Intel Xeon E5-1660 v3 vs Intel Xeon E5-2676 v3 Comparison

Where Each One Wins

The benchmark record splits cleanly between these two Haswell-EP Xeon parts. The Intel Xeon E5-2676 v3 wins every recorded Cinebench test, both single-core and multi-core, across R15, R20, and R23. The Intel Xeon E5-1660 v3 does not win a single head-to-head benchmark in the database.

The E5-2676 v3 takes the multi-core workloads decisively. In Cinebench R15 multi-core, it scores 1102 against 1074 for the E5-1660 v3, a 2.6% advantage. The gap holds steady in R20 multi-core at 4592 versus 4475, again 2.6%. In R23 multi-core, the E5-2676 v3 reaches 10935 while the E5-1660 v3 manages 10656, maintaining the same 2.6% margin.

Single-core results follow the same pattern, though the margins are surprisingly consistent. The E5-2676 v3 leads in R15 single-core with 155 versus 151, a 2.6% delta. In R20 single-core, it scores 648 against 631, a 2.7% edge. R23 single-core shows 1543 versus 1504, back to 2.6%. This uniformity across all six tests suggests a fundamental clock-for-clock efficiency advantage for the 12-core part, not a workload-specific quirk.

The E5-1660 v3 does have one structural advantage: a higher base clock of 3.00 GHz versus 2.40 GHz, and a higher boost clock of 3.50 GHz versus 3.00 GHz. Yet the recorded benchmarks show that this frequency lead does not translate into a single win. The 12-core E5-2676 v3 compensates with two additional cores and 50% more shared L3 cache, 30 MB versus 20 MB.

For users running heavily threaded rendering or simulation workloads, the E5-2676 v3 is the clear choice from the data. The E5-1660 v3 offers no measurable advantage in any of the six recorded Cinebench tests, despite its higher clocks and higher TDP of 140 watts versus 120 watts for the E5-2676 v3.

Architecture Differences

Both processors share the same Haswell-EP architecture, built on Intel's 22 nm process node. The die size is identical at 356 mm², and both integrate 2,600 million transistors. The foundry is Intel for both parts. The socket is the same Intel Socket 2011-3, and both support PCI Express Gen 3 with 40 CPU lanes.

The core counts differ substantially. The E5-2676 v3 provides 12 cores and 24 threads, while the E5-1660 v3 provides 8 cores and 16 threads. This 50% core advantage for the E5-2676 v3 is the primary architectural differentiator. The L3 cache scales with the core count: 30 MB shared on the E5-2676 v3 versus 20 MB shared on the E5-1660 v3. Per-core L1 and L2 caches are identical at 64 KB and 256 KB respectively.

Clock speeds favor the smaller part. The E5-1660 v3 runs at 3.00 GHz base and 3.50 GHz boost, while the E5-2676 v3 runs at 2.40 GHz base and 3.00 GHz boost. Despite this 0.60 GHz deficit in both base and boost clocks, the E5-2676 v3 still wins every benchmark, indicating that the additional cores and cache more than offset the frequency disadvantage.

Memory support shows a minor difference. Both parts use quad-channel memory with 68.3 GB/s bandwidth and ECC support. The E5-2676 v3 supports both DDR3 and DDR4, while the E5-1660 v3 supports DDR4 only. This gives the E5-2676 v3 broader platform compatibility for systems with older DDR3 memory.

Power characteristics differ meaningfully. The E5-1660 v3 has a TDP of 140 watts despite having fewer cores, reflecting its higher operating frequencies. The E5-2676 v3 draws 120 watts, 20 watts less, while delivering more benchmark performance. The unlocked multiplier on the E5-1660 v3 is a notable feature, allowing overclocking, whereas the E5-2676 v3 has a locked multiplier. The part numbers distinguish the two: SR20N for the E5-1660 v3 and SR1Y5 for the E5-2676 v3.

Both parts are marked as end-of-life production status. The E5-1660 v3 has a release date of September 7, 2014. The E5-2676 v3 does not have a recorded release date in the database. Both target the server and workstation market segment, and neither includes integrated graphics.

Head-to-Head Benchmarks

The head-to-head results are uniform and unambiguous. Across all six Cinebench tests, the E5-2676 v3 wins with deltas ranging from 2.6% to 2.7%. The E5-1660 v3 has zero wins.

In Cinebench R15 multi-core, the E5-2676 v3 scores 1102 against 1074 for the E5-1660 v3. The 2.6% delta is modest but consistent. Single-core R15 shows 155 versus 151, also 2.6%. This is notable because the E5-1660 v3 has a substantially higher boost clock of 3.50 GHz versus 3.00 GHz, yet still loses the single-threaded test.

Cinebench R20 multi-core widens the absolute gap: 4592 versus 4475, a 117-point difference at 2.6%. Single-core R20 shows 648 versus 631, a 17-point difference at 2.7%, the largest percentage delta recorded. This single-core margin is surprising given the E5-1660 v3's clock advantage.

Cinebench R23 multi-core shows 10935 versus 10656, a 279-point gap at 2.6%. Single-core R23 records 1543 versus 1504, a 39-point gap at 2.6%. The consistency of the 2.6% delta across R15, R20, and R23 in both single and multi-core tests indicates a stable per-clock efficiency advantage for the E5-2676 v3 architecture implementation.

The average benchmark score reinforces this picture. The E5-2676 v3 has an average benchmark score of 3163, placing it at the 53rd percentile of all CPUs. The E5-1660 v3 averages 3082, at the 52nd percentile. The 81-point average difference corresponds to a 2.6% advantage for the E5-2676 v3, in line with the individual test deltas.

In the nearest rivals comparison, the E5-2676 v3 sits alongside the AMD Ryzen 5 5560U at 3162 (delta 0%), the AMD Ryzen 3 5300GE at 3160 (delta 0.1%), the AMD Ryzen 7 PRO 5850U at 3159 (delta 0.1%), and the Intel Xeon E5-2650 v4 at 3154 (delta 0.3%). The E5-1660 v3 sits near the Intel Core i7-6850K at 3079 (delta 0.1%), the Intel Core i7-8700B at 3086 (delta -0.1%), the Intel Xeon W-2133 at 3087 (delta -0.2%), and the AMD Ryzen 7 1800X at 3068 (delta 0.4%). Both processors are clustered tightly with their direct competitors, with deltas under 0.5% in all cases.

FAQ

Q: Which processor has more cores?

A: The Intel Xeon E5-2676 v3 has 12 cores and 24 threads, while the Intel Xeon E5-1660 v3 has 8 cores and 16 threads.

Q: Does the higher clock speed of the E5-1660 v3 help it win any benchmarks?

A: No. Despite a base clock of 3.00 GHz and boost clock of 3.50 GHz, the E5-1660 v3 loses all six recorded Cinebench tests to the E5-2676 v3, which runs at 2.40 GHz base and 3.00 GHz boost.

Q: What is the largest percentage margin in the head-to-head results?

A: The largest margin is 2.7% in Cinebench R20 single-core, where the E5-2676 v3 scores 648 versus 631 for the E5-1660 v3.

Q: Which processor has more L3 cache?

A: The E5-2676 v3 has 30 MB of shared L3 cache, compared to 20 MB on the E5-1660 v3. Both have 64 KB L1 and 256 KB L2 per core.

Q: Do both processors support ECC memory?

A: Yes, both support ECC memory with quad-channel memory buses and 68.3 GB/s bandwidth. The E5-2676 v3 supports both DDR3 and DDR4, while the E5-1660 v3 supports DDR4 only.

Q: Which processor has a higher TDP?

A: The E5-1660 v3 has a TDP of 140 watts, higher than the 120 watts of the E5-2676 v3, despite the E5-2676 v3 having more cores and winning every benchmark.

The Verdict

The data directs server and workstation buyers toward the Intel Xeon E5-2676 v3 for almost any workload represented by Cinebench. It wins all six recorded tests with a consistent 2.6% to 2.7% margin over the E5-1660 v3, while drawing 20 fewer watts of power. The 12-core, 24-thread configuration with 30 MB of L3 cache provides a structural advantage that the E5-1660 v3's higher clocks cannot overcome.

The E5-1660 v3 retains one practical appeal: its unlocked multiplier allows overclocking, which could potentially close the gap in single-threaded tasks if the user is willing to push the 140-watt part beyond its stock 3.50 GHz boost. The database does not record any overclocked results, so this remains speculative. At stock settings, the E5-1660 v3 offers no measured performance benefit.

For users with existing DDR3 memory infrastructure, the E5-2676 v3 is the only choice of the two, as it supports both DDR3 and DDR4 while the E5-1660 v3 supports DDR4 only. Both parts are end-of-life, so platform longevity is not a differentiator.

The percentile rankings place both processors close together, with the E5-2676 v3 at the 53rd percentile and the E5-1660 v3 at the 52nd percentile. The average benchmark scores of 3163 and 3082 respectively confirm that the E5-2676 v3 is the stronger part, but the margin is narrow enough that the E5-1660 v3 remains a viable alternative in scenarios where its higher clocks and unlocked multiplier might be leveraged.

In summary, the E5-2676 v3 is the better stock processor by every recorded metric, with lower power consumption and broader memory compatibility. The E5-1660 v3 is the choice only for users who specifically want an unlocked multiplier and are willing to accept lower multi-threaded performance and higher power draw in exchange.