Intel Core Ultra 5 235 vs Intel Core Ultra X9 388H Comparison

Head-to-Head Benchmarks

The benchmark data splits these two Intel processors almost evenly, with the Core Ultra 5 235 taking 9 wins and the Core Ultra X9 388H taking 8. The decisive factor is workload type: the desktop part dominates in specialized single-thread and data-oriented tasks, while the mobile part crushes in the Cinebench rendering suite.

The single largest margin belongs to the Core Ultra X9 388H in Cinebench R20 multicore, where it scores 13,101 versus 6,202 for the Core Ultra 5 235, a 52.7% advantage. The same pattern appears in Cinebench R15 multicore (2,955 vs 1,488, a 49.6% lead) and Cinebench R20 singlecore (1,849 vs 875, another 52.7% lead). In Cinebench R23 multicore, the X9 388H extends its winning streak with 18,911 versus 14,769, a 21.9% gap. The single-core Cinebench R23 margin narrows to just 5.2%, with the X9 388H posting 2,200.5 against 2,085.

The Core Ultra 5 235 answers in the PassMark suite. Its most decisive victory comes in random string sorting, where it scores 48,980 versus 44,010, an 11.3% edge. Extended instructions follow at 9.4% (32,752 vs 29,943), and data compression at 8% (390,711 vs 361,763). The desktop chip also wins in single-thread PassMark by 5.5% (4,516 vs 4,280), floating point math by 4.8% (117,951 vs 112,550), find prime numbers by 3.6% (371 vs 358), and data encryption by 2.8% (29,293 vs 28,490). The multithread PassMark score goes to the Core Ultra 5 235 as well, 37,816 versus 36,811, a 2.7% margin.

The X9 388H takes two additional PassMark disciplines: integer math (90,882 vs 87,948, a 3.2% lead) and physics (3,226 vs 2,570, a 20.3% lead). The physics result is particularly notable because it is the second-largest proportional win for either chip, trailing only the Cinebench multicore blowouts.

Overall average benchmark scores reflect the split: the Core Ultra 5 235 averages 46,062, placing it in the 89th percentile of all CPUs, while the X9 388H averages 44,466, sitting in the 88th percentile. The nearest rivals for the Core Ultra 5 235 include the AMD Ryzen AI 9 HX 375 (average 46,030, a 0.1% delta) and the Intel Core i9-13900HX (average 46,098, a 0.1% delta). The X9 388H's closest competitors are the AMD Ryzen 5 7500X3D (average 44,573, a 0.2% delta) and the Intel Core i9-13950HX (average 44,342, a 0.3% delta).

Architecture Differences

The two processors come from different generations and target different platforms. The Core Ultra 5 235 belongs to Core Ultra Series 2, built on Arrow Lake-S architecture for desktop systems with an Intel Socket 1851. The Core Ultra X9 388H is from Core Ultra Series 3, using Panther Lake architecture for mobile devices with an Intel BGA 2540 socket.

Both use a 3 nm process node, but the foundries differ. The Core Ultra 5 235 is fabricated by TSMC, while the X9 388H is fabricated by Intel. The desktop chip carries 17,800 million transistors on a 243 mm² die, whereas the database records no transistor count or die size for the mobile part.

Core configuration favors the X9 388H in raw count: 16 cores and 16 threads versus 14 cores and 14 threads. Neither chip supports hyperthreading, as threads equal cores in both cases. The L1 cache is identical at 192 KB per core, and L2 is also the same at 3 MB per core. The L3 cache differs: the Core Ultra 5 235 has 24 MB shared, while the X9 388H has 18 MB shared.

Clock speeds tell a different story. The Core Ultra 5 235 has a base clock of 3.40 GHz and a boost clock of 5.00 GHz. The X9 388H starts lower at 2.10 GHz base but boosts slightly higher to 5.10 GHz. The thermal design power (TDP) is dramatically different: 65 watts for the desktop part versus 25 watts for the mobile part, reflecting their intended environments.

Memory support diverges as well. The Core Ultra 5 235 uses DDR5 with dual-channel memory and a bandwidth of 102.4 GB/s. The X9 388H uses LPDDR5X, also dual-channel, but with a higher bandwidth of 153.6 GB/s. Neither supports ECC memory.

PCIe lanes are a major differentiator. The Core Ultra 5 235 provides Gen 5 with 20 lanes from the CPU, while the X9 388H offers only Gen 5 with 4 lanes. Integrated graphics also differ: the Core Ultra 5 235 has Arc Xe-LPG Graphics with 24 execution units, while the X9 388H has Arc B390.

Release dates are separated by roughly a year. The Core Ultra 5 235 launched on January 6, 2025, with a launch MSRP of $257. The X9 388H launched on January 4, 2026, and the database has no launch MSRP for it. Both are listed as Active in production, and neither has an unlocked multiplier. The part numbers are SRQAS for the desktop chip and SA4QWQ9EK for the mobile chip.

FAQ

Q: Which processor wins in Cinebench multicore tests?

A: The Intel Core Ultra X9 388H wins every Cinebench multicore test. The margins are 49.6% in R15, 52.7% in R20, and 21.9% in R23.

Q: Does the Core Ultra 5 235 win any benchmark at all?

A: Yes, it wins 9 of the 17 head-to-head tests. Its wins include PassMark data compression, data encryption, extended instructions, find prime numbers, floating point math, multithread, random string sorting, and single-thread tests.

Q: How close are the average benchmark scores?

A: The Core Ultra 5 235 averages 46,062, and the X9 388H averages 44,466. That is a difference of 1,596 points, roughly 3.5% in favor of the desktop chip.

Q: What is the TDP difference?

A: The Core Ultra 5 235 has a TDP of 65 watts. The Core Ultra X9 388H has a TDP of 25 watts. The mobile part draws less than half the power budget.

Q: Which chip has more cores?

A: The Core Ultra X9 388H has 16 cores and 16 threads. The Core Ultra 5 235 has 14 cores and 14 threads. Neither supports simultaneous multithreading.

Q: How does memory bandwidth compare?

A: The X9 388H has higher memory bandwidth at 153.6 GB/s, while the Core Ultra 5 235 achieves 102.4 GB/s. Both use dual-channel memory, but the X9 388H uses LPDDR5X and the Core Ultra 5 235 uses DDR5.

Specification Differences

| Specification | Core Ultra 5 235 | Core Ultra X9 388H |

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

| Cores | 14 | 16 |

| Threads | 14 | 16 |

| Base clock | 3.40 GHz | 2.10 GHz |

| Boost clock | 5.00 GHz | 5.10 GHz |

| TDP | 65 W | 25 W |

| Socket | Intel Socket 1851 | Intel BGA 2540 |

| Architecture | Arrow Lake | Panther Lake |

| Generation | Ultra 5 (Arrow Lake) | Ultra X9 (Panther Lake-H) |

| Foundry | TSMC | Intel |

| Transistors | 17,800 million | Not recorded |

| Die size | 243 mm² | Not recorded |

| L3 cache | 24 MB shared | 18 MB shared |

| Memory support | DDR5 | LPDDR5X |

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

| PCIe lanes | Gen 5, 20 lanes | Gen 5, 4 lanes |

| Integrated graphics | Arc Xe-LPG Graphics 24EU | Arc B390 |

| Market segment | Desktop | Mobile |

| Release date | 2025-01-06 | 2026-01-04 |

| Launch MSRP | $257 | Not recorded |

| Part number | SRQAS | SA4QWQ9EK |

The Verdict

The data points to a clear split based on intended usage. For rendering workloads, the Core Ultra X9 388H is the definitive choice. Its Cinebench margins are overwhelming, particularly in multicore tests where it leads by 49.6% to 52.7% depending on the test version. The physics benchmark also favors the mobile chip by 20.3%, and integer math goes its way by 3.2%. Anyone whose primary tasks involve Cinebench-class rendering or physics simulation should select the X9 388H.

For general computing tasks measured by PassMark, the Core Ultra 5 235 holds the advantage. It wins in the majority of PassMark disciplines, including the overall multithread score (37,816 vs 36,811) and single-thread score (4,516 vs 4,280). Its wins in data compression, encryption, extended instructions, and random string sorting indicate strength in data-heavy workloads. The desktop chip also posts a higher average benchmark score (46,062 vs 44,466) and a higher percentile ranking (89th vs 88th).

The architecture differences reinforce the use-case split. The X9 388H packs more cores (16 vs 14) and a slightly higher boost clock (5.10 GHz vs 5.00 GHz), which explains its rendering dominance. The Core Ultra 5 235 compensates with a much higher base clock (3.40 GHz vs 2.10 GHz), more L3 cache (24 MB vs 18 MB), and drastically more PCIe lanes (20 vs 4). The desktop chip also has a higher TDP envelope, 65 watts versus 25 watts, allowing sustained performance at the cost of power efficiency.

The X9 388H's higher memory bandwidth (153.6 GB/s vs 102.4 GB/s) does not translate into PassMark victories in memory-sensitive tests, as the Core Ultra 5 235 wins data compression and random string sorting. The mobile chip's lower TDP makes it suitable for systems where power is constrained, but the desktop part's performance in the majority of PassMark tests suggests better all-around throughput when power is available.

Where Each One Wins

The Core Ultra X9 388H wins in Cinebench R15 singlecore (309.5 vs 210, a 32.1% lead), Cinebench R15 multicore (2,955 vs 1,488, a 49.6% lead), Cinebench R20 singlecore (1,849 vs 875, a 52.7% lead), Cinebench R20 multicore (13,101 vs 6,202, a 52.7% lead), Cinebench R23 singlecore (2,200.5 vs 2,085, a 5.2% lead), Cinebench R23 multicore (18,911 vs 14,769, a 21.9% lead), PassMark integer math (90,882 vs 87,948, a 3.2% lead), and PassMark physics (3,226 vs 2,570, a 20.3% lead). These are the tests where the mobile part's extra cores and higher boost clock deliver measurable advantages.

The Core Ultra 5 235 wins in PassMark data compression (390,711 vs 361,763, an 8% lead), PassMark data encryption (29,293 vs 28,490, a 2.8% lead), PassMark extended instructions (32,752 vs 29,943, a 9.4% lead), PassMark find prime numbers (371 vs 358, a 3.6% lead), PassMark floating point math (117,951 vs 112,550, a 4.8% lead), PassMark multithread (37,816 vs 36,811, a 2.7% lead), PassMark random string sorting (48,980 vs 44,010, an 11.3% lead), and PassMark single-thread (4,516 vs 4,280, a 5.5% lead). These wins cover a broader range of everyday computation, from encryption to sorting to floating point operations.

The final tally favors the Core Ultra 5 235 by a single win, 9 against 8. But the magnitude of the X9 388H's Cinebench victories outweighs the count in rendering scenarios, while the Core Ultra 5 235's consistent PassMark edge makes it the stronger general-purpose processor. Users should choose based on their dominant workload: rendering and physics favor the X9 388H, while data processing, encryption, and general multithreaded tasks favor the Core Ultra 5 235.