Intel Xeon D-1567 vs Intel Xeon E5-2628L v3 Comparison

Head-to-Head Benchmarks

The data is unambiguous: the Intel Xeon D-1567 wins every single benchmark in the head-to-head comparison, taking all 6 recorded tests. The margins are consistent but narrow, ranging from 1.6% to 2.3% across the Cinebench suite. This is not a rout in the traditional sense, but it is a clean sweep. The largest advantage comes in Cinebench R15 single-core, where the D-1567 scores 136 against 133 for the E5-2628L v3, a delta of 2.3%. The smallest gap is in Cinebench R20 single-core, at 1.6%, with scores of 567 and 558 respectively.

In multi-threaded workloads, the pattern holds. Cinebench R15 multi-core shows the D-1567 at 965 versus 949 for the E5-2628L v3, a 1.7% lead. Cinebench R20 multi-core repeats the exact same delta of 1.7%, with 4022 against 3955. The most demanding test, Cinebench R23 multi-core, also sees a 1.7% advantage, with the D-1567 scoring 9578 and the E5-2628L v3 scoring 9417. The single-core results follow suit in R23, where the D-1567 posts 1352 versus 1329, again a 1.7% delta.

What these percentages mean in practice is that the D-1567 is consistently ahead, but by a margin that would be noticeable in sustained rendering or repeated multi-core tasks, not by a transformative amount. The average benchmark score in the database reinforces this: the D-1567 has an average of 2770, while the E5-2628L v3 sits at 2724. That is a difference of roughly 1.7%, matching the head-to-head deltas. Both processors land within 1 percentile point of each other globally, with the D-1567 at the 51st percentile and the E5-2628L v3 at the 50th, meaning they occupy nearly identical tiers of overall performance in the database.

The nearest rivals for each chip also tell a story of parity. The D-1567 is essentially tied with the AMD Ryzen 3 PRO 4350GE (delta 0.1%), the AMD Ryzen 3 PRO 4350G (delta -0.2%), the Intel Core i7-5930K (delta -0.2%), and the AMD Ryzen 3 5400U (delta 0.2%). The E5-2628L v3 is likewise within a hair of the AMD Ryzen 7 4700U (delta 0.1%), the Intel Core i7-11390H (delta 0.3%), the Intel Xeon D-1577 (delta 0.3%), and the AMD Ryzen 5 PRO 5650U (delta 0.4%). Neither chip breaks away from its peer group; they are both mid-pack performers in the 50th percentile range.

Where Each One Wins

Given that the D-1567 wins all six head-to-head tests, the use-case split is straightforward for benchmark performance. The D-1567 is the choice for single-threaded responsiveness, as it leads in Cinebench R15 single-core by 2.3%, and in R20 and R23 single-core by 1.6% and 1.7% respectively. For applications that rely on one or two cores, such as light scripting, some database queries, or older software that does not scale, the D-1567 holds a slight edge.

For multi-threaded workloads, the D-1567 also wins, but the margin narrows to a consistent 1.7% across R15, R20, and R23 multi-core tests. This covers rendering, video encoding, and virtual machine hosting. The 12 cores and 24 threads of the D-1567, against 10 cores and 20 threads on the E5-2628L v3, provide the structural advantage that shows up in these scores. The E5-2628L v3 does not win any category in the recorded data. Its only relative strengths are outside the benchmark suite: it carries a larger shared L3 cache of 25 MB versus the per-core L3 of the D-1567, and it supports quad-channel memory with a stated bandwidth of 59.7 GB/s, whereas the D-1567 is dual-channel and has no recorded bandwidth figure. But in actual Cinebench performance, the E5-2628L v3 is consistently second.

The database also shows that the E5-2628L v3 is end-of-life in production status, while the D-1567 is still active. For new deployments, the D-1567 is the only one of the two that the database lists as currently available. That alone is a deciding factor for procurement, regardless of the small benchmark deltas.

The Verdict

The verdict from the recorded data is clear: the Intel Xeon D-1567 is the superior processor in every measured benchmark. It wins all six Cinebench tests, has a higher average benchmark score (2770 versus 2724), and sits one percentile point higher in the global ranking. The margins are modest, between 1.6% and 2.3%, so the performance difference is not dramatic, but it is consistent and never favors the E5-2628L v3.

The D-1567 also offers more cores (12 versus 10) and more threads (24 versus 20), which explains its multi-core advantage. It is built on a 14 nm process versus the 22 nm of the E5-2628L v3, and it has a lower TDP of 65 watts against 75 watts. It is also the only one of the two still in active production. The E5-2628L v3, with its end-of-life status, is the older design.

Who should pick which? For any new server or workstation build, the D-1567 is the only sensible choice from these two, based on the benchmarks and active status. The E5-2628L v3 might be considered for a legacy system replacement, where its quad-channel memory bandwidth of 59.7 GB/s and larger shared L3 cache of 25 MB could matter in memory-bound tasks that are not captured by Cinebench. But the database does not record any benchmark where the E5-2628L v3 wins, so such a choice would be speculative. The data says D-1567, and the margin, while small, is a clean sweep.

FAQ

Q: Which CPU has the higher single-core performance?

A: The Intel Xeon D-1567 wins all three single-core tests: Cinebench R15 (136 vs 133), R20 (567 vs 558), and R23 (1352 vs 1329). The largest margin is 2.3% in R15.

Q: What is the multi-core performance difference?

A: The D-1567 leads in Cinebench R15 multi-core (965 vs 949), R20 multi-core (4022 vs 3955), and R23 multi-core (9578 vs 9417). Each of these is a 1.7% advantage.

Q: How do these CPUs compare to their nearest rivals?

A: The D-1567 is within 0.2% of the AMD Ryzen 3 PRO 4350G and the Intel Core i7-5930K, and within 0.1% of the AMD Ryzen 3 PRO 4350GE. The E5-2628L v3 is within 0.4% of the AMD Ryzen 5 PRO 5650U and within 0.1% of the AMD Ryzen 7 4700U.

Q: Which CPU has more cores and threads?

A: The Intel Xeon D-1567 has 12 cores and 24 threads. The Intel Xeon E5-2628L v3 has 10 cores and 20 threads.

Q: What is the production status of each chip?

A: The Intel Xeon D-1567 is listed as Active. The Intel Xeon E5-2628L v3 is listed as End-of-life.

Q: Which CPU has a higher average benchmark score?

A: The D-1567 has an average benchmark score of 2770, while the E5-2628L v3 has 2724. The D-1567 is at the 51st percentile of all CPUs, the E5-2628L v3 at the 50th.

Architecture Differences

The two processors come from different architectural generations. The Intel Xeon D-1567 is based on Broadwell, specifically the Broadwell-DE variant, and is fabricated on a 14 nm process at Intel. It contains 3,200 million transistors on a die size of 246 mm². The Intel Xeon E5-2628L v3 is based on Haswell, specifically Haswell-EP, on a 22 nm process. It has 2,600 million transistors on a larger die of 356 mm². The newer process node of the D-1567 allows for more transistors in a smaller area, which is part of why it achieves better performance with a lower TDP.

The cache layouts differ significantly. The D-1567 has 64 KB of L1 per core, 256 KB of L2 per core, and 1.5 MB of L3 per core. The E5-2628L v3 has the same per-core L1 and L2 sizes, but its L3 is 25 MB shared across all cores. This is a fundamental difference: per-core L3 on the D-1567 versus shared L3 on the E5-2628L v3. The total L3 for the D-1567 is not listed as a single figure, but with 12 cores at 1.5 MB per core, the architecture favors per-core locality, while the E5-2628L v3 favors a large shared pool.

Memory support also diverges. The D-1567 supports both DDR3 and DDR4, but only in a dual-channel configuration. The E5-2628L v3 supports DDR4 only, but in a quad-channel setup, with a recorded memory bandwidth of 59.7 GB/s. The D-1567 has no recorded memory bandwidth figure. The E5-2628L v3 also has more PCIe lanes: Gen 3 with 40 lanes, versus Gen 3 with 24 lanes on the D-1567. Both support ECC memory.

The sockets are incompatible. The D-1567 uses Intel BGA 1667, which is a soldered, embedded-style socket. The E5-2628L v3 uses Intel Socket 2011-3, a standard LGA socket for server motherboards. This means platform choice is binary: you cannot swap these two chips on the same board.

Specification Differences

The key specification differences between the two are as follows. The D-1567 has 12 cores and 24 threads; the E5-2628L v3 has 10 cores and 20 threads. The D-1567 has a base clock of 2.10 GHz and a boost clock of 2.70 GHz. The E5-2628L v3 has a base clock listed as 2000.00 MHz (2.00 GHz) and a boost clock of 2.50 GHz. The D-1567 is clocked higher in both states.

The D-1567 has a TDP of 65 watts, while the E5-2628L v3 has a TDP of 75 watts. The process nodes differ: 14 nm for the D-1567, 22 nm for the E5-2628L v3. The transistor counts are 3,200 million versus 2,600 million, and the die sizes are 246 mm² versus 356 mm².

The cache difference is notable: the D-1567 uses 1.5 MB L3 per core, while the E5-2628L v3 uses 25 MB shared L3. Memory support is DDR3 and DDR4 dual-channel for the D-1567, versus DDR4 quad-channel for the E5-2628L v3, which also has a recorded bandwidth of 59.7 GB/s. The D-1567 has 24 PCIe lanes (Gen 3), the E5-2628L v3 has 40 PCIe lanes (Gen 3). The sockets are BGA 1667 versus Socket 2011-3.

The production status is Active for the D-1567 and End-of-life for the E5-2628L v3. The release dates differ: the D-1567 launched in 2016, the E5-2628L v3 in 2014. The part numbers are SR2M3 for the D-1567 and SR1XZ for the E5-2628L v3. The launch MSRP for the D-1567 is 1199, and for the E5-2628L v3 is $1364. Both are multiplier-unlocked false, both lack integrated graphics, and both are in the server/workstation market segment.