies on the host system's RAM.

Q: Which card is more power-efficient?

A: The Radeon 860M has a 15 W TDP and requires no power connectors, making it suitable for integrated mobile use. The GTX 980 Ti has a 250 W TDP, needs a 1x 6-pin plus 1x 8-pin power connector setup, and recommends a 600 W power supply.

Architecture Differences

The two GPUs represent fundamentally different design philosophies separated by a decade of semiconductor progress. The GTX 980 Ti is built on NVIDIA's Maxwell 2.0 architecture, manufactured on a 28 nm process at TSMC, with 8,000 million transistors packed into a 601 mm² die — a transistor density of 13.3 million transistors per square millimeter. This is a large, power-hungry discrete GPU designed for maximum throughput in a desktop chassis. Its compute configuration includes 2,816 shading units, 176 texture mapping units, and 96 raster operations pipelines. The card relies on a 384-bit memory bus connected to 6 GB of GDDR5 memory, delivering 336.6 GB/s of bandwidth, with pixel rate of 103.3 GPixel/s and texture rate of 189.4 GTexel/s. FP32 performance is listed at 6,060 TFLOPS, and there are no dedicated ray tracing or tensor cores.

The Radeon 860M, in contrast, uses AMD's RDNA 3.5 architecture on a 4 nm TSMC process, representing the Navi III IGP generation for Strix Point Mobile systems. Its transistor count and die size are listed as unknown, but the architectural approach is completely different: this is an integrated GPU that shares system memory rather than having its own dedicated VRAM. The 860M has 512 shading units, 32 TMUs, and 16 ROPs — far fewer than the GTX 980 Ti — but it includes 8 ray tracing cores, a feature the older NVIDIA card cannot offer. Its FP32 throughput is 3,072 TFLOPS, and it also supports FP16 at 3,072 TFLOPS with a 1:1 ratio, whereas the GTX 980 Ti has no listed FP16 capability. The 860M's pixel rate is 48 GPixel/s and texture rate is 96 GTexel/s, both roughly half of the GTX 980 Ti's figures. The RDNA 3.5 architecture brings support for DirectX 12 Ultimate (12_2), while Maxwell 2.0 is limited to DirectX 12 (12_1). Both support OpenGL 4.6 and Vulkan 1.4, but the 860M's Vulkan implementation benefits from a more modern hardware design.

Specification Differences

The specification gap between these two cards is vast. The GTX 980 Ti uses the GM200 chip, while the Radeon 860M uses the Krackan Point chip. Process node differs dramatically: 28 nm for the GTX 980 Ti versus 4 nm for the Radeon 860M. The GTX 980 Ti has a base clock of 1000 MHz and boost clock of 1076 MHz; the Radeon 860M has a much lower base clock of 600 MHz but a significantly higher boost clock of 3000 MHz. Memory is another fundamental split: the GTX 980 Ti has 6 GB of GDDR5 on a 384-bit bus with 336.6 GB/s bandwidth and 1753 MHz memory clock (7 Gbps effective), while the Radeon 860M uses system-shared memory with system-dependent bandwidth. Shading units are 2,816 vs 512, TMUs are 176 vs 32, and ROPs are 96 vs 16. FP32 performance is 6,060 TFLOPS vs 3,072 TFLOPS. The GTX 980 Ti's TDP is 250 W, while the 860M draws only 15 W. Power connectors are required only on the GTX 980 Ti (1x 6-pin + 1x 8-pin); the 860M has none. The GTX 980 Ti suggests a 600 W power supply, while the 860M has no such requirement. Bus interface differs: PCIe 3.0 x16 for the GTX 980 Ti versus PCIe 4.0 x8 for the 860M. Display outputs on the GTX 980 Ti include 1x DVI, 1x HDMI 2.0, and 3x DisplayPort 1.2, while the 860M's outputs are portable-device dependent. The GTX 980 Ti measures 267 mm in length, 111 mm in height, and 40 mm in width, occupying a dual-slot form factor; the 860M is an IGP with no listed dimensions. Production status shows the GTX 980 Ti as end-of-life versus the 860M as active. Release dates are 2015-06-01 for the GTX 980 Ti and 2025-02-28 for the Radeon 860M. The GTX 980 Ti had a launch MSRP of 649 USD; the 860M has no listed launch MSRP.

Head-to-Head Benchmarks

The two shared benchmark results tell a consistent story of NVIDIA dominance, though the magnitude varies significantly between APIs. In Geekbench OpenCL, the GTX 980 Ti scores 43,513 against the Radeon 860M's 22,759, a delta of 91.2%. This is the largest gap between the two cards in any test, reflecting the GTX 980 Ti's massive advantage in raw compute resources: 2,816 shading units versus 512, and 6,060 TFLOPS FP32 versus 3,072 TFLOPS. OpenCL workloads tend to scale well with shading unit count and memory bandwidth, both of which heavily favor the older NVIDIA card. The GTX 980 Ti's 336.6 GB/s of dedicated GDDR5 bandwidth versus the 860M's system-shared memory further widens the gap in memory-intensive compute tasks.

In Geekbench Vulkan, the margin narrows considerably but still favors the GTX 980 Ti: 47,724 vs 30,043, a 58.9% advantage. Vulkan's lower-level API overhead may help the Radeon 860M close some of the gap, as modern architectures like RDNA 3.5 are designed with Vulkan efficiency in mind. The 860M's higher boost clock of 3000 MHz, compared to the GTX 980 Ti's 1076 MHz, also helps it in latency-sensitive workloads. Still, the GTX 980 Ti's superior hardware resources — more than five times the shading units and three times the ROPs — prove decisive even in a modern API. The win count stands at 2-0 in favor of the GTX 980 Ti, with no benchmark victories for the Radeon 860M in this head-to-head comparison.

Looking at the broader benchmark landscape, the GTX 980 Ti also records a 1,321 score in 3DMark Steel Nomad DX12 and a 19,520 in Geekbench Metal, tests that the Radeon 860M does not share. The nearest rivals for each card provide additional context: the GTX 980 Ti's closest competitor is the AMD FirePro S7150 at 28,117 average score (0.3% behind), while the Radeon 860M's nearest rival is the NVIDIA GeForce MX550 at 26,421 (0.1% ahead). These rival comparisons reinforce that the GTX 980 Ti, despite its age, performs in a higher overall tier than the 860M. Yet the 860M's 72nd percentile ranking, achieved at 15 W with no dedicated memory, underscores how far integrated graphics have come. The GTX 980 Ti's 250 W power draw buys roughly 6.1% higher average benchmark performance — a tradeoff that makes sense only in desktop contexts where power and space are not constraints.