NVIDIA's recently released GeForce MX550 discrete GPU has been tested and it barely outperforms an AMD RDNA 2 'Radeon 680M' integrated GPU.
AMD's Integrated RDNA 2 GPU Is On Par With NVIDIA's Discrete GeForce MX550 35W GPU While Offering Superior Efficiency
The NVIDIA GeForce MX550 uses the Turing TU117 GPU. Based on the information we had earlier, the chip is a cut-down configuration featuring 1024 CUDA cores, a clock rate of up to 1320 MHz, a 2 GB GDDR6 memory interface running across a 64-bit bus at 12 Gbps clocks, and a TDP that's rated at 25W. While the GPU is based on the Turing chip, it lacks RT and DLSS capabilities since its using a GPU that's a step below the RTX offerings.
NVIDIA GeForce MX Series GPU Family Specifications:
| GPU Name | GPU Architecture | CUDA Cores | GPU Clock | Memory Speed | Memory Bus | TDP |
|---|---|---|---|---|---|---|
| GeForce MX570 | Ampere GA107 | 2048 | TBA - 1155 MHz | 12 Gbps GDDR6 | 64-bit | 25W |
| GeForce MX550 | Turing TU117 | 1024 | TBA - 1320 MHz | 12 Gbps GDDR6 | 64-bit | 25W |
| GeForce MX 450 | Turing TU117 | 896 | 540 - 1575 MHz | 10 Gbps GDDR6 | 64-bit | 25W |
| GeForce MX 430 | Turing TU117 | 896 | TBA - TBA MHz | 10 Gbps GDDR6 7 Gbps GDDR5 | 64-bit | 25W |
| GeForce MX 350 | Pascal GP107 | 640 | 1354-1468 MHz | 7 Gbps GDDR5 | 64-bit | 25W |
| GeForce MX 350 | Pascal GP107 | 640 | 746-937 MHz | 7 Gbps GDDR5 | 64-bit | 15W |
| GeForce MX 330 | Pascal GP108 | 384 | 1531-1594 MHz | 6/7 Gbps GDDR5 | 64-bit | 25W |
| GeForce MX 310 | Pascal GP108 | 256 | 1341-1379 MHz | 6/7 Gbps GDDR5 | 64-bit | 25W |
| GeForce MX 250 | Pascal GP108 | 384 | 1518-1582 MHz | 6/7 Gbps GDDR5 | 64-bit | 25W |
| GeForce MX 250 | Pascal GP108 | 384 | 937-1038 MHz | 6 Gbps GDDR5 | 64-bit | 10.5W |
| GeForce MX 230 | Pascal GP108 | 256 | 1518-1531 MHz | 6/7 Gbps GDDR5 | 64-bit | 25W |
| GeForce MX 150 | Pascal GP108 | 384 | 1468-1531 MHz | 6 Gbps GDDR5 | 64-bit | 25W |
| GeForce MX 150 | Pascal GP108 | 384 | 937-1038 MHz | 5 Gbps GDDR5 | 64-bit | 10W |
| GeForce MX 130 | Maxwell GM108 | 384 | 1122-1242 MHz | 6 Gbps GDDR5 | 64-bit | 25W |
| GeForce MX 110 | Maxwell GM108 | 256 | 963-993 MHz | 1.8 Gbps DDR3 | 64-bit | 10W |
The AMD Radeon 600M series integrated GPUs, based on the RDNA 2 graphics architecture, are what's adopted by the Ryzen 6000 APUs, offering up to 12 Compute Units for 768 cores and up to 2.4 GHz GPU frequencies. The Radeon 600M GPUs pack a 50% large compute engine than the early Vega iGPUs, 50% higher bandwidth, twice the L2 cache, and twice the render backends (RB+).
The Radeon 600M is split into two SKUs, the Radeon 680M which is featured on Ryzen 9 & Ryzen 7 chips, packing the full 12 CU & 2.4 GHz configuration (4 RB+). This particular review included AMD's Ryzen 7 6800H which has the Radeon 680M running at 2.2 GHz instead of the full 2.4 GHz on high-end HX parts. The Radeon 660M powers the Ryzen 5 APUs with up to 6 CUs, 1.9 GHz clocks and 2 Render backends. Both chips were tested within these benchmarks.
AMD Ryzen 6000H 'Rembrandt' APU Lineup For Notebooks:
| APU Name | APU Family | Architecture | Process | Cores / Threads | Base Clock | Boost Clock | L3 Cache | Graphics | TDP |
|---|---|---|---|---|---|---|---|---|---|
| Ryzen 9 6980HX | Rembrandt H | Zen 3+ | 6nm | 8 / 16 | 3.3 GHz | 5.00 GHz | 16 MB | 12 CU RDNA 2 (2400 MHz) | 45W+ |
| Ryzen 9 6980HS | Rembrandt H | Zen 3+ | 6nm | 8 / 16 | 3.3 GHz | 5.00 GHz | 16 MB | 12 CU RDNA 2 (2400 MHz) | 35W |
| Ryzen 9 6900HX | Rembrandt H | Zen 3+ | 6nm | 8 / 16 | 3.3 GHz | 4.90 GHz | 16 MB | 12 CU RDNA 2 (2400 MHz) | 45W+ |
| Ryzen 9 6900HS | Rembrandt H | Zen 3+ | 6nm | 8 / 16 | 3.3 GHz | 4.90 GHz | 16 MB | 12 CU RDNA 2 (2400 MHz) | 35W |
| Ryzen 7 6800H | Rembrandt H | Zen 3+ | 6nm | 8 / 16 | 3.2 GHz | 4.70 GHz | 16 MB | 12 CU RDNA 2 (2200 MHz) | 45W |
| Ryzen 7 6800HS | Rembrandt H | Zen 3+ | 6nm | 8 / 16 | 3.2 GHz | 4.70 GHz | 16 MB | 12 CU RDNA 2 (2200 MHz) | 35W |
| Ryzen 5 6600H | Rembrandt H | Zen 3+ | 6nm | 6 / 12 | 3.3 GHz | 4.50 GHz | 16 MB | 6 CU RDNA 2 (1900 MHz) | 45W |
| Ryzen 5 6600HS | Rembrandt H | Zen 3+ | 6nm | 6 / 12 | 3.3 GHz | 4.50 GHz | 16 MB | 6 CU RDNA 2 (1900 MHz) | 35W |
AMD RDNA 2 Radeon 600M iGPU vs NVIDIA GeForce MX550 dGPU (Synthetic Benchmarks):
So coming to the performance metrics, the AMD Radeon 680M ended up 25% faster in 3DMark Fire Strike, 5% faster in Time Spy, and since it features ray tracing capabilities unlike the NVIDIA discrete GPU, it won the Port Royal benchmark uncontested. Meanwhile, the Radeon 660M which has half the compute units was only able to match in performance to an Intel Alder Lake Core i7-1280P CPU with Iris Xe graphics.
AMD RDNA 2 Radeon 600M iGPU vs NVIDIA GeForce MX550 dGPU (Gaming Benchmarks):
Moving from synthetic to gaming benchmarks, a range of titles were tested at 1080p which include both AAA and eSports games. In 9 of the 14 games, the AMD Radeon 680M iGPU was able to out-perform the NVIDIA GeForce MX550. The Radeon 660M was able to beat the Intel Iris Xe iGPU in a total of 13 out of 14 titles.
Overall, the NVIDIA GeForce MX550, despite its discrete and higher TDP design, was only able to offer 5% higher performance than the AMD Ryzen 7 6800H's integrated solution, the Radeon 680M. The Ryzen 5 6600H was able to deliver a 39% uplift over Intel's Iris Xe iGPU. Once again, the 2.4 GHz Radeon 680M has the potential to even surpass the GeForce MX550, and it's impressive coming from an integrated chip that is competing against a discrete GPU that consumes around 35W of power.
We have also seen the AMD Radeon 680M iGPU with RDNA 2 graphics deliver spectacular gaming performance across several titles in detailed testing by TechEpiphany on his new channel known as AMD APU Gaming. We covered a few of his tests in our post here but do make sure to check out the new channel for more performance tests of the most powerful integrated GPU available on PCs at the moment.
News Source: HXL
