Intel Core Ultra 9 285 vs Intel Xeon 6724P Comparison

Where Each One Wins

The benchmark record splits these two Intel processors into clearly different roles. The Intel Core Ultra 9 285 wins 13 of the 17 head-to-head tests, while the Intel Xeon 6724P takes 4. That raw count, however, hides the nature of each victory.

The Core Ultra 9 285 dominates every Cinebench iteration, both single-core and multi-core. Its smallest Cinebench margin is 12.1%, and that appears consistently across R15, R20, and R23. In single-threaded PassMark testing, the desktop part leads by 48.9%, a massive gap that reflects its 5.60 GHz boost clock versus the Xeon's 4.30 GHz. The Ultra 9 also wins floating point math by 44%, data encryption by 36.7%, prime number finding by 23.4%, and multithreaded PassMark by 10.2%. Random string sorting goes to the Ultra 9 by 7.6%.

The Xeon 6724P's four wins are concentrated in workloads that favor its server-oriented design. Data compression leads by 4%, integer math by 4.3%, and extended instructions by 16.2%. Its largest victory is in physics simulation, where it beats the Ultra 9 by 28.1%. These are not trivial margins, but they are fewer in number and clustered in specific compute patterns.

Looking at the broader database context, the Ultra 9 sits at the 95th percentile of all CPUs, with an average benchmark score of 75488. The Xeon sits at the 94th percentile with 72396. The nearest rivals for the Ultra 9 are all within 0.3% of its average score, including the AMD EPYC 8224P at -0.1% and the AMD Ryzen 7 PRO 9755 at -0.3%. The Xeon's nearest rivals are led by the Intel Xeon 6517P at +0.1%, with the Core Ultra 7 265KF at +0.7%. Both processors are effectively at the top of their respective peer groups, but the Ultra 9 edges slightly higher in aggregate.

The Verdict

Choose the Intel Core Ultra 9 285 for single-thread responsiveness, encryption, floating-point workloads, and general multithreaded rendering. The data shows a 48.9% advantage in PassMark single-thread, a 36.7% edge in data encryption, and a 44% lead in floating point math. Its Cinebench R23 multi-core score of 48945 versus 43643 means the desktop chip renders faster across the board. The 24-core, 24-thread configuration with a 5.60 GHz boost clock handles latency-sensitive tasks with clear superiority.

Choose the Intel Xeon 6724P for physics simulation, compression, integer-heavy math, and extended instruction workloads. The 28.1% physics win and 16.2% extended instructions lead are substantial. Its 16 cores with 32 threads, plus 72 MB of shared L3 cache, serve workloads that benefit from higher thread counts per core and larger aggregate cache. The Xeon's eight-channel memory bus at 409.6 GB/s bandwidth, versus the Ultra 9's dual-channel 102.4 GB/s, supports memory-hungry server tasks, though the benchmark data does not directly measure memory throughput.

The average benchmark score favors the Ultra 9 at 75488 versus 72396, a 4.1% difference. The percentile ranking also favors the desktop part, 95th versus 94th. For mixed workloads that include rendering, encryption, and general computation, the Ultra 9 is the stronger choice. For specialized server tasks involving physics, compression, and extended instructions, the Xeon holds specific advantages.

Head-to-Head Benchmarks

The largest single-test margin belongs to the Core Ultra 9 285 in PassMark single-thread, where it scores 4881 against the Xeon's 3279, a 48.9% delta. This is the clearest signal of the Ultra 9's architectural advantage in per-core performance. The same 4881 score appears in both single_thread and singlethread entries, confirming consistency.

Floating point math shows the second-largest gap. The Ultra 9 posts 194988 versus 135404, a 44% advantage. This suggests the desktop chip's FPU pipeline is significantly more efficient for this workload. Data encryption follows at 46949 versus 34332, a 36.7% lead. Prime number finding shows a 23.4% edge, 459 versus 372.

The Xeon's biggest win is physics simulation, where it scores 5004 against 3598, a 28.1% delta. This is the only test where the Xeon leads by more than 20%. Extended instructions give the Xeon a 16.2% win, 54101 versus 45357. Integer math and data compression are closer, with the Xeon leading by 4.3% and 4% respectively.

Cinebench results are remarkably uniform. The Ultra 9 wins R15 multi-core by 12.1% (4933 vs 4399), R15 single-core by 12.3% (696 vs 620), R20 multi-core by 12.1% (20556 vs 18330), R20 single-core by 12.1% (2901 vs 2587), R23 multi-core by 12.1% (48945 vs 43643), and R23 single-core by 12.1% (6909 vs 6161). This consistency indicates a stable per-core advantage across all Cinebench versions.

PassMark multithread shows the Ultra 9 at 56602 versus 51345, a 10.2% win. Random string sorting goes to the Ultra 9 by 7.6%, 73651 versus 68477. These moderate wins complement the larger single-thread and floating-point margins.

FAQ

Q: Which processor has a higher boost clock?

A: The Intel Core Ultra 9 285 boosts to 5.60 GHz, while the Intel Xeon 6724P boosts to 4.30 GHz. The base clocks are 2.50 GHz for the Ultra 9 and 3.60 GHz for the Xeon.

Q: How do the core and thread counts differ?

A: The Core Ultra 9 285 has 24 cores and 24 threads. The Xeon 6724P has 16 cores and 32 threads. Despite fewer cores, the Xeon has more threads due to simultaneous multithreading.

Q: Which chip has more L3 cache?

A: The Xeon 6724P has 72 MB shared L3 cache, while the Core Ultra 9 285 has 36 MB shared L3 cache. The Xeon also has 112 KB L1 per core and 2 MB L2 per core, versus the Ultra 9's 192 KB L1 and 3 MB L2 per core.

Q: What memory channel configurations do these processors support?

A: The Xeon 6724P supports eight-channel DDR5 with 409.6 GB/s bandwidth. The Core Ultra 9 285 supports dual-channel DDR5 with 102.4 GB/s bandwidth. Both support ECC memory.

Q: Which processor has integrated graphics?

A: The Core Ultra 9 285 includes Arc Xe-LPG Graphics with 64 execution units. The Xeon 6724P has no integrated graphics (N/A).

Q: What are the launch MSRP values?

A: The Core Ultra 9 285 has a launch MSRP of $579. The Xeon 6724P has a launch MSRP of $3622.

Architecture Differences

The two processors represent fundamentally different Intel designs. The Core Ultra 9 285 uses the Arrow Lake-S architecture on a 3 nm process fabricated by TSMC. It contains 17,800 million transistors on a 243 mm² die. The Xeon 6724P uses the Granite Rapids architecture on a 5 nm process fabricated by Intel. The Xeon's transistor count and die size are not recorded in the database.

The Ultra 9 is a desktop part with 24 cores and 24 threads, meaning no hyperthreading. The Xeon is a server/workstation part with 16 cores and 32 threads. The Xeon's L1 cache is 112 KB per core, its L2 is 2 MB per core, and its L3 is 72 MB shared. The Ultra 9 has 192 KB L1 per core, 3 MB L2 per core, and 36 MB shared L3. The Xeon's larger L3 cache aligns with its server role, while the Ultra 9's larger per-core L1 and L2 support its single-thread performance.

Process node differences matter here. The 3 nm TSMC process for the Ultra 9 versus the 5 nm Intel process for the Xeon contributes to the desktop chip's clock advantage and power efficiency. The Ultra 9 has a 65 W TDP, while the Xeon has a 210 W TDP. The Xeon's higher power budget supports its eight-channel memory controller and 88 PCIe Gen 5 lanes. The Ultra 9 offers 20 PCIe Gen 5 lanes.

The Xeon's socket is Intel Socket 4710, while the Ultra 9 uses Intel Socket 1851. These are not interchangeable platforms. The Xeon's release date is February 24, 2025, while the Ultra 9 was released December 31, 2024. Both are currently Active in production.

Specification Differences

| Specification | Intel Core Ultra 9 285 | Intel Xeon 6724P |

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

| Cores | 24 | 16 |

| Threads | 24 | 32 |

| Base clock | 2.50 GHz | 3.60 GHz |

| Boost clock | 5.60 GHz | 4.30 GHz |

| TDP | 65 W | 210 W |

| Socket | Intel Socket 1851 | Intel Socket 4710 |

| Architecture | Arrow Lake | Granite Rapids |

| Process node | 3 nm | 5 nm |

| Foundry | TSMC | Intel |

| L1 cache | 192 KB (per core) | 112 KB (per core) |

| L2 cache | 3 MB (per core) | 2 MB (per core) |

| L3 cache | 36 MB (shared) | 72 MB (shared) |

| Memory bus | Dual-channel | Eight-channel |

| Memory bandwidth | 102.4 GB/s | 409.6 GB/s |

| PCIe | Gen 5, 20 lanes | Gen 5, 88 lanes |

| Integrated graphics | Arc Xe-LPG 64EU | N/A |

| Market segment | Desktop | Server/Workstation |

| Launch MSRP | $579 | $3622 |

| Part number | SRQD4 | SRVUA |

The Xeon's eight-channel memory bus and 409.6 GB/s bandwidth represent a fourfold advantage over the Ultra 9's dual-channel 102.4 GB/s. The Xeon also offers 88 PCIe lanes versus 20, making it the clear choice for systems with many expansion cards or NVMe drives. The Ultra 9 counters with a 5.60 GHz boost clock, integrated graphics, and a substantially lower TDP. Both processors support DDR5 and ECC memory. Neither has an unlocked multiplier.