Intel Core Ultra 5 235A vs Intel Core Ultra X9 378H Comparison

The Intel Core Ultra 5 235A and the Intel Core Ultra X9 378H are two very different processors that end up in a statistical dead heat. The desktop-oriented Ultra 5 235A wins the majority of benchmark comparisons, but the mobile Ultra X9 378H counters with significant victories in specific workloads. The data reveals a 15-to-2 win split in favor of the Ultra 5 235A, yet the margins are often razor-thin, making the choice dependent on use case rather than raw dominance.

Head-to-Head Benchmarks

The most striking pattern in the head-to-head data is how closely these two CPUs match each other in standard rendering workloads. In Cinebench R23 multicore, the Ultra 5 235A scores 32,633 against the Ultra X9 378H’s 32,553, a lead of just 0.2%. The single-core results tell the same story: 4,607 versus 4,595, a 0.3% edge for the desktop chip. This near-parity extends across the entire Cinebench suite, with R15 and R20 multicore deltas of 0.2% and single-core deltas of 0.3-0.4%. For all practical purposes, these two processors deliver identical performance in CPU-bound rendering tasks.

The Passmark suite reveals where the Ultra 5 235A builds its advantage. The largest single win comes in random string sorting, where it scores 49,489 against 44,648—a commanding 10.8% lead. Prime number finding shows a 9.8% gap (392 vs 357), and floating point math favors the desktop chip by 3.7% (118,778 vs 114,500). Data compression also goes to the Ultra 5 235A with a 1.9% margin (393,800 vs 386,591), while data encryption and extended instructions each show a 1% advantage. In single-threaded Passmark tests, the Ultra 5 235A leads by 2.3% (4,557 vs 4,453), reinforcing its slight per-core superiority.

The Ultra X9 378H does not go down without a fight. Its most dramatic victory comes in Passmark physics, where it scores 3,404 against the Ultra 5 235A’s 2,437—a massive 28.4% advantage. This is the single largest delta in the entire comparison and suggests a fundamentally different performance profile. The mobile chip also wins Passmark integer math with 92,603 versus 88,626, a 4.3% lead. These two wins, while few in number, are substantial in magnitude.

Overall average benchmark scores tell a slightly different story than the head-to-head matchups. The Ultra 5 235A averages 48,201 points, placing it in the 90th percentile of all CPUs, while the Ultra X9 378H averages 47,468 points at the 89th percentile. The desktop chip’s nearest rival is the AMD Ryzen AI Max PRO 380 at 48,171 (0.1% behind), while the mobile chip’s closest competitor is the AMD Ryzen 9 PRO 5945 at 47,527 (0.1% ahead of the Ultra X9). These percentile rankings confirm that both processors sit in the upper echelon of available hardware, with the Ultra 5 235A holding a slight overall edge.

Architecture Differences

The two processors represent different generations and manufacturing philosophies. The Ultra 5 235A belongs to the Core Ultra Series 2, built on the Arrow Lake architecture with the Arrow Lake-S codename. It uses a 3 nm process node from TSMC, packs 17,800 million transistors onto a 243 mm² die, and fits the Intel Socket 1851. The Ultra X9 378H, by contrast, is from the Core Ultra Series 3 with the Panther Lake codename and Panther Lake-H generation. It also uses a 3 nm process node, but the foundry is Intel rather than TSMC, and transistor count and die size are not disclosed.

Core configurations differ meaningfully. The Ultra 5 235A offers 14 cores and 14 threads, while the Ultra X9 378H provides 16 cores and 16 threads. Neither chip supports simultaneous multithreading, so thread counts equal core counts. The base clock speeds diverge sharply: the desktop chip runs at 3.40 GHz base, while the mobile chip sits at 2.00 GHz base. Both boost to 5.00 GHz, which explains why single-threaded performance is so similar despite the base clock disparity.

Cache hierarchies show structural differences. Both chips share the same 192 KB L1 cache per core, but the L2 cache differs: 3 MB per core on the Ultra 5 235A versus 2.5 MB per core on the Ultra X9 378H. The L3 cache also favors the desktop chip with 24 MB shared versus 18 MB shared on the mobile part. These cache advantages likely contribute to the Ultra 5 235A’s wins in memory-sensitive workloads like random string sorting and data compression.

Memory support presents a clear split. The Ultra 5 235A supports DDR5 with dual-channel memory and 102.4 GB/s bandwidth, while the Ultra X9 378H uses LPDDR5X with dual-channel memory and a higher 153.6 GB/s bandwidth. The mobile chip’s faster memory bandwidth may explain its integer math and physics victories, which can benefit from rapid data movement. PCIe connectivity also differs: the desktop chip offers Gen 5 with 20 CPU lanes, while the mobile chip provides Gen 5 with only 4 CPU lanes.

Integrated graphics take different paths. The Ultra 5 235A features Arc Xe-LPG Graphics with 24 execution units, while the Ultra X9 378H includes the more capable Arc B390. The mobile part’s superior iGPU aligns with its laptop-oriented market segment, while the desktop chip assumes a discrete graphics card. Power envelopes differ dramatically as well: the Ultra 5 235A carries a 65 W TDP, while the Ultra X9 378H draws just 25 W.

The Verdict

The benchmark data points to the Intel Core Ultra 5 235A as the better overall processor for most users. It wins 15 of 17 head-to-head comparisons, holds a higher average benchmark score (48,201 vs 47,468), and ranks in the 90th percentile versus the Ultra X9 378H’s 89th. The desktop chip’s advantages in single-threaded performance (2.3% in Passmark), cache-heavy workloads, and rendering tasks make it the safer choice for general productivity and content creation.

However, the Ultra X9 378H cannot be dismissed. Its 28.4% lead in Passmark physics is extraordinary, and its 4.3% edge in integer math shows real strength in specific computational patterns. The mobile chip also achieves this with a 25 W TDP versus 65 W, making it dramatically more power-efficient. For workloads that stress physics calculations or integer operations—particularly in constrained thermal environments—the Ultra X9 378H is the data-supported pick.

The near-identical Cinebench scores complicate the narrative. A 0.2% difference in multicore and 0.3% in single-core is within run-to-run variance, meaning these chips are effectively tied for rendering performance. The choice ultimately hinges on whether the user prioritizes the Ultra 5 235A’s broader set of wins or the Ultra X9 378H’s targeted strengths and lower power draw.

FAQ

Q: Which processor has more cores?

A: The Intel Core Ultra X9 378H has 16 cores and 16 threads, while the Intel Core Ultra 5 235A has 14 cores and 14 threads.

Q: What is the biggest performance gap between the two?

A: The largest delta is in Passmark physics, where the Ultra X9 378H leads by 28.4% with a score of 3,404 versus 2,437.

Q: How do the chips compare in single-core performance?

A: The Ultra 5 235A leads in Cinebench R23 single-core with 4,607 versus 4,595 (0.3% ahead) and in Passmark single-thread with 4,557 versus 4,453 (2.3% ahead).

Q: Do both processors support the same memory type?

A: No. The Ultra 5 235A supports DDR5 with 102.4 GB/s bandwidth, while the Ultra X9 378H uses LPDDR5X with 153.6 GB/s bandwidth.

Q: What are the power consumption figures?

A: The Ultra 5 235A has a 65 W TDP, while the Ultra X9 378H has a 25 W TDP.

Q: Which chip has a higher overall benchmark percentile?

A: The Ultra 5 235A sits in the 90th percentile of all CPUs, while the Ultra X9 378H is in the 89th percentile.

Where Each One Wins

The Intel Core Ultra 5 235A dominates in most measurable categories. It wins all six Cinebench tests across R15, R20, and R23, though by narrow margins of 0.2-0.4%. In Passmark, it takes data compression (1.9% lead), data encryption (1%), extended instructions (1%), prime number finding (9.8%), floating point math (3.7%), multithreaded performance (0.2%), random string sorting (10.8%), and single-threaded tests (2.3%). The desktop chip’s cache advantage—3 MB L2 per core and 24 MB L3 versus 2.5 MB and 18 MB—likely drives its wins in sorting and compression tasks. Its higher base clock of 3.40 GHz versus 2.00 GHz contributes to better responsiveness in latency-sensitive workloads.

The Intel Core Ultra X9 378H wins two specific tests, but they are notable. Passmark physics shows a 28.4% advantage, indicating superior performance in simulation and physics-heavy applications. Passmark integer math gives it a 4.3% lead, suggesting strength in encryption, compression, and other integer-based algorithms. The mobile chip’s higher memory bandwidth (153.6 GB/s vs 102.4 GB/s) may explain these wins, as physics calculations and integer math often require rapid data access. Its 16 cores versus 14 also provide more parallel processing capacity for these workloads.

For power-sensitive deployments, the Ultra X9 378H is the clear winner. Its 25 W TDP versus 65 W means it delivers comparable Cinebench scores at less than half the power draw. This makes it the logical choice for laptops, compact systems, or any environment where thermal output and battery life matter. The Ultra 5 235A, with its Socket 1851 and desktop market segment, targets traditional tower builds where power is less constrained.

Specification Differences

The two processors differ across nearly every major specification category. Core and thread counts diverge: 14/14 for the Ultra 5 235A versus 16/16 for the Ultra X9 378H. Base clocks are starkly different at 3.40 GHz versus 2.00 GHz, though both boost to 5.00 GHz. The TDP gap is substantial at 65 W versus 25 W. The desktop chip uses Intel Socket 1851, while the mobile chip uses Intel BGA 2540.

Architecture and manufacturing separate the two. The Ultra 5 235A uses Arrow Lake with the Arrow Lake-S codename, fabricated by TSMC. The Ultra X9 378H uses Panther Lake with the Panther Lake-H codename, fabricated by Intel. Both use a 3 nm process, but the desktop chip discloses 17,800 million transistors and a 243 mm² die size, while the mobile chip lists neither. Cache sizes differ in L2 (3 MB vs 2.5 MB per core) and L3 (24 MB vs 18 MB). Memory support splits between DDR5 and LPDDR5X, with bandwidth of 102.4 GB/s versus 153.6 GB/s. PCIe lanes differ significantly at 20 versus 4. Integrated graphics range from Arc Xe-LPG 24EU to Arc B390. The Ultra 5 235A has a launch MSRP of $269, while the Ultra X9 378H has no listed launch MSRP. Both are currently active production parts, and neither has an unlocked multiplier.