Intel Core i7-8705G vs Intel Xeon E3-1585 v5 Comparison

The Intel Xeon E3-1585 v5 and Intel Core i7-8705G are both 14 nm, 4-core/8-thread processors with 8 MB of shared L3 cache, but they target fundamentally different markets. The Xeon is a server/workstation part on the Skylake-H architecture, while the i7 is a mobile chip on Kaby Lake-G. Benchmark data shows a clear performance hierarchy: the Xeon E3-1585 v5 wins all six head-to-head Cinebench comparisons, with margins ranging from 10.4% to 11%. However, the i7-8705G counters with a higher boost clock, broader memory bandwidth, and a newer architecture, making the choice less about raw compute and more about platform fit.

Where Each One Wins

The Intel Xeon E3-1585 v5 is the undisputed winner in every CPU-bound benchmark recorded. Across all six Cinebench tests—R15, R20, and R23, both single-core and multi-core—the Xeon takes the top spot. Its most significant victory is in Cinebench R15 multicore, where it scores 720 against the i7-8705G’s 651, a 10.6% lead. The single-core R15 result is even more pronounced in relative terms: 101 versus 91, an 11% advantage. This pattern holds in newer workloads; in Cinebench R23 multicore, the Xeon posts 7146 versus 6464, a 10.6% gap, and in single-core R23 it leads 1008 to 912, a 10.5% margin. The data shows a consistent, roughly 10-11% performance edge for the Xeon across all tested scenarios.

The Core i7-8705G, by contrast, does not win a single benchmark in this comparison. Its wins are not in raw CPU throughput but in platform features. It supports DDR4 memory with a 38.4 GB/s memory bandwidth, which is 12.6% higher than the Xeon’s 34.1 GB/s. The i7 also has a higher boost clock of 4.10 GHz versus the Xeon’s 3.90 GHz, suggesting it can reach higher single-thread frequencies when thermal headroom allows, though the benchmark results do not reflect this advantage. Additionally, the i7 is a mobile part with a smaller die size of 125 mm² compared to the Xeon’s 171 mm², indicating a more power-efficient design for compact systems.

The Xeon’s wins are entirely compute-oriented, making it the superior choice for sustained multi-threaded workloads like rendering or simulation. The i7’s strengths are architectural: newer Kaby Lake core design, higher memory bandwidth, and a more recent release date. For users prioritizing peak clock speed or memory throughput, the i7 offers those specific advantages, but they do not translate into better benchmark scores in the tested Cinebench suite.

The Verdict

The data is unambiguous: the Intel Xeon E3-1585 v5 is the faster processor in every measured benchmark. Its average benchmark score of 2067 edges out the i7-8705G’s 2058, a 0.4% difference, but the head-to-head results show a much larger and consistent margin of roughly 10.5% across all Cinebench tests. If the decision rests solely on CPU performance, the Xeon E3-1585 v5 is the clear pick.

However, the choice is rarely that simple. The Xeon targets the server/workstation segment with ECC memory support and a launch MSRP of $556, making it a premium part for professional environments. The i7-8705G is a mobile processor with no ECC support, designed for laptops and compact devices. The i7’s 4.10 GHz boost clock is higher than the Xeon’s 3.90 GHz, and its memory bandwidth is greater, which could benefit memory-intensive tasks despite the lower benchmark scores. The i7 also uses the newer Kaby Lake architecture, which may offer better instruction-level efficiency in untested workloads.

For a workstation user running Cinebench-style render loops, the Xeon E3-1585 v5 is the obvious choice—it is 10.6% faster in multicore R23 and comes with ECC memory for data integrity. For a mobile user who needs a compact, lower-power solution, the i7-8705G is the only viable option between the two, given its BGA 2270 socket and mobile market segment. The Xeon’s BGA 1440 socket and server classification make it unsuitable for typical laptops. In short: pick the Xeon for raw compute and reliability, pick the i7 for mobility and higher memory bandwidth.

Head-to-Head Benchmarks

The most one-sided result appears in Cinebench R15 single-core, where the Xeon E3-1585 v5 scores 101 against the i7-8705G’s 91, an 11% lead. This is the largest percentage margin in the entire comparison, and it is notable because the i7 has a higher boost clock of 4.10 GHz. Despite that clock advantage, the Xeon’s 3.90 GHz base clock and Skylake-H architecture deliver better single-thread performance in this test.

The multicore results tell a similar story. In Cinebench R15 multicore, the Xeon scores 720 versus 651, a 10.6% win. The margin stays remarkably consistent across newer benchmarks: R20 multicore shows 3001 versus 2714 (10.6%), and R23 multicore shows 7146 versus 6464 (10.6%). This consistency suggests the performance gap is structural—likely due to the Xeon’s higher base clock (3.50 GHz vs 3.10 GHz) and possibly better thermal or power delivery characteristics in its workstation platform.

Single-core results in newer tests follow the same trend. Cinebench R20 single-core gives the Xeon 423 points against the i7’s 383, a 10.4% advantage. In R23 single-core, the Xeon scores 1008 versus 912, a 10.5% lead. The i7-8705G’s only benchmark not in the head-to-head list is Geekbench, where it scores 3968 multicore and 1277 single-core, but these are not compared against the Xeon. Overall, the Xeon’s victory is comprehensive and uniform, with no single test where the i7 comes close to matching it.

FAQ

Q: Which processor has a higher average benchmark score?

A: The Intel Xeon E3-1585 v5 has an average benchmark score of 2067, while the Intel Core i7-8705G scores 2058. The Xeon leads by 0.4%.

Q: Does the Core i7-8705G win any head-to-head benchmark?

A: No. The head-to-head data shows the Xeon E3-1585 v5 wins all 6 recorded benchmarks, with the i7-8705G recording zero wins.

Q: What is the largest performance gap between the two processors?

A: The largest gap is in Cinebench R15 single-core, where the Xeon E3-1585 v5 scores 101 versus the i7-8705G’s 91, an 11% difference.

Q: Which processor supports ECC memory?

A: The Intel Xeon E3-1585 v5 supports ECC memory, while the Intel Core i7-8705G does not.

Q: How do their memory bandwidths compare?

A: The Core i7-8705G has a memory bandwidth of 38.4 GB/s, which is higher than the Xeon E3-1585 v5’s 34.1 GB/s.

Q: What are their boost clock speeds?

A: The Core i7-8705G has a boost clock of 4.10 GHz, which is higher than the Xeon E3-1585 v5’s 3.90 GHz.

Architecture Differences

The two processors share the same 14 nm process node and Intel foundry, but they diverge in almost every other architectural aspect. The Xeon E3-1585 v5 uses the Skylake architecture (codename Skylake-H) and is built on a die size of 171 mm² with 2,300 million transistors. Its base clock is 3.50 GHz with a boost of 3.90 GHz, and it runs on the Intel BGA 1440 socket. The i7-8705G uses the newer Kaby Lake architecture (codename Kaby Lake G) with a smaller die size of 125 mm², a base clock of 3.10 GHz, and a higher boost clock of 4.10 GHz, on the Intel BGA 2270 socket.

Cache configurations are identical: both have 64 KB L1 and 256 KB L2 per core, plus 8 MB of shared L3. Memory support differs significantly. The Xeon supports both DDR3 and DDR4 in dual-channel mode with a memory bandwidth of 34.1 GB/s, while the i7 supports only DDR4 with a higher bandwidth of 38.4 GB/s. The Xeon also supports ECC memory, a critical feature for server/workstation reliability, which the i7 lacks.

PCIe connectivity is another differentiator. The Xeon offers 16 PCIe Gen 3 lanes (CPU only), while the i7 provides only 8 lanes. This makes the Xeon better suited for multi-GPU or high-bandwidth expansion in workstation builds. Integrated graphics also differ: the Xeon packs Iris Pro Graphics P580, while the i7 has the more common Intel HD Graphics 630. The market segments reflect these choices—the Xeon is classified as Server/Workstation, the i7 as Mobile. Both processors are end-of-life products, with the Xeon releasing in May 2016 and the i7 in January 2018. Neither has an unlocked multiplier, so overclocking is not an option for either part.