AMD Radeon 660M vs AMD Radeon RX 7900 XT Comparison

AMD
RADEON

AMD Radeon 660M

CORE STATE Rembrandt
VRAM System Shared
CLOCK SPEED 1900 MHz
TDP 40 W
BUS WIDTH System Shared
ARCHITECTURE RDNA 2.0
nm
PROCESS 6 nm
LAUNCH DATE 2022
VS
AMD
RADEON

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

PERFORMANCE BENCHMARKS

geekbench_opencl
12,876
12,756
geekbench_vulkan
14,748
71,068
3dmark_3dmark_steel_nomad_dx12
N/A
5,623
passmark_directx_10
N/A
158
passmark_directx_11
N/A
340
passmark_directx_12
N/A
110
passmark_directx_9
N/A
322
passmark_g2d
N/A
1,232
passmark_g3d
N/A
29,009
passmark_gpu_compute
N/A
16,828

Analysis: AMD Radeon 660M vs AMD Radeon RX 7900 XT

The benchmark data presents a striking paradox: the integrated AMD Radeon 660M and the flagship discrete AMD Radeon RX 7900 XT are separated by an enormous gulf in hardware resources, yet their average benchmark scores are nearly identical. The 660M averages 13,812 points, while the RX 7900 XT averages 13,745 points, a difference of just 0.5%. This places both cards at the 55th percentile of all GPUs, and they are each other's closest rivals. However, a closer look at the individual tests reveals that this parity is an artifact of averaging, as the two cards win in completely different and decisive ways.

Head-to-Head Benchmarks

The most significant victory for the AMD Radeon RX 7900 XT is in the Geekbench Vulkan test, where it delivers a crushing performance. The RX 7900 XT scores 71,068 points, which is 79.2% higher than the 660M's score of 14,748. This is not a marginal win; it is a total rout that demonstrates the immense compute advantage of the discrete card when leveraging the Vulkan API. The 660M's score in this test is closer to its OpenCL result, suggesting it is hitting a performance ceiling that the RX 7900 XT simply ignores.

In contrast, the AMD Radeon 660M wins the Geekbench OpenCL test, albeit by a much narrower margin. The 660M scores 12,876 points, which is 0.9% ahead of the RX 7900 XT's score of 12,756. While this is a win for the integrated part, the delta is small enough to be considered a statistical tie. It is notable that the RX 7900 XT scores lower in OpenCL than in Vulkan, a divergence that likely reflects driver optimization or workload scheduling differences between the two APIs.

The broader benchmark suite for the RX 7900 XT shows its true range. In the Passmark G3D test, it scores 29,009 points, a figure that dwarfs the 660M's entire benchmark portfolio. The RX 7900 XT also achieves 16,828 points in Passmark GPU Compute and 1,232 points in Passmark G2D. These results contrast sharply with the 660M, which has no equivalent scores in the data. The only other direct comparison available is the average benchmark score, where the 660M's 13,812 points edges out the RX 7900 XT's 13,745 points, giving the integrated GPU a 0.5% lead in that specific metric.

Where Each One Wins

The data indicates a clear split based on workload type and API. The AMD Radeon RX 7900 XT is the definitive winner in any scenario that utilizes the Vulkan API. Its Geekbench Vulkan score of 71,068 is more than 4.8 times higher than the 660M's score of 14,748. This suggests that for modern games and applications that rely heavily on Vulkan for low-level hardware access, the RX 7900 XT is in a completely different performance class. The RX 7900 XT also excels in raw compute and rendering throughput, as evidenced by its Passmark G3D score of 29,009 and GPU Compute score of 16,828, which are far beyond what the 660M can achieve in the tests it participated in.

The AMD Radeon 660M wins in the Geekbench OpenCL test, scoring 12,876 versus the RX 7900 XT's 12,756. This is a narrow 0.9% margin, but it does indicate that the 660M is not universally slower. The 660M's performance in this test is commensurate with its average score, suggesting it provides consistent, if modest, compute performance. For workloads that are specifically optimized for OpenCL, the 660M holds its own against the much larger RX 7900 XT. However, this is a niche victory; the RX 7900 XT's sheer number of shading units (5,376 vs 384) and texture mapping units (336 vs 24) implies that its raw compute ceiling is far higher, even if the OpenCL test does not reflect it.

The specification sheets reinforce this dichotomy. The RX 7900 XT has a dedicated 20 GB of GDDR6 memory on a 320-bit bus with 800.0 GB/s of bandwidth. The 660M uses system shared memory, with bandwidth described as system dependent. This architectural advantage allows the RX 7900 XT to feed its 5,376 shading units at a rate that the 660M cannot match, explaining the massive Vulkan score. The 660M's win in OpenCL is likely a result of test-specific conditions, but it is the only head-to-head benchmark where the integrated part comes out on top.

The Verdict

The performance data presents a clear verdict: the AMD Radeon RX 7900 XT is the superior graphics processor for demanding tasks, while the AMD Radeon 660M is a capable integrated solution that happens to win a single benchmark. The RX 7900 XT's Geekbench Vulkan score of 71,068 is the single most dominant metric in the comparison, showing a 79.2% advantage over the 660M. This is the metric that matters for modern gaming and compute, and it is where the RX 7900 XT's 84 ray tracing cores and 51.48 TFLOPS of FP32 performance are put to use. Any user seeking high-end graphics performance should choose the RX 7900 XT without hesitation.

The AMD Radeon 660M is not without merit. Its Geekbench OpenCL score of 12,876 is higher than the RX 7900 XT's 12,756, and its average benchmark score of 13,812 is 0.5% higher than the RX 7900 XT's average. For users whose primary workload is OpenCL-based and who are confined to an integrated GPU, the 660M is a solid choice. However, these wins are narrow and do not offset the massive disadvantage in Vulkan performance. The data shows that the 660M is best suited for basic computing and light graphics tasks, while the RX 7900 XT is for those who need maximum performance. The choice is not about value; it is about the required performance level.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The AMD Radeon 660M has a higher average benchmark score of 13,812, compared to the AMD Radeon RX 7900 XT's score of 13,745. This gives the 660M a 0.5% advantage in this specific metric.

Q: What is the largest performance difference in the head-to-head tests?

A: The largest difference is in the Geekbench Vulkan test, where the AMD Radeon RX 7900 XT scores 71,068, which is 79.2% higher than the AMD Radeon 660M's score of 14,748.

Q: Does the AMD Radeon 660M win any benchmark against the RX 7900 XT?

A: Yes, the AMD Radeon 660M wins the Geekbench OpenCL test with a score of 12,876, which is 0.9% higher than the RX 7900 XT's score of 12,756.

Q: How do the two GPUs compare in terms of memory bandwidth?

A: The AMD Radeon RX 7900 XT has a dedicated 20 GB of GDDR6 memory with a 320-bit bus and 800.0 GB/s bandwidth. The AMD Radeon 660M uses system shared memory, and its bandwidth is system dependent.

Q: What is the percentile ranking for both GPUs?

A: Both the AMD Radeon 660M and the AMD Radeon RX 7900 XT are ranked at the 55th percentile of all GPUs.

Q: Which GPU has more shading units?

A: The AMD Radeon RX 7900 XT has 5,376 shading units, while the AMD Radeon 660M has 384 shading units.

Architecture Differences

The two GPUs represent distinct generations of AMD graphics architecture. The AMD Radeon 660M is based on the RDNA 2.0 architecture, using the Rembrandt chip, and is classified as a Navi II IGP for mobile platforms. The AMD Radeon RX 7900 XT is based on the newer RDNA 3.0 architecture, using the Navi 31 chip with the codename "Plum Bonito," and belongs to the Navi III generation. The process nodes also differ significantly: the 660M is built on a 6 nm process, while the RX 7900 XT is built on a 5 nm process. This translates to a substantial difference in transistor count, with the RX 7900 XT packing 57,700 million transistors on a 529 mm² die, compared to the 660M's 13,100 million transistors on a 208 mm² die. The RX 7900 XT also has a higher transistor density of 109.1M per mm², versus 63.0M per mm² for the 660M.

The core configurations are vastly different. The RX 7900 XT features 5,376 shading units, 336 texture mapping units, and 192 render output units, along with 84 ray tracing cores. The 660M has 384 shading units, 24 texture mapping units, 16 render output units, and 6 ray tracing cores. The RX 7900 XT's clock speeds are also higher, with a boost clock of 2394 MHz and a game clock of 2025 MHz, compared to the 660M's boost clock of 1900 MHz. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, but the RX 7900 XT's additional hardware resources give it a clear advantage in compute-intensive tasks.

Specification Differences

The most obvious specification difference is power consumption. The AMD Radeon RX 7900 XT has a TDP of 300 W and requires two 8-pin power connectors along with a suggested power supply of 700 W. The AMD Radeon 660M has a TDP of 40 W and uses no power connectors, as it is an integrated graphics processor. The physical form factors differ as well: the RX 7900 XT is a dual-slot card measuring 276 mm in length, 110 mm in height, and 51 mm in width, while the 660M is an IGP with no dedicated slot width.

Memory specifications are also starkly different. The RX 7900 XT has 20 GB of GDDR6 memory on a 320-bit bus with 800.0 GB/s of bandwidth. The 660M uses system shared memory with a system dependent bandwidth and bus width. The RX 7900 XT's memory clock is listed as 2500 MHz, with 20 Gbps effective speed. The bus interface also differs, with the RX 7900 XT using PCIe 4.0 x16 and the 660M using PCIe 4.0 x8. The RX 7900 XT has a launch MSRP of 899 USD. Display outputs also vary, with the RX 7900 XT featuring 1x HDMI 2.1a, 2x DisplayPort 2.1, and 1x USB Type-C, while the 660M's outputs are listed as portable device dependent.

DETAILED SPECIFICATIONS

SPECIFICATION
660M
RX 7900 XT
Core Specs
Shading Units
384
5,376 +1300.0%
Shaders
384
5,376 +1300.0%
TMUs
24
336 +1300.0%
ROPs
16
192 +1100.0%
Compute Units
6
84 +1300.0%
Clocks
Base Clock
1500 MHz
1387 MHz
Boost Clock
1900 MHz
2394 MHz
Game Clock
2025 MHz
Shader Clock
2025 MHz
Memory Clock
System Shared
2500 MHz 20 Gbps effective
Memory
Memory Size
System Shared
20 GB
VRAM (MB)
20,480
Memory Type
System Shared
GDDR6
Memory Bus
System Shared
320 bit
Bandwidth
System Dependent
800.0 GB/s
Cache
L1 Cache
128 KB per Array
256 KB per Array
L2 Cache
2 MB
6 MB
L3 Cache
80 MB
L0 Cache
32 KB per WGP
64 KB per WGP
Performance
Pixel Rate
30.40 GPixel/s
459.6 GPixel/s
Texture Rate
45.60 GTexel/s
804.4 GTexel/s
FP32 (TFLOPS)
1,459.2 GFLOPS
51.48 TFLOPS
FP64 (TFLOPS)
91.20 GFLOPS (1:16)
1.609 TFLOPS (1:32)
FP16 (TFLOPS)
2.918 TFLOPS (2:1)
103.0 TFLOPS (2:1)
AI/RT
RT Cores
6
84 +1300.0%
Matrix Cores
168
Power
TDP
40 W
300 W
TDP (W)
40
300 +650.0%
Suggested PSU
700 W
Power Connectors
None
2x 8-pin
Architecture
Architecture
RDNA 2.0
RDNA 3.0
GPU Name
Rembrandt
Navi 31
Codename
Plum Bonito
Generation
Navi II IGP (Rembrandt Mobile)
Navi III (RX 7000)
Process Size
6 nm
5 nm
Transistors
13,100 million
57,700 million
Die Size
208 mm²
529 mm²
Foundry
TSMC
TSMC
Density
63.0M / mm²
109.1M / 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 Ultimate (12_2)
OpenGL
4.6
4.6
Vulkan
1.4
1.4
OpenCL
2.0
2.2
Shader Model
6.8
6.8
Physical
Slot Width
IGP
Dual-slot
Length
276 mm 10.9 inches
Height
110 mm 4.3 inches
Outputs
Portable Device Dependent
1x HDMI 2.1a2x DisplayPort 2.11x USB Type-C
Bus Interface
PCIe 4.0 x8
PCIe 4.0 x16
Other
Launch Price
899 USD
Production
End-of-life
End-of-life
Predecessor
Vega II IGP
Navi II
Successor
Navi III IGP
Navi IV
View Radeon 660M Details View Radeon RX 7900 XT Details