Intel Xeon E3-1285 v6 vs Intel Xeon E5-2629 v3 Comparison

Where Each One Wins

The recorded data splits cleanly along workload type. The Intel Xeon E3-1285 v6 wins all six head-to-head benchmark comparisons, but the size of each win tells a more interesting story about what each processor is optimized for.

In single-threaded workloads, the E3-1285 v6 shows a modest but consistent edge. Its Cinebench R15 single-core score of 115 beats the E5-2629 v3's 112 by 2.7%. The R20 single-core test shows a 3% advantage, with scores of 481 versus 467. The R23 single-core result repeats the pattern: 1146 versus 1114, a 2.9% lead. These are narrow margins, but they appear in every generation of the Cinebench test suite, which suggests a genuine architectural advantage in lightly threaded tasks.

The multi-core picture is similar in percentage terms but different in character. The E3-1285 v6 scores 818 in Cinebench R15 multi-core versus 795 for the E5-2629 v3, a 2.9% win. R20 multi-core shows 3409 versus 3315, a 2.8% margin. R23 multi-core records 8119 versus 7893, again a 2.9% difference. The fact that a 4-core, 8-thread processor beats an 8-core, 16-thread processor in multi-core tests is notable. The E5-2629 v3 has double the core count, yet it cannot translate that into a win, even in fully parallel workloads.

The average benchmark scores reinforce this. The E3-1285 v6 averages 2348 across its benchmark suite, while the E5-2629 v3 averages 2283. The E3-1285 v6 sits at the 48th percentile among all CPUs in the database, and the E5-2629 v3 sits at the 47th percentile. They are nearly peers in overall standing, separated by less than one percentile point.

Where does each one genuinely win? The E3-1285 v6 wins on raw speed in every measured test, but the E5-2629 v3 wins on structural advantages: it has more cores, more threads, a larger shared L3 cache, and a quad-channel memory bus. These are not benchmark victories, but they matter for specific deployment scenarios. The E5-2629 v3 also carries an end-of-life production status, while the E3-1285 v6 does not have such a status recorded.

The Verdict

The benchmark data points to a straightforward conclusion: the Intel Xeon E3-1285 v6 is the faster processor in every Cinebench test recorded, and it should be the choice for anyone prioritizing per-thread performance or lightly threaded workloads. Its wins range from 2.7% to 3% across six tests, which is a narrow but consistent margin.

However, the E5-2629 v3 has structural merits that the benchmarks do not capture. It offers 8 cores and 16 threads compared to the E3-1285 v6's 4 cores and 8 threads. Its 20 MB shared L3 cache dwarfs the 8 MB on the E3-1285 v6. Its quad-channel memory bus, paired with a recorded 59.7 GB/s memory bandwidth, supports memory-heavy workloads far better than the E3-1285 v6's dual-channel configuration. For server workloads that scale with core count, memory capacity, and memory bandwidth, the E5-2629 v3 remains a viable option despite losing every benchmark in this comparison.

The production status also matters. The E5-2629 v3 is marked as end-of-life, which suggests limited availability for new deployments. The E3-1285 v6 has no such status recorded, implying it may still be in production or at least not formally discontinued.

Who should pick which? A user running single-threaded database queries, development builds, or lightly threaded applications should choose the E3-1285 v6. Its higher clock speeds, 4.10 GHz base and 4.50 GHz boost, deliver better per-core performance. A user running heavily parallel workloads that need more threads, larger cache, and higher memory bandwidth should consider the E5-2629 v3, provided they can source the part despite its end-of-life status. The data does not show the E5-2629 v3 winning any benchmark, so the choice for it rests entirely on architectural features rather than measured performance.

Head-to-Head Benchmarks

The six recorded comparisons span three versions of the Cinebench suite, each with a single-core and multi-core variant. The E3-1285 v6 wins all six, with deltas ranging from 2.7% to 3%.

The largest single win is in Cinebench R20 single-core, where the E3-1285 v6 scores 481 against the E5-2629 v3's 467, a 3% advantage. This is the only test where the margin reaches the 3% threshold. The R15 single-core test shows the smallest win: 115 versus 112, a 2.7% margin. The R23 single-core test records 1146 versus 1114, a 2.9% gap.

Multi-core tests are uniform. R15 multi-core: 818 versus 795, 2.9%. R20 multi-core: 3409 versus 3315, 2.8%. R23 multi-core: 8119 versus 7893, 2.9%. The consistency is striking. The E3-1285 v6 maintains nearly the same percentage advantage whether the workload uses one thread or all available threads.

This consistency suggests the E3-1285 v6's advantage comes from its higher clock speeds and newer architecture, not from any workload-specific optimization. The E5-2629 v3's 8 cores cannot overcome the clock speed deficit. Its base clock is 2.40 GHz and boost clock is 3.20 GHz, compared to 4.10 GHz and 4.50 GHz on the E3-1285 v6. The E3-1285 v6 also uses a 14 nm process node, while the E5-2629 v3 uses 22 nm. The newer process node and higher clocks appear to deliver a per-core performance advantage that the E5-2629 v3's additional cores cannot offset.

FAQ

Q: Which processor has a higher single-core performance?

A: The Intel Xeon E3-1285 v6 wins all three single-core Cinebench tests. It scores 115 in R15, 481 in R20, and 1146 in R23, compared to 112, 467, and 1114 for the E5-2629 v3.

Q: Does the E5-2629 v3 win any multi-core test?

A: No. The E5-2629 v3 loses all three multi-core tests. The E3-1285 v6 scores 818, 3409, and 8119 in R15, R20, and R23 multi-core, while the E5-2629 v3 scores 795, 3315, and 7893.

Q: How much larger is the L3 cache on the E5-2629 v3?

A: The E5-2629 v3 has 20 MB of shared L3 cache, while the E3-1285 v6 has 8 MB. The E5-2629 v3's cache is 12 MB larger, which is 2.5 times the capacity.

Q: What memory configurations do these processors support?

A: The E3-1285 v6 supports DDR4 memory with a dual-channel bus. The E5-2629 v3 supports both DDR3 and DDR4 memory with a quad-channel bus, and its memory bandwidth is recorded as 59.7 GB/s.

Q: Do both processors support ECC memory?

A: Yes, both the E3-1285 v6 and the E5-2629 v3 support ECC memory.

Q: Which processor has more PCIe lanes?

A: The E5-2629 v3 provides 40 PCIe Gen 3 lanes from the CPU, while the E3-1285 v6 provides 16 PCIe Gen 3 lanes. The E5-2629 v3 offers 24 additional lanes.

Architecture Differences

The two processors come from different architectural generations. The E3-1285 v6 uses the Kaby Lake architecture, specifically Kaby Lake-DT, built on a 14 nm process node at Intel's foundry. It integrates 1,400 million transistors on a 160 mm² die. The E5-2629 v3 uses the Haswell architecture, specifically Haswell-EP, built on a 22 nm process node. It integrates 2,600 million transistors on a 356 mm² die.

The cache hierarchy differs substantially. Both processors have 64 KB of L1 cache per core and 256 KB of L2 cache per core. The L3 cache diverges: the E3-1285 v6 has 8 MB shared, while the E5-2629 v3 has 20 MB shared. This makes sense given the core counts: the E3-1285 v6 has 4 cores, and the E5-2629 v3 has 8 cores.

The E3-1285 v6 includes integrated graphics in the form of HD Graphics P630. The E5-2629 v3 has no integrated graphics. This is a meaningful difference for systems that rely on a discrete GPU versus those that can use the built-in display output.

The process node difference is significant. The 14 nm process on the E3-1285 v6 is two generations ahead of the 22 nm process on the E5-2629 v3 in terms of lithography. This partially explains the clock speed advantage: 4.10 GHz base and 4.50 GHz boost on the E3-1285 v6 versus 2.40 GHz base and 3.20 GHz boost on the E5-2629 v3.

Specification Differences

The two processors differ in nearly every major specification category.

Core and thread counts: the E3-1285 v6 has 4 cores and 8 threads, while the E5-2629 v3 has 8 cores and 16 threads. The E5-2629 v3 doubles the core and thread count.

Clock speeds: the E3-1285 v6 runs at 4.10 GHz base and 4.50 GHz boost. The E5-2629 v3 runs at 2.40 GHz base and 3.20 GHz boost. The E3-1285 v6 has a 1.70 GHz higher base clock and a 1.30 GHz higher boost clock.

Thermal design power: the E3-1285 v6 has a TDP of 79 watts, while the E5-2629 v3 has a TDP of 85 watts. The difference is 6 watts.

Sockets: the E3-1285 v6 uses Intel Socket 1151, while the E5-2629 v3 uses Intel Socket 2011-3. These sockets are not compatible with each other.

Memory support: the E3-1285 v6 supports DDR4 with a dual-channel bus. The E5-2629 v3 supports both DDR3 and DDR4 with a quad-channel bus. The E5-2629 v3 has a recorded memory bandwidth of 59.7 GB/s.

PCIe lanes: the E3-1285 v6 provides 16 Gen 3 lanes from the CPU. The E5-2629 v3 provides 40 Gen 3 lanes from the CPU.

Integrated graphics: the E3-1285 v6 includes HD Graphics P630. The E5-2629 v3 has no integrated graphics.

Process node and die: the E3-1285 v6 uses a 14 nm process with 1,400 million transistors on a 160 mm² die. The E5-2629 v3 uses a 22 nm process with 2,600 million transistors on a 356 mm² die.

Release dates: the E3-1285 v6 was released on July 31, 2017. The E5-2629 v3 was released on September 7, 2014. The E3-1285 v6 is nearly three years newer.

Production status: the E5-2629 v3 is marked as end-of-life, while the E3-1285 v6 has no production status recorded.