AMD Ryzen Embedded V1605B vs Intel Xeon E3-1235L v5 Comparison

The Intel Xeon E3-1235L v5 and AMD Ryzen Embedded V1605B are both 4-core, 14nm processors aimed at compact systems, but they diverge sharply in thread count, cache layout, and benchmark behavior. The database shows the AMD part winning three of four head-to-head tests, while the Intel part takes a decisive victory in Cinebench R23 multi-core. Both sit at the 34th percentile of all CPUs, with average scores of 1232 for Intel and 1196 for AMD, so the overall performance envelope is close, but the distribution of wins tells a more nuanced story.

FAQ

Q: Which processor has more threads?

A: The AMD Ryzen Embedded V1605B has 8 threads, while the Intel Xeon E3-1235L v5 has 4 threads.

Q: Which processor has a higher boost clock?

A: The AMD Ryzen Embedded V1605B boosts to 3.60 GHz, compared to 3.00 GHz for the Intel Xeon E3-1235L v5.

Q: Which processor supports ECC memory?

A: The Intel Xeon E3-1235L v5 supports ECC memory; the AMD Ryzen Embedded V1605B does not.

Q: Which processor has a lower TDP?

A: The AMD Ryzen Embedded V1605B has a 15W TDP, while the Intel Xeon E3-1235L v5 has a 25W TDP.

Q: Which processor wins Cinebench R23 multi-core?

A: The Intel Xeon E3-1235L v5 scores 4261 versus 3160 for the AMD, a 34.8% lead.

Q: Which processor has a higher average benchmark score?

A: The Intel Xeon E3-1235L v5 averages 1232, slightly above the AMD's 1196, though both are at the 34th percentile.

Architecture Differences

The two chips share a 14nm process node, but the similarities end there. Intel fabricates its Skylake-DT die in-house, while AMD uses GlobalFoundries for its Zen (Great Horned Owl) design. The transistor counts differ dramatically: AMD packs 4,950 million transistors on a 210 mm² die, while Intel uses 1,750 million on a 122 mm² die. That larger transistor budget shows up in the cache hierarchy. AMD provides 128 KB of L1 and 512 KB of L2 per core, plus a 2 MB shared L3. Intel offers 64 KB L1 and 256 KB L2 per core, but a much larger 8 MB shared L3.

Memory support also diverges. Intel accepts both DDR3 and DDR4 in dual-channel mode and lists a memory bandwidth of 34.1 GB/s. AMD is limited to DDR4 dual-channel, with no bandwidth figure recorded. ECC is a differentiator: Intel supports it, AMD does not. Intel also exposes PCIe Gen 3 with 16 lanes from the CPU, while AMD's PCIe configuration is not listed. Integrated graphics differ as well: Intel uses HD Graphics P530, AMD uses Radeon Vega 8. The sockets are incompatible (Intel Socket 1151 versus AMD Socket FP5), and the market segments differ: Intel is classified as Server/Workstation, AMD as Desktop. Intel is end-of-life, while AMD remains active. Release dates are 2015-10-18 for Intel and 2018-02-20 for AMD. Intel's launch MSRP is $250; AMD has no listed MSRP.

The Verdict

The benchmark data points to a clear split. The AMD Ryzen Embedded V1605B wins three of the four head-to-head tests: Cinebench R15 multi-core, R15 single-core, and R23 single-core. The Intel Xeon E3-1235L v5 wins only Cinebench R23 multi-core, but that win is substantial, a 34.8% margin. For workloads that resemble Cinebench R23 multi-threaded rendering, the Intel part is the stronger choice. For single-threaded tasks or older Cinebench R15 multi-threaded workloads, the AMD part leads. The AMD also offers a lower TDP (15W versus 25W) and is still in active production, while Intel is end-of-life. However, Intel brings ECC support, a listed memory bandwidth, and PCIe Gen 3 lanes, which matter for server or workstation use. The choice depends on whether the target application scales like R23 multi-core or like the other tested scenarios.

Specification Differences

| Field | Intel Xeon E3-1235L v5 | AMD Ryzen Embedded V1605B |

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

| Threads | 4 | 8 |

| Boost clock | 3.00 GHz | 3.60 GHz |

| TDP | 25 W | 15 W |

| Socket | Intel Socket 1151 | AMD Socket FP5 |

| Architecture | Skylake | Zen |

| Foundry | Intel | GlobalFoundries |

| Transistors | 1,750 million | 4,950 million |

| Die size | 122 mm² | 210 mm² |

| L1 cache | 64 KB per core | 128 KB per core |

| L2 cache | 256 KB per core | 512 KB per core |

| L3 cache | 8 MB shared | 2 MB shared |

| Memory support | DDR3, DDR4 | DDR4 |

| Memory bandwidth | 34.1 GB/s | Not listed |

| ECC | Yes | No |

| PCIe | Gen 3, 16 lanes | Not listed |

| Integrated graphics | HD Graphics P530 | Radeon Vega 8 |

| Market segment | Server/Workstation | Desktop |

| Production status | End-of-life | Active |

| Release date | 2015-10-18 | 2018-02-20 |

| Launch MSRP | $250 | Not listed |

| Part number | SR2LM | Not listed |

Base clock (2.00 GHz), core count (4), process node (14 nm), memory bus (dual-channel), and unlocked multiplier (false) are identical, so they are omitted.

Head-to-Head Benchmarks

The four recorded Cinebench tests show a consistent pattern with one notable exception. In Cinebench R15 multi-core, the AMD Ryzen Embedded V1605B scores 654 against Intel's 429, a 34.4% advantage for AMD. The single-core R15 test is even more lopsided: AMD scores 128, Intel 60, giving AMD a 53.1% lead. Moving to Cinebench R23, the single-core result still favors AMD, 841 versus 601, a 28.5% margin. But the R23 multi-core test flips the script: Intel scores 4261, AMD 3160, and Intel leads by 34.8%. That is Intel's only win, but it is the largest absolute score difference in the set. The R15 multi-core result is surprising given AMD's 8 threads versus Intel's 4, yet the R23 multi-core result shows Intel pulling ahead despite the thread deficit. The data does not explain the reversal, but it is consistent across the recorded runs.

Where Each One Wins

The AMD Ryzen Embedded V1605B wins in three of the four head-to-head benchmarks. It takes Cinebench R15 multi-core (654 vs 429), R15 single-core (128 vs 60), and R23 single-core (841 vs 601). That makes it the better choice for single-threaded workloads and for multi-threaded tasks that behave like Cinebench R15. It also has a lower TDP (15W) and is still in active production, which may simplify system integration.

The Intel Xeon E3-1235L v5 wins only Cinebench R23 multi-core, but that win is decisive: 4261 versus 3160, a 34.8% lead. For applications that scale similarly to Cinebench R23 multi-threaded rendering, Intel is clearly ahead. Intel also offers ECC memory support, a listed memory bandwidth of 34.1 GB/s, and PCIe Gen 3 with 16 lanes, which are relevant for server or workstation deployments. Its end-of-life status and higher TDP are drawbacks, but the R23 multi-core performance and platform features give it a distinct niche.