AMD Ryzen 9 7900X vs Intel Core Ultra 5 235HX Comparison
AMD Ryzen 9 7900X
Core Ultra 5 235HX
PERFORMANCE BENCHMARKS
Analysis: AMD Ryzen 9 7900X vs Intel Core Ultra 5 235HX
Where Each One Wins
The recorded data splits these two processors into distinct roles. The AMD Ryzen 9 7900X wins 9 of the 15 head-to-head benchmark comparisons, while the Intel Core Ultra 5 235HX takes 6. The pattern is clear: AMD dominates heavily threaded workload types, while Intel wins the single-thread and floating-point categories.
For multi-threaded integer work, the AMD part is decisively ahead. In PassMark integer math, the 7900X scores 169273 against 97177 for the 235HX, a 74.2% advantage. That is the largest margin in the entire comparison. Data compression also favors AMD strongly: 632505 versus 426417, a 48.3% gap. Random string sorting shows a 46.6% lead for AMD, and extended instructions land at 36.8% ahead. These are workload types that scale with core count and memory subsystem throughput.
The Intel part wins where latency and per-thread efficiency matter most. In Cinebench R23 single-core, the 235HX scores 4903 against 2016.5 for AMD, a 58.9% advantage. That is a massive single-thread gap, almost three times the score. Cinebench R15 single-core shows a similar story with a 34.6% lead for Intel. PassMark single-thread sits at 4683 versus 4238, a 9.5% edge. Floating-point math also goes to Intel: 129819 versus 103952, a 19.9% lead.
The multi-core Cinebench results are mixed. The 7900X wins Cinebench R15 multi-core by 37.7%, but the 235HX wins Cinebench R23 multi-core by 15.6%. That reversal suggests the Intel part scales better in the newer, more demanding Cinebench R23 workload despite having fewer threads. The AMD part has 24 threads versus 14 for Intel, yet R23 still favors Intel, indicating a significant architectural efficiency difference.
Architecture Differences
The two processors come from different design philosophies. The AMD Ryzen 9 7900X uses Zen 4 architecture, codenamed Raphael, built on a 5 nm process at TSMC. It has 12 cores and 24 threads. The Intel Core Ultra 5 235HX uses Arrow Lake architecture, codenamed Arrow Lake-HX, built on a 3 nm process at TSMC. It has 14 cores but only 14 threads, meaning no hyperthreading or simultaneous multithreading.
Cache layouts differ substantially. The AMD part has 64 KB of L1 per core, 1 MB of L2 per core, and 64 MB of shared L3 cache. The Intel part has 192 KB of L1 per core, 3 MB of L2 per core, but only 24 MB of shared L3. Intel has more L1 and L2 per core, while AMD has much larger L3 capacity. That larger L3 cache may explain AMD's strong showing in data compression and integer workloads, which benefit from larger working sets residing on-chip.
Transistor counts and die sizes also differ. The AMD chip has 13,140 million transistors across a 2x 71 mm² die design. The Intel chip has 17,800 million transistors on a single 243 mm² die. The Intel process node is more advanced at 3 nm versus 5 nm, yet the Intel die is physically larger.
Memory support both use DDR5 with dual-channel buses. AMD achieves 83.2 GB/s memory bandwidth, while Intel reaches 102.4 GB/s. Intel has a 23% bandwidth advantage, which may contribute to its Cinebench R23 multi-core win. ECC memory is supported on AMD but not on Intel.
PCIe lane counts differ: AMD provides Gen 5 with 24 lanes (CPU only), while Intel provides Gen 5 with 20 lanes. Both have unlocked multipliers. The AMD part ships with Radeon Graphics integrated, while Intel includes Arc Xe-LPG Graphics with 48 execution units.
Market segments differ: the 7900X is a desktop part on AMD Socket AM5, while the 235HX is a mobile part on Intel BGA 2114. The AMD part has a TDP of 170 watts, the Intel part 55 watts. Release dates also differ: the 7900X launched in September 2022, the 235HX in January 2025.
The Verdict
The data points to a workload-based choice. For anyone running heavily threaded integer workloads, data compression, encryption, or sorting tasks, the AMD Ryzen 9 7900X is the pick. Its leads in PassMark integer math (74.2%), data compression (48.3%), random string sorting (46.6%), and extended instructions (36.8%) are substantial. It also wins PassMark multithread by 25.8% and physics by 20%.
For single-thread-heavy workloads, the Intel Core Ultra 5 235HX is clearly better. The Cinebench R23 single-core margin of 58.9% is the largest single-thread gap in the comparison. Cinebench R15 single-core also favors Intel by 34.6%. PassMark single-thread and floating-point math both go to Intel as well.
The average benchmark scores tell a nuanced story. The AMD part has an average benchmark score of 53288, slightly higher than Intel's 52073. Both sit at the 91st percentile versus all CPUs. The AMD part's nearest rival list includes the AMD EPYC 7313P (0.2% ahead), Intel Xeon Phi 7290 (0.3% behind), Intel Xeon 634 (0.6% ahead), and Intel Core i7-14700F (0.6% behind). The Intel part's nearest rivals include the AMD EPYC 8124P (0.1% behind), AMD Ryzen 9 5950X (0.2% ahead), Intel Core i7-14700 (0.4% behind), and Intel Core i9-13900F (0.7% ahead). Both processors cluster closely with their nearest rivals, with deltas under 1%.
For desktop builders needing raw multi-threaded throughput, the 7900X offers 12 cores with 24 threads and a 5.60 GHz boost clock. For mobile systems prioritizing single-thread responsiveness and energy efficiency, the 235HX offers 14 cores, a 5.10 GHz boost clock, and a 55 watt TDP. The data does not show an overall winner, only a clear split by workload type.
FAQ
Q: Which processor has more threads?
A: The AMD Ryzen 9 7900X has 24 threads from 12 cores. The Intel Core Ultra 5 235HX has 14 threads from 14 cores. Intel's part does not use simultaneous multithreading.
Q: Why does Intel win Cinebench R23 multi-core despite having fewer threads?
A: The Intel part scores 34731 in Cinebench R23 multi-core versus 29300 for AMD, a 15.6% lead. Intel has higher per-core L1 (192 KB) and L2 (3 MB) cache, a more advanced 3 nm process, and 102.4 GB/s memory bandwidth versus 83.2 GB/s for AMD. These factors contribute to better scaling in that specific workload.
Q: What is the largest performance gap in either direction?
A: The largest gap is AMD's 74.2% lead in PassMark integer math. The second largest is Intel's 58.9% lead in Cinebench R23 single-core.
Q: Do both processors support DDR5 memory?
A: Yes, both support DDR5 with dual-channel memory buses. AMD has 83.2 GB/s bandwidth, Intel has 102.4 GB/s. AMD also supports ECC memory, Intel does not.
Q: Which processor has more L3 cache?
A: The AMD Ryzen 9 7900X has 64 MB of shared L3 cache. The Intel Core Ultra 5 235HX has 24 MB of shared L3 cache. AMD has nearly three times the L3 capacity.
Q: Are both processors currently in production?
A: Yes, both are listed as active production parts. The AMD part launched in September 2022, the Intel part in January 2025.
Head-to-Head Benchmarks
The AMD Ryzen 9 7900X wins several benchmarks by wide margins. PassMark integer math shows the largest gap: 169273 versus 97177, a 74.2% advantage. This workload heavily favors AMD's 24 threads and 64 MB L3 cache. Data compression follows at 48.3%, with scores of 632505 versus 426417. Random string sorting shows a 46.6% lead (74658 versus 50929). Extended instructions land at 36.8% (47619 versus 34808). Cinebench R15 multi-core gives AMD a 37.7% win (4820.5 versus 3500). PassMark multithread shows 25.8% (51406 versus 40849). Physics goes to AMD by 20% (3060 versus 2550). Data encryption favors AMD by 14% (37263 versus 32697). Prime number finding is close, with AMD ahead by 7.5% (388 versus 361).
The Intel Core Ultra 5 235HX wins the single-thread and floating-point categories. Cinebench R23 single-core is its biggest win: 4903 versus 2016.5, a 58.9% margin. That is a remarkable single-thread advantage. Cinebench R15 single-core follows at 34.6% (494 versus 323). Floating-point math goes to Intel by 19.9% (129819 versus 103952). PassMark single-thread shows a 9.5% lead (4683 versus 4238). Cinebench R23 multi-core also goes to Intel by 15.6% (34731 versus 29300), despite Intel having 14 threads versus AMD's 24.
The wins tally at 9 for AMD and 6 for Intel. AMD's wins are concentrated in integer, compression, sorting, encryption, and general multithreaded workloads. Intel's wins are concentrated in single-thread performance, floating-point math, and the newer Cinebench R23 multi-core test. The overall average benchmark scores are very close: 53288 for AMD versus 52073 for Intel, a difference of about 2.3%. Both processors sit at the 91st percentile among all CPUs in the database.
Specification Differences
The two processors differ across nearly every major specification category.
| Specification | AMD Ryzen 9 7900X | Intel Core Ultra 5 235HX |
|---|---|---|
| Cores | 12 | 14 |
| Threads | 24 | 14 |
| Base clock | 4.70 GHz | 2.90 GHz |
| Boost clock | 5.60 GHz | 5.10 GHz |
| TDP | 170 W | 55 W |
| Socket | AMD Socket AM5 | Intel BGA 2114 |
| Architecture | Zen 4 | Arrow Lake |
| Codename | Raphael | Arrow Lake-HX |
| Process node | 5 nm | 3 nm |
| Foundry | TSMC | TSMC |
| Transistors | 13,140 million | 17,800 million |
| Die size | 2x 71 mm² | 243 mm² |
| L1 cache | 64 KB per core | 192 KB per core |
| L2 cache | 1 MB per core | 3 MB per core |
| L3 cache | 64 MB shared | 24 MB shared |
| Memory bandwidth | 83.2 GB/s | 102.4 GB/s |
| ECC support | Yes | No |
| PCIe lanes | Gen 5, 24 lanes | Gen 5, 20 lanes |
| Integrated graphics | Radeon Graphics | Arc Xe-LPG Graphics 48EU |
| Market segment | Desktop | Mobile |
| Release date | September 2022 | January 2025 |
| Launch MSRP | $549 | Not recorded |
| Multiplier unlocked | Yes | Yes |
Both support DDR5 memory with dual-channel buses. Both use TSMC as the foundry. Both have unlocked multipliers. The AMD part has more threads, more L3 cache, higher clock speeds, higher TDP, ECC support, and more PCIe lanes. The Intel part has more cores, more L1 and L2 per core, higher memory bandwidth, a smaller process node, more transistors, larger die, and a significantly lower TDP.