Intel Xeon E5-1630 v4 vs Intel Xeon E5-2620 v3 Comparison

FAQ

Q: Which processor has more cores and threads?

A: The Intel Xeon E5-2620 v3 has 6 cores and 12 threads, while the Intel Xeon E5-1630 v4 has 4 cores and 8 threads.

Q: How do the two compare in multi-core performance?

A: The E5-2620 v3 wins all multi-core benchmarks in the head-to-head data. In Cinebench R23 multi-core, it scores 6595 versus 6464 for the E5-1630 v4, a 2% lead. The same 2% margin appears in R15 multi-core (664 vs 651) and R20 multi-core (2769 vs 2714).

Q: Which chip has higher clock speeds?

A: The E5-1630 v4 has a base clock of 3.70 GHz and a boost clock of 4.00 GHz. The E5-2620 v3 runs at 2.40 GHz base and 3.20 GHz boost. Despite this clock disadvantage, the E5-2620 v3 still wins every single-core benchmark in the head-to-head data.

Q: What are the process nodes and architectures?

A: The E5-1630 v4 is built on Intel's 14 nm Broadwell-EP process, while the E5-2620 v3 uses the older 22 nm Haswell-EP process. The E5-1630 v4 packs 3,400 million transistors on a 246 mm² die; the E5-2620 v3 has 2,600 million transistors on a 356 mm² die.

Q: Do they support the same memory configuration?

A: Both support DDR4 memory with a quad-channel memory bus and ECC memory. The E5-1630 v4 has a higher memory bandwidth at 76.8 GB/s, compared to 59.7 GB/s for the E5-2620 v3.

Q: What is the average benchmark score for each?

A: The E5-2620 v3 has an average benchmark score of 1907, placing it in the 43rd percentile of all CPUs. The E5-1630 v4 averages 1869, sitting in the 42nd percentile.

Architecture Differences

The two Xeon processors come from different Intel microarchitectures. The E5-1630 v4 is a Broadwell-EP part on a 14 nm process node, while the E5-2620 v3 is Haswell-EP on a 22 nm node. That process shrink allows the E5-1630 v4 to fit 3,400 million transistors into a 246 mm² die. The E5-2620 v3, despite having more cores, uses fewer transistors at 2,600 million and requires a larger 356 mm² die.

Core counts also differ substantially. The E5-1630 v4 offers 4 cores with 8 threads, whereas the E5-2620 v3 provides 6 cores and 12 threads. This gives the E5-2620 v3 a two-core, four-thread advantage in threaded workloads.

Cache configurations follow the core counts. Both chips have 64 KB of L1 cache per core and 256 KB of L2 cache per core. The L3 cache differs: the E5-1630 v4 has 10 MB shared, while the E5-2620 v3 has 15 MB shared. The larger shared cache on the E5-2620 v3 aligns with its higher core count.

Clock speed is where the E5-1630 v4 leads. Its base clock of 3.70 GHz and boost clock of 4.00 GHz are well above the E5-2620 v3's 2.40 GHz base and 3.20 GHz boost. This is a significant frequency gap, though the benchmark data shows it does not translate into a performance win.

Memory bandwidth also differs. The E5-1630 v4 supports 76.8 GB/s of memory bandwidth versus 59.7 GB/s for the E5-2620 v3. Both use DDR4 over a quad-channel memory bus, and both support ECC memory. PCIe support is identical: Gen 3 with 40 lanes (CPU only).

Power draw is notably different. The E5-1630 v4 has a TDP of 140 watts, while the E5-2620 v3 is rated at 85 watts. The E5-2620 v3 delivers more cores and better benchmark scores while consuming less power, though the E5-1630 v4's higher clocks and smaller process node explain its higher rating.

Both processors are end-of-life parts using the Intel Socket 2011-3. Neither has integrated graphics, and neither has an unlocked multiplier. The E5-1630 v4 launched in 2016, while the E5-2620 v3 arrived in 2014.

Head-to-Head Benchmarks

The head-to-head data covers six Cinebench tests, and the E5-2620 v3 wins all of them. The margins are consistent and small, ranging from 2% to 2.2%.

In Cinebench R15 multi-core, the E5-2620 v3 scores 664 against 651 for the E5-1630 v4. That is a 2% lead. The single-core R15 test shows a 2.2% gap: 93 to 91. This is the largest margin in any test.

Cinebench R20 repeats the pattern. Multi-core results are 2769 for the E5-2620 v3 and 2714 for the E5-1630 v4, a 2% difference. Single-core results are 391 versus 383, again a 2% gap.

Cinebench R23 shows the same story. Multi-core scores are 6595 for the E5-2620 v3 and 6464 for the E5-1630 v4. Single-core scores are 931 versus 912. Both differences sit at 2%.

The most notable aspect of these results is that the E5-2620 v3 wins single-core tests despite having a substantially lower boost clock. The E5-1630 v4 boosts to 4.00 GHz, yet scores 912 in R23 single-core, while the E5-2620 v3 boosts to only 3.20 GHz and scores 931. This suggests the newer architecture's clock advantage is not producing the expected single-thread performance edge in these measurements.

The average benchmark scores reinforce the trend. The E5-2620 v3 averages 1907 across all recorded benchmarks, placing it in the 43rd percentile of all CPUs. The E5-1630 v4 averages 1869, in the 42nd percentile. The closest rival to the E5-2620 v3 is the Intel Xeon E3-1285 v4 with an average score of 1908 and a 0% delta. The E5-1630 v4's nearest rival is the Intel Core i3-9350K at 1867, just 0.1% behind.

In short, the E5-2620 v3 wins every benchmark in the head-to-head comparison. The margins are modest, but they are consistent across all three Cinebench versions and both multi-core and single-core workloads.

The Verdict

The data points clearly to the Intel Xeon E5-2620 v3 as the stronger performer. It wins all six head-to-head benchmarks, holds a higher average benchmark score (1907 vs 1869), and sits one percentile point higher in the overall CPU rankings. It does this with 6 cores and 12 threads, 15 MB of L3 cache, and a lower TDP of 85 watts.

The E5-1630 v4 does have advantages in the specification sheet. Its 3.70 GHz base clock and 4.00 GHz boost clock are far above the E5-2620 v3's frequencies. Its memory bandwidth is higher at 76.8 GB/s versus 59.7 GB/s. Its 14 nm process is newer than the 22 nm node used by the E5-2620 v3. None of these advantages translate into a benchmark win, however.

The E5-2620 v3's wins are not large. At 2% in most tests, the practical difference between the two in everyday workloads will be small. But a consistent win across every recorded benchmark, combined with a higher average score and better percentile placement, makes the E5-2620 v3 the clear choice from a performance standpoint.

The E5-1630 v4 is not without a role. Its higher clocks and smaller process node may appeal in situations where the benchmark data does not capture the full picture, such as lightly threaded legacy applications or specific latency-sensitive tasks. But based strictly on the recorded measurements, the E5-2620 v3 is the better processor.

Specification Differences

The two processors differ in several key specifications. The E5-1630 v4 has 4 cores and 8 threads; the E5-2620 v3 has 6 cores and 12 threads. Base clocks are 3.70 GHz for the E5-1630 v4 and 2.40 GHz for the E5-2620 v3. Boost clocks are 4.00 GHz and 3.20 GHz, respectively.

TDP differs significantly: 140 watts for the E5-1630 v4 versus 85 watts for the E5-2620 v3. The architecture and process node also differ: Broadwell-EP on 14 nm for the E5-1630 v4, Haswell-EP on 22 nm for the E5-2620 v3. Transistor counts are 3,400 million versus 2,600 million, and die sizes are 246 mm² versus 356 mm².

L3 cache is 10 MB shared on the E5-1630 v4 and 15 MB shared on the E5-2620 v3. Memory bandwidth is 76.8 GB/s for the E5-1630 v4 and 59.7 GB/s for the E5-2620 v3. Release dates are 2016 for the E5-1630 v4 and 2014 for the E5-2620 v3. The E5-1630 v4 has a launch MSRP of $406; the E5-2620 v3 has a launch MSRP of $417.

Identical specifications include the socket (Intel Socket 2011-3), memory type (DDR4), memory bus (quad-channel), ECC support (yes), PCIe (Gen 3, 40 lanes CPU only), integrated graphics (none), multiplier lock (locked), L1 cache (64 KB per core), and L2 cache (256 KB per core).

Where Each One Wins

The Intel Xeon E5-2620 v3 wins every head-to-head benchmark in the database. Its six-core, twelve-thread configuration gives it an advantage in multi-threaded workloads like rendering and content creation, where Cinebench scores are the relevant metric. It also wins the single-core tests despite lower clock speeds, which is a more surprising result. Its lower TDP of 85 watts makes it the more efficient choice for sustained workloads, and its larger 15 MB L3 cache may help in data-heavy tasks.

The Intel Xeon E5-1630 v4 does not win any recorded benchmark, but its specification sheet points to specific strengths. The 3.70 GHz base clock and 4.00 GHz boost clock are the highest frequencies in this comparison, which could benefit workloads that are not captured in the Cinebench suite. Its 76.8 GB/s memory bandwidth is substantially higher than the E5-2620 v3's 59.7 GB/s, a potential advantage for memory-bound applications. The 14 nm process node and smaller die size suggest a more modern design overall.

For a builder choosing between these two end-of-life Xeon parts, the recorded data favors the E5-2620 v3 on performance and efficiency. The E5-1630 v4's case rests on clock speed, memory bandwidth, and process technology, but none of those advantages appear as benchmark wins in the database. The E5-2620 v3 is the safer pick for anyone prioritizing measured performance.