GPU Comparison

The GeForce 920M and Quadro P600 are both end-of-life NVIDIA mobile/compact GPUs, but they sit on opposite sides of a massive generational and architectural divide. The 920M is a 2015-era entry-level Kepler part designed for basic laptops, while the P600 is a 2017-era professional Pascal card built for compact workstations. The benchmark data shows a stark performance gap, with the Quadro P600 winning every head-to-head test by a wide margin, though the GeForce 920M's average benchmark score is actually higher than the P600's due to differing test sets.

Head-to-Head Benchmarks

The two direct comparisons available are Geekbench OpenCL and Geekbench Vulkan, and the Quadro P600 dominates both. In the OpenCL test, the P600 scores 11,181 against the 920M's 3,725. That is a delta of -66.7% for the 920M, meaning the P600 is roughly three times faster in raw compute throughput. The Vulkan result is even more lopsided: the P600 hits 9,755 while the 920M manages only 2,849, a -70.8% delta. This is not a close contest; the P600 is in a different performance class entirely.

Looking at the average benchmark scores, the picture becomes more complex. The 920M posts an average of 3,287 across its two tests, while the P600 averages 2,923 across nine tests. The P600's average is dragged down by its Passmark DirectX scores, which are extremely low (14 in DirectX 10, 24 in DirectX 11, 14 in DirectX 12, and 56 in DirectX 9). These are likely synthetic legacy tests that do not reflect the card's real-world capability. The P600's Passmark G3D score of 3,317 and GPU compute score of 1,415 are far more representative of its actual performance, and both are in line with a card that scores 11,181 in OpenCL.

The nearest-rival data reinforces the 920M's position. It sits at the 20th percentile of all GPUs, with its closest comparables being the GeForce GT 730M (delta -0.9%), Intel HD Graphics 530 (delta -1.4%), and GeForce GT 640 (delta +2.4%). These are all entry-level or integrated parts from the same era. The P600, by contrast, is at the 19th percentile, but its nearest rivals are the GeForce RTX 4060 Ti 8 GB (delta +0.3%), RTX PRO 4000 Blackwell SFF (delta +0.4%), and RTX 4060 Ti 16 GB (delta +0.6%). This is misleading at first glance, the P600's average is being compared to much newer cards, but it shows that the P600's average score is not a measure of its compute strength, which is clearly far higher than the 920M's.

Architecture Differences

The 920M is built on Kepler 2.0 architecture using the GK208B chip, manufactured on a 28 nm process at TSMC. It packs 1,020 million transistors into an 87 mm² die, giving a transistor density of 11.7M per mm². The P600 uses Pascal architecture with the GP107 chip, made on Samsung's 14 nm process. It holds 3,300 million transistors on a 132 mm² die, with a density of 25.0M per mm². The P600's process node is two generations newer, which explains its ability to double the transistor count and density while only increasing TDP by 7 W.

Clock speeds tell a similar story. The 920M runs at a fixed 954 MHz for both base and boost, with memory clocked at 900 MHz (1,800 Mbps effective). The P600 starts at 1,329 MHz base and boosts to 1,557 MHz, with memory at 1,002 MHz (4 Gbps effective). The P600's higher clocks, combined with its newer architecture, produce dramatically higher fill rates: 24.91 GPixel/s pixel rate and 37.37 GTexel/s texture rate versus the 920M's 7.632 GPixel/s and 30.53 GTexel/s. The P600's FP32 throughput is 1,195.8 GFLOPS compared to the 920M's 732.7 GFLOPS. The P600 also has a listed FP16 rate of 18.68 GFLOPS (1:64), while the 920M has no FP16 data.

Memory subsystems are fundamentally different. The 920M uses 2 GB of DDR3 on a 64-bit bus, yielding 14.40 GB/s of bandwidth. The P600 also has 2 GB, but it is GDDR5 on a 128-bit bus, delivering 64.13 GB/s, over four times the bandwidth. This is critical for any workload that streams data, and it is a major reason the P600's compute scores are so much higher. The P600 also has 16 ROPs versus the 920M's 8, and 24 TMUs versus 32, though the P600's higher clocks compensate for the TMU deficit in texture rate.

Where Each One Wins

The GeForce 920M has no benchmark wins in the head-to-head tests, but its average score of 3,287 is higher than the P600's 2,923. This is a statistical artifact of the test sets, not a real performance advantage. The 920M is competitive with integrated graphics like the Intel HD Graphics 530 and older discrete parts like the GT 730M, which means it is suited for basic 2D desktop work, older or light 3D games, and video playback. It is a "better than nothing" GPU for a budget laptop.

The Quadro P600 wins every meaningful compute test. Its OpenCL score of 11,181 and Vulkan score of 9,755 make it suitable for GPU-accelerated rendering, CAD, video encoding, and other professional workloads that leverage compute. Its Passmark G3D score of 3,317 indicates solid DirectX 11-era gaming performance as well, though the DirectX 12 score of 14 is anomalously low and should be ignored. The P600's four mini-DisplayPort 1.4a outputs and 150 mm length make it a practical choice for a compact workstation where multiple monitors are needed.

The 920M's only real advantage is its 33 W TDP versus the P600's 40 W, and its IGP form factor, which means it is soldered onto a laptop motherboard. The P600 requires a PCIe 3.0 x16 slot and a 200 W suggested PSU, though it draws no power from a connector. For a builder, the P600 is the clear choice if you need compute performance; the 920M is only viable if you are stuck with it in a laptop and cannot upgrade.

FAQ

Q: Which GPU has a higher OpenCL score?

A: The Quadro P600 scores 11,181 in Geekbench OpenCL, while the GeForce 920M scores 3,725, a -66.7% delta in favor of the P600.

Q: Are these GPUs good for modern gaming?

A: The P600's Passmark G3D score of 3,317 suggests it can handle older or lighter 3D titles, but its DirectX 12 score of 14 is extremely low. The 920M's average score of 3,287 places it near the Intel HD Graphics 530 (delta -1.4%), which is not a gaming part.

Q: What is the memory bandwidth difference?

A: The P600 has 64.13 GB/s of bandwidth from its GDDR5 memory on a 128-bit bus. The 920M has 14.40 GB/s from DDR3 on a 64-bit bus. The P600 has over four times the bandwidth.

Q: Which card is more power-efficient?

A: The 920M has a 33 W TDP versus the P600's 40 W TDP. However, the P600 delivers far more performance per watt, given its 1,195.8 GFLOPS FP32 versus the 920M's 732.7 GFLOPS at a 7 W higher power draw.

Q: Can the P600 be used in a small form factor PC?

A: Yes, it is a single-slot card measuring 150 mm in length and 69 mm in height, with no power connectors required. It needs a PCIe 3.0 x16 slot and a 200 W suggested PSU.

Q: What is the transistor count comparison?

A: The P600 has 3,300 million transistors on a 132 mm² die, while the 920M has 1,020 million on an 87 mm² die. The P600 uses a 14 nm Samsung process; the 920M uses a 28 nm TSMC process.

Specification Differences

| Field | GeForce 920M | Quadro P600 |

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

| Chip | GK208B | GP107 |

| Architecture | Kepler 2.0 | Pascal |

| Process Node | 28 nm | 14 nm |

| Foundry | TSMC | Samsung |

| Transistors | 1,020 million | 3,300 million |

| Die Size | 87 mm² | 132 mm² |

| Transistor Density | 11.7M / mm² | 25.0M / mm² |

| Base Clock | 954 MHz | 1329 MHz |

| Boost Clock | 954 MHz | 1557 MHz |

| Memory Clock | 900 MHz (1800 Mbps effective) | 1002 MHz (4 Gbps effective) |

| Memory Type | DDR3 | GDDR5 |

| Memory Bus | 64 bit | 128 bit |

| Memory Bandwidth | 14.40 GB/s | 64.13 GB/s |

| TMUs | 32 | 24 |

| ROPs | 8 | 16 |

| Pixel Rate | 7.632 GPixel/s | 24.91 GPixel/s |

| Texture Rate | 30.53 GTexel/s | 37.37 GTexel/s |

| FP32 | 732.7 GFLOPS | 1,195.8 GFLOPS |

| FP16 | N/A | 18.68 GFLOPS (1:64) |

| TDP | 33 W | 40 W |

| Slot Width | IGP | Single-slot |

| Suggested PSU | N/A | 200 W |

| Bus Interface | PCIe 3.0 x8 | PCIe 3.0 x16 |

| Display Outputs | Portable Device Dependent | 4x mini-DisplayPort 1.4a |

| DirectX | 12 (11_0) | 12 (12_1) |

| Vulkan | 1.2.175 | 1.4 |

| Dimensions | N/A | 150 mm x 69 mm |

| Release Date | 2015-03-12 | 2017-02-06 |

The Verdict

The data is unambiguous. The Quadro P600 outperforms the GeForce 920M in every head-to-head benchmark by a factor of three or more. The P600's 11,181 OpenCL score versus 3,725 for the 920M, and its 9,755 Vulkan score versus 2,849, make it the only choice for any compute-oriented task. The P600's 64.13 GB/s memory bandwidth, 16 ROPs, and 1,195.8 GFLOPS FP32 throughput are all in a different league from the 920M's 14.40 GB/s, 8 ROPs, and 732.7 GFLOPS.

The 920M is a 28 nm Kepler part from 2015, soldered into laptops as an IGP. Its only advantage is a lower 33 W TDP and the fact that it does not require a PCIe slot. It sits at the 20th percentile of all GPUs, competing with Intel integrated graphics and decade-old discrete parts like the GT 640. If you are building a system, there is no reason to choose the 920M over the P600 unless you are constrained to a laptop motherboard with no upgrade path.

The P600, despite its lower average benchmark score (2,923 versus 3,287) due to anomalous Passmark DirectX results, is the clear winner. Its 19th percentile ranking is misleading because its nearest rivals are modern RTX 4060 Ti variants, which indicates the average score is not representative of its compute strength. For a professional workload requiring OpenCL or Vulkan acceleration, the P600 is the only rational pick. For gaming, its G3D score of 3,317 is workable, but the 920M is not a gaming GPU by any measure. Pick the P600 if you need performance; avoid the 920M unless it is already in your laptop.