The first unofficial benchmarks of the AMD Radeon RX 6600M RDNA 2 GPU based on the Navi 23 SKU have landed. The benchmarks were posted over at Zhuanlan (via HXL) and show how well the latest RDNA 2 GPU competes against mainstream laptop offerings from NVIDIA.
AMD Radeon RX 6600M 'Navi 23' Laptop GPU With RDNA 2 Architecture Benchmarked, Slower Than The NVIDIA RTX 3060
The AMD Radeon RX 6600M is the first graphics chip to make use of the Navi 23 GPU SKU. The chip has yet to make an entrance on the desktop front so this is the first time we are actually looking at its real benchmarks. The chip was tested on an AMD Ryzen 7 5800H laptop which featured 16 GB of DDR4-3200 (Single Rank x8) memory. The same test platform was also used to test the competition which included the NVIDIA RTX 3060 and the RTX 3050 Ti. Before we get into the benchmarks, let's quickly recap the specifications for the Radeon RX 6600M.
AMD Radeon RX 6600M 'RDNA 2' Mobility Graphics Chip Specifications
When it comes to specifications, the AMD Navi 23 RDNA 2 GPU for the Radeon RX 6600M & Radeon Pro W6600M is expected to feature up to 28 CUs or 1792 stream processors. It will also come with up to 8 GB GDDR6 memory running across a 128-bit bus interface and with TGPs ranging up to 100 Watts. The AMD Navi 23 GPUs are additionally going to feature 32 MB of Infinity Cache and will be the smallest Navi chip to house the brand new cache technology. Discrete GPUs below the Navi 23 such as Navi 24 and the integrated RDNA 2 chips on Van Gogh APUs are not going to feature the tech.

AMD Radeon RX 7000/6000 (RDNA 3/ RDNA 2) Mobility GPUs
| GPU Name | GPU Architecture | Process Node | Die Size | GPU Cores | GPU Clock (Max) | Memory Capacity | Memory Bus | TGP |
|---|---|---|---|---|---|---|---|---|
| Radeon RX 7600M XT | RDNA 3 Navi 33 | 6nm | 204mm2 | 2048 | 2300 MHz | 8 GB GDDR6 (32 MB Infinity Cache) | 128-bit / 288 GB/s | 75-120W |
| Radeon RX 7600M | RDNA 3 Navi 33 | 6nm | 204mm2 | 1792 | 2070 MHz | 8 GB GDDR6 (32 MB Infinity Cache) | 128-bit / 256 GB/s | 50-90W |
| Radeon RX 7700S | RDNA 3 Navi 33 | 6nm | 204mm2 | 2048 | 2200 MHz | 8 GB GDDR6 (32 MB Infinity Cache) | 128-bit / 288 GB/s | 75-100W |
| Radeon RX 7600S | RDNA 3 Navi 33 | 6nm | 204mm2 | 1792 | 1865 MHz | 8 GB GDDR6 (32 MB Infinity Cache) | 128-bit / 256 GB/s | 50-75W |
| Radeon RX 6850M XT | Navi 22 RDNA 2 | 7nm | 334.5mm2 | 2560 | 2463 MHz | 12 GB GDDR6 (96 MB Infinity Cache) | 192-bit / 432 GB/s | 165W+ |
| Radeon RX 6800M | Navi 22 RDNA 2 | 7nm | 334.5mm2 | 2560 | 2300 MHz | 12 GB GDDR6 (96 MB Infinity Cache) | 192-bit / 384 GB/s | 145W+ |
| Radeon RX 6700M | Navi 22 RDNA 2 | 7nm | 334.5mm2 | 2304 | 2300 MHz | 10 GB GDDR6 (80 MB Infinity Cache) | 160-bit / 320 GB/s | 135W |
| Radeon RX 6650M XT | Navi 23 RDNA 2 | 7nm | 237mm2 | 2048 | 2162 MHz | 8 GB GDDR6 (32 MB Infinity Cache) | 128-bit / 256 GB/s | 120W |
| Radeon RX 6650M | Navi 23 RDNA 2 | 7nm | 237mm2 | 1792 | 2222 MHz | 8 GB GDDR6 (32 MB Infinity Cache) | 128-bit / 256 GB/s | 120W |
| Radeon RX 6600M | Navi 23 RDNA 2 | 7nm | 237mm2 | 1792 | 2177 MHz | 8 GB GDDR6 (32 MB Infinity Cache) | 128-bit / 224 GB/s | 100W |
| Radeon RX 6550M | Navi 23 RDNA 2 | 7nm | 237mm2 | 2048 | 2560 MHz | 4 GB GDDR6 (32 MB Infinity Cache) | 64-bit / 144 GB/s | 80W |
| Radeon RX 6500M | Navi 24 RDNA 2 | 6nm | 141mm2 | 1024 | 2191 MHz | 4 GB GDDR6 (16 MB Infinity Cache) | 64-bit / 128 GB/s | 50W |
| Radeon RX 6300M | Navi 24 RDNA 2 | 6nm | 141mm2 | 768 | 1512 MHz | 2 GB GDDR6 (8 MB Infinity Cache) | 64-bit / 107.8 GB/s | 25W |
| Radeon RX 6800S | Navi 22 RDNA 2 | 7nm | 334.5mm2 | 2048 | 1975 MHz | 12 GB GDDR6 (96 MB Infinity Cache) | 128-bit / 256 GB/s | 100W |
| Radeon RX 6700S | Navi 22 RDNA 2 | 7nm | 334.5mm2 | 1792 | 1890 MHz | 8 GB GDDR6 (32 MB Infinity Cache) | 128-bit / 224 GB/s | 80W |
| Radeon RX 6600S | Navi 23 RDNA 2 | 7nm | 237mm2 | 1792 | 1881 MHz | 4 GB GDDR6 (32 MB Infinity Cache) | 128-bit / 224 GB/s | 80W |
AMD's official Performance data shows the Radeon RX 6600M outperforming the NVIDIA RTX 3060 6 GB mobility chip in several current-generation titles. When it comes to actual benchmarks, AMD's Radeon RX 6600M (100W with 120W VR losses) seems to offer slower performance than the NVIDIA RTX 3060 (115W+15W). While the RTX 3060 does feature higher power consumption, at the same power limit of 100W, the RTX 3060 delivers better performance.
Radeon RX 6600M 'Navi 23' GPU Synthetic Benchmarks:
Radeon RX 6600M 'Navi 23' GPU Gaming Benchmarks:
Radeon RX 6600M 'Navi 23' GPU Power & Content Creation Benchmarks:
There's no difference in performance between the two laptop GPUs in synthetic benchmarks but as soon as we shift to gaming at 1080p, the RTX 3060 takes an 8% average performance lead. The ray-tracing performance is even higher with the RTX 3060 being up to 36% faster than the Radeon RX 6600M. AMD's Advantage laptop also comes with SmartShift technology which can deliver a 3-4% performance boost. As for content creation performance, NVIDIA's RTX GPUs take a huge lead due to impressive driver support and optimizations for content creator applications.
Now not all is bad, the Radeon RX 6600M delivers a 20-30% performance boost over the GeForce RTX 3050 Ti so if AMD can manage to price the mainstream Navi 23 chip accordingly, then it could serve in a really well position between the RTX 3050 Ti and the RTX 3060. The main issue for the RX 6600M is reported to be its 128-bit bus interface. The tester states that it would have been a better choice if AMD had gone with a 4 GB GDDR6 memory capacity which would have further toned the price and power down. The HP Omen 16 is one of the many premium options featuring the chip that will be available soon and is currently listed for a price of around $1500 US.
Follow Wccftech on Google to get more of our news coverage in your feeds.