AMD Ryzen 5 40 vs Intel Core Ultra 5 115U Comparison

Head-to-Head Benchmarks

The head-to-head data is lopsided, but not entirely one-sided. Intel's Core Ultra 5 115U claims 13 wins across the 15 recorded comparisons, while AMD's Ryzen 5 40 manages 2 wins, one of which is a statistical tie. The most dramatic gap appears in Cinebench R23 multi-core, where Intel scores 10855 against AMD's 4841, a delta of -55.4% from AMD's perspective. That is more than double the multi-threaded render performance, and it sets the tone for most of the heavy compute workloads.

Single-core performance tells a different story in one specific test. In Cinebench R15 single-core, AMD wins with 165.5 against Intel's 154, a 7.5% advantage. This is the only head-to-head test where AMD leads by a meaningful margin, and it is surprising given that Intel dominates the newer Cinebench R23 single-core test with 1532 versus 1150, a 24.9% gap. The R15 result looks like an outlier in an otherwise consistent Intel sweep, but it is a recorded data point and worth noting for legacy single-threaded workloads.

The PassMark suite reinforces Intel's dominance across the board. In floating point math, Intel scores 30538 against AMD's 15194, a 50.2% deficit for AMD. Physics simulation shows a similar pattern: Intel at 898, AMD at 432, a 51.9% gap. Prime number finding is the largest relative difference at -62.3%, with Intel scoring 53 versus AMD's 20. Integer math is less extreme but still decisive: Intel 41043 versus AMD 31598, a 23% gap. Data encryption shows Intel ahead at 8606 versus 6646, a 22.8% difference, and extended instructions follow with Intel at 8620 versus 6437, a 25.3% gap.

The multithreaded PassMark score puts Intel at 13024 versus AMD's 9341, a 28.3% gap, which aligns with the Cinebench R23 multi-core result. The single-thread PassMark test gives Intel 3057 against AMD's 2477, a 19% gap. Data compression is nearly even: Intel at 141945 versus AMD at 141533, a marginal 0.3% difference. Random string sorting is exactly tied at 15124 for both chips, with the database assigning the win to AMD due to the zero delta. That tie is the only place where AMD matches Intel in absolute output.

Overall average benchmark scores reflect the same hierarchy. Intel's average is 16753, AMD's is 15882, a difference of about 5.5%. Both chips sit at the 70th percentile among all CPUs, so they occupy the same performance tier, but Intel is consistently toward the top of that tier while AMD hovers near the middle.

Architecture Differences

The two processors come from fundamentally different design philosophies. AMD's Ryzen 5 40 uses the Zen 2 architecture, codenamed Mendocino, built on a 6 nm process at TSMC. Intel's Core Ultra 5 115U uses the Meteor Lake architecture on Intel's 7 nm process. The process node difference is small on paper, but the architectural generation gap is significant: Zen 2 debuted years before Meteor Lake, and that shows in the benchmark results.

Core configuration is a major differentiator. AMD offers 4 cores and 8 threads, while Intel offers 8 cores and 10 threads. Intel's hybrid arrangement explains the odd thread count, mixing performance and efficiency cores. The extra cores give Intel a structural advantage in any workload that scales with thread count, which is precisely what the multi-core benchmarks show. AMD's base clock is listed at 2.80 GHz with a boost of 4.30 GHz, while Intel's base clock is listed at 1500.00 MHz with a boost of 4.20 GHz. The base clock figures are not directly comparable because Intel's listed value likely reflects the efficiency core base, but the boost clocks are close, with AMD slightly higher at 4.30 versus 4.20.

Cache hierarchies differ substantially. AMD has 64 KB of L1 per core, 512 KB of L2 per core, and 4 MB of shared L3. Intel has 112 KB of L1 per core, 2 MB of L2 per core, and 10 MB of shared L3. Intel's L3 cache is 2.5 times larger, which helps with repeated data access patterns and partially explains the single-thread performance advantage in most tests.

Memory support also diverges. AMD uses LPDDR5 with 88.0 GB/s of bandwidth, while Intel uses DDR5 with 89.6 GB/s. The bandwidth figures are close, within 1.8%, so memory throughput is not the primary differentiator. Both support dual-channel memory, and neither supports ECC. PCIe connectivity is another split: AMD provides Gen 3 with 4 lanes, Intel provides Gen 4 with 12 lanes. Intel's PCIe implementation is more modern and offers more connectivity, which matters for external devices and storage expansion.

Integrated graphics differ as well. AMD pairs the CPU with a Radeon 610M, while Intel integrates Arc Xe-LPG with 48 execution units. The recorded data does not include graphics benchmarks, but the architectural difference is clear: Intel's Arc-class GPU is a newer design with more execution resources.

Where Each One Wins

The use-case split follows the benchmark data almost perfectly. Intel wins in every multi-threaded workload recorded. Cinebench R23 multi-core, Cinebench R15 multi-core, PassMark multithread, floating point math, integer math, physics, prime numbers, encryption, and extended instructions all favor Intel, often by wide margins. This makes the Core Ultra 5 115U the clear choice for rendering, simulation, mathematical computation, and any productivity task that can use more than a few threads.

Intel also wins most single-threaded tests. Cinebench R23 single-core shows a 24.9% lead, and PassMark single-thread shows a 19% lead. Only Cinebench R15 single-core favors AMD, with a 7.5% edge. For modern single-threaded applications, the data points to Intel. For legacy software that behaves like the R15 workload, AMD holds a narrow advantage.

AMD's wins are limited but not meaningless. The random string sorting tie at 15124 shows that AMD can match Intel in specific memory-heavy, pointer-chasing workloads. The Cinebench R15 single-core win suggests that certain older code paths respond better to AMD's Zen 2 implementation. Data compression is nearly tied at a 0.3% difference, so AMD is competitive in compression workloads despite losing the overall benchmark.

For users running modern multi-threaded applications, Intel is the obvious pick. For users with legacy single-threaded workloads or applications that mirror the R15 pattern, AMD offers a small but real advantage. The data compression near-tie means AMD is not embarrassing itself in that niche.

Specification Differences

The key specification differences between the two chips are as follows. AMD has 4 cores and 8 threads, while Intel has 8 cores and 10 threads. AMD's base clock is 2.80 GHz, Intel's is 1500.00 MHz, though the boost clocks are closer at 4.30 GHz for AMD and 4.20 GHz for Intel. Both have a 15 TDP.

AMD uses Zen 2 architecture on TSMC's 6 nm process with a 100 mm² die size. Intel uses Meteor Lake on Intel's 7 nm process with no recorded die size. AMD's socket is AMD Socket FT6, Intel's is Intel BGA 2049. AMD's cache consists of 64 KB L1 per core, 512 KB L2 per core, and 4 MB shared L3. Intel's cache consists of 112 KB L1 per core, 2 MB L2 per core, and 10 MB shared L3.

Memory support differs: AMD uses LPDDR5, Intel uses DDR5 dependent on the motherboard. Memory bandwidth is 88.0 GB/s for AMD and 89.6 GB/s for Intel. Both are dual-channel. PCIe support is Gen 3 with 4 lanes for AMD, Gen 4 with 12 lanes for Intel. Integrated graphics are Radeon 610M for AMD and Arc Xe-LPG 48EU for Intel.

Release dates are far apart: AMD is dated 2025-09-30, Intel is dated 2023-12-13. Both are active in production, both are mobile segment parts, both have locked multipliers, and neither supports ECC memory. Intel has a recorded part number of SRN6F, while AMD's is unknown.

FAQ

Q: Which processor has more cores and threads?

A: Intel's Core Ultra 5 115U has 8 cores and 10 threads, while AMD's Ryzen 5 40 has 4 cores and 8 threads.

Q: How much larger is Intel's L3 cache compared to AMD's?

A: Intel has 10 MB of shared L3 cache, while AMD has 4 MB of shared L3 cache, making Intel's L3 2.5 times larger.

Q: In which test does AMD actually beat Intel?

A: AMD wins the Cinebench R15 single-core test with 165.5 against Intel's 154, a 7.5% advantage. AMD also receives the win in random string sorting, but that test is tied at 15124 for both chips.

Q: What is the biggest performance gap between the two?

A: The largest relative gap is in PassMark's find prime numbers test, where Intel scores 53 against AMD's 20, a 62.3% deficit for AMD.

Q: Do both processors support the same memory type?

A: No, AMD supports LPDDR5 while Intel supports DDR5 that depends on the motherboard. Both use dual-channel memory, and their bandwidth figures are close at 88.0 GB/s for AMD and 89.6 GB/s for Intel.

Q: Are both processors in the same performance percentile?

A: Yes, both are at the 70th percentile among all CPUs, but Intel's average benchmark score is 16753 versus AMD's 15882.

The Verdict

The data is unambiguous for most workloads. Intel's Core Ultra 5 115U is the stronger processor in nearly every recorded benchmark. It wins 13 of 15 head-to-head comparisons, including all multi-core tests and the majority of single-threaded tests. The Cinebench R23 multi-core result alone, 10855 versus 4841, makes Intel the clear choice for rendering and threaded compute. The PassMark multithread score of 13024 versus 9341 reinforces that conclusion.

AMD's Ryzen 5 40 does have a niche. The Cinebench R15 single-core win at 7.5% and the exact tie in random string sorting show that AMD can hold its own in specific legacy or memory-latency-sensitive tasks. The data compression result is within 0.3%, so AMD is competitive there as well. But those are narrow victories in a dataset where Intel leads by double digits in most categories.

Who should pick which? Users running modern multi-threaded applications, compilation workloads, physics simulations, or encryption tasks should choose Intel. The 50.2% lead in floating point math and the 51.9% lead in physics are decisive. Users with legacy single-threaded software that matches the R15 pattern, or those who specifically care about random string sorting and see the tie as sufficient, could choose AMD without losing that specific test. But the overall average benchmark score, 16753 for Intel versus 15882 for AMD, tells the story: Intel is the more capable chip across the recorded measurement suite. AMD remains a viable option only in narrow, well-defined scenarios.