Intel Core Ultra 9 285K vs Intel Core Ultra X7 358H Comparison

Head-to-Head Benchmarks

The recorded data shows a comprehensive sweep. The Intel Core Ultra 9 285K wins all 17 head-to-head benchmark comparisons against the Intel Core Ultra X7 358H. No recorded test favors the mobile chip.

The largest single-core margin appears in Cinebench R20, where the 285K scores 3388 against 1695, a 99.9% advantage. That is nearly double the output. Cinebench R15 single-core shows a smaller but still decisive gap: 359 versus 301.5, a 19.1% lead. The R23 single-core test records 2377 against 2080, a 14.3% edge. PassMark single-thread results show 5087 versus 4124, a 23.4% lead. The pattern across all single-threaded workloads is consistent: the desktop part holds a meaningful but not overwhelming advantage in raw per-core speed.

Multi-core results tell a different story, with far larger deltas. Cinebench R23 multi-core shows 42522 versus 18747, a 126.8% lead. R15 multi-core delivers 6494 versus 3027, a 114.5% gap. R20 multi-core records 24003 versus 12011, a 99.8% difference. PassMark multithread shows 67260 versus 33802, a 99% lead. The desktop chip roughly doubles the mobile chip in most heavily threaded workloads.

Specialized workloads amplify the gap further. PassMark data compression shows 790052 versus 332508, a 137.6% advantage. Random string sorting records 94927 versus 40357, a 135.2% lead. Extended instructions deliver 62277 versus 27274, a 128.3% gap. Data encryption shows 57745 versus 26046, a 121.7% edge. Floating point math records 224324 versus 103842, a 116% lead. Integer math shows 172379 versus 83147, a 107.3% advantage. Prime number finding delivers 541 versus 337, a 60.5% lead. PassMark physics shows 3938 versus 3021, a 30.4% gap.

The closest margins appear in single-core tests, where the 285K leads by roughly 14% to 23%. The widest margins appear in data compression and random string sorting, where the desktop chip more than doubles the mobile chip. The data suggests the 285K scales far better with additional cores and cache.

Architecture Differences

The two processors come from different segments and generations. The Intel Core Ultra 9 285K belongs to the Core Ultra Series 2, uses the Arrow Lake architecture with the Arrow Lake-S codename, and targets the desktop market. The Intel Core Ultra X7 358H belongs to the Core Ultra Series 3, uses the Panther Lake codename with the Panther Lake-H generation label, and targets the mobile market.

Both use a 3 nm process node, but they come from different foundries. The 285K is fabricated by TSMC, while the 358H is fabricated by Intel. The 285K packages 17,800 million transistors on a 243 mm² die. The 358H has no recorded transistor count or die size in the database.

Core counts differ substantially. The 285K provides 24 cores and 24 threads. The 358H provides 16 cores and 16 threads. Neither chip uses simultaneous multithreading, so threads equal cores in both cases. The 285K runs a base clock of 3.70 GHz with a boost clock of 5.70 GHz. The 358H runs a base clock of 1.90 GHz with a boost clock of 4.80 GHz. The desktop chip also carries a 125 W TDP against 25 W for the mobile part, and the 285K has an unlocked multiplier while the 358H does not.

Cache hierarchies differ in the L3 level only. Both processors record 192 KB of L1 per core and 3 MB of L2 per core. The 285K shares 36 MB of L3, while the 358H shares 18 MB. That is exactly half the L3 capacity.

Memory support also diverges. The 285K uses DDR5 with a dual-channel bus and 102.4 GB/s of bandwidth. The 358H uses LPDDR5X with a dual-channel bus and 153.6 GB/s of bandwidth. The mobile chip has a higher recorded memory bandwidth despite its lower power envelope. The 285K supports ECC memory; the 358H does not.

PCIe connectivity differs sharply. The 285K provides Gen 5 with 20 lanes from the CPU. The 358H provides Gen 5 with 4 lanes from the CPU. The desktop chip has five times the CPU-attached lane count.

Integrated graphics also differ. The 285K carries Arc Xe-LPG Graphics with 64 execution units. The 358H carries Arc B390. The database records no benchmark comparison for these iGPUs, so no performance conclusion is possible from the data.

The 285K uses Intel Socket 1851, while the 358H uses Intel BGA 2540. The desktop part launched on 2024-10-23 with a launch MSRP of $589. The mobile part launched on 2026-01-04 and has no recorded launch MSRP.

The Verdict

The benchmark data points to a clear split by use case. The Intel Core Ultra 9 285K dominates every recorded workload in this comparison. Its average benchmark score of 83807 places it at the 96th percentile of all CPUs in the database. The Intel Core Ultra X7 358H records an average score of 40967 at the 87th percentile. The desktop chip outperforms the mobile chip by roughly 105% in average score.

The nearest rivals for the 285K are all server or high-end desktop parts. The Intel Core Ultra 9 290K Plus scores 84003, a 0.2% advantage over the 285K. The AMD EPYC 4584PX scores 83090, a 0.9% deficit. The AMD EPYC 9135 scores 82980, a 1% deficit. The AMD EPYC 7F72 scores 85072, a 1.5% advantage. The 285K sits in a tight cluster of high-end processors.

The nearest rivals for the 358H are mobile and lower-power parts. The AMD Ryzen AI 5 PRO 440 scores 41208, a 0.6% advantage. The Intel Core Ultra 7 356H scores 41215, a 0.6% advantage. The AMD Ryzen AI 5 PRO 435G scores 40718, a 0.6% deficit. The Intel Core Ultra 7 366H scores 41263, a 0.7% advantage. The 358H sits in a cluster of mid-range mobile processors.

The 285K belongs in a desktop system where power draw, socket type, and cooling are not constraints. The 358H belongs in a mobile system where a 25 W TDP and BGA packaging are requirements. The data does not show any workload where the 358H wins.

FAQ

Q: Which processor wins in Cinebench R23 multi-core?

A: The Intel Core Ultra 9 285K scores 42522 against 18747 for the Intel Core Ultra X7 358H, a 126.8% lead.

Q: How close are the two in single-threaded performance?

A: The closest margins appear in single-core tests. The 285K leads by 14.3% in Cinebench R23 single-core (2377 versus 2080), by 19.1% in R15 single-core (359 versus 301.5), by 23.4% in PassMark single-thread (5087 versus 4124), and by 99.9% in R20 single-core (3388 versus 1695).

Q: Do both processors have the same number of cores?

A: No. The 285K has 24 cores and 24 threads. The 358H has 16 cores and 16 threads.

Q: Which processor supports ECC memory?

A: The 285K supports ECC memory. The 358H does not.

Q: What memory types do the two processors support?

A: The 285K supports DDR5 with 102.4 GB/s of bandwidth. The 358H supports LPDDR5X with 153.6 GB/s of bandwidth.

Q: What is the 358H's market segment?

A: The 358H is a mobile processor with a 25 W TDP and Intel BGA 2540 socket. The 285K is a desktop processor with a 125 W TDP and Intel Socket 1851.

Where Each One Wins

The Intel Core Ultra 9 285K wins every recorded benchmark, so the use-case split comes from platform characteristics rather than benchmark victories.

For desktop workloads, the 285K delivers 24 cores, 24 threads, a 5.70 GHz boost clock, and 36 MB of shared L3. It supports DDR5 memory, ECC memory, Gen 5 PCIe with 20 CPU lanes, and an unlocked multiplier. It has a 243 mm² die with 17,800 million transistors. These characteristics suit heavy multi-threaded rendering, data compression, encryption, and floating-point workloads. The data shows it doubles or nearly doubles the mobile chip in most of those tests.

For mobile workloads, the 358H offers a 25 W TDP, which fits thin-and-light systems. It uses LPDDR5X memory with 153.6 GB/s of bandwidth, higher than the desktop chip's 102.4 GB/s. It has an integrated Arc B390 GPU. The database records no iGPU benchmark, but the memory bandwidth advantage suggests the integrated graphics have access to faster memory. The 358H also has a later launch date of 2026-01-04.

The 285K has a higher average benchmark score of 83807 against 40967, and a higher percentile ranking of 96 against 87. The 358H's nearest rivals are all mobile or low-power processors, which confirms its positioning in the mobile segment. The 285K's nearest rivals are server processors, which confirms its high-end desktop positioning.

Specification Differences

| Specification | Intel Core Ultra 9 285K | Intel Core Ultra X7 358H |

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

| Series | Core Ultra Series 2 | Core Ultra Series 3 |

| Cores | 24 | 16 |

| Threads | 24 | 16 |

| Base clock | 3.70 GHz | 1.90 GHz |

| Boost clock | 5.70 GHz | 4.80 GHz |

| TDP | 125 W | 25 W |

| Socket | Intel Socket 1851 | Intel BGA 2540 |

| Codename | Arrow Lake-S | Panther Lake |

| Generation | Ultra 9 (Arrow Lake) | Ultra X7 (Panther Lake-H) |

| Foundry | TSMC | Intel |

| Transistors | 17,800 million | Not recorded |

| Die size | 243 mm² | Not recorded |

| L3 cache | 36 MB (shared) | 18 MB (shared) |

| Memory support | DDR5 | LPDDR5X |

| Memory bandwidth | 102.4 GB/s | 153.6 GB/s |

| ECC memory | Yes | No |

| PCIe | Gen 5, 20 lanes (CPU only) | Gen 5, 4 lanes (CPU only) |

| Integrated graphics | Arc Xe-LPG Graphics 64EU | Arc B390 |

| Market segment | Desktop | Mobile |

| Release date | 2024-10-23 | 2026-01-04 |

| Multiplier unlocked | Yes | No |

| Part number | SRQD5 | SA4RAQ9ET |

| Average benchmark score | 83807 | 40967 |

| Percentile vs all CPUs | 96 | 87 |

The two chips share the same L1 and L2 cache sizes per core (192 KB and 3 MB respectively), the same 3 nm process node, the same dual-channel memory bus, and the same manufacturer. Everything else in the recorded data differs.