Intel Core 9 273PQE vs Intel Core Ultra 5 235H Comparison

Where Each One Wins

The benchmark data splits cleanly along workload type. The Intel Core 9 273PQE wins 16 of 17 recorded head-to-head tests, while the Intel Core Ultra 5 235H takes a single victory. That one win, however, is instructive: the Ultra 5 235H leads in PassMark find prime numbers by 17.2%, scoring 239 against 198. This is a workload that rewards the Ultra 5's architecture and cache layout, and it stands as the only category where the mobile chip outpaces the desktop part.

Every Cinebench test, both single-core and multi-core, goes to the Core 9 273PQE by roughly 53%. The Cinebench R23 multi-core score of 39190 versus 25598 represents a 53.1% advantage, and the single-core gap in the same test is identical at 53.1%, with 5532 against 3613. The consistency of that delta across all six Cinebench runs (R15, R20, and R23, each in single and multi-core) indicates a fundamental throughput advantage that scales linearly across rendering workloads.

PassMark results reinforce the split. The Core 9 273PQE dominates integer math by 121.7% (164629 versus 74247), data compression by 94% (585752 versus 301979), and extended instructions by 65.9% (38743 versus 23354). Floating point math favors the desktop chip by 34.3% (125546 versus 93509). Multi-thread performance shows a 53.2% lead (46107 versus 30091), while physics simulation trails at a 38.7% edge (2754 versus 1985). Random string sorting, a memory-latency-sensitive test, goes to the Core 9 by 46.8% (53167 versus 36208).

The narrowest desktop win is in PassMark single-thread, where the Core 9 273PQE leads by only 4.9% (4573 versus 4359). This is the closest contest in the entire dataset, suggesting that the two chips have similar per-core efficiency at lower thread counts. The data encryption test shows a moderate 28.2% gap (29636 versus 23121).

For use-case planning, the Core 9 273PQE is the clear choice for rendering, compression, encryption, and math-heavy parallel workloads. The Ultra 5 235H's single win in prime number search indicates an edge in certain integer iteration patterns, but that does not offset the 53% multi-core deficit in Cinebench or the 94% gap in data compression.

Architecture Differences

The two processors come from different Intel design families with distinct manufacturing and packaging strategies. The Core 9 273PQE is a desktop part on Intel Socket 1700, built on a 10 nm process at Intel's own foundry, and carries the Bartlett Lake codename. The Ultra 5 235H is a mobile processor on Intel BGA 2049, fabricated on a 3 nm process at TSMC, and uses the Arrow Lake architecture with the Arrow Lake-H codename. The 3 nm node gives the mobile chip a density advantage, but the desktop chip compensates with higher power delivery and clock headroom.

Core and thread counts differ substantially. The Core 9 273PQE has 12 cores and 24 threads, enabled by Hyper-Threading-style simultaneous multithreading. The Ultra 5 235H has 14 cores but only 14 threads, meaning it lacks multithreading on its performance cores. Despite having two more physical cores, the Ultra 5 cannot match the desktop part's thread count, which explains much of the multi-core benchmark gap.

Cache hierarchies are structured differently. The Core 9 273PQE allocates 80 KB of L1 per core, 2 MB of L2 per core, and 36 MB of shared L3. The Ultra 5 235H uses 192 KB of L1 per core, 3 MB of L2 per core, and only 18 MB of shared L3. The desktop chip has double the L3 capacity, which benefits large working sets in compression and rendering. The mobile chip has larger per-core L1 and L2 allocations, which may explain its win in the prime number search where per-core iteration state fits in private caches.

Clock speeds favor the desktop part. The Core 9 273PQE has a 3.40 GHz base clock and a 5.90 GHz boost clock. The Ultra 5 235H runs at 2.40 GHz base and 5.00 GHz boost. The 0.9 GHz boost advantage for the desktop chip contributes to its single-thread lead, though the modest 4.9% PassMark single-thread gap suggests the Ultra 5's newer process node narrows the per-clock efficiency difference.

Memory support diverges. Both use dual-channel buses, but the Core 9 273PQE supports DDR4 and DDR5, with 89.6 GB/s bandwidth and ECC memory enabled. The Ultra 5 235H supports DDR5 and LPDDR5X, with 102.4 GB/s bandwidth and no ECC. The mobile chip's higher memory bandwidth is notable, yet it does not translate into a benchmark win in memory-sensitive tests like random string sorting, which the desktop chip wins by 46.8%.

PCIe lanes also differ: the Core 9 273PQE provides Gen 5 with 16 CPU lanes, while the Ultra 5 235H provides Gen 5 with 8 CPU lanes. Integrated graphics are distinct as well: the desktop part uses UHD Graphics 770, while the mobile part uses Arc Graphics 140T. The desktop chip carries a 125 W TDP versus 28 W for the mobile chip, a power envelope difference that explains the desktop's clock and throughput advantages.

Head-to-Head Benchmarks

The most decisive result in the dataset is PassMark integer math, where the Core 9 273PQE scores 164629 against 74247, a 121.7% lead. This is the largest percentage gap in any test and reflects the desktop chip's combination of 24 threads, higher clocks, and larger L3 cache.

Data compression shows a 94% gap (585752 versus 301979). The Core 9 273PQE's 36 MB of shared L3 and 24 threads give it a clear edge in workloads that stream large data blocks. Extended instructions follow at 65.9% (38743 versus 23354), indicating better SIMD and vector processing throughput.

Cinebench results are uniform across all six variants. Each test, whether R15, R20, or R23, shows a delta of 53% or 53.1% in favor of the Core 9 273PQE. The R23 multi-core score of 39190 versus 25598 and the R23 single-core score of 5532 versus 3613 both land at 53.1%. This uniformity suggests a fixed throughput ratio between the two chips rather than workload-specific scaling.

PassMark multi-thread shows a 53.2% lead (46107 versus 30091), closely matching the Cinebench multi-core deltas. Floating point math favors the desktop chip by 34.3% (125546 versus 93509). Physics scores show a 38.7% edge (2754 versus 1985). Random string sorting goes to the Core 9 by 46.8% (53167 versus 36208).

Data encryption is the closest multi-core test, with the Core 9 273PQE leading 29636 to 23121, a 28.2% gap. Single-thread PassMark shows the narrowest overall margin at 4.9% (4573 versus 4359). The only reversal is PassMark find prime numbers, where the Ultra 5 235H wins 239 to 198, a 17.2% margin.

The overall average benchmark score places the Core 9 273PQE at 66099, which ranks in the 93rd percentile of all CPUs. The Ultra 5 235H averages 37522, ranking in the 85th percentile. The Core 9 273PQE sits within 0.1% of the Intel Core Ultra 5 250KF Plus (66159) and 0.3% ahead of the AMD Ryzen 9 7950X3D (65914). The Ultra 5 235H sits within 0.3% of the Intel Core i9-13900HK (37425) and 0.6% ahead of the Intel Core Ultra 5 225F (37313).

FAQ

Q: Which processor is faster in multi-core rendering?

A: The Intel Core 9 273PQE leads in every Cinebench multi-core test by 53.1%. Its R23 multi-core score is 39190 versus 25598 for the Ultra 5 235H.

Q: Does the Ultra 5 235H win any benchmark?

A: Yes, it wins PassMark find prime numbers with a score of 239 against 198, a 17.2% advantage for the mobile chip.

Q: How close is single-thread performance?

A: The Core 9 273PQE leads PassMark single-thread by only 4.9% (4573 versus 4359). Cinebench single-core tests show a larger 53% gap (5532 versus 3613 in R23).

Q: What explains the Ultra 5's prime number win despite losing other tests?

A: The Ultra 5 235H has larger per-core L1 and L2 caches (192 KB and 3 MB per core versus 80 KB and 2 MB), which likely benefits the prime number search workload's per-core iteration pattern.

Q: How do the two chips compare in memory bandwidth?

A: The Ultra 5 235H has higher memory bandwidth at 102.4 GB/s versus 89.6 GB/s, but the Core 9 273PQE still wins random string sorting by 46.8% (53167 versus 36208).

Q: Which chip has more threads?

A: The Core 9 273PQE has 24 threads from 12 cores, while the Ultra 5 235H has 14 threads from 14 cores, meaning the mobile chip has no SMT on its cores.

Specification Differences

| Field | Intel Core 9 273PQE | Intel Core Ultra 5 235H |

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

| Cores | 12 | 14 |

| Threads | 24 | 14 |

| Base clock | 3.40 GHz | 2.40 GHz |

| Boost clock | 5.90 GHz | 5.00 GHz |

| TDP | 125 W | 28 W |

| Socket | Intel Socket 1700 | Intel BGA 2049 |

| Architecture | Bartlett Lake | Arrow Lake |

| Codename | Bartlett Lake | Arrow Lake-H |

| Process node | 10 nm | 3 nm |

| Foundry | Intel | TSMC |

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

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

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

| Memory support | DDR4, DDR5 | DDR5, LPDDR5X |

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

| ECC memory | Yes | No |

| PCIe | Gen 5, 16 Lanes (CPU only) | Gen 5, 8 Lanes (CPU only) |

| Integrated graphics | UHD Graphics 770 | Arc Graphics 140T |

| Market segment | Desktop | Mobile |

| Release date | 2026-03-08 | 2025-01-12 |

| Launch MSRP | $589 | None |

The Verdict

The recorded data supports a clear performance hierarchy. The Intel Core 9 273PQE is the stronger processor in 16 of 17 benchmarks, with leads ranging from 4.9% in single-thread PassMark to 121.7% in integer math. Its 93rd percentile ranking, 24 threads, 5.90 GHz boost clock, and 36 MB of L3 cache make it the appropriate choice for desktop workloads that demand parallel throughput, including rendering, compression, encryption, and vector math.

The Intel Core Ultra 5 235H, ranking in the 85th percentile, is the better fit for mobile systems where the 28 W TDP and BGA 2049 socket matter. Its single benchmark win in prime number search, larger per-core caches, and higher memory bandwidth (102.4 GB/s) indicate strengths in specific iteration-heavy and memory-latency-sensitive tasks. Its 14 cores without multithreading limit its multi-core ceiling, as shown by the 53.1% Cinebench deficit.

The choice between these two depends entirely on platform and workload profile. Desktop builders with Socket 1700 motherboards and access to the Core 9 273PQE's 125 W power envelope receive a processor that outperforms the mobile chip by roughly half in most threaded benchmarks. Mobile users who need the Ultra 5 235H's integrated Arc Graphics 140T and LPDDR5X memory support receive a capable processor that wins only in the prime number test, but does so with far lower power consumption. The data does not suggest any scenario where the Ultra 5 235H matches the Core 9 273PQE in multi-core throughput.