AMD Athlon 300U vs Intel Core i5-11260H vs Intel Core i9-11950H

AMD Athlon 300U

The AMD Athlon 300U is a mobile entry level dual core SoC that was announced in January 2019. It combines two Zen cores (with SMT / Hyperthreading so running 4 threads) clocked at 2.4 - 3.3 GHz with a Radeon RX Vega 3 graphics card with 3 CUs (192 Shaders) clocked at up to 1000 MHz. Specified at 15 Watt TDP, the SoC is intended for thin mid-range laptops. It is quite similar to the AMD Ryzen 3 3200U (2.6 - 3.5 GHz).

Compared to the Picasso APUs of the Ryzen 3000 series, the Athlon 300U is still based on the first generation of the 2000 series. Only the PRO variant is based on the new Zen+ microarchitecture that should lead to a 3% IPS (performance per clock) improvement.

The integrated dual-channel memory controller supports up to DDR4-2400 memory. As the features of the Picasso APUs are the same compared to the Raven Ridge predecessors, we point to our Raven Ridge launch article.

In contrast to the faster quad-core Picasso APUs, the Athlon 300U only supports 3 instead of 4 displays in total.

Performance wise, the Athlon 300 should be slightly slower than the Ryzen 3 2200U (2.5 - 3.4 GHz Dual Core with SMT).

Intel Core i5-11260H

The Intel Core i5-11260H is a mid range hexa core SoC for gaming laptops and mobile workstations. It is based on the Tiger Lake H45 generation and was announced in mid 2021. It integrates six Willow Cove processor cores (12 threads thanks to HyperThreading). The base clock speed depends on the TDP setting and at 45 Watt is at 2.6 GHz. The single and dual core boost speed can reach up to 4.4 GHz, all cores can reach up to 4 GHz. The CPU offers 12 MB level 3 cache and supports DDR4-3200 memory.

Thanks to the new Tiger Lake architecture, the i5-11260H should be able to reach the level of the high clocked Intel Core i7-10850H (Comet Lake-H, 6 cores, 2.7 - 5.1 GHz, 16 MB L3) in single and multi threaded performance. Compared, to current AMD offerings, the 11260H should perform similar to a Ryzen 5 5600H (6 cores, 2.4 - 3.3. GHz, 16 MB L3 cache) or the 5600HS when the TDP is set to 35W.

The SoC also features a much improved UHD Graphics (16 EUs) iGPU based on the Xe architecture.

Furthermore, Tiger Lake SoCs add PCIe 4 support (20 lanes in the H45 series), AI hardware acceleration, and the partial integration of Thunderbolt 4/USB 4 and Wi-Fi 6E in the chip.

The chip is produced in the improved 10nm process (called 10nm SuperFin) at Intel, which should be comparable to the 7nm process at TSMC (e.g. Ryzen 4000 series). The default TDP is rated at 45 W at 2.6 GHz base speed, at 35 Watt the base clock speed decreases to 2.1 GHz (cTDP down).

All articles on Tiger Lake can be found on our Tiger Lake architecture hub.

Intel Core i9-11950H

The Intel Core i9-11950H is a high end octa core SoC for laptops and mobile workstations. It is based on the Tiger Lake H45 generation and was announced in mid 2021. It integrates eight Willow Cove processor cores (16 threads thanks to HyperThreading). The base clock speed depends on the TDP setting and at 45 Watt is at 2.6 GHz. The single core boost speed can reach up to 5 GHz (ITBM 3.0, also two cores), all cores can reach up to 4.5 GHz. The CPU offers 24 MB level 3 cache and supports DDR4-3200 memory. Compared to the i9-11980HK, the 11950H offers professional management features like Intel vPro, SIPP or TXT.

Thanks to the new Tiger Lake architecture, the i9-11980HK should be significantly faster than the old Intel Core i9-10980HK (Comet Lake-H, 2.4 - 5.3 GHz, 16 MB L3) in single and multi threaded performance and similar to the Ryzen 9 5980HX (Zen 3, 3.3 - 4.8 GHz, 16 MB L3). Therefore, the i9 should be one of the fastest laptop CPUs in 2021 and well suited for even very demanding tasks.

The SoC also included the improved Xe graphics card called UHD graphics with all 32 EUs.

Furthermore, Tiger Lake SoCs add PCIe 4 support (20 lanes in the H45 series), AI hardware acceleration, and the partial integration of Thunderbolt 4/USB 4 and Wi-Fi 6E in the chip.

The chip is produced on the improved 10nm process (called 10nm SuperFin) at Intel, which should be comparable to the 7nm process at TSMC (e.g. Ryzen 4000 series). The default TDP is rated at 45 W, at 65 Watt the base clock speed increases to 3.3 GHz (cTDP up).

All articles on Tiger Lake can be found on our Tiger Lake architecture hub.