AMD Radeon HD 8950M vs NVIDIA GeForce GTX 1650 SUPER Comparison

Head-to-Head Benchmarks

The recorded data contains two direct comparisons between the AMD Radeon HD 8950M and the NVIDIA GeForce GTX 1650 SUPER, and in both instances the NVIDIA part dominates decisively.

In Geekbench OpenCL, the GTX 1650 SUPER scores 43,875 points against the HD 8950M’s 13,772. That is a delta of 68.6% in favor of NVIDIA, meaning the newer card delivers roughly three times the compute throughput in this workload. The gap widens further in Geekbench Vulkan, where the GTX 1650 SUPER posts 50,519 versus 12,979 for the AMD part, a 74.3% advantage. The Vulkan result is particularly telling because it reflects not just raw shading power but also driver efficiency and architectural maturity under a modern low-level API.

Looking at the aggregate picture, the HD 8950M’s average benchmark score of 13,376 places it in the 54th percentile of all GPUs in the database, with its nearest rivals being the Radeon RX 5500M (13,356, 0.2% slower), FirePro M6100 (13,354, 0.2% slower), Radeon Pro 555 (13,407, 0.2% faster), and Radeon Pro 555X (13,321, 0.4% slower). The GTX 1650 SUPER, by contrast, sits at the 50th percentile with an average score of 11,047, though its nearest rivals include the Radeon RX 550 (11,075, 0.2% faster) and the MX350 (10,883, 1.5% slower). Note that the percentile rankings are computed across entirely different benchmark suites, so they are not directly comparable; what matters is the head-to-head delta, which is unambiguous.

The wins tally is stark: the GTX 1650 SUPER wins 2 out of 2 head-to-head tests, while the HD 8950M records zero victories. There is no scenario in the available data where the AMD mobile chip comes out ahead. The margin is not narrow either; it is a generational gap in compute performance that no driver optimization or workload tuning could plausibly close.

FAQ

Q: Which GPU wins in Geekbench OpenCL?

A: The NVIDIA GeForce GTX 1650 SUPER wins with a score of 43,875 against 13,772 for the AMD Radeon HD 8950M, a 68.6% lead.

Q: How large is the Vulkan performance gap?

A: In Geekbench Vulkan, the GTX 1650 SUPER scores 50,519 versus 12,979, which is a 74.3% advantage for NVIDIA.

Q: What are the average benchmark scores for each card?

A: The HD 8950M has an average benchmark score of 13,376, while the GTX 1650 SUPER has an average of 11,047. These averages come from different benchmark sets, so they should not be used for direct comparison; the head-to-head tests are the authoritative measure.

Q: How does the HD 8950M compare to its nearest rivals?

A: The HD 8950M is essentially tied with the Radeon RX 5500M (0.2% faster), FirePro M6100 (0.2% faster), and Radeon Pro 555X (0.4% faster), while the Radeon Pro 555 is 0.2% faster than it. All deltas are within a fraction of a percent.

Q: How does the GTX 1650 SUPER compare to its nearest rivals?

A: The GTX 1650 SUPER is 0.2% slower than the Radeon RX 550 and the RTX PRO 6000D Blackwell Max-Q, and 1.5% faster than the GeForce MX350. The deltas are again tiny, indicating a tight cluster around its average score.

Q: Does the HD 8950M win any benchmark?

A: No. The recorded head-to-head results show zero wins for the AMD card and two wins for the NVIDIA card.

Architecture Differences

The two GPUs come from fundamentally different eras and design philosophies. The AMD Radeon HD 8950M is built on GCN 2.0 architecture, with the chip codenamed Saturn, fabricated on a 28 nm process at TSMC. It packs 2,080 million transistors into a 160 mm² die, yielding a transistor density of 13.0 million per square millimeter. The NVIDIA GeForce GTX 1650 SUPER uses the Turing architecture with the TU116 chip, also from TSMC but on a 12 nm process. It contains 6,600 million transistors across a 284 mm² die, for a density of 23.2 million per square millimeter. The process advantage is clear: NVIDIA fits more than three times the transistors into less than double the die area.

Shader resources follow the same pattern. The HD 8950M has 768 shading units, 48 texture mapping units, and 16 raster output pipelines. The GTX 1650 SUPER fields 1,280 shading units, 80 TMUs, and 32 ROPs. That is a 67% increase in shader count, a 67% increase in TMUs, and a doubling of ROPs. Pixel rate goes from 17.20 GPixel/s to 55.20 GPixel/s, and texture rate jumps from 51.60 GTexel/s to 138.0 GTexel/s. Floating-point performance scales similarly: the AMD card delivers 1.651 TFLOPS of FP32, while the NVIDIA card delivers 4.416 TFLOPS, with an additional FP16 peak of 8.832 TFLOPS at a 2:1 ratio, a capability the AMD part does not list.

Memory configuration also diverges sharply. The HD 8950M uses 2 GB of GDDR5 on a 128-bit bus, running at 1375 MHz with 5.5 Gbps effective speed, for a bandwidth of 88.00 GB/s. The GTX 1650 SUPER doubles the capacity to 4 GB of GDDR6 on the same 128-bit bus, but at 1500 MHz with 12 Gbps effective speed, yielding 192.0 GB/s of bandwidth. The NVIDIA card thus offers twice the capacity and more than twice the bandwidth.

API support differs at the margins. Both support DirectX 12, but the HD 8950M is limited to feature level 12_0 while the GTX 1650 SUPER reaches 12_1. OpenGL support is identical at 4.6. Vulkan support shows a gap: the AMD card lists Vulkan 1.2.170, while the NVIDIA card lists Vulkan 1.4, reflecting the newer driver stack and hardware capabilities.

Clock behavior is another separator. The HD 8950M has no base or boost clock listed, only memory clocks. The GTX 1650 SUPER has a base clock of 1530 MHz and a boost clock of 1725 MHz.

The Verdict

The data points to a clear verdict: the NVIDIA GeForce GTX 1650 SUPER is the superior GPU in every measurable head-to-head test. For workloads that rely on OpenCL compute or Vulkan rendering, the NVIDIA card is roughly three to four times faster, with deltas of 68.6% and 74.3% respectively. The HD 8950M cannot compete on raw throughput, memory bandwidth, or API feature level.

Who should pick the AMD Radeon HD 8950M? The data suggests only a buyer constrained to the older mobile MXM form factor, as the HD 8950M uses an MXM Module slot while the GTX 1650 SUPER is a Dual-slot card with a 1x 6-pin power connector and a suggested PSU of 300 W. If the system requires an MXM module and the benchmark suite is limited to the two Geekbench tests recorded, the AMD part is the only option. However, there is no performance justification in the database for choosing it.

Who should pick the NVIDIA GeForce GTX 1650 SUPER? Anyone who needs compute performance, modern API support, or memory capacity. The GTX 1650 SUPER offers 4 GB of GDDR6 versus 2 GB of GDDR5, Vulkan 1.4 versus 1.2.170, DirectX 12_1 versus 12_0, and a launch MSRP of 159 USD. Its average benchmark score of 11,047 places it in a tight cluster with the Radeon RX 550 and MX350, but its head-to-head results against the HD 8950M are not remotely close.

The percentile data adds nuance. The HD 8950M sits at the 54th percentile of all GPUs, while the GTX 1650 SUPER sits at the 50th. This does not contradict the head-to-head results; it reflects different benchmark populations and the fact that the HD 8950M’s nearest rivals are all within 0.4% of its average score, indicating a crowded field of similar-performance parts. The GTX 1650 SUPER’s cluster is similarly tight, but its absolute head-to-head scores dwarf those of the AMD card.

In short, the GTX 1650 SUPER is the better GPU by every recorded metric. The HD 8950M is an end-of-life product from 2013, and the benchmark data reflects six years of architectural advancement. There is no workload in the database where the AMD part emerges victorious.

Specification Differences

| Specification | AMD Radeon HD 8950M | NVIDIA GeForce GTX 1650 SUPER |

|---|---|---|

| Architecture | GCN 2.0 | Turing |

| Chip | Saturn | TU116 |

| Process node | 28 nm | 12 nm |

| Transistors | 2,080 million | 6,600 million |

| Die size | 160 mm² | 284 mm² |

| Transistor density | 13.0M / mm² | 23.2M / mm² |

| Base clock | Not listed | 1530 MHz |

| Boost clock | Not listed | 1725 MHz |

| Memory clock | 1375 MHz, 5.5 Gbps effective | 1500 MHz, 12 Gbps effective |

| Memory size | 2 GB | 4 GB |

| Memory type | GDDR5 | GDDR6 |

| Memory bus width | 128 bit | 128 bit |

| Memory bandwidth | 88.00 GB/s | 192.0 GB/s |

| Shading units | 768 | 1280 |

| TMUs | 48 | 80 |

| ROPs | 16 | 32 |

| Pixel rate | 17.20 GPixel/s | 55.20 GPixel/s |

| Texture rate | 51.60 GTexel/s | 138.0 GTexel/s |

| FP32 performance | 1.651 TFLOPS | 4.416 TFLOPS |

| FP16 performance | Not listed | 8.832 TFLOPS (2:1) |

| TDP | 100 W | 100 W |

| Slot width | MXM Module | Dual-slot |

| Power connectors | Not listed | 1x 6-pin |

| Suggested PSU | Not listed | 300 W |

| Display outputs | Portable Device Dependent | 1x DVI, 1x HDMI 2.0, 1x DisplayPort 1.4a |

| DirectX support | 12 (12_0) | 12 (12_1) |

| OpenGL support | 4.6 | 4.6 |

| Vulkan support | 1.2.170 | 1.4 |

| Release date | 2013-05-13 | 2019-11-21 |

| Production status | End-of-life | End-of-life |

| Predecessor | London | GeForce 10 |

| Successor | Gem System | GeForce 20 |

| Launch MSRP | Not listed | 159 USD |