The Intel Core i7-1065G7 is a power efficient quad-core SoC of the Ice Lake U product family designed for thin laptops and Ultrabooks. It was announced in May 2019 (Computex). The CPU has four Sunnycove processor cores (8 threads thanks to Hyper-Threading) clocked at 1.3 (base) - 3.9 (single core Turbo) GHz. 2 cores can reach 3.8 GHz and all four 3.5 GHz using Turbo Boost. According to Intel the Sunnycove cores achieve 18% more IPCs (Instructions per Clock).
Other improvements for Ice Lake are the AI hardware acceleration and the partial integration of Thunderbolt and Wi-Fi 6 in the chip. The integrated DDR4 memory controller supports modules with up to 3200 MHz (and LPDDDR4 3733).
The biggest improvement of Ice Lake is the integrated Gen 11 graphics adapter called Iris Plus Graphics. The Core i7-1065G7 integrates the biggest G7 variant with 64 EUs clocked at 300 - 1100 MHz. The Iris Plus G7 should be twice as fast as the predecessors and best the AMD Vega 10 GPU from current Ryzen APUs.
Performance
The average 1065G7 in our database is most comparable to the AMD Ryzen 7 2700U and the Intel Core i5-8259U, as far as multi-thread benchmark scores are concerned. This is a so-so result and a direct consequence of Intel's inability to make the best of its brand-new 10 nm node. The Core i7-10710U, a 10th generation processor built with an older 14 nm process, will easily rip a 1065G7 to pieces. That said, the 1065G7 will surely beat a 10710U in terms of energy efficiency.
Thanks to its decent cooling solution and sufficiently high CPU power limits, the Galaxy Book Flex 15-NP950 is among the fastest laptops powered by the 1065G7 that we know of. It can be around 40% faster in CPU-bound workloads than the slowest system featuring the same chip in our database, as of August 2023.
Power consumption
This Core i7 has a default TDP, also known as the long-term power limit, of 15 W. Laptop makers are free to change that value to anything between 12 W and 25 W with clock speeds and performance changing accordingly as a result. Either way, that's a tad too high to allow for passively cooled designs.
The quad-core Intel CPU is built with Intel's second-gen 10 nm process (not 10 nm SuperFin or Intel 7) for average, as of late 2022, energy efficiency.
The AMD Ryzen 5 4600H is a mobile SoC for big laptops based on the Renoir architecture. The 4600H integrates six of the eight cores based on the Zen 2 microarchitecture. They are clocked at 3 (guaranteed base clock) to 4 GHz (Turbo) and support SMT / Hyperthreading (12 threads).
In addition to the six CPU cores, the APU also integrates a Radeon RX Vega 6 integrated graphics adapter with 6 CUs and up to 1500 MHz. The dual channel memory controller supports DDR4-3200 and energy efficient LPDDR4-4266 RAM. Furthermore, 8 MB level 3 cache can be found on the chip. See our hub page on the Renoir Processors for more information.
Performance
The average 4600H in our database matches the Intel Core i7-10875H and even the Core i9-10885H, both significantly more costly chips, in multi-thread performance, giving this Ryzen an outrageously high price-to-performance ratio.
Your mileage may vary depending on how high the CPU power limits are and how competent the cooling solution of your system is.
Power consumption
The Ryzen 5 chip has a default TDP (also known as the long-term power limit) of 45 W, a value that laptop makers are free to change to anything between 35 W and 54 W with clock speed and performance changing accordingly as a result. Either way, a high-performance cooling solution is a must for a CPU like this.
The 7 nm TSMC process this APU is built with makes for better-than-average, as of early 2022, energy efficiency.
The AMD Ryzen 9 4900HS is a mobile SoC for big laptops based on the Renoir architecture. The 4900HS integrates all eight cores based on the Zen 2 microarchitecture. They are clocked at 3 (guaranteed base clock) to 4.3 GHz (Turbo) and support SMT / Hyperthreading (16 threads). The chip is manufactured in the modern 7 nm process at TSMC and partly thanks to it AMD advertises a 2x improved performance per Watt for the Renoir chips
According to our information, the Ryzen 9 4900HS is only slightly faster than the Ryzen 7 4800HS. The CPU is clocked 100 MHz (base and boost) higher, but the gains greatly depend on the cooling and TDP settings. Compared to Intel chips, the i9-9980HK top model should be comparable. That means the Ryzen 9 4900HS should be able to handle all demanding tasks and gaming. Compared to the similar named 45W-version, the Ryzen 9 4900H, the 4900HS is differs in 100 MHz boost and 300 MHz base clock.
In addition to the eight CPU cores, the APU also integrates a Radeon RX Vega 7 integrated graphics card with most likely 7 CUs. The dual channel memory controller supports DDR4-3200 and energy efficient LPDDR4-4266 RAM. Furthermore, 8 MB level 3 cache can be found on the chip. See our hub page on the Renoir Processors for more information.
The TDP of the APU is specified at 35 Watt (default). That means the chip is intended for big and relatively heavy laptops. The Ryzen 7 4900HS is the 35 Watt version of the R9 4900H with a 10 Watt lower TDP.
Average Benchmarks Intel Core i7-1065G7 → 100%n=32
Average Benchmarks AMD Ryzen 5 4600H → 137%n=32
Average Benchmarks AMD Ryzen 9 4900HS → 167%n=32
- Range of benchmark values for this graphics card - Average benchmark values for this graphics card * Smaller numbers mean a higher performance 1 This benchmark is not used for the average calculation
v1.26
log 26. 16:02:33
#0 checking url part for id 11399 +0s ... 0s
#1 checking url part for id 11679 +0s ... 0s
#2 checking url part for id 11882 +0s ... 0s
#3 not redirecting to Ajax server +0s ... 0s
#4 did not recreate cache, as it is less than 5 days old! Created at Thu, 25 Apr 2024 05:38:32 +0200 +0.001s ... 0.001s
#5 composed specs +0.062s ... 0.063s
#6 did output specs +0s ... 0.063s
#7 getting avg benchmarks for device 11399 +0.017s ... 0.08s
#8 got single benchmarks 11399 +0.125s ... 0.205s
#9 getting avg benchmarks for device 11679 +0.018s ... 0.223s
#10 got single benchmarks 11679 +0.048s ... 0.271s
#11 getting avg benchmarks for device 11882 +0.017s ... 0.288s
#12 got single benchmarks 11882 +0.013s ... 0.301s
#13 got avg benchmarks for devices +0s ... 0.301s
#14 min, max, avg, median took s +0.49s ... 0.791s