AMD Ryzen Embedded V1605B vs Intel Core i7-8665UE Comparison

# AMD Ryzen Embedded V1605B vs Intel Core i7-8665UE

The AMD Ryzen Embedded V1605B and Intel Core i7-8665UE are both 15-watt, 4-core/8-thread processors aimed at compact and embedded systems, yet they deliver strikingly different performance profiles across benchmark generations. The AMD part, built on the Zen architecture for desktop-class embedded use, wins three of four head-to-head Cinebench comparisons, while the Intel part, a Whiskey Lake-U mobile chip, takes one decisive victory. Their average benchmark scores are nearly identical — 1196 for the AMD versus 1193 for the Intel — placing both in the 33rd to 34th percentile of all CPUs, but the distribution of those scores reveals opposite strengths in legacy versus modern workloads.

FAQ

Q: Which processor has the higher average benchmark score?

A: The AMD Ryzen Embedded V1605B edges out the Intel Core i7-8665UE with an average benchmark score of 1196, compared to 1193 for the Intel part, a difference of just 3 points (0.3%).

Q: How do the two compare in Cinebench R23 multicore performance?

A: The Intel Core i7-8665UE is the clear winner here, scoring 4124 versus 3160 for the AMD Ryzen Embedded V1605B, a 23.4% advantage for Intel in the R23 multicore test.

Q: Which chip wins the Cinebench R15 multicore test?

A: The AMD Ryzen Embedded V1605B dominates, scoring 654 versus 415 for the Intel Core i7-8665UE, a 57.6% lead for AMD in the older R15 multicore benchmark.

Q: What is the difference in single-core performance?

A: The AMD Ryzen Embedded V1605B wins both single-core tests decisively — 120.7% ahead in Cinebench R15 single-core (128 vs 58) and 44.5% ahead in R23 single-core (841 vs 582).

Q: Do both processors have the same core and thread count?

A: Yes, both feature 4 cores and 8 threads, and both have a 15-watt TDP, but they differ significantly in base and boost clocks, cache layout, and integrated graphics.

Q: What is the production status of each chip?

A: The AMD Ryzen Embedded V1605B is listed as "Active," while the Intel Core i7-8665UE is marked "End-of-life" in the database.

Architecture Differences

The two processors come from fundamentally different design philosophies within the same 14 nm process generation. The AMD Ryzen Embedded V1605B is built on the Zen architecture, codenamed "Great Horned Owl," fabricated by GlobalFoundries. It packs 4,950 million transistors onto a 210 mm² die, using a 14 nm process node. The Intel Core i7-8665UE, in contrast, uses the Whiskey Lake architecture (Whiskey Lake-U) fabricated by Intel's own fabs on a 14 nm process, though transistor count and die size are not disclosed in the data.

Cache hierarchies diverge sharply. The AMD part allocates 128 KB of L1 cache per core and 512 KB of L2 per core, with only 2 MB of shared L3 cache. The Intel part uses smaller per-core caches — 64 KB L1 and 256 KB L2 — but compensates with a much larger 8 MB shared L3 cache, four times the AMD's last-level capacity. This difference likely explains why the Intel part excels in newer, more cache-sensitive workloads like Cinebench R23 multicore.

Memory support also differs. Both support dual-channel DDR4, but the Intel part additionally supports DDR3, and its memory bandwidth is rated at 38.4 GB/s, a figure not provided for the AMD chip. The Intel part also exposes PCIe Gen 3 with 16 lanes (CPU only), while PCIe details are absent from the AMD's specification sheet. Integrated graphics differ as well: the AMD uses Radeon Vega 8, while the Intel uses UHD Graphics 620. Neither supports ECC memory, and both have locked multipliers.

The AMD processor is classified as a desktop-market embedded chip on AMD Socket FP5, while the Intel part is a mobile-market processor on Intel BGA 1528. Their release dates are close — the AMD launched in February 2018, the Intel in September 2018 — but their fates differ: AMD remains active, Intel is end-of-life.

Head-to-Head Benchmarks

The head-to-head results paint a complex picture that rewards looking at each benchmark generation separately. In Cinebench R15 multicore, the AMD Ryzen Embedded V1605B scores 654 against the Intel Core i7-8665UE's 415, a 57.6% advantage. The magnitude here is substantial — the AMD chip delivers nearly 58% more throughput in this legacy test, a gap that suggests the Zen core's architecture was particularly well-suited to the R15 workload's threading and memory patterns.

The single-core R15 test is even more lopsided. AMD scores 128 versus Intel's 58, a 120.7% lead for the AMD part. That is more than double the performance, and it is a striking result for a chip with a lower boost clock (3.60 GHz versus 4.40 GHz for Intel). The data shows that raw clock speed did not translate into R15 single-core performance for the Intel part, likely due to thermal or power constraints within the 15-watt envelope.

Moving to Cinebench R23, the tables turn dramatically in multicore. The Intel Core i7-8665UE scores 4124, while the AMD manages 3160, giving Intel a 23.4% winning margin. This is a substantial reversal — the newer benchmark, which is more demanding on memory bandwidth and cache, appears to favor the Intel part's 8 MB L3 cache and 38.4 GB/s memory bandwidth. The AMD's smaller 2 MB L3 cache may be a bottleneck here.

In Cinebench R23 single-core, however, AMD reasserts dominance with a score of 841 versus 582 for Intel, a 44.5% lead. This keeps the AMD part ahead in three of the four head-to-head tests, with wins in both single-core tests and one multicore test. The Intel part's sole victory, while significant in percentage terms, is confined to the newest multicore workload.

The overall win tally is 3-1 in favor of the AMD Ryzen Embedded V1605B, yet the average benchmark scores remain nearly tied (1196 vs 1193). This discrepancy underscores how benchmark selection can skew the narrative — the AMD wins by large margins in some tests, while the Intel wins by a smaller but still substantial margin in another.

Specification Differences

The two processors differ across nearly every specification field except core count, thread count, TDP, and ECC support. Here are the key divergences:

• Base Clock: AMD at 2000.00 MHz, Intel at 1700.00 MHz — AMD has a 300 MHz base-clock advantage.
• Boost Clock: AMD at 3.60 GHz, Intel at 4.40 GHz — Intel has an 800 MHz boost-clock advantage.
• Socket: AMD Socket FP5 versus Intel BGA 1528.
• Architecture/Codename: Zen (Great Horned Owl) versus Whiskey Lake (Whiskey Lake-U).
• Generation: Ryzen Embedded (Zen) versus Core i7 (Whiskey Lake).
• Foundry: GlobalFoundries versus Intel.
• Transistors: 4,950 million for AMD; not disclosed for Intel.
• Die Size: 210 mm² for AMD; not disclosed for Intel.
• L1 Cache: 128 KB per core (AMD) versus 64 KB per core (Intel).
• L2 Cache: 512 KB per core (AMD) versus 256 KB per core (Intel).
• L3 Cache: 2 MB shared (AMD) versus 8 MB shared (Intel).
• Memory Support: DDR4 only (AMD) versus DDR3 and DDR4 (Intel).
• Memory Bandwidth: Not provided for AMD; 38.4 GB/s for Intel.
• PCIe: Not provided for AMD; Gen 3, 16 Lanes (CPU only) for Intel.
• Integrated Graphics: Radeon Vega 8 (AMD) versus UHD Graphics 620 (Intel).
• Market Segment: Desktop (AMD) versus Mobile (Intel).
• Production Status: Active (AMD) versus End-of-life (Intel).
• Release Date: 2018-02-20 (AMD) versus 2018-09-30 (Intel).
• Part Number: Not provided for AMD; SRF9W for Intel.
• Launch MSRP: None for AMD; $409 for Intel (stated once as launch MSRP).

Where Each One Wins

The AMD Ryzen Embedded V1605B is the better choice for workloads that rely on older or single-threaded Cinebench performance. It wins three of the four head-to-head tests, including both single-core tests by margins of 120.7% and 44.5%. Its base clock advantage of 300 MHz and larger per-core L1 and L2 caches likely contribute to these wins. For applications that resemble Cinebench R15 — whether that is legacy rendering pipelines, single-threaded legacy software, or lightly threaded tasks — the AMD part has a clear edge. Its desktop-market classification also suggests it may be better suited for always-on embedded systems where sustained performance at 15 watts matters.

The Intel Core i7-8665UE, despite its end-of-life status, takes the decisive win in Cinebench R23 multicore, scoring 23.4% higher than the AMD. This is the most modern benchmark in the comparison, and its 8 MB L3 cache, 38.4 GB/s memory bandwidth, and higher 4.40 GHz boost clock appear to give it an advantage in cache-heavy, multi-threaded modern workloads. For applications built around contemporary rendering engines or multithreaded compute that stress memory bandwidth and last-level cache, the Intel part is the stronger performer. Its mobile-market segment and support for both DDR3 and DDR4 also offer flexibility in memory selection, and its launch MSRP of $409 positions it as a premium part.

In practical terms, the choice hinges on the workload profile. If the target application is older, single-thread-constrained, or matches R15-era performance expectations, the AMD Ryzen Embedded V1605B wins. If the workload is modern, multithreaded, and cache-sensitive, the Intel Core i7-8665UE is the safer bet despite its end-of-life status. The near-identical average benchmark scores (1196 vs 1193) and percentile rankings (34th vs 33rd) suggest that for mixed workloads, neither chip offers a decisive overall advantage — the right answer depends entirely on which benchmarks mirror the real application.