AMD Ryzen 9 9900X3D vs Intel Core Ultra 5 225T Comparison

The AMD Ryzen 9 9900X3D and the Intel Core Ultra 5 225T are desktop processors aimed at different performance levels. The recorded data shows a decisive advantage for the AMD part: it wins all 17 head-to-head comparisons, posts an average benchmark score of 54,762 against 30,468, and sits at the 91st percentile of all CPUs, while the Intel part sits at the 82nd percentile. These are not close results, and the gap appears across every measured workload.

Head-to-Head Benchmarks

All 17 recorded head-to-head tests favor the AMD Ryzen 9 9900X3D. The largest margin is PassMark integer math: 179,727 versus 59,543, a delta of 201.8%. Data compression follows at 685,074 versus 233,998, for a 192.8% advantage. Extended instructions show a 176% gap (55,426 versus 20,083), physics a 160.1% gap (5,340 versus 2,053), and random string sorting a 149.8% gap (71,864 versus 28,774). These are workloads where the AMD part roughly doubles or more than doubles the Intel result.

Multithreaded rendering tests reinforce the pattern. PassMark multithread scores 56,154 against 25,358, a 121.4% delta. Cinebench R23 multicore records 47,737 versus 21,971, and the R20 and R15 multicore tests show the same 117.3% delta at 20,049 versus 9,227 and 4,811 versus 2,214, respectively. The consistency of the Cinebench deltas across the R15, R20, and R23 versions indicates a stable performance ratio rather than a workload-specific anomaly.

Single-core results are also lopsided, though the smallest margin of the entire comparison appears here. PassMark single-thread scores 4,646 versus 4,348, a 6.9% delta. Cinebench R23 single-core records 6,739 versus 3,101, a 117.3% delta; R20 single-core records 2,830 versus 1,302, a 117.4% delta; and R15 single-core records 679 versus 312, a 117.6% delta. The PassMark single-thread result is the only test where the processors are close, making it an outlier within an otherwise lopsided comparison.

Encryption and prime-number workloads show narrower but still substantial gaps: PassMark data encryption scores 33,282 versus 18,289, a delta of 82%, and find prime numbers scores 544 versus 284, a delta of 91.5%. Floating-point math is the most moderate of the multithreaded deltas at 44.6%, with 119,622 versus 82,751.

In the broader database, the AMD part's average score of 54,762 is 0.6% below the Intel Core Ultra 5 245KF, 1.3% above the Intel Core Ultra 5 245K, 2% above the Intel Xeon 6505P, and 2.1% above the Intel Core i7-14700F. The Intel Core Ultra 5 225T's average of 30,468 is 0.2% above the Intel Core i9-11980HK, 0.3% above the AMD Ryzen 5 7640HS and AMD Ryzen 7 7736U, and 0.3% below the Intel Core i5-13600H. The AMD part's nearest rivals sit within 2.1% of its score, while the Intel part's nearest rivals sit within 0.3%.

The Verdict

The benchmark data leaves no ambiguity about performance: the AMD Ryzen 9 9900X3D is faster in every recorded comparison. A user choosing strictly on measured throughput would select the AMD part, as it leads by 117.3% or more in all recorded Cinebench tests and by 121.4% in PassMark multithread. Its average score of 54,762 places it near the Intel Core Ultra 5 245KF, which is only 0.6% ahead, and above the Intel Core Ultra 5 245K by 1.3%.

The Intel Core Ultra 5 225T does hold specification advantages that matter in constrained designs. Its TDP is 65 W versus 120 W, its process node is 3 nm versus 4 nm, and its memory bandwidth is 102.4 GB/s versus 89.6 GB/s. It also provides more L1 and L2 cache per core. None of those advantages translate into a single benchmark win in the recorded data, but they indicate a lower-power part with a different target.

The percentile positions summarize the class gap. The AMD part sits at the 91st percentile of all CPUs, while the Intel part sits at the 82nd percentile. The Intel part's nearest rivals are the Intel Core i9-11980HK, AMD Ryzen 5 7640HS, Intel Core i5-13600H, and AMD Ryzen 7 7736U, with deltas of 0.2%, 0.3%, and -0.3%. That places it in a different competitive class from the AMD part, whose nearest rivals record average scores from 53,620 to 55,093. The data indicates the AMD part for users who need maximum throughput and the Intel part for users bound by a 65 W power envelope.

Architecture Differences

Both processors use TSMC fabrication, but on different nodes: the AMD chip uses a 4 nm process and the Intel chip uses 3 nm. Transistor counts are close, with AMD at 16,630 million and Intel at 17,800 million, although the physical layout differs sharply. AMD uses a 2x 70.6 mm² die configuration, while Intel uses a single 243 mm² die.

The core configurations differ by design. AMD provides 12 cores and 24 threads, while Intel provides 10 cores and 10 threads. The AMD architecture is Zen 5 under the Granite Ridge codename; the Intel architecture is Arrow Lake under the Arrow Lake-S codename. The AMD part is a Ryzen 9 processor from the 9000 series, while the Intel part is an Ultra 5 processor from the Core Ultra Series 2.

Cache hierarchies are structured differently. AMD allocates 80 KB of L1 and 1 MB of L2 per core, with a large 128 MB L3. Intel allocates 192 KB of L1 and 3 MB of L2 per core, with a shared 20 MB L3. The larger L3 on the AMD part is the most conspicuous difference and contributes to its multithreaded and compression results. The Intel part's larger per-core L1 and L2 allocations do not offset the L3 deficit in the recorded benchmarks. The 128 MB L3 on the AMD part is the largest cache allocation in the comparison, and it aligns with the largest benchmark margins in data compression and integer math.

Memory support is DDR5 dual-channel on both, but bandwidth differs: 89.6 GB/s on AMD versus 102.4 GB/s on Intel. AMD supports ECC memory; Intel does not. PCIe connectivity also differs, with Gen 5 on both but 24 CPU lanes on AMD and 20 on Intel. Integrated graphics differ as well: Radeon Graphics on AMD versus Arc Xe-LPG Graphics 16EU on Intel. Both parts are listed as Active in the database and target the Desktop market segment.

Specification Differences

The two processors differ in cores, threads, base and boost clocks, TDP, socket, architecture, codename, generation, process node, transistor count, die size, cache, memory bandwidth, ECC support, PCIe lanes, integrated graphics, release date, multiplier lock, and part number. They share DDR5 memory support, a dual-channel memory bus, the TSMC foundry, the Desktop market segment, and Active production status.

| Field | AMD Ryzen 9 9900X3D | Intel Core Ultra 5 225T |

|---|---|---|

| Cores | 12 | 10 |

| Threads | 24 | 10 |

| Base clock | 4.40 GHz | 2.50 GHz |

| Boost clock | 5.50 GHz | 4.90 GHz |

| TDP | 120 W | 65 W |

| Socket | AMD Socket AM5 | Intel Socket 1851 |

| Architecture | Zen 5 | Arrow Lake |

| Codename | Granite Ridge | Arrow Lake-S |

| Generation | Ryzen 9 (Zen 5 (Granite Ridge)) | Ultra 5 (Arrow Lake) |

| Process node | 4 nm | 3 nm |

| Transistors | 16,630 million | 17,800 million |

| Die size | 2x 70.6 mm² | 243 mm² |

| L1 cache | 80 KB (per core) | 192 KB (per core) |

| L2 cache | 1 MB (per core) | 3 MB (per core) |

| L3 cache | 128 MB | 20 MB (shared) |

| Memory bandwidth | 89.6 GB/s | 102.4 GB/s |

| ECC memory | Yes | No |

| PCIe | Gen 5, 24 Lanes (CPU only) | Gen 5, 20 Lanes (CPU only) |

| Integrated graphics | Radeon Graphics | Arc Xe-LPG Graphics 16EU |

| Release date | 2025-01-05 | 2024-12-31 |

| Multiplier unlocked | Yes | No |

| Part number | 100-000001368 | unknown |

The AMD processor has a launch MSRP of $599.

FAQ

Q: Which processor wins the most head-to-head benchmark comparisons?

A: The AMD Ryzen 9 9900X3D wins all 17 recorded head-to-head tests, while the Intel Core Ultra 5 225T wins none.

Q: How large is the average benchmark score gap?

A: The AMD part records an average benchmark score of 54,762, while the Intel part records 30,468. The AMD part sits at the 91st percentile of all CPUs, and the Intel part sits at the 82nd percentile.

Q: What is the Cinebench R23 multicore difference?

A: The AMD Ryzen 9 9900X3D scores 47,737, and the Intel Core Ultra 5 225T scores 21,971, a delta of 117.3%.

Q: Does the Intel part have any specification advantage?

A: Yes. It has a TDP of 65 W versus 120 W, a 3 nm process versus 4 nm, 102.4 GB/s memory bandwidth versus 89.6 GB/s, 192 KB L1 per core versus 80 KB, and 3 MB L2 per core versus 1 MB.

Q: Which processor has more L3 cache?

A: The AMD Ryzen 9 9900X3D has 128 MB of L3 cache, while the Intel Core Ultra 5 225T has 20 MB shared.

Q: Are both multipliers unlocked?

A: No. The AMD part has an unlocked multiplier; the Intel part does not.