AMD Ryzen 5 PRO 4655GE vs Intel Core Ultra 5 228V Comparison
AMD Ryzen 5 PRO 4655GE
Core Ultra 5 228V
PERFORMANCE BENCHMARKS
Analysis: AMD Ryzen 5 PRO 4655GE vs Intel Core Ultra 5 228V
Head-to-Head Benchmarks
The head-to-head comparison between the Intel Core Ultra 5 228V and the AMD Ryzen 5 PRO 4655GE reveals a lopsided contest. Out of 17 recorded benchmark comparisons, the Intel part takes 13 wins, while the AMD processor secures 4. The average benchmark score for the Intel chip is 21440, placing it in the 75th percentile of all CPUs. The AMD Ryzen 5 PRO 4655GE posts an average score of 20900, which lands it in the 74th percentile. The aggregate picture is close, but the distribution of wins is not.
The biggest single victory for the Intel Core Ultra 5 228V comes in the PassMark find prime numbers test. Here the Intel part scores 168 against 28 for the AMD chip, a delta of 500%. That is the largest percentage gap in the entire dataset. The PassMark physics test also shows a dramatic Intel advantage: 1538 versus 592, a 159.8% difference. Floating point math follows the same pattern, with the Intel chip scoring 53310 against 27947, a 90.8% lead. These three tests highlight a consistent theme: the Lunar Lake architecture is exceptionally strong in compute-heavy, single-threaded or physics-oriented workloads.
Single-core Cinebench results reinforce this theme. In Cinebench R15 single-core, the Intel part scores 267 versus 171, a 56.1% advantage. In Cinebench R20 single-core, the margin is 916 versus 714, a 28.3% gap. The PassMark single-thread test shows 3836 versus 2660, a 44.2% lead for Intel. Even in Cinebench R23 single-core, where the margin is smallest, the Intel chip still wins: 1758 versus 1700, a 3.4% edge.
The AMD Ryzen 5 PRO 4655GE does have its own victories, and they are meaningful. The most significant is Cinebench R23 multi-core, where the AMD chip scores 12045 against 9932 for Intel, a 17.5% lead. This is the only Cinebench multi-core test that AMD wins. In the older Cinebench R15 multi-core test, the Intel chip wins with 1502.5 against 1213, a 23.9% margin. In Cinebench R20 multi-core, Intel wins again with 6491 against 5058, a 28.3% margin. The flip in R23 is curious, but the data is clear.
AMD also wins PassMark integer math with 48513 versus 39679, an 18.2% advantage. Data compression goes to AMD as well: 193067 versus 173924, a 9.9% lead. Random string sorting is essentially a tie, with AMD winning by a razor-thin 0.6% margin: 21382 versus 21254.
The remaining Intel wins include PassMark multithread (18227 versus 14171, a 28.6% gap), PassMark data encryption (13032 versus 11586, a 12.5% lead), and PassMark extended instructions (14801 versus 11792, a 25.5% margin). The pattern is clear: Intel dominates single-threaded performance, floating point, encryption, and physics, while AMD counters in integer math, compression, and one generation of Cinebench multi-core.
Where Each One Wins
The Intel Core Ultra 5 228V is the choice for workloads that depend heavily on single-thread performance. Every single-core benchmark in the dataset favors Intel, from Cinebench R15 single-core (56.1% ahead) to Cinebench R23 single-core (3.4% ahead). The PassMark single-thread score of 3836 versus 2660 confirms that this is not a narrow edge; it is a substantial lead in lightly threaded tasks.
Floating point work is also firmly in Intel territory. The PassMark floating point math score of 53310 is 90.8% higher than the AMD chip's 27947. This suggests that scientific computing, simulation, and other floating-point-heavy applications will see a significant benefit from the Intel part. The same applies to physics calculations, where the Intel chip is 159.8% ahead.
Encryption and extended instruction workloads also favor Intel. The data encryption score of 13032 is 12.5% higher, and the extended instructions score of 14801 is 25.5% higher. These are tasks that reward modern instruction set support and efficient execution.
The AMD Ryzen 5 PRO 4655GE wins in integer math and data compression. The integer math score of 48513 is 18.2% ahead of Intel, and the data compression score of 193067 is 9.9% ahead. These are workloads that often benefit from more threads and higher sustained throughput. The AMD part also wins Cinebench R23 multi-core by 17.5%, which suggests that certain heavily threaded rendering workloads can favor the AMD chip.
Random string sorting is effectively a wash, with AMD ahead by only 0.6%. This is not a meaningful differentiator for either platform.
Architecture Differences
The two processors come from fundamentally different design philosophies. The Intel Core Ultra 5 228V is built on the Lunar Lake architecture, manufactured on a 3 nm process by TSMC. It is a mobile part with a 17 watt TDP, designed for efficiency in thin-and-light systems. The AMD Ryzen 5 PRO 4655GE uses the Zen 2 architecture, codenamed Renoir, on a 7 nm process also from TSMC. It is a desktop part with a 35 watt TDP, aimed at mainstream AM4 platforms.
Core counts differ significantly. The Intel chip has 8 cores and 8 threads, meaning no simultaneous multithreading. The AMD chip has 6 cores and 12 threads, using SMT to double thread count. Despite having fewer physical cores, the AMD part offers more threads, which helps in certain multi-threaded workloads like Cinebench R23 multi-core.
Clock speeds tell a different story. The Intel chip has a base clock of 2.10 GHz and a boost clock of 4.50 GHz. The AMD chip has a higher base clock of 3.30 GHz but a lower boost clock of 4.20 GHz. The higher boost clock on the Intel part contributes to its single-thread dominance.
Cache hierarchies are also distinct. Intel uses 192 KB of L1 cache per core, 2.5 MB of L2 cache per core, and 8 MB of shared L3 cache. AMD uses 64 KB of L1 per core, 512 KB of L2 per core, and 8 MB of shared L3 cache. The larger per-core L1 and L2 on the Intel side likely explain some of its performance advantages in latency-sensitive workloads.
Memory support diverges as well. The Intel chip's memory support is listed as dependent on the motherboard, with dual-channel memory bus support. The AMD chip explicitly supports DDR4 with a dual-channel bus and a memory bandwidth of 51.2 GB/s. The AMD part also supports ECC memory, while the Intel part does not.
PCIe connectivity differs. The Intel chip provides Gen 5 with 4 lanes from the CPU. The AMD chip provides Gen 3 with 20 lanes from the CPU. This makes the AMD part more suitable for systems with multiple PCIe devices, while the Intel part offers newer but fewer lanes.
The integrated graphics also differ. Intel uses Arc 130V graphics, while AMD uses Radeon Vega 7. The Intel part is a mobile component on the Intel BGA 2833 socket, while the AMD part is a desktop component on the AMD Socket AM4. The AMD chip has a transistor count of 9,800 million and a die size of 156 mm²; the Intel chip's transistor count and die size are not recorded.
The AMD part was released on 2022-11-10, while the Intel part was released on 2024-09-23. The AMD chip is built for desktop platforms with a 35 watt TDP, while the Intel chip is a low-power mobile part at 17 watts. Both are currently active in production.
The Verdict
The data points to the Intel Core Ultra 5 228V as the stronger overall performer for most workloads. It wins 13 of 17 head-to-head comparisons, and its wins are often by large margins. The PassMark single-thread score of 3836 is 44.2% ahead of the AMD chip, and the Cinebench R15 single-core margin of 56.1% is decisive. For users who prioritize single-threaded responsiveness, floating point math, encryption, physics, or extended instruction workloads, the Intel part is the clear choice based on the recorded data.
The AMD Ryzen 5 PRO 4655GE should be selected for workloads that specifically reward integer math, data compression, or certain multi-threaded rendering scenarios. Its Cinebench R23 multi-core win of 17.5% is notable, as is its 18.2% lead in integer math and 9.9% lead in data compression. The AMD part also offers ECC memory support and a desktop socket with 20 PCIe Gen 3 lanes, which may matter for system builders who need those features.
The overall average benchmark scores are close: 21440 for Intel versus 20900 for AMD. But the distribution of wins is not close. Intel wins the majority of tests and wins most of them by wide margins. The AMD chip's wins are concentrated in a narrower set of workloads. For a general-purpose processor, the Intel Core Ultra 5 228V is the more balanced and higher-performing option in this comparison. For specialized integer or compression workloads, the AMD Ryzen 5 PRO 4655GE has a defensible position.
FAQ
Q: Which processor has the higher single-thread score in PassMark?
A: The Intel Core Ultra 5 228V scores 3836 in the PassMark single-thread test, compared to 2660 for the AMD Ryzen 5 PRO 4655GE, a 44.2% advantage for Intel.
Q: Does the AMD Ryzen 5 PRO 4655GE win any multi-core benchmarks?
A: Yes. The AMD chip wins Cinebench R23 multi-core with a score of 12045 versus 9932 for the Intel chip, a 17.5% lead. However, Intel wins Cinebench R15 multi-core (1502.5 versus 1213) and Cinebench R20 multi-core (6491 versus 5058).
Q: How do the core and thread counts differ?
A: The Intel Core Ultra 5 228V has 8 cores and 8 threads. The AMD Ryzen 5 PRO 4655GE has 6 cores and 12 threads, using simultaneous multithreading to exceed the Intel thread count.
Q: What is the largest performance gap in any benchmark?
A: The PassMark find prime numbers test shows the largest gap. The Intel chip scores 168, while the AMD chip scores 28, a 500% difference in favor of Intel.
Q: Which processor supports ECC memory?
A: The AMD Ryzen 5 PRO 4655GE supports ECC memory. The Intel Core Ultra 5 228V does not support ECC memory.
Q: What are the TDP ratings for each processor?
A: The Intel Core Ultra 5 228V has a TDP of 17 watts and is a mobile part. The AMD Ryzen 5 PRO 4655GE has a TDP of 35 watts and is a desktop part.