NVIDIA GeForce 920M vs NVIDIA GeForce RTX 4060 Ti 16 GB Comparison

NVIDIA
GEFORCE

NVIDIA GeForce 920M

CORE STATE GK208B
VRAM 2 GB
CLOCK SPEED 954 MHz
TDP 33 W
BUS WIDTH 64 bit
ARCHITECTURE Kepler 2.0
nm
PROCESS 28 nm
LAUNCH DATE 2015
VS
NVIDIA
GEFORCE

GeForce RTX 4060 Ti 16 GB

CORE STATE AD106
VRAM 16 GB
CLOCK SPEED 2535 MHz
TDP 165 W
BUS WIDTH 128 bit
ARCHITECTURE Ada Lovelace
nm
PROCESS 5 nm
LAUNCH DATE 2023

PERFORMANCE BENCHMARKS

geekbench_opencl
3,725
N/A
geekbench_vulkan
2,849
N/A
3dmark_3dmark_steel_nomad_dx12
N/A
2,907

Analysis: NVIDIA GeForce 920M vs NVIDIA GeForce RTX 4060 Ti 16 GB

# NVIDIA GeForce 920M vs NVIDIA GeForce RTX 4060 Ti 16 GB

The GeForce 920M is a 2015-era mobile IGP built on 28 nm Kepler 2.0, while the RTX 4060 Ti 16 GB is a 2023 desktop card on 5 nm Ada Lovelace. Benchmark data shows the 920M averaging 3287 points across its two tests, while the RTX 4060 Ti 16 GB averages 2907 points in its single benchmark — but these scores come from different test suites and cannot be directly compared. The 920M sits in the 20th percentile of all GPUs, with rivals like the GT 730M (3316) and Intel HD Graphics 530 (3332) within a few percentage points. The RTX 4060 Ti 16 GB sits in the 18th percentile, with its nearest rival being the RTX PRO 4000 Blackwell SFF (2910, -0.1% delta), making it statistically indistinguishable from that card in the Steel Nomad DX12 test.

FAQ

Q: Which card has more raw compute power?

A: The RTX 4060 Ti 16 GB delivers 22.06 TFLOPS of FP32 performance versus 732.7 GFLOPS for the 920M — a roughly 30x advantage. The Ada Lovelace card also matches that figure in FP16 (22.06 TFLOPS 1:1), while the Kepler part has no listed FP16 capability.

Q: How do the memory subsystems compare?

A: The RTX 4060 Ti 16 GB has 16 GB of GDDR6 on a 128-bit bus, yielding 288.0 GB/s of bandwidth. The 920M has 2 GB of DDR3 on a 64-bit bus, providing just 14.40 GB/s. That is a 20x difference in bandwidth and an 8x difference in capacity.

Q: Which card supports modern APIs better?

A: The RTX 4060 Ti 16 GB supports DirectX 12 Ultimate (12_2) and Vulkan 1.4, while the 920M supports DirectX 12 (11_0) and Vulkan 1.2.175. Both support OpenGL 4.6. The Ada card also has 34 RT cores and 136 tensor cores; the Kepler part has neither.

Q: Are these cards comparable in power consumption?

A: No. The 920M has a 33 W TDP and uses no power connectors, drawing power from the motherboard as an integrated-style part. The RTX 4060 Ti 16 GB has a 165 W TDP, requires a 1x 16-pin connector, and suggests a 450 W power supply.

Q: What is the launch MSRP of the RTX 4060 Ti 16 GB?

A: The launch MSRP is 499 USD. The 920M has no listed launch MSRP in the data.

Q: How do their benchmark scores compare to their closest rivals?

A: The 920M's average of 3287 is 0.9% below the GT 730M (3316) and 1.4% below the Intel HD Graphics 530 (3332), but 2.4% above the GT 640 (3210). The RTX 4060 Ti 16 GB's 2907 is 0.1% below the RTX PRO 4000 Blackwell SFF (2910) and 0.2% below the RTX 4060 Ti 8 GB (2913).

Where Each One Wins

The 920M wins in scenarios where power draw and physical footprint are the limiting factors. Its 33 W TDP and IGP form factor mean it can be integrated into thin laptops without dedicated cooling or power delivery. It also wins on portability — the display outputs are "Portable Device Dependent," meaning it relies on the laptop's built-in panel. The 920M's 954 MHz base and boost clocks are identical, suggesting a stable, low-power operation profile suitable for basic 2D workloads and light media playback.

The RTX 4060 Ti 16 GB wins in every performance category where dedicated rendering power matters. Its 16 GB memory capacity is critical for large textures, high-resolution assets, and AI workloads that exceed 8 GB. The 288.0 GB/s bandwidth allows it to feed its 4352 shading units and 136 TMUs at rates the 920M cannot approach. The inclusion of RT cores and tensor cores gives it hardware acceleration for ray-traced effects and DLSS-style upscaling, neither of which the Kepler architecture supports. The dual-slot form factor and 240 mm length indicate it is designed for desktop cases with proper airflow.

Architecture Differences

The 920M uses the GK208B chip on Kepler 2.0 architecture, manufactured on a 28 nm process at TSMC. It packs 1,020 million transistors into an 87 mm² die, yielding a transistor density of 11.7M per mm². The RTX 4060 Ti 16 GB uses the AD106 chip on Ada Lovelace architecture, manufactured on a 5 nm process at TSMC. It contains 22,900 million transistors in a 188 mm² die, achieving 121.8M per mm² — a 10x improvement in density.

The Kepler part has 384 shading units, 32 TMUs, and 8 ROPs. The Ada part has 4352 shading units, 136 TMUs, and 48 ROPs. The 920M has no RT cores or tensor cores; the RTX 4060 Ti 16 GB has 34 RT cores and 136 tensor cores. This is the defining architectural split: Kepler 2.0 predates hardware ray tracing and tensor processing, while Ada Lovelace is built around them.

API support differs meaningfully. The 920M lists DirectX 12 (11_0), which means it meets the minimum feature level but lacks the full DirectX 12 Ultimate feature set. The RTX 4060 Ti 16 GB supports DirectX 12 Ultimate (12_2), which includes mesh shaders, variable rate shading, and other modern rendering features. Vulkan support is also newer on the Ada card (1.4 vs 1.2.175).

The memory architecture is fundamentally different. The 920M uses DDR3 with a 64-bit bus; the RTX 4060 Ti 16 GB uses GDDR6 with a 128-bit bus. The 920M's memory runs at 900 MHz (1800 Mbps effective), while the Ada card runs at 2250 MHz (18 Gbps effective). Pixel rate tells the story: 7.632 GPixel/s for the 920M versus 121.7 GPixel/s for the RTX 4060 Ti 16 GB.

Specification Differences

| Specification | GeForce 920M | RTX 4060 Ti 16 GB |

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

| Architecture | Kepler 2.0 | Ada Lovelace |

| Process node | 28 nm | 5 nm |

| Transistors | 1,020 million | 22,900 million |

| Die size | 87 mm² | 188 mm² |

| Base clock | 954 MHz | 2310 MHz |

| Boost clock | 954 MHz | 2535 MHz |

| Memory size | 2 GB | 16 GB |

| Memory type | DDR3 | GDDR6 |

| Memory bus | 64 bit | 128 bit |

| Memory bandwidth | 14.40 GB/s | 288.0 GB/s |

| Shading units | 384 | 4352 |

| TMUs | 32 | 136 |

| ROPs | 8 | 48 |

| RT cores | None | 34 |

| Tensor cores | None | 136 |

| Pixel rate | 7.632 GPixel/s | 121.7 GPixel/s |

| Texture rate | 30.53 GTexel/s | 344.8 GTexel/s |

| FP32 | 732.7 GFLOPS | 22.06 TFLOPS |

| TDP | 33 W | 165 W |

| Slot width | IGP | Dual-slot |

| Power connectors | None | 1x 16-pin |

| Suggested PSU | None | 450 W |

| Bus interface | PCIe 3.0 x8 | PCIe 4.0 x8 |

| DirectX | 12 (11_0) | 12 Ultimate (12_2) |

| Vulkan | 1.2.175 | 1.4 |

| Release date | 2015-03-12 | 2023-05-17 |

Head-to-Head Benchmarks

The FACT PACK lists no direct head-to-head benchmark results between these two cards. The 920M has two benchmark entries: Geekbench OpenCL (3725) and Geekbench Vulkan (2849), averaging 3287. The RTX 4060 Ti 16 GB has one entry: 3DMark Steel Nomad DX12 (2907). These tests measure different workloads — Geekbench OpenCL and Vulkan target compute and general-purpose GPU tasks, while 3DMark Steel Nomad DX12 is a modern game-like DirectX 12 workload.

Because the test suites do not overlap, no direct comparison is statistically valid. What the data does show is the relative standing of each card within its own benchmark ecosystem. The 920M's Geekbench scores put it in the 20th percentile of all GPUs, with its nearest rival the GT 730M at 3316 (-0.9%). The RTX 4060 Ti 16 GB sits in the 18th percentile, with its nearest rival the RTX PRO 4000 Blackwell SFF at 2910 (-0.1%). Both cards are near the bottom of their respective benchmark pools, but the RTX 4060 Ti 16 GB's single score is tightly clustered with its rivals — deltas of -0.1%, -0.2%, and +0.5% — while the 920M's rivals span a slightly wider range of -3% to +2.4%.

The architectural data provides the more meaningful comparison. The RTX 4060 Ti 16 GB's 22.06 TFLOPS FP32 is 30x the 920M's 732.7 GFLOPS. Its texture rate of 344.8 GTexel/s is 11x the 920M's 30.53 GTexel/s. The pixel rate gap is even larger: 121.7 GPixel/s versus 7.632 GPixel/s, a 16x difference. These are the numbers that matter for real workloads, even if the benchmark scores themselves cannot be directly compared.

The Verdict

The data presents two cards with no overlapping use cases. The GeForce 920M is an end-of-life integrated-class part from 2015, built for portability and minimal power draw. Its 33 W TDP, IGP slot width, and lack of power connectors make it suitable for basic laptops where battery life and heat are priorities. Its 2 GB DDR3 memory and 14.40 GB/s bandwidth limit it to light 2D work, video playback, and older or indie games at low settings. The 20th percentile ranking and close rivalry with Intel HD Graphics 530 confirm it belongs in the integrated-graphics tier.

The RTX 4060 Ti 16 GB is a dual-slot desktop card requiring a 450 W power supply, designed for serious gaming and content creation. Its 16 GB GDDR6 memory and 288.0 GB/s bandwidth handle large textures and high-resolution assets without swapping. The 34 RT cores and 136 tensor cores enable hardware ray tracing and AI-accelerated features that the 920M cannot execute at all. The 18th percentile ranking in the Steel Nomad DX12 test shows it is not a top-tier card in that specific workload, but its nearest rival delta of -0.1% versus the RTX PRO 4000 Blackwell SFF indicates it is competitive with professional-class hardware in that test.

Buyers should choose based on the platform, not the benchmark scores. If the system is a laptop from 2015 or earlier, the 920M is what it has, and it will handle basic tasks. If building a new desktop with a 450 W PSU and dual-slot clearance, the RTX 4060 Ti 16 GB is the clear choice for anything beyond 2D workloads. The 30x FP32 gap, 20x bandwidth gap, and the presence of RT and tensor cores make the Ada card objectively superior in every measurable performance category. The 920M's only advantages are its 33 W power draw and IGP form factor, which are irrelevant in a desktop context. The verdict is unambiguous: the RTX 4060 Ti 16 GB is the only one of these two worth considering for any modern compute or gaming workload.

DETAILED SPECIFICATIONS

SPECIFICATION
920M
RTX 4060 Ti 16 GB
Core Specs
Shading Units
384
4,352 +1033.3%
Shaders
384
4,352 +1033.3%
TMUs
32
136 +325.0%
ROPs
8
48 +500.0%
SM Count
—
34
Clocks
Base Clock
954 MHz
2310 MHz
Boost Clock
954 MHz
2535 MHz
Memory Clock
900 MHz 1800 Mbps effective
2250 MHz 18 Gbps effective
Memory
Memory Size
2 GB
16 GB
VRAM (MB)
2,048
16,384 +700.0%
Memory Type
DDR3
GDDR6
Memory Bus
64 bit
128 bit
Bandwidth
14.40 GB/s
288.0 GB/s
Cache
L1 Cache
16 KB (per SMX)
128 KB (per SM)
L2 Cache
512 KB
32 MB
Performance
Pixel Rate
7.632 GPixel/s
121.7 GPixel/s
Texture Rate
30.53 GTexel/s
344.8 GTexel/s
FP32 (TFLOPS)
732.7 GFLOPS
22.06 TFLOPS
FP64 (TFLOPS)
30.53 GFLOPS (1:24)
344.8 GFLOPS (1:64)
FP16 (TFLOPS)
—
22.06 TFLOPS (1:1)
AI/RT
RT Cores
—
34
Tensor Cores
—
136
Power
TDP
33 W
165 W
TDP (W)
33
165 +400.0%
Suggested PSU
—
450 W
Power Connectors
None
1x 16-pin
Architecture
Architecture
Kepler 2.0
Ada Lovelace
GPU Name
GK208B
AD106
Generation
GeForce 900M
GeForce 40
Process Size
28 nm
5 nm
Transistors
1,020 million
22,900 million
Die Size
87 mm²
188 mm²
Foundry
TSMC
TSMC
Density
11.7M / mm²
121.8M / mm²
API Support
DirectX
12 (11_0)
12 Ultimate (12_2)
OpenGL
4.6
4.6
Vulkan
1.2.175
1.4
OpenCL
3.0
3.0
CUDA
3.5
8.9
Shader Model
6.5 (5.1)
6.8
Physical
Slot Width
IGP
Dual-slot
Length
—
240 mm 9.4 inches
Height
—
111 mm 4.4 inches
Outputs
Portable Device Dependent
1x HDMI 2.13x DisplayPort 1.4a
Bus Interface
PCIe 3.0 x8
PCIe 4.0 x8
Other
Launch Price
—
499 USD
Production
End-of-life
End-of-life
Predecessor
GeForce 800M
GeForce 30
Successor
GeForce 10 Mobile
GeForce 50
View GeForce 920M Details View GeForce RTX 4060 Ti 16 GB Details