AMD Ryzen Embedded 9900X3D vs Intel Core Ultra 5 228V Comparison

AMD Ryzen Embedded 9900X3D and Intel Core Ultra 5 228V occupy opposite ends of the processor spectrum, one a 12-core desktop part built for sustained throughput, the other an 8-thread mobile chip engineered for efficiency. The recorded data shows a clear split: the AMD part leads in raw multi-threaded work while the Intel part counters in single-thread tasks and carries a significantly lower TDP. The database places the Intel chip in the 75th percentile of all CPUs, while the AMD chip sits at the 50th percentile, though the AMD processor arrives without a recorded average benchmark score, limiting direct percentile comparisons.

The Verdict

The AMD Ryzen Embedded 9900X3D is the choice for workloads that scale across many cores. Its 12 cores and 24 threads, combined with a 128 MB L3 cache, give it a structural advantage in rendering, compilation, and server-style parallel tasks. The part boosts to 5.50 GHz, runs on DDR5 memory with ECC support, and uses a 4 nm TSMC process. It draws 120 W, a figure that reflects its capacity for heavy sustained loads. The Intel Core Ultra 5 228V, by contrast, uses 8 cores and 8 threads, boosts to 4.50 GHz, and operates at 17 W. Its benchmark results show a single-thread score of 3836 in PassMark, which is a strong result for a mobile part, and its Cinebench R23 single-core score of 1758 confirms that efficiency does not come at the cost of per-thread speed.

For a desktop system where power delivery and cooling are not constraints, the AMD processor is the clear pick. The data shows it has 24 PCIe Gen 5 lanes, a larger cache pool, and a higher boost clock. For a thin-and-light laptop or a fanless embedded system, the Intel chip is the only viable option, given its 17 W TDP and BGA socket. The Intel part also has a higher percentile ranking (75th versus 50th), suggesting that its overall performance profile places it above a larger share of the CPU population in the database, even though it has fewer threads. The verdict depends entirely on the use case: the AMD part for compute density, the Intel part for mobility and low power draw.

Where Each One Wins

The Intel Core Ultra 5 228V wins in every recorded single-thread benchmark. Its PassMark single-thread score of 3836 and Cinebench R23 single-core score of 1758 demonstrate strong per-core performance for a chip with an 8-thread limit. The Cinebench R20 single-core result of 916 and R15 single-core result of 267 reinforce this pattern. The AMD processor has no recorded benchmarks in the database, so a direct comparison of scores is not possible, but the Intel chip's single-thread results are competitive with desktop processors in its nearest rival group, such as the Intel Core i9-11900H, which sits just 0.3% behind the Intel chip's average score.

The AMD Ryzen Embedded 9900X3D wins on architectural capacity. It has 12 cores versus 8, 24 threads versus 8, and 128 MB of L3 cache versus 8 MB. The L3 cache difference is substantial: 16 times the shared cache of the Intel part. The AMD chip also supports ECC memory, a feature absent on the Intel chip, and offers 24 PCIe Gen 5 lanes versus 4. The AMD part has a higher base clock (4.40 GHz versus 2.10 GHz) and a higher boost clock (5.50 GHz versus 4.50 GHz), which should translate into faster multi-threaded execution when all cores are active. The Intel chip counters with a more advanced 3 nm process node versus AMD's 4 nm, a smaller TDP envelope, and a higher percentile ranking.

Architecture Differences

The AMD Ryzen Embedded 9900X3D uses the Granite Ridge codename and belongs to the Zen 5 generation. It is built on a 4 nm process at TSMC and packs 16,630 million transistors across two 70.6 mm² dies. The cache hierarchy features 80 KB of L1 per core, 1 MB of L2 per core, and 128 MB of L3. The part supports DDR5 memory on a dual-channel bus with 89.6 GB/s of bandwidth and includes ECC memory support. It uses the AMD Socket AM5 and has an unlocked multiplier, meaning the user can adjust clock speeds. The integrated graphics are Radeon Graphics, and the chip connects via PCIe Gen 5 with 24 lanes from the CPU.

The Intel Core Ultra 5 228V uses the Lunar Lake architecture and belongs to the Core Ultra Series 2. It is fabricated on a 3 nm process at TSMC. The cache layout differs significantly: 192 KB of L1 per core, 2.5 MB of L2 per core, and only 8 MB of shared L3. The memory support is listed as dependent on the motherboard, with a dual-channel bus but no recorded bandwidth figure. ECC memory is not supported. The chip uses the Intel BGA 2833 socket, is not multiplier-unlocked, and has integrated Arc 130V graphics. It provides PCIe Gen 5 with just 4 lanes from the CPU. The Intel part has a TDP of 17 W compared to the AMD part's 120 W, a 103 W difference that explains their different market segments: mobile versus desktop.

The release dates in the database show the Intel chip launched earlier, on 2024-09-23, while the AMD chip arrived on 2025-10-06. Both parts are marked as active in production. The AMD chip's part number is 100-000001368E, while the Intel chip has the part number SRPMVSRPMU. Neither part has a launch MSRP recorded.

FAQ

Q: Which processor has more cores and threads?

A: The AMD Ryzen Embedded 9900X3D has 12 cores and 24 threads. The Intel Core Ultra 5 228V has 8 cores and 8 threads.

Q: What is the TDP difference between the two chips?

A: The AMD part has a TDP of 120 W. The Intel part has a TDP of 17 W. The difference is 103 W.

Q: Which chip supports ECC memory?

A: The AMD Ryzen Embedded 9900X3D supports ECC memory. The Intel Core Ultra 5 228V does not support ECC memory.

Q: What are the boost clock speeds?

A: The AMD part boosts to 5.50 GHz. The Intel part boosts to 4.50 GHz. The AMD chip also has a higher base clock at 4.40 GHz versus 2.10 GHz.

Q: How much L3 cache does each processor have?

A: The AMD processor has 128 MB of L3 cache. The Intel processor has 8 MB of shared L3 cache. The AMD part has 16 times more L3 cache.

Q: Which processor has a higher percentile ranking in the database?

A: The Intel Core Ultra 5 228V ranks in the 75th percentile of all CPUs. The AMD Ryzen Embedded 9900X3D ranks in the 50th percentile.

Head-to-Head Benchmarks

The database contains no head-to-head benchmark results between these two specific processors. The AMD Ryzen Embedded 9900X3D has an empty benchmarks array and no average score. The Intel Core Ultra 5 228V, however, has a full set of recorded scores that place it in the 75th percentile with an average benchmark score of 21440.

The Intel chip's nearest rivals in the database provide context. The AMD Ryzen 5 2600 scores 21484, a delta of -0.2% relative to the Intel chip, meaning the Intel chip is 0.2% slower. The Intel Core i9-11900H scores 21367, a delta of 0.3%, meaning the Intel chip is 0.3% faster. The Intel Xeon D-1746TER scores 21635, a delta of -0.9%, making it 0.9% faster than the Core Ultra 5 228V. The AMD EPYC 9454 scores 21223, a delta of 1%, meaning the Intel chip is 1% faster. These deltas show that the Core Ultra 5 228V performs in a tight band around these established desktop and server parts, despite its mobile segment classification.

Looking at the Intel chip's specific benchmark scores, the multi-threaded results show a 9932 in Cinebench R23, 6491 in Cinebench R20, and 1502.5 in Cinebench R15. The PassMark suite shows a multithread score of 18227, integer math at 39679, floating point math at 53310, and data compression at 173924. The single-thread PassMark score of 3836 and the Cinebench R23 single-core score of 1758 indicate that the Intel part does not sacrifice per-core speed for its low power draw. The physics score of 1538 and the extended instructions score of 14801 round out the picture of a capable mobile processor.

The AMD part's lack of recorded benchmarks means the analysis must rely on its specifications. The 12-core, 24-thread configuration, the 5.50 GHz boost clock, and the 128 MB L3 cache all point to a part designed for throughput. The 89.6 GB/s memory bandwidth and 24 PCIe Gen 5 lanes support that interpretation. The 4 nm process and 16,630 million transistors indicate a modern, dense design. Without benchmark scores, the database cannot quantify its performance against the Intel chip, but the architectural differences are stark: the AMD part has 50% more cores, 200% more threads, and 16 times more L3 cache. The Intel part has a 3 nm process advantage, a 75th percentile ranking, and a 17 W TDP that the AMD part cannot match.