Intel Core 9 273PQE vs Intel Core Ultra X7 358H Comparison

Where Each One Wins

The benchmark split is heavily one-sided. The Intel Core 9 273PQE wins 15 of the 17 recorded head-to-head tests, while the Intel Core Ultra X7 358H takes only 2. This is not a close contest by aggregate score either: the Core 9 273PQE posts an average benchmark score of 66099, placing it in the 93rd percentile of all CPUs, while the Ultra X7 358H averages 40967, landing in the 87th percentile.

The Core 9 273PQE dominates everything that stresses raw throughput and sustained multi-core work. In Cinebench R23 multicore, it scores 39190 against 18747 for the Ultra X7 358H, a 109% advantage. That is the single largest margin in the entire comparison. PassMark integer math shows a similar story: 164629 versus 83147, a 98% gap. Data compression also goes decisively to the desktop part, 585752 versus 332508, a 76.2% delta. Extended instructions favor the Core 9 273PQE by 42.1%, and multithread performance by 36.4%.

Single-core performance is also not close. The Core 9 273PQE leads Cinebench R23 single-core by 166% (5532 versus 2080) and R20 single-core by 37.1% (2323 versus 1695). PassMark single-thread shows 4573 versus 4124, a 10.9% edge. Even in floating-point math, where mobile parts sometimes close gaps, the Core 9 273PQE stays ahead by 20.9% (125546 versus 103842).

The Ultra X7 358H wins exactly two tests, and both are narrow. It leads PassMark find prime numbers by 337 versus 198, a 41.2% margin in its favor. It also edges PassMark physics, 3021 versus 2754, an 8.8% advantage. These are isolated wins, not patterns. The physics result is notable because the Ultra X7 358H achieves it despite having far lower aggregate performance, but the prime number test is a specialized workload where the mobile part's architecture appears better suited.

Architecture Differences

The two processors come from different Intel families and target different market segments. The Core 9 273PQE is a desktop part built on Bartlett Lake, using a 10 nm process node with 12 cores and 24 threads. It is a desktop chip on Intel Socket 1700. The Ultra X7 358H is a mobile processor from the Core Ultra Series 3 family, codenamed Panther Lake, using a 3 nm node with 16 cores and 16 threads. It sits in Intel BGA 2540, a mobile socket.

The core and thread counts diverge sharply. The Core 9 273PQE has 12 cores but 24 threads, meaning it supports Hyper-Threading (or an equivalent simultaneous multithreading implementation). The Ultra X7 358H has 16 cores but only 16 threads, so it lacks that capability. This explains much of the multi-core benchmark gap: the desktop part can process two threads per core, effectively doubling its logical core count to 24, while the mobile part stays at 16 logical threads.

Cache layouts also differ. The Core 9 273PQE has 80 KB of L1 per core, 2 MB of L2 per core, and 36 MB of shared L3. The Ultra X7 358H has 192 KB of L1 per core, 3 MB of L2 per core, but only 18 MB of shared L3. The desktop part carries twice the total L3 cache, which helps in workloads with large working sets. The mobile part has larger per-core L1 and L2, which can benefit latency-sensitive single-threaded tasks, but the total cache capacity is far smaller.

Memory support is another major divider. The Core 9 273PQE supports both DDR4 and DDR5 memory in dual-channel configuration, with 89.6 GB/s bandwidth and ECC memory support. The Ultra X7 358H uses LPDDR5X, also dual-channel, but with 153.6 GB/s bandwidth and no ECC support. Despite the mobile part's higher memory bandwidth figure, the desktop part's larger cache and higher clock speeds compensate in most benchmarks.

Clock speeds are starkly different. The Core 9 273PQE has a base clock of 3.40 GHz and a boost clock of 5.90 GHz. The Ultra X7 358H starts at 1.90 GHz and boosts to 4.80 GHz. That 1.10 GHz boost advantage for the desktop part is the primary driver of its single-core wins. The desktop part also has a 125 W TDP against 25 W for the mobile chip, reflecting its intended use in power-unconstrained desktop systems.

PCIe lane availability differs: the Core 9 273PQE provides Gen 5 with 16 lanes (CPU only), while the Ultra X7 358H provides Gen 5 with 4 lanes (CPU only). Integrated graphics also differs: the desktop part uses UHD Graphics 770, while the mobile part uses Arc B390.

FAQ

Q: Which processor has more threads?

A: The Intel Core 9 273PQE has 24 threads from 12 cores, while the Intel Core Ultra X7 358H has 16 threads from 16 cores. The desktop part's simultaneous multithreading doubles its logical thread count.

Q: Why does the Core 9 273PQE win so many multi-core benchmarks despite having fewer physical cores?

A: The Core 9 273PQE has 24 threads versus 16 for the Ultra X7 358H, plus a much higher boost clock at 5.90 GHz versus 4.80 GHz. The combination of more logical threads and higher clock speeds produces a 109% lead in Cinebench R23 multicore.

Q: Does the Ultra X7 358H win any benchmark?

A: Yes, it wins two head-to-head tests: PassMark find prime numbers (337 versus 198, a 41.2% advantage) and PassMark physics (3021 versus 2754, an 8.8% advantage). These are isolated workloads where the mobile part's architecture is more efficient.

Q: What is the difference in process node?

A: The Core 9 273PQE uses a 10 nm process, while the Ultra X7 358H uses a 3 nm process. The mobile part's smaller node allows for more cores (16 versus 12) and higher memory bandwidth (153.6 GB/s versus 89.6 GB/s) within a much lower 25 W TDP.

Q: Do both processors support ECC memory?

A: No. The Core 9 273PQE supports ECC memory, while the Ultra X7 358H does not. The desktop part also supports both DDR4 and DDR5, whereas the mobile part is limited to LPDDR5X.

Q: Which processor has the higher average benchmark score?

A: The Core 9 273PQE averages 66099 across all recorded benchmarks, compared to 40967 for the Ultra X7 358H. The desktop part sits in the 93rd percentile of all CPUs, while the mobile part sits in the 87th percentile.

Specification Differences

| Specification | Intel Core 9 273PQE | Intel Core Ultra X7 358H |

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

| Cores | 12 | 16 |

| Threads | 24 | 16 |

| Base Clock | 3.40 GHz | 1.90 GHz |

| Boost Clock | 5.90 GHz | 4.80 GHz |

| TDP | 125 W | 25 W |

| Socket | Intel Socket 1700 | Intel BGA 2540 |

| Codename | Bartlett Lake | Panther Lake |

| Process Node | 10 nm | 3 nm |

| 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 | LPDDR5X |

| Memory Bandwidth | 89.6 GB/s | 153.6 GB/s |

| ECC Memory | Yes | No |

| PCIe | Gen 5, 16 Lanes (CPU only) | Gen 5, 4 Lanes (CPU only) |

| Integrated Graphics | UHD Graphics 770 | Arc B390 |

| Market Segment | Desktop | Mobile |

| Release Date | 2026-03-08 | 2026-01-04 |

| Launch MSRP | $589 | Not provided |

Head-to-Head Benchmarks

The largest win for the Core 9 273PQE comes in Cinebench R23 single-core, where it scores 5532 against 2080, a 166% margin. This is the most extreme single-core gap in the dataset and reflects the combined effect of a 1.10 GHz boost clock advantage and the desktop part's larger L3 cache. The same test in multicore shows a 109% lead (39190 versus 18747), confirming that the desktop part is not merely faster per core but also scales better across threads.

Cinebench R20 tells a similar but slightly compressed story. Single-core goes 2323 versus 1695, a 37.1% margin, and multicore goes 16459 versus 12011, a 37% margin. Cinebench R15 shows 557 versus 301.5 in single-core (84.7% delta) and 3950 versus 3027 in multicore (30.5% delta). The R15 multicore gap is the smallest Cinebench margin, but the desktop part still wins clearly.

PassMark integer math delivers the second-largest overall margin: 164629 versus 83147, a 98% difference. This workload heavily rewards the desktop part's thread count and clock speed. Data compression follows at 76.2% (585752 versus 332508), and extended instructions at 42.1% (38743 versus 27274). Floating-point math is closer but still one-sided: 125546 versus 103842, a 20.9% gap.

The Core 9 273PQE also wins passmark multithread by 36.4% (46107 versus 33802), random string sorting by 31.7% (53167 versus 40357), and data encryption by 13.8% (29636 versus 26046). PassMark single-thread shows the smallest desktop win at 10.9% (4573 versus 4124), indicating that the Ultra X7 358H's larger per-core L1 and L2 caches partially offset the clock deficit in lightly threaded workloads.

The Ultra X7 358H's two wins are worth examining closely. In PassMark find prime numbers, it scores 337 versus 198, a 41.2% advantage. This test is sensitive to integer throughput per clock and possibly to the mobile part's larger L1 cache (192 KB per core versus 80 KB). In PassMark physics, it scores 3021 versus 2754, an 8.8% edge. Physics simulations often benefit from high memory bandwidth, and the Ultra X7 358H's LPDDR5X provides 153.6 GB/s versus 89.6 GB/s for the desktop part, which may explain this result.

The Verdict

The data is unambiguous. The Intel Core 9 273PQE is the faster processor in nearly every measurable dimension, winning 15 of 17 head-to-head benchmarks and holding a 61.3% higher average benchmark score (66099 versus 40967). Its 93rd percentile ranking versus the 87th percentile for the Ultra X7 358H reflects a processor that sits a full performance tier above its mobile counterpart.

For workloads that stress multi-threaded throughput, the choice is obvious. Cinebench R23 multicore shows a 109% advantage, PassMark integer math shows 98%, and data compression shows 76.2%. The Core 9 273PQE's 24 threads, 5.90 GHz boost clock, and 36 MB of L3 cache combine to produce results that the Ultra X7 358H cannot approach. The desktop part also wins every single-core test, from Cinebench R23's 166% margin to PassMark single-thread's 10.9% edge, meaning it is not merely a multi-core specialist.

The Ultra X7 358H is a 25 W mobile processor, and its two wins reflect its architectural strengths in specific niches. PassMark find prime numbers and PassMark physics are the only tests where it leads, and the physics win likely stems from its 153.6 GB/s memory bandwidth, which is 71.4% higher than the desktop part's. The mobile part's larger per-core L1 and L2 caches (192 KB and 3 MB versus 80 KB and 2 MB) also help in latency-sensitive tasks, but these advantages do not translate into broad benchmark dominance.

Market segment is the deciding factor. The Core 9 273PQE is a desktop processor with a 125 W TDP and a $589 launch MSRP, designed for high-performance systems where power draw is not a constraint. The Ultra X7 358H is a mobile processor with a 25 W TDP, designed for laptops and compact devices where efficiency matters more than absolute performance. Users requiring maximum compute throughput should select the Core 9 273PQE. Users constrained by power or form factor should consider the Ultra X7 358H, but they must accept a substantial performance tradeoff: the desktop part is 109% faster in Cinebench R23 multicore and 166% faster in single-core. The recorded data shows no scenario where the mobile part outperforms the desktop part in general-purpose computing.