AMD Radeon RX 7900 XT vs NVIDIA GeForce GTX 1660 SUPER Comparison

AMD
RADEON

AMD Radeon RX 7900 XT

CORE STATE Navi 31
VRAM 20 GB
CLOCK SPEED 2394 MHz
TDP 300 W
BUS WIDTH 320 bit
ARCHITECTURE RDNA 3.0
nm
PROCESS 5 nm
LAUNCH DATE 2022
VS
NVIDIA
GEFORCE

GeForce GTX 1660 SUPER

CORE STATE TU116
VRAM 6 GB
CLOCK SPEED 1785 MHz
TDP 125 W
BUS WIDTH 192 bit
ARCHITECTURE Turing
nm
PROCESS 12 nm
LAUNCH DATE 2019

PERFORMANCE BENCHMARKS

3dmark_3dmark_steel_nomad_dx12
5,623
1,278
geekbench_opencl
12,756
52,490
geekbench_vulkan
71,068
57,102
passmark_directx_10
158
65
passmark_directx_11
340
104
passmark_directx_12
110
50
passmark_directx_9
322
189
passmark_g2d
1,232
805
passmark_g3d
29,009
12,699
passmark_gpu_compute
16,828
5,076

Analysis: AMD Radeon RX 7900 XT vs NVIDIA GeForce GTX 1660 SUPER

The AMD Radeon RX 7900 XT and NVIDIA GeForce GTX 1660 SUPER represent two very different generations of GPU design, and the benchmark data quantifies the chasm between them. Across ten head-to-head tests, the RX 7900 XT claims nine victories, while the GTX 1660 SUPER manages a single, surprising win in one compute workload. The average benchmark scores tell a similar story: the RX 7900 XT averages 13,745 points, while the GTX 1660 SUPER averages 12,986 points, a modest overall gap that belies the extreme divergence in individual test categories. This analysis breaks down exactly where each card excels, what architectural choices drive those results, and which user should consider each product.

Head-to-Head Benchmarks

The most decisive margin in the entire comparison appears in the 3DMark Steel Nomad DX12 test. Here, the RX 7900 XT scores 5,623 points against the GTX 1660 SUPER's 1,278 points, a delta of 340%. This is not a marginal improvement; it is a generational obliteration. The RX 7900 XT delivers more than four times the raw DX12 performance, a result that aligns with its positioning as a modern high-end part versus a card from the GTX 16 era.

The pattern of dominance continues across the Passmark suite, though with varying intensity. In Passmark G3D, the RX 7900 XT scores 29,009 versus 12,699, a 128.4% advantage. The Passmark GPU Compute test shows an even larger gap, with the RX 7900 XT at 16,828 and the GTX 1660 SUPER at 5,076, representing a 231.5% lead. The DirectX 11 test shows a 226.9% delta (340 vs. 104), while DirectX 10 shows 143.1% (158 vs. 65). Even legacy DirectX 9 performance favors the newer card, with a 70.4% delta (322 vs. 189). The 2D graphics test is the closest of the RX 7900 XT's wins, with a 53% delta (1,232 vs. 805).

The Vulkan results are closer than the DX12 numbers but still favor the RX 7900 XT. In Geekbench Vulkan, the AMD card scores 71,068 against 57,102, a 24.5% lead. This shows that while the RX 7900 XT is faster, the GTX 1660 SUPER remains competitive in this particular API, likely due to the efficiency of NVIDIA's driver stack on older architectures.

The single victory for the GTX 1660 SUPER is in Geekbench OpenCL, where it scores 52,490 versus the RX 7900 XT's 12,756. This is a 75.7% deficit for the AMD card, a remarkable inversion of the other results. This outlier suggests that the OpenCL workload in this test is not optimized for the RDNA 3.0 architecture, or that the test's specific compute patterns favor the older Turing design's execution units. It is a stark reminder that synthetic benchmarks do not always correlate across different APIs.

Architecture Differences

The two GPUs are built on fundamentally different foundations. The RX 7900 XT uses the Navi 31 chip, fabricated on a 5 nm process at TSMC, putting 57,700 million transistors into a 529 mm² die. The GTX 1660 SUPER uses the TU116 chip on a 12 nm process, also from TSMC, with 6,600 million transistors on a 284 mm² die. The transistor density difference is stark: 109.1M / mm² for the RX 7900 XT versus 23.2M / mm² for the GTX 1660 SUPER. This density advantage is a direct result of the process node difference, enabling the RX 7900 XT to pack nearly nine times more transistors.

The compute resources scale accordingly. The RX 7900 XT features 5,376 shading units, 336 texture mapping units, and 192 ROPs. The GTX 1660 SUPER has 1,408 shading units, 88 TMUs, and 48 ROPs. The RX 7900 XT also brings 84 ray tracing cores, a feature entirely absent from the GTX 1660 SUPER, which has no RT cores listed. This hardware RT support is a defining differentiator for modern workloads.

Memory configurations diverge sharply. The RX 7900 XT ships with 20 GB of GDDR6 on a 320-bit bus, providing 800.0 GB/s of bandwidth. The GTX 1660 SUPER has 6 GB of GDDR6 on a 192-bit bus, yielding 336.0 GB/s. The clock speeds also differ: the RX 7900 XT has a base of 1,387 MHz and a boost of 2,394 MHz, while the GTX 1660 SUPER has a higher base of 1,530 MHz but a lower boost of 1,785 MHz. The memory clocks show 20 Gbps effective for the AMD card versus 14 Gbps effective for the NVIDIA card.

The architectural families are also distinct. The RX 7900 XT uses RDNA 3.0 with the codename Plum Bonito, part of the Navi III generation. The GTX 1660 SUPER uses Turing, part of the GeForce 16 generation. The API support reflects the newer design: the RX 7900 XT supports DirectX 12 Ultimate (12_2), while the GTX 1660 SUPER supports DirectX 12 (12_1). Both support OpenGL 4.6 and Vulkan 1.4, but the feature level within DirectX 12 is a clear separation point.

Power and interface requirements scale with performance. The RX 7900 XT has a TDP of 300 W, requires a 700 W suggested PSU, and uses 2x 8-pin power connectors. The GTX 1660 SUPER has a TDP of 125 W, a 300 W suggested PSU, and a single 8-pin connector. The RX 7900 XT uses PCIe 4.0 x16, while the GTX 1660 SUPER uses PCIe 3.0 x16. Physical dimensions also differ, with the RX 7900 XT being longer at 276 mm versus 229 mm, and thicker at 51 mm versus 35 mm.

Where Each One Wins

The RX 7900 XT is the clear winner for any workload that leverages modern graphics APIs or heavy compute. Its 340% lead in 3DMark Steel Nomad DX12 makes it the obvious choice for current-generation gaming at high resolutions, where the 20 GB frame buffer and 800.0 GB/s bandwidth provide ample headroom. The 128.4% lead in Passmark G3D and the 231.5% lead in GPU compute reinforce this, indicating strong performance in both rendering and general-purpose compute tasks. The 24.5% lead in Vulkan shows it also handles cross-platform APIs well.

The GTX 1660 SUPER's only win is in Geekbench OpenCL, where it beats the RX 7900 XT by 75.7%. This makes it a niche pick for specific OpenCL compute workloads, such as certain physics simulations or legacy applications that are not optimized for RDNA 3.0. However, this single win is offset by its losses in every other category, including a 120% deficit in DirectX 12 and a 53% deficit in 2D graphics.

For users prioritizing low power draw, the GTX 1660 SUPER has a clear advantage, with a TDP of 125 W versus the RX 7900 XT's 300 W. It also requires a smaller PSU (300 W vs. 700 W) and a single 8-pin connector instead of two. The GTX 1660 SUPER is also physically smaller, making it easier to fit in compact cases. However, these are efficiency and form-factor wins, not performance wins.

The Verdict

The data is unambiguous: the AMD Radeon RX 7900 XT is the superior performer in nine of ten benchmarks, with an average score 5.8% higher than the GTX 1660 SUPER. The average benchmark scores place the RX 7900 XT at 13,745, while the GTX 1660 SUPER sits at 12,986. This places the RX 7900 XT at the 55th percentile of all GPUs, while the GTX 1660 SUPER sits at the 53rd percentile. The RX 7900 XT's nearest rivals include the AMD Radeon 660M (avg score 13,812, delta -0.5%) and the NVIDIA RTX A2000 Mobile (avg score 13,821, delta -0.5%), while the GTX 1660 SUPER competes with the NVIDIA GeForce RTX 3050 Ti Mobile (avg score 12,940, delta 0.4%) and the AMD Radeon RX 580 (avg score 12,928, delta 0.4%).

For a user building a high-end gaming rig or a workstation for modern 3D rendering, the RX 7900 XT is the only rational choice. Its 340% advantage in DX12 and 128.4% advantage in G3D make it a generation ahead. The GTX 1660 SUPER, with its launch MSRP of 229 USD versus the RX 7900 XT's launch MSRP of 899 USD, is a legacy part that should only be considered for OpenCL-specific compute tasks or for systems where power and size constraints are paramount. The benchmark data shows no scenario where the GTX 1660 SUPER outperforms the RX 7900 XT in gaming or general compute, aside from that single OpenCL test. The verdict is clear: the RX 7900 XT wins on performance, while the GTX 1660 SUPER wins only on efficiency and the specific OpenCL workload.

FAQ

Q: Which GPU is faster in 3DMark Steel Nomad DX12?

A: The AMD Radeon RX 7900 XT scores 5,623 points versus the GTX 1660 SUPER's 1,278 points, a 340% lead for the AMD card.

Q: Does the GTX 1660 SUPER beat the RX 7900 XT in any benchmark?

A: Yes, in Geekbench OpenCL, the GTX 1660 SUPER scores 52,490 against the RX 7900 XT's 12,756, a 75.7% advantage for the NVIDIA card.

Q: What are the memory configurations of the two cards?

A: The RX 7900 XT has 20 GB of GDDR6 on a 320-bit bus with 800.0 GB/s bandwidth. The GTX 1660 SUPER has 6 GB of GDDR6 on a 192-bit bus with 336.0 GB/s bandwidth.

Q: How do their power requirements differ?

A: The RX 7900 XT has a TDP of 300 W and a suggested PSU of 700 W, while the GTX 1660 SUPER has a TDP of 125 W and a suggested PSU of 300 W.

Q: Which card supports ray tracing hardware?

A: Only the RX 7900 XT, which has 84 ray tracing cores. The GTX 1660 SUPER has no RT cores listed.

Q: What is the average benchmark score difference?

A: The RX 7900 XT averages 13,745 points, while the GTX 1660 SUPER averages 12,986 points, a difference of 759 points in favor of the AMD card.

DETAILED SPECIFICATIONS

SPECIFICATION
RX 7900 XT
GTX 1660 SUPER
Core Specs
Shading Units
5,376
1,408 -73.8%
Shaders
5,376
1,408 -73.8%
TMUs
336
88 -73.8%
ROPs
192
48 -75.0%
Compute Units
84
—
SM Count
—
22
Clocks
Base Clock
1387 MHz
1530 MHz
Boost Clock
2394 MHz
1785 MHz
Game Clock
2025 MHz
—
Shader Clock
2025 MHz
—
Memory Clock
2500 MHz 20 Gbps effective
1750 MHz 14 Gbps effective
Memory
Memory Size
20 GB
6 GB
VRAM (MB)
20,480
6,144 -70.0%
Memory Type
GDDR6
GDDR6
Memory Bus
320 bit
192 bit
Bandwidth
800.0 GB/s
336.0 GB/s
Cache
L1 Cache
256 KB per Array
64 KB (per SM)
L2 Cache
6 MB
1536 KB
L3 Cache
80 MB
—
L0 Cache
64 KB per WGP
—
Performance
Pixel Rate
459.6 GPixel/s
85.68 GPixel/s
Texture Rate
804.4 GTexel/s
157.1 GTexel/s
FP32 (TFLOPS)
51.48 TFLOPS
5.027 TFLOPS
FP64 (TFLOPS)
1.609 TFLOPS (1:32)
157.1 GFLOPS (1:32)
FP16 (TFLOPS)
103.0 TFLOPS (2:1)
10.05 TFLOPS (2:1)
AI/RT
RT Cores
84
—
Matrix Cores
168
—
Power
TDP
300 W
125 W
TDP (W)
300
125 -58.3%
Suggested PSU
700 W
300 W
Power Connectors
2x 8-pin
1x 8-pin
Architecture
Architecture
RDNA 3.0
Turing
GPU Name
Navi 31
TU116
Codename
Plum Bonito
—
Generation
Navi III (RX 7000)
GeForce 16
Process Size
5 nm
12 nm
Transistors
57,700 million
6,600 million
Die Size
529 mm²
284 mm²
Foundry
TSMC
TSMC
Density
109.1M / mm²
23.2M / mm²
AMD MCM
GCD Transistors
45,400 million
—
GCD Die Size
304.35 mm²
—
MCD Transistors
2,050 million x6
—
API Support
DirectX
12 Ultimate (12_2)
12 (12_1)
OpenGL
4.6
4.6
Vulkan
1.4
1.4
OpenCL
2.2
3.0
CUDA
—
7.5
Shader Model
6.8
6.8
Physical
Slot Width
Dual-slot
Dual-slot
Length
276 mm 10.9 inches
229 mm 9 inches
Height
110 mm 4.3 inches
111 mm 4.4 inches
Outputs
1x HDMI 2.1a2x DisplayPort 2.11x USB Type-C
1x DVI1x HDMI 2.01x DisplayPort 1.4a
Bus Interface
PCIe 4.0 x16
PCIe 3.0 x16
Other
Launch Price
899 USD
229 USD
Production
End-of-life
End-of-life
Predecessor
Navi II
GeForce 10
Successor
Navi IV
GeForce 20
View Radeon RX 7900 XT Details View GeForce GTX 1660 SUPER Details