AMD Ryzen Embedded 9900X vs Intel Core Ultra 5 228V Comparison

Head-to-Head Benchmarks

The database contains no shared benchmark results for the AMD Ryzen Embedded 9900X and the Intel Core Ultra 5 228V. The AMD part has no recorded benchmark entries, while the Intel part has a full set of Cinebench and PassMark scores. This makes a direct, same-test comparison impossible. What can be analyzed is the Intel part's recorded performance profile and its position relative to nearby CPUs in the database, alongside the AMD part's architectural specifications and its overall percentile ranking.

The Intel Core Ultra 5 228V delivers a Cinebench R23 multi-core score of 9932 and a single-core score of 1758. In Cinebench R20, it records 6491 multi-core and 916 single-core. Cinebench R15 shows 1502.5 multi-core and 267 single-core. The PassMark suite shows a multi-thread score of 18227 and a single-thread score of 3836. The data indicates a processor oriented toward balanced throughput in both lightly threaded and heavily threaded workloads.

The Intel part's average benchmark score is 21440, placing it in the 75th percentile of all CPUs in the database. Its nearest rivals include the AMD Ryzen 5 2600 with an average score of 21484 and a delta of -0.2%, meaning the Intel part trails by 0.2%. The Intel Core i9-11900H scores 21367, which is 0.3% behind the Intel Core Ultra 5 228V. The Intel Xeon D-1746TER scores 21635, putting it 0.9% ahead. The AMD EPYC 9454 scores 21223, which is 1% behind. The Intel Core Ultra 5 228V sits within a tight 1.9 percentage point band across these four rivals, indicating that its aggregate performance is closely matched to a range of desktop, mobile, and server processors from different generations.

The AMD Ryzen Embedded 9900X holds a 50th percentile ranking across all CPUs, with an average benchmark score of 0. The absence of recorded benchmark results means the database cannot verify its real-world performance. Its specifications, however, suggest a high-throughput design. It has 12 cores and 24 threads, a base clock of 4.40 GHz and a boost clock of 5.60 GHz, with a TDP of 120 W. The Intel part has 8 cores and 8 threads, a base clock of 2.10 GHz and a boost clock of 4.50 GHz, with a TDP of 17 W. The core and thread disparity is substantial: the AMD part offers 50% more cores and 200% more threads. Clock speeds also favor AMD, with a 1.10 GHz advantage in boost and a 2.30 GHz advantage at base.

The Verdict

The data supports a clear split by workload type and platform constraints. For multi-threaded, sustained compute, the AMD Ryzen Embedded 9900X is the stronger candidate on paper. Its 12 cores, 24 threads, 64 MB of L3 cache, and 120 W TDP indicate a processor designed for parallel workloads that can tolerate higher power draw. The Intel Core Ultra 5 228V, with 8 cores, 8 threads, 8 MB of L3 cache, and a 17 W TDP, is built for efficiency and mobile deployment. Its recorded benchmark results confirm it can handle mainstream productivity tasks, but its thread count limits heavy parallel scaling.

For single-threaded performance, the Intel part's recorded scores show strength. A Cinebench R23 single-core score of 1758 and a PassMark single-thread score of 3836 are competitive figures. The AMD part has no recorded scores, so no direct comparison is possible. The AMD part's higher boost clock of 5.60 GHz suggests strong single-thread potential, but the database does not confirm it.

The percentile gap is notable: the Intel part sits at the 75th percentile, while the AMD part sits at the 50th. This reflects the AMD part's lack of benchmark data rather than its measured performance. The 50th percentile is the default position for a CPU with no recorded scores. The Intel part's 75th percentile is earned from its average benchmark score of 21440.

Platform choice is another deciding factor. The AMD part uses AMD Socket AM5 and supports DDR5 memory with ECC. The Intel part uses Intel BGA 2833, a soldered mobile socket, and its memory support is listed as dependent on the motherboard. The AMD part targets desktop and embedded deployments, while the Intel part targets mobile systems. Users with an AM5 platform can install the AMD part; users with a Lunar Lake mobile board are limited to the Intel part.

Architecture Differences

The AMD Ryzen Embedded 9900X belongs to the Ryzen Embedded line within the 9000 series, using the Granite Ridge codename and Zen 5 architecture. It is manufactured on a 4 nm process at TSMC. The transistor count is 16,630 million, spread across two chiplets with a die size of 2x 70.6 mm². The cache hierarchy includes 80 KB of L1 per core, 1 MB of L2 per core, and 64 MB of L3. It supports DDR5 memory over a dual-channel bus with a bandwidth of 89.6 GB/s. ECC memory is supported. PCIe connectivity is Gen 5 with 24 lanes from the CPU. Integrated graphics are listed as Radeon Graphics. The multiplier is unlocked, and the part number is 100-000000662E. It was released on 2025-10-06 and remains in active production.

The Intel Core Ultra 5 228V belongs to the Core Ultra Series 2, using the Lunar Lake architecture. It is manufactured on a 3 nm process at TSMC. The cache hierarchy includes 192 KB of L1 per core, 2.5 MB of L2 per core, and 8 MB of shared L3. Memory support is listed as unknown and dependent on the motherboard, though the memory bus is dual-channel. ECC memory is not supported. PCIe connectivity is Gen 5 with 4 lanes from the CPU. Integrated graphics are Arc 130V. The multiplier is locked. The part number is SRPMVSRPMU. It was released on 2024-09-23 and remains in active production.

The process node difference is small: 4 nm for AMD versus 3 nm for Intel. The larger difference is in cache allocation. The AMD part provides 64 MB of L3, eight times the Intel part's 8 MB. Per-core L2 also differs: 1 MB per core for AMD versus 2.5 MB per core for Intel. The Intel part has a larger L1 per core at 192 KB versus 80 KB. These cache layouts reflect different design philosophies: AMD favors a large shared L3 for multi-threaded data sharing, while Intel provides more private cache per core for latency-sensitive single-threaded work.

The TDP gap is extreme. The AMD part draws 120 W, while the Intel part draws 17 W. This is a 103 W difference. The Intel part is designed for fanless or low-power mobile systems, while the AMD part requires active cooling and a desktop-class power delivery system. The socket types confirm the deployment targets: AMD Socket AM5 is a desktop socket, while Intel BGA 2833 is a ball-grid array soldered to the motherboard.

Memory bandwidth is recorded only for the AMD part at 89.6 GB/s. The Intel part has no recorded memory bandwidth figure. The AMD part supports ECC, which is relevant for embedded and server-adjacent workloads where data integrity matters. The Intel part does not support ECC.

FAQ

Q: Which processor has more cores and threads?

A: The AMD Ryzen Embedded 9900X has 12 cores and 24 threads. The Intel Core Ultra 5 228V has 8 cores and 8 threads.

Q: What are the boost clock speeds of each processor?

A: The AMD Ryzen Embedded 9900X boosts to 5.60 GHz. The Intel Core Ultra 5 228V boosts to 4.50 GHz.

Q: How do the cache sizes compare?

A: The AMD Ryzen Embedded 9900X has 80 KB of L1 per core, 1 MB of L2 per core, and 64 MB of L3. The Intel Core Ultra 5 228V has 192 KB of L1 per core, 2.5 MB of L2 per core, and 8 MB of shared L3.

Q: What is the TDP of each processor?

A: The AMD Ryzen Embedded 9900X has a TDP of 120 W. The Intel Core Ultra 5 228V has a TDP of 17 W.

Q: Which processor supports ECC memory?

A: The AMD Ryzen Embedded 9900X supports ECC memory. The Intel Core Ultra 5 228V does not.

Q: What benchmark scores are recorded for the Intel Core Ultra 5 228V?

A: The Intel part records a Cinebench R23 multi-core score of 9932 and single-core score of 1758. Its PassMark multi-thread score is 18227 and single-thread score is 3836. The AMD part has no recorded benchmark scores in the database.

Where Each One Wins

The AMD Ryzen Embedded 9900X wins in raw multi-threaded capability. Its 12 cores and 24 threads, combined with 64 MB of L3 cache and a 120 W TDP, point to sustained performance in heavily parallel workloads such as compilation, rendering, virtualization, and data processing. The dual-channel DDR5 support with 89.6 GB/s bandwidth and ECC capability strengthen its position for embedded and server-adjacent tasks where memory integrity is critical. The 24 PCIe Gen 5 lanes provide substantial I/O headroom for expansion cards, NVMe storage, and accelerators. Its unlocked multiplier allows tuning for users who need to adjust clocks, though the database records no measured results to confirm overclocking gains.

The Intel Core Ultra 5 228V wins in efficiency and mobility. Its 17 W TDP makes it suitable for compact, battery-powered, or passively cooled systems. The recorded benchmark scores confirm it delivers competitive single-thread performance: a Cinebench R23 single-core score of 1758 and a PassMark single-thread score of 3836. Its 3 nm process node is one step ahead of the AMD part's 4 nm node, which likely contributes to its lower power draw. The 8 MB of shared L3 and 2.5 MB of L2 per core provide a cache hierarchy optimized for single-threaded responsiveness. The Arc 130V integrated graphics add GPU capability without a discrete card, which is an advantage for compact mobile systems.

The Intel part's aggregate performance sits at the 75th percentile of all CPUs, with an average benchmark score of 21440. Its nearest rivals, the AMD Ryzen 5 2600, Intel Core i9-11900H, Intel Xeon D-1746TER, and AMD EPYC 9454, all fall within 1.9 percentage points of its score. This indicates that the Intel part, despite its low TDP, delivers performance comparable to a range of desktop and server processors from prior generations. The AMD part holds a 50th percentile ranking, but this is based on zero recorded benchmarks, so it should not be interpreted as a measured performance level.

For users on an AM5 platform, the AMD part is the only socket-compatible choice. For users on a Lunar Lake mobile platform, the Intel part is the only option. The AMD part's ECC support and 24 PCIe Gen 5 lanes make it the stronger fit for storage-heavy or memory-sensitive embedded workloads. The Intel part's lower TDP and recorded benchmark results make it the stronger fit for mobile productivity, where battery life and thermal limits constrain sustained performance.

The data shows no overlap in benchmark results, so the verdict rests on specifications and the Intel part's measured scores. The AMD part's higher core count, thread count, clock speeds, cache size, and memory bandwidth position it as the higher-performance part on paper. The Intel part's recorded scores and efficiency position it as the more portable and power-efficient part. The choice depends on whether the workload demands parallel throughput or low power consumption.