Intel Core i5-6350HQ vs Intel Xeon E5620 Comparison

Where Each One Wins

The benchmark data presents a remarkably one-sided picture. The Intel Core i5-6350HQ wins all five recorded head-to-head Cinebench comparisons against the Intel Xeon E5620. There is no benchmark in the database where the Xeon comes out ahead. This is not a close contest with trade-offs; it is a consistent sweep across both multi-core and single-core workloads.

For the i5-6350HQ, the wins are broad. It leads in Cinebench R15 multi-core, R20 multi-core and single-core, and R23 multi-core and single-core. The margins are nearly uniform, hovering around 21% in every test. That consistency tells you the advantage is architectural, not workload-specific. Whether the software uses one thread or all available threads, the mobile Skylake part simply executes better.

The Xeon E5620 does have one structural advantage: it supports 8 threads versus the i5's 4 threads. Yet that extra thread count does not translate into a single benchmark victory. In multi-core tests where those extra threads should help most, the i5 still wins by roughly 21%. The Xeon's older Westmere cores are so far behind in per-thread performance that even having twice the thread count cannot close the gap.

In percentile terms, the i5-6350HQ sits at the 29th percentile of all CPUs in the database, while the Xeon E5620 sits at the 28th percentile. These are adjacent rankings, but the average benchmark scores differ slightly: the i5 averages 1046, the Xeon averages 1029. That places them near the same performance tier overall, but the head-to-head data shows the i5 is consistently the stronger chip in Cinebench workloads.

Architecture Differences

These two processors come from completely different eras and design philosophies. The Intel Core i5-6350HQ is a Skylake-H mobile part built on Intel's 14 nm process. It contains 2,300 million transistors on a 171 mm² die. The Xeon E5620 is a Westmere-EP server chip built on 32 nm, with 1,170 million transistors on a larger 239 mm² die. The newer process node gives the i5 a significant density and efficiency advantage.

Both have 4 physical cores, but the thread counts differ. The i5-6350HQ has 4 threads, meaning no Hyper-Threading. The Xeon E5620 has 8 threads, enabling twice as many concurrent threads. The cache layouts also differ substantially. The i5 has 64 KB of L1 per core and 256 KB of L2 per core, with 6 MB of shared L3. The Xeon has the same 64 KB L1 and 256 KB L2 per core, but doubles the shared L3 to 12 MB.

Memory support diverges as well. The i5 supports both DDR3 and DDR4 over a dual-channel bus, with a recorded memory bandwidth of 34.1 GB/s. The Xeon supports only DDR3 over a triple-channel bus, and the database does not list a bandwidth figure. The Xeon does support ECC memory, which the i5 does not. That makes the Xeon the only one of the two suitable for error-corrected memory configurations.

The i5 includes integrated graphics, specifically Intel Iris Pro 580. The Xeon has no integrated graphics at all, which is typical for a server part. PCIe support also differs: the i5 has Gen 3 with 16 lanes from the CPU, while the Xeon has Gen 2 only. The i5 uses an Intel BGA 1440 socket, while the Xeon uses Intel Socket 1366. The i5 was released in early 2016 with a launch MSRP of $306; the Xeon was released in March 2010 and has no recorded launch MSRP.

Head-to-Head Benchmarks

The Cinebench R15 multi-core test shows the i5-6350HQ scoring 364 against the Xeon's 301. That is a 20.9% advantage for the i5. In a rendering workload that can use multiple cores, the i5's newer architecture more than compensates for the Xeon's extra threads.

Moving to Cinebench R20, the pattern holds. In multi-core, the i5 scores 1518 versus 1255, a 21% lead. In single-core, the i5 scores 214 versus 177, also a 20.9% lead. The single-core result is particularly telling because the Xeon's base clock of 2.40 GHz is actually higher than the i5's 2.30 GHz base clock. The Xeon's boost clock of 2.67 GHz is also lower than the i5's 3.20 GHz boost, but the single-core gap is not just about clock speed. The Skylake core simply executes instructions far more efficiently per clock than the Westmere core.

Cinebench R23 results confirm the same story. Multi-core: i5 scores 3616, Xeon scores 2989, a 21% lead. Single-core: i5 scores 510, Xeon scores 422, a 20.9% lead. Every single recorded benchmark shows the i5 winning by approximately 21%. There is no test where the Xeon's 8 threads or 12 MB L3 cache provide any measurable benefit in Cinebench.

The consistency of the delta is noteworthy. Whether the load is single-threaded or multi-threaded, the i5 wins by essentially the same percentage. This suggests the advantage is not tied to a specific workload characteristic like memory bandwidth or thread scaling, but rather to the fundamental per-core performance delta between a 14 nm Skylake core and a 32 nm Westmere core.

FAQ

Q: Which CPU has more cores?

A: Both have 4 physical cores. The Xeon E5620 has 8 threads due to Hyper-Threading, while the i5-6350HQ has only 4 threads.

Q: Does the extra thread count on the Xeon help it win any benchmarks?

A: No. The Xeon loses all five recorded head-to-head Cinebench benchmarks. Despite having 8 threads versus 4, it trails by roughly 21% in every test, including multi-core workloads where extra threads should help most.

Q: Which CPU has better single-core performance?

A: The i5-6350HQ. In Cinebench R20 single-core, it scores 214 versus the Xeon's 177, a 20.9% lead. In R23 single-core, it scores 510 versus 422, also a 20.9% lead.

Q: Do these CPUs support ECC memory?

A: Only the Xeon E5620 supports ECC memory. The i5-6350HQ does not.

Q: Which CPU has integrated graphics?

A: Only the i5-6350HQ, which includes Intel Iris Pro 580. The Xeon E5620 has no integrated graphics.

Q: How do their average benchmark scores compare?

A: The i5-6350HQ has an average benchmark score of 1046, while the Xeon E5620 has 1029. The i5 sits at the 29th percentile of all CPUs, the Xeon at the 28th.

The Verdict

If you are choosing between these two based on the recorded data, the Intel Core i5-6350HQ is the clear pick for general compute performance. It wins every Cinebench test in the database, both single-core and multi-core, by a consistent margin of about 21%. Its newer 14 nm Skylake architecture delivers superior per-clock performance, and its high boost clock of 3.20 GHz gives it a substantial frequency advantage when workloads are light.

The Xeon E5620 is the better choice only if you specifically need ECC memory support or a triple-channel memory bus for a server or workstation build on Socket 1366. It also offers more threads, which could theoretically help in heavily parallel software that scales beyond 4 threads, but the database shows no such benefit in Cinebench. Its 12 MB L3 cache is larger, but that does not translate into any recorded benchmark win.

For a mobile or compact system, the i5-6350HQ also brings integrated Iris Pro 580 graphics, saving the need for a discrete GPU. The Xeon requires external graphics entirely. The i5 has a much lower TDP of 45 watts versus the Xeon's 80 watts, making it easier to cool in smaller chassis. The Xeon's triple-channel memory and ECC support are its only genuine advantages, and those matter only in specific server contexts.

The average benchmark scores place them only 17 points apart (1046 versus 1029), which suggests they occupy a similar performance tier in the overall database. But in direct head-to-head testing, the i5 dominates. Any user prioritizing raw Cinebench performance should take the i5-6350HQ without hesitation. The Xeon E5620 is a legacy server part that, while still functional, cannot match the architectural efficiency of the Skylake mobile chip.

Specification Differences

| Specification | Intel Core i5-6350HQ | Intel Xeon E5620 |

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

| Cores | 4 | 4 |

| Threads | 4 | 8 |

| Base Clock | 2.30 GHz | 2.40 GHz |

| Boost Clock | 3.20 GHz | 2.67 GHz |

| TDP | 45 W | 80 W |

| Socket | Intel BGA 1440 | Intel Socket 1366 |

| Architecture | Skylake | Westmere |

| Codename | Skylake-H | Westmere-EP |

| Process Node | 14 nm | 32 nm |

| Transistors | 2,300 million | 1,170 million |

| Die Size | 171 mm² | 239 mm² |

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

| L2 Cache | 256 KB (per core) | 256 KB (per core) |

| L3 Cache | 6 MB (shared) | 12 MB (shared) |

| Memory Support | DDR3, DDR4 | DDR3 |

| Memory Bus | Dual-channel | Triple-channel |

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

| ECC Memory | No | Yes |

| PCIe | Gen 3, 16 Lanes (CPU only) | Gen 2 |

| Integrated Graphics | Intel Iris Pro 580 | None |

| Market Segment | Mobile | Server/Workstation |

| Release Date | 2016-01-23 | 2010-03-15 |

| Launch MSRP | $306 | Not listed |

| Multiplier Unlocked | No | No |

| Part Number | SR2QZ | SLBV4 |