AMD Ryzen 5 PRO 4655GE vs Intel Core Ultra 7 155U Comparison

FAQ

Q: Which processor has the higher average benchmark score?

A: The Intel Core Ultra 7 155U holds a marginal edge with an average benchmark score of 21,174, compared to the AMD Ryzen 5 PRO 4655GE's 20,900. That difference is roughly 1.3%, placing both CPUs in the same performance tier.

Q: How do the two chips rank against all CPUs?

A: The Intel part sits at the 75th percentile, while the AMD part sits at the 74th percentile. Both are effectively in the same overall performance bracket, with the Intel chip being one percentile point higher.

Q: Which processor wins more head-to-head benchmark tests?

A: The Intel Core Ultra 7 155U wins 11 of the 17 head-to-head tests, while the AMD Ryzen 5 PRO 4655GE wins 6. However, the AMD chip's wins include the heavy multi-threaded Cinebench R23 test, which carries significant weight.

Q: What is the single biggest performance gap in either direction?

A: The largest margin is Intel's 132.1% lead in the PassMark find prime numbers test (65 vs 28). The second-largest is AMD's 19.7% deficit in Cinebench R23 multi-core, where AMD actually wins by scoring 12,045 against Intel's 9,674.5.

Q: Do both processors support ECC memory?

A: No. The AMD Ryzen 5 PRO 4655GE supports ECC memory, while the Intel Core Ultra 7 155U does not. This is a notable differentiator for workstation or server-oriented builds.

Q: Are these chips from the same manufacturing node?

A: Both are listed as 7 nm parts, but they come from different foundries. Intel fabricates its chip at Intel's own 7 nm process, while AMD uses TSMC's 7 nm process.

Architecture Differences

The Intel Core Ultra 7 155U is built on the Meteor Lake architecture, part of the Core Ultra Series 1. It is a mobile-segment processor with 12 cores and 14 threads, using a hybrid design that pairs performance and efficiency cores. The chip integrates Arc Xe-LPG graphics with 64 execution units. Its cache hierarchy includes 112 KB of L1 per core, 2 MB of L2 per core, and a 12 MB shared L3 cache. Memory support is DDR5, with bandwidth rated at 89.6 GB/s. The processor is built on Intel's 7 nm process and uses the Intel BGA 2049 socket.

The AMD Ryzen 5 PRO 4655GE is a desktop part from the 4000 series, based on the Zen 2 architecture (codename Renoir). It has 6 cores and 12 threads, with a simpler uniform core layout. The integrated graphics are Radeon Vega 7. Cache amounts are 64 KB L1 per core, 512 KB L2 per core, and 8 MB shared L3. Memory support is DDR4, with bandwidth at 51.2 GB/s. AMD lists the transistor count at 9,800 million on a 156 mm² die, fabricated by TSMC at 7 nm. It uses the AMD Socket AM4.

A key architectural difference is the memory interface. Intel's chip supports newer DDR5 memory, while AMD's part uses DDR4. The AMD chip also has ECC memory support, which Intel lacks. The Intel processor's hybrid core arrangement likely explains its higher core count (12 vs 6) despite having only two more threads (14 vs 12). The AMD chip's higher TDP of 35 watts, versus Intel's 15 watts, suggests AMD allocates more power headroom for sustained desktop workloads.

Where Each One Wins

The Intel Core Ultra 7 155U dominates in single-threaded and lightly threaded workloads. It wins the Cinebench R15 single-core test by 39.5%, the Cinebench R20 single-core test by 13.9%, and the PassMark single-thread test by 26.8%. It also leads in integer math (18.7%), floating-point math (44%), and multithread performance (15.6%). The physics test shows a massive 83.1% lead, and the prime number finding test shows an even larger 132.1% advantage. These results point to a processor that handles bursty, latency-sensitive tasks exceptionally well.

The AMD Ryzen 5 PRO 4655GE wins in specific sustained workloads. Its most significant victory is Cinebench R23 multi-core, where it scores 12,045 versus Intel's 9,674.5, a 19.7% lead. It also wins data compression (8.3% higher), data encryption (2.2% higher), extended instructions (16% higher), and random string sorting (7.6% higher). These wins suggest AMD's six full Zen 2 cores sustain throughput better in certain compute-heavy, multi-threaded scenarios, despite having fewer cores.

The split is clear: Intel excels at short, single-threaded bursts and math-heavy operations; AMD excels at sustained multi-threaded rendering and data manipulation tasks. The Cinebench R23 result is particularly telling, as it flips the multi-core narrative established by the R15 and R20 tests.

Specification Differences

| Specification | Intel Core Ultra 7 155U | AMD Ryzen 5 PRO 4655GE |

|---|---|---|

| Cores | 12 | 6 |

| Threads | 14 | 12 |

| Base Clock | 1700.00 MHz | 3.30 GHz |

| Boost Clock | 4.80 GHz | 4.20 GHz |

| TDP | 15 W | 35 W |

| Socket | Intel BGA 2049 | AMD Socket AM4 |

| Architecture | Meteor Lake | Zen 2 |

| Process Node | 7 nm (Intel) | 7 nm (TSMC) |

| L1 Cache | 112 KB (per core) | 64 KB (per core) |

| L2 Cache | 2 MB (per core) | 512 KB (per core) |

| L3 Cache | 12 MB (shared) | 8 MB (shared) |

| Memory Support | DDR5 | DDR4 |

| Memory Bandwidth | 89.6 GB/s | 51.2 GB/s |

| ECC Memory | No | Yes |

| PCIe | Gen 4, 12 Lanes | Gen 3, 20 Lanes |

| Integrated Graphics | Arc Xe-LPG 64EU | Radeon Vega 7 |

| Market Segment | Mobile | Desktop |

| Release Date | 2023-12-13 | 2022-11-10 |

The most consequential differences are core count, TDP, memory generation, and PCIe version. Intel has twice the cores, but AMD has a significantly higher base clock (3.30 GHz vs 1.70 GHz) and a lower boost clock gap. Intel's L3 cache is 50% larger, but AMD's per-core L2 cache is smaller. Intel supports PCIe Gen 4, while AMD supports PCIe Gen 3 but with more lanes (20 vs 12). AMD's launch MSRP is not listed; Intel's launch MSRP is $490. Both are on 7 nm processes, but from different foundries.

Head-to-Head Benchmarks

The head-to-head data reveals a complex picture. Starting with Cinebench R15, Intel wins both tests decisively: multi-core at 1585 vs 1213 (a 30.7% lead) and single-core at 238.5 vs 171 (39.5% lead). In Cinebench R20, Intel's margins narrow but remain solid: multi-core 5765 vs 5058 (14% lead) and single-core 813 vs 714 (13.9% lead).

The story flips in Cinebench R23. AMD wins multi-core with 12,045 against Intel's 9,674.5, a 19.7% margin. The single-core result is nearly identical, with AMD at 1700 and Intel at 1666, a 2% lead for AMD. This reversal suggests that the R23 workload scales differently across the two architectures, favoring AMD's lower core count but higher per-core sustained performance.

In PassMark tests, Intel wins six of eight. The largest margins are in physics (1084 vs 592, an 83.1% lead) and floating-point math (40256 vs 27947, a 44% lead). Intel also wins integer math (57564 vs 48513, 18.7%), multithread (16382 vs 14171, 15.6%), and single-thread (3372 vs 2660, 26.8%). The prime number test is a blowout at 65 vs 28.

AMD's PassMark wins are narrower: data compression (193067 vs 177122, 8.3% lead), data encryption (11586 vs 11333, 2.2% lead), extended instructions (11792 vs 9903, 16% lead), and random string sorting (21382 vs 19767, 7.6% lead). These wins are concentrated in data-oriented tasks, while Intel's wins are in math and physics simulations.

The Verdict

The data indicates two distinct usage profiles. For mobile users seeking a balanced, power-efficient processor with strong single-threaded performance, the Intel Core Ultra 7 155U is the clear choice. It wins 11 of 17 tests, has a 75th percentile ranking, and delivers decisive leads in physics, floating-point math, and single-core workloads. Its 15-watt TDP and mobile socket make it suitable for thin-and-light laptops, while its DDR5 support and PCIe Gen 4 connectivity provide a modern platform foundation.

For desktop users running sustained multi-threaded rendering or data-intensive tasks, the AMD Ryzen 5 PRO 4655GE offers specific advantages. Its Cinebench R23 multi-core win by 19.7% is significant for long-duration rendering. It also wins data compression and encryption tests, which matter for file archiving and security workloads. The ECC memory support and 20 PCIe Gen 3 lanes could appeal to small-scale server or workstation builds. Its 35-watt TDP and desktop socket are appropriate for always-on operation.

The average benchmark scores are nearly identical (21,174 vs 20,900), so neither chip dominates overall. The decision hinges on workload. Users prioritizing burst performance, math-heavy code, and modern memory standards should choose Intel. Users prioritizing sustained multi-threaded rendering, data throughput, and ECC support should choose AMD. The Intel chip's higher launch MSRP of $490 may also factor into platform-level decisions, though the AMD chip's pricing is not listed. Both processors remain active production parts, indicating continued availability for their respective markets.