AMD Ryzen 5 PRO 5655G vs Intel Core 5 220H Comparison
AMD Ryzen 5 PRO 5655G
Core 5 220H
PERFORMANCE BENCHMARKS
Analysis: AMD Ryzen 5 PRO 5655G vs Intel Core 5 220H
Head-to-Head Benchmarks
The benchmark data reveals a clear split between these two processors: the Intel Core 5 220H wins 12 of the 17 recorded head-to-head tests, while the AMD Ryzen 5 PRO 5655G takes 5. The Intel part dominates in raw compute and physics workloads, while AMD counters with strong results in Cinebench R23 and extended instruction tests.
The largest margin in the entire comparison belongs to the Intel Core 5 220H in PassMark physics, where it scores 1478 against AMD's 686, a 115.5% advantage. This is a massive gap and indicates that the Intel chip's architecture handles the physics simulation workload far more efficiently. Floating point math also heavily favors Intel, with a score of 51671 versus 38649, a 33.7% lead. Prime number finding shows a 67.3% advantage for Intel (82 versus 49), which reflects a strong integer throughput in that specific test.
Intel also leads in the other PassMark workloads. Multithread performance sits at 21884 versus 19129, a 14.4% advantage. Integer math shows 73555 versus 67561, an 8.9% margin. Random string sorting goes to Intel by 12.6% (28438 versus 25253). Data compression, however, is closer, with AMD edging ahead at 255608 versus 247921, a 3% difference. Data encryption is nearly identical: 15253 for AMD versus 15216 for Intel, a 0.2% gap that is effectively negligible.
In Cinebench tests, the results are contradictory across versions. The R15 and R20 suites favor Intel. Cinebench R15 multicore gives Intel 1835 against AMD's 1738, a 5.6% lead. R15 singlecore shows Intel at 262 versus 245, a 6.9% advantage. R20 multicore has Intel at 7812 versus 7244, a 7.8% margin, and R20 singlecore likewise shows Intel ahead by 7.8% (1102 versus 1022).
The picture flips in Cinebench R23. AMD wins multicore by a wide margin, scoring 17249 versus Intel's 11198, a 35.1% difference. AMD also takes R23 singlecore with 2435 versus 1853, a 23.9% lead. This inconsistency across Cinebench versions suggests that the R23 workload interacts differently with the two architectures, favoring AMD substantially. Extended instructions in PassMark also go to AMD, with 17944 versus 14642, an 18.4% advantage.
Overall, the Intel Core 5 220H posts an average benchmark score of 28574, placing it in the 80th percentile of all CPUs. The AMD Ryzen 5 PRO 5655G averages 28032, also in the 80th percentile. Their nearest rivals cluster closely: Intel sits within 0.3% of the AMD EPYC 7203P, AMD Ryzen 7 PRO 6850HS, Intel Xeon E-2436, and Intel Core i5-12600. AMD's nearest rivals include the AMD Ryzen 5 PRO 8500GE, Intel Core i9-12900H, Intel Core i5-14500T, and AMD Ryzen AI 5 435, all within 0.3%.
FAQ
Q: Which processor is faster in Cinebench R23 multicore?
A: The AMD Ryzen 5 PRO 5655G scores 17249, which is 35.1% higher than the Intel Core 5 220H's 11198. This is the largest multicore gap in the comparison.
Q: How do the two chips compare in PassMark physics performance?
A: The Intel Core 5 220H scores 1478, which is 115.5% higher than the AMD Ryzen 5 PRO 5655G's 686. This is the largest single-test margin in the entire dataset.
Q: Does the AMD processor win any single-threaded tests?
A: Yes, AMD wins Cinebench R23 singlecore with 2435 versus 1853, a 23.9% advantage. However, Intel wins Cinebench R15 and R20 singlecore tests, and PassMark single-thread tests, where Intel scores 3405 versus 3241, a 5.1% lead.
Q: What is the difference in average benchmark score?
A: The Intel Core 5 220H has an average benchmark score of 28574, while the AMD Ryzen 5 PRO 5655G averages 28032. Both sit in the 80th percentile of all CPUs.
Q: Are there any tests where the two processors are nearly tied?
A: PassMark data encryption is nearly identical. AMD scores 15253 and Intel scores 15216, a difference of only 0.2%.
Q: How many head-to-head tests does each processor win?
A: The Intel Core 5 220H wins 12 tests, while the AMD Ryzen 5 PRO 5655G wins 5 tests.
Architecture Differences
The two processors come from entirely different design philosophies. The Intel Core 5 220H is built on Raptor Lake architecture, specifically the Raptor Lake-H refresh, fabricated on Intel's 10 nm process at Intel's own foundry. It has 12 cores and 16 threads, with a base clock of 2.70 GHz and a boost clock of 4.90 GHz. Its thermal design power is 45 watts, and it uses the Intel BGA 1744 socket, indicating a mobile-focused design.
The AMD Ryzen 5 PRO 5655G belongs to the 5000 series, using Zen 3 architecture under the codename Cezanne. It is built on TSMC's 7 nm process, with 10,700 million transistors on a 180 mm² die. It has 6 cores and 12 threads, a base clock of 3.90 GHz, and a boost clock of 4.40 GHz. Its thermal design power is 65 watts, and it uses the AMD Socket AM4, a desktop platform.
Cache layouts differ significantly. Intel provides 80 KB of L1 cache per core, 2 MB of L2 cache per core, and 18 MB of shared L3 cache. AMD provides 64 KB of L1 per core, 512 KB of L2 per core, and 16 MB of L3. The Intel design's larger per-core L2 cache is notable, while AMD's total L3 cache is slightly smaller at 16 MB.
Memory support diverges as well. Intel supports both DDR4 and DDR5 in a dual-channel configuration, while AMD supports only DDR4, also dual-channel. AMD's memory bandwidth is listed at 51.2 GB/s, while no bandwidth figure is recorded for Intel. AMD also supports ECC memory, while Intel does not.
PCIe connectivity differs substantially. Intel offers PCIe Gen 5 with 8 lanes from the CPU, while AMD offers PCIe Gen 3 with 16 lanes from the CPU. This means Intel has a newer PCIe standard but fewer lanes, while AMD has more lanes on an older standard.
Integrated graphics also differ. Intel uses Iris Xe Graphics with 80 execution units, while AMD uses Radeon Vega 7. Both are integrated solutions, but they belong to different graphics architectures.
The market segments reflect their intended use. Intel is marked as Mobile, with a release date of December 2024 and a launch MSRP of $342. AMD is marked as Desktop, released in May 2024, and has no recorded launch MSRP. Both are currently Active in production, and neither has an unlocked multiplier.
The Verdict
The data points to a straightforward conclusion: the Intel Core 5 220H is the stronger overall performer in the majority of recorded workloads, winning 12 of 17 tests. Its advantages are especially pronounced in physics simulation, floating point math, and multithreaded PassMark workloads. Anyone prioritizing those areas should choose Intel.
The AMD Ryzen 5 PRO 5655G, however, has its own clear strengths. Its Cinebench R23 results, both multicore and singlecore, are dramatically better, with a 35.1% multicore advantage and a 23.9% singlecore advantage. It also leads in data compression and extended instructions. For workloads that mirror Cinebench R23's demands, AMD is the better fit.
The average benchmark scores are close, with Intel at 28574 and AMD at 28032, a difference of roughly 1.9%. Both processors sit in the 80th percentile of all CPUs. This means that for general-purpose use, the two are comparable in overall standing, but their behavior in specific workloads varies widely.
The physical and platform differences matter. Intel is a mobile chip with a 45 watt TDP and a BGA socket, while AMD is a desktop chip with a 65 watt TDP and an AM4 socket. These are not interchangeable. The choice should be driven by the intended platform first, then by the workload profile. On a mobile platform, Intel's physics and math strengths are compelling. On a desktop platform where Cinebench-style rendering and extended instruction workloads matter, AMD's results are superior.
Specification Differences
The recorded specifications show the following differences between the two processors:
- Cores: Intel 12, AMD 6
- Threads: Intel 16, AMD 12
- Base clock: Intel 2.70 GHz, AMD 3.90 GHz
- Boost clock: Intel 4.90 GHz, AMD 4.40 GHz
- TDP: Intel 45 W, AMD 65 W
- Socket: Intel BGA 1744, AMD Socket AM4
- Architecture: Intel Raptor Lake, AMD Zen 3
- Codename: Intel Raptor Lake-H, AMD Cezanne
- Process node: Intel 10 nm, AMD 7 nm
- Foundry: Intel, TSMC
- Transistors: AMD 10,700 million, not recorded for Intel
- Die size: AMD 180 mm², not recorded for Intel
- L1 cache: Intel 80 KB per core, AMD 64 KB per core
- L2 cache: Intel 2 MB per core, AMD 512 KB per core
- L3 cache: Intel 18 MB shared, AMD 16 MB
- Memory support: Intel DDR4 and DDR5, AMD DDR4 only
- Memory bandwidth: AMD 51.2 GB/s, not recorded for Intel
- ECC memory: AMD true, Intel false
- PCIe: Intel Gen 5 with 8 lanes, AMD Gen 3 with 16 lanes
- Integrated graphics: Intel Iris Xe Graphics 80EU, AMD Radeon Vega 7
- Market segment: Intel Mobile, AMD Desktop
- Release date: Intel December 2024, AMD May 2024
- Launch MSRP: Intel $342, AMD not recorded
Where Each One Wins
The Intel Core 5 220H is the clear winner in compute-heavy and physics-oriented tasks. Its 115.5% lead in PassMark physics makes it the obvious pick for workloads that involve simulation or physics calculations. It also wins floating point math by 33.7%, prime number finding by 67.3%, and integer math by 8.9%. Its multithread PassMark score is 14.4% higher, and it wins random string sorting by 12.6%. In Cinebench R15 and R20, Intel leads in both singlecore and multicore, with margins ranging from 5.6% to 7.8%. For users running these specific benchmarks, Intel is the stronger choice.
The AMD Ryzen 5 PRO 5655G wins in Cinebench R23, with a 35.1% multicore advantage and a 23.9% singlecore advantage. It also wins extended instructions by 18.4% and data compression by 3%. Data encryption is essentially tied, with AMD ahead by only 0.2%. For rendering workloads that align with Cinebench R23, or for tasks that leverage extended instruction sets, AMD should be preferred.
The use-case split is clear. Intel wins the majority of tests, particularly in math, physics, and multithreaded PassMark workloads. AMD wins the tests where its Zen 3 architecture and higher base clock appear to provide an edge, especially in the R23 benchmark and extended instruction processing. Neither chip dominates universally, so the correct choice depends on which workloads matter most.