NVIDIA GeForce RTX 3060 Ti vs NVIDIA GeForce RTX 5090 D V2 Comparison

NVIDIA
GEFORCE

NVIDIA GeForce RTX 3060 Ti

CORE STATE GA104
VRAM 8 GB
CLOCK SPEED 1665 MHz
TDP 200 W
BUS WIDTH 256 bit
ARCHITECTURE Ampere
nm
PROCESS 8 nm
LAUNCH DATE 2020
VS
NVIDIA
GEFORCE

GeForce RTX 5090 D V2

CORE STATE GB202
VRAM 24 GB
CLOCK SPEED 2407 MHz
TDP 575 W
BUS WIDTH 384 bit
ARCHITECTURE Blackwell 2.0
nm
PROCESS 5 nm
LAUNCH DATE 2025

PERFORMANCE BENCHMARKS

3dmark_3dmark_steel_nomad_dx12
2,626
16,504
geekbench_opencl
78,927
N/A
geekbench_vulkan
47,784
N/A
passmark_directx_10
132
N/A
passmark_directx_11
163
N/A
passmark_directx_12
78
N/A
passmark_directx_9
234
N/A
passmark_g2d
989
N/A
passmark_g3d
20,349
N/A
passmark_gpu_compute
10,006
N/A

Analysis: NVIDIA GeForce RTX 3060 Ti vs NVIDIA GeForce RTX 5090 D V2

The benchmark data is unambiguous: the NVIDIA GeForce RTX 5090 D V2 is in a completely different performance class than the RTX 3060 Ti. In the single available head-to-head test, the 5090 D V2 outperforms the 3060 Ti by a staggering 528.5%, a margin that far exceeds any generational gap seen in recent history. This is not a close contest; it is a demonstration of how far flagship silicon has moved in roughly five years.

Head-to-Head Benchmarks

The only direct comparison in the data is the 3DMark Steel Nomad DX12 test, and it is a decisive rout. The RTX 5090 D V2 scores 16,504 points, while the RTX 3060 Ti manages just 2,626 points. That is a delta of 528.5% in favor of the newer card. To put that in perspective, the RTX 5090 D V2’s score is over six times higher than the 3060 Ti’s. The result is so lopsided that it effectively sets the tone for every other section of this analysis.

The RTX 5090 D V2’s absolute score also places it in the top tier of all GPUs. Its average benchmark score of 16,504 matches its 3DMark result, and its percentile ranking is 59th among all GPUs. Interestingly, its nearest rivals are not other high-end gaming cards but rather professional and older workstation parts: the NVIDIA T400 (16,508, 0% delta), the AMD Radeon PRO W7500 (16,415, 0.5% behind), the NVIDIA RTX PRO 6000 Blackwell (16,408, 0.6% behind), and the AMD Radeon RX 5700 XT (16,361, 0.9% behind). This suggests that at this score level, the 5090 D V2 is competing in a niche where synthetic test results cluster tightly.

The RTX 3060 Ti, by contrast, has a broader benchmark profile but a much lower peak. Its average score across all tests is 16,129, which is higher than its single 3DMark result due to other tests like Geekbench OpenCL (78,927) and Passmark G3D (20,349). In terms of nearest rivals, the 3060 Ti sits alongside the AMD Radeon RX 9060 (16,014, 0.7% ahead of the 3060 Ti), the AMD Radeon Pro 5600M (16,351, 1.4% ahead), the AMD Radeon RX 5700 XT (16,361, 1.4% ahead), and the AMD Radeon R9 370X (15,862, 1.7% behind). The 3060 Ti’s percentile ranking is also 59, but that is misleading; it is the same percentile because the distribution of scores is not linear, and the 3060 Ti’s average is dragged up by non-gaming compute tests.

When you isolate the gaming-relevant DX12 test, the 5090 D V2 is not just ahead; it is in another stratosphere. The 528.5% delta is the single most important number in this dataset.

Where Each One Wins

The RTX 5090 D V2 wins in every metric that matters for modern gaming and ray tracing. It has more of everything: more shading units (21,760 vs. 4,864), more texture mapping units (680 vs. 152), more render output units (176 vs. 80), more ray tracing cores (170 vs. 38), and more tensor cores (680 vs. 152). Its pixel rate is 423.6 GPixel/s versus 133.2 GPixel/s, and its texture rate is 1,636.8 GTexel/s versus 253.1 GTexel/s. In raw FP32 compute, the 5090 D V2 delivers 104.8 TFLOPS, which is 6.5 times the 3060 Ti’s 16.20 TFLOPS.

The RTX 3060 Ti wins in none of the direct specification comparisons. However, it does have one advantage: it is a much more modest card. Its TDP is 200 W, compared to 575 W for the 5090 D V2. Its suggested PSU is 550 W, versus 950 W. It is also physically smaller (242 mm length vs. 304 mm). For users with older power supplies or smaller cases, the 3060 Ti is a drop-in upgrade. But in terms of pure performance, there is no category where the 3060 Ti comes out ahead.

The 3060 Ti’s benchmark suite includes compute tests, and its Geekbench OpenCL score of 78,927 and Vulkan score of 47,784 are respectable, but they do not translate into gaming frame rates. Its Passmark DirectX 12 score is a mere 78, which is pitiful compared to its own DirectX 11 score of 163, suggesting that the card is not well optimized for modern APIs. The 5090 D V2 has no such deficiency; its single benchmark is a DX12 test.

Architecture Differences

The two cards are built on entirely different architectures and fabrication processes. The RTX 5090 D V2 uses the Blackwell 2.0 architecture on a 5 nm TSMC process. The RTX 3060 Ti uses the Ampere architecture on an 8 nm Samsung process. This is a two-node jump in process technology, which explains a significant portion of the performance gap.

The chip sizes tell the story too. The 5090 D V2’s GB202 die is 750 mm², containing 92,200 million transistors. The 3060 Ti’s GA104 die is 392 mm², with 17,400 million transistors. The newer card packs over five times the transistor count into less than twice the die area, resulting in a transistor density of 122.9M per mm² versus 44.4M per mm². That is a 2.8x density advantage, which is a direct result of the more advanced 5 nm process.

Memory technology has also advanced. The 5090 D V2 uses 24 GB of GDDR7 on a 384-bit bus, delivering 1.34 TB/s of bandwidth. The 3060 Ti uses 8 GB of GDDR6 on a 256-bit bus, delivering 448.0 GB/s. The memory clock is nominally the same (1750 MHz), but the effective data rate is double: 28 Gbps for GDDR7 versus 14 Gbps for GDDR6. The bandwidth advantage of the 5090 D V2 is 3x, which is critical for high-resolution textures and ray tracing workloads.

Other architectural differences include the bus interface. The 5090 D V2 uses PCIe 5.0 x16, while the 3060 Ti uses PCIe 4.0 x16. The newer card also supports DisplayPort 2.1b outputs, whereas the 3060 Ti is limited to DisplayPort 1.4a. Both support HDMI 2.1, but the 5090 D V2 has three DisplayPort outputs versus the 3060 Ti’s three (the pack lists the 5090 D V2 as having 3x DisplayPort 2.1b and 1x HDMI 2.1b; the 3060 Ti has 1x HDMI 2.1 and 3x DisplayPort 1.4a).

FAQ

Q: Is the RTX 5090 D V2 worth upgrading to from an RTX 3060 Ti?

A: Based on the 3DMark Steel Nomad DX12 score, the 5090 D V2 is 528.5% faster. That is a massive upgrade, but it also requires a 950 W PSU versus 550 W, and a larger case.

Q: What is the memory capacity difference?

A: The 5090 D V2 has 24 GB of GDDR7, while the 3060 Ti has 8 GB of GDDR6. The newer card also has a 384-bit bus versus 256-bit, giving it 1.34 TB/s bandwidth versus 448.0 GB/s.

Q: Which card has more ray tracing cores?

A: The 5090 D V2 has 170 RT cores, while the 3060 Ti has 38. This is a 4.5x difference, which directly impacts ray-traced performance.

Q: Are there any tests where the RTX 3060 Ti scores higher than the RTX 5090 D V2?

A: No. In the only head-to-head benchmark (3DMark Steel Nomad DX12), the 5090 D V2 wins. The 3060 Ti has no wins in the dataset.

Q: What is the process node difference?

A: The 5090 D V2 is built on 5 nm TSMC, while the 3060 Ti is on 8 nm Samsung. The newer process allows for 92,200 million transistors on a 750 mm² die, versus 17,400 million on 392 mm².

Q: What is the power consumption?

A: The 5090 D V2 has a TDP of 575 W and requires a 950 W PSU. The 3060 Ti has a TDP of 200 W and requires a 550 W PSU.

Specification Differences

The following specifications differ between the two cards:

  • Architecture: Blackwell 2.0 vs. Ampere
  • Process Node: 5 nm (TSMC) vs. 8 nm (Samsung)
  • Transistors: 92,200 million vs. 17,400 million
  • Die Size: 750 mm² vs. 392 mm²
  • Transistor Density: 122.9M / mm² vs. 44.4M / mm²
  • Base Clock: 2017 MHz vs. 1410 MHz
  • Boost Clock: 2407 MHz vs. 1665 MHz
  • Memory Speed: 28 Gbps effective vs. 14 Gbps effective
  • Memory Size: 24 GB vs. 8 GB
  • Memory Type: GDDR7 vs. GDDR6
  • Memory Bus: 384 bit vs. 256 bit
  • Memory Bandwidth: 1.34 TB/s vs. 448.0 GB/s
  • Shading Units: 21,760 vs. 4,864
  • TMUs: 680 vs. 152
  • ROPs: 176 vs. 80
  • RT Cores: 170 vs. 38
  • Tensor Cores: 680 vs. 152
  • Pixel Rate: 423.6 GPixel/s vs. 133.2 GPixel/s
  • Texture Rate: 1,636.8 GTexel/s vs. 253.1 GTexel/s
  • FP32 (and FP16) Compute: 104.8 TFLOPS vs. 16.20 TFLOPS
  • TDP: 575 W vs. 200 W
  • Power Connectors: 1x 16-pin vs. 1x 12-pin
  • Suggested PSU: 950 W vs. 550 W
  • Bus Interface: PCIe 5.0 x16 vs. PCIe 4.0 x16
  • Display Outputs: 1x HDMI 2.1b, 3x DisplayPort 2.1b vs. 1x HDMI 2.1, 3x DisplayPort 1.4a
  • Dimensions: 304 mm x 137 mm x 48 mm vs. 242 mm x 112 mm (width not specified)
  • Production Status: Active vs. End-of-life
  • Release Date: 2025-08-14 vs. 2020-11-30
  • Predecessor: GeForce 40 vs. GeForce 20
  • Successor: GeForce 60 vs. GeForce 40
  • Launch MSRP: 2,299 USD vs. 399 USD

The Verdict

The data-driven verdict is straightforward. For any user seeking maximum performance in modern DX12 workloads, the RTX 5090 D V2 is the only logical choice. Its 528.5% lead in 3DMark Steel Nomad is not an incremental improvement; it is a generational leap that redefines expectations. It offers 24 GB of GDDR7 memory, 170 RT cores, and 104.8 TFLOPS of compute power, making it suitable for 8K gaming, heavy ray tracing, and compute-intensive tasks.

The RTX 3060 Ti, however, is not without a role. Its 200 W TDP and 550 W PSU requirement make it a viable option for existing systems with limited power delivery. Its smaller 242 mm length fits in compact cases where the 304 mm 5090 D V2 will not. For users who primarily play older titles or do not require ray tracing, the 3060 Ti remains functional, but its 8 GB VRAM and 448.0 GB/s bandwidth are bottlenecks for modern high-resolution textures.

The choice depends entirely on budget and use case, but from a pure performance standpoint, the RTX 5090 D V2 wins every measurable category. The RTX 3060 Ti is an end-of-life product with a 399 USD launch MSRP, while the 5090 D V2 is an active product with a 2,299 USD launch MSRP. The price gap is significant, but so is the performance gap. For those who can afford it, the 5090 D V2 is the definitive choice. For those on a strict budget, the 3060 Ti remains a capable, low-power option, but it is not a competitor to the 5090 D V2 in any benchmark.

DETAILED SPECIFICATIONS

SPECIFICATION
RTX 3060 Ti
RTX 5090 D V2
Core Specs
Shading Units
4,864
21,760 +347.4%
Shaders
4,864
21,760 +347.4%
TMUs
152
680 +347.4%
ROPs
80
176 +120.0%
SM Count
38
170 +347.4%
Clocks
Base Clock
1410 MHz
2017 MHz
Boost Clock
1665 MHz
2407 MHz
Memory Clock
1750 MHz 14 Gbps effective
1750 MHz 28 Gbps effective
Memory
Memory Size
8 GB
24 GB
VRAM (MB)
8,192
24,576 +200.0%
Memory Type
GDDR6
GDDR7
Memory Bus
256 bit
384 bit
Bandwidth
448.0 GB/s
1.34 TB/s
Cache
L1 Cache
128 KB (per SM)
128 KB (per SM)
L2 Cache
4 MB
96 MB
Performance
Pixel Rate
133.2 GPixel/s
423.6 GPixel/s
Texture Rate
253.1 GTexel/s
1,636.8 GTexel/s
FP32 (TFLOPS)
16.20 TFLOPS
104.8 TFLOPS
FP64 (TFLOPS)
253.1 GFLOPS (1:64)
1.637 TFLOPS (1:64)
FP16 (TFLOPS)
16.20 TFLOPS (1:1)
104.8 TFLOPS (1:1)
AI/RT
RT Cores
38
170 +347.4%
Tensor Cores
152
680 +347.4%
Power
TDP
200 W
575 W
TDP (W)
200
575 +187.5%
Suggested PSU
550 W
950 W
Power Connectors
1x 12-pin
1x 16-pin
Architecture
Architecture
Ampere
Blackwell 2.0
GPU Name
GA104
GB202
Generation
GeForce 30
GeForce 50
Process Size
8 nm
5 nm
Transistors
17,400 million
92,200 million
Die Size
392 mm²
750 mm²
Foundry
Samsung
TSMC
Density
44.4M / mm²
122.9M / mm²
API Support
DirectX
12 Ultimate (12_2)
12 Ultimate (12_2)
OpenGL
4.6
4.6
Vulkan
1.4
1.4
OpenCL
3.0
3.0
CUDA
8.6
12.0
Shader Model
6.8
6.9
Physical
Slot Width
Dual-slot
Dual-slot
Length
242 mm 9.5 inches
304 mm 12 inches
Height
112 mm 4.4 inches
137 mm 5.4 inches
Outputs
1x HDMI 2.13x DisplayPort 1.4a
1x HDMI 2.1b3x DisplayPort 2.1b
Bus Interface
PCIe 4.0 x16
PCIe 5.0 x16
Other
Launch Price
399 USD
2,299 USD
Production
End-of-life
Active
Predecessor
GeForce 20
GeForce 40
Successor
GeForce 40
GeForce 60
View GeForce RTX 3060 Ti Details View GeForce RTX 5090 D V2 Details