AMD Ryzen 9 7900 vs Intel Core Ultra 7 265T Comparison
AMD Ryzen 9 7900
Core Ultra 7 265T
PERFORMANCE BENCHMARKS
Analysis: AMD Ryzen 9 7900 vs Intel Core Ultra 7 265T
FAQ
Q: Which processor has the higher average benchmark score?
A: The AMD Ryzen 9 7900 records an average benchmark score of 49228, while the Intel Core Ultra 7 265T sits at 47697. The AMD part lands 3.2% higher overall, and both sit at the 90th percentile among all CPUs in the database.
Q: How do the two compare in single-threaded Cinebench R23 performance?
A: The Intel Core Ultra 7 265T dominates, scoring 4455 versus 1966 for the AMD Ryzen 9 7900. That is a 55.9% advantage for Intel, the largest single-test margin in the head-to-head comparison.
Q: Which processor wins in multithreaded workloads?
A: It depends on the test. The Intel part wins Cinebench R23 multicore with 31558 against 24776, a 21.5% lead. However, the AMD part wins PassMark multithread with 48347 versus 37084, a 30.4% lead.
Q: What are the core and thread counts for each processor?
A: The AMD Ryzen 9 7900 has 12 cores and 24 threads. The Intel Core Ultra 7 265T has 20 cores and 20 threads, meaning it does not support simultaneous multithreading.
Q: Which chip has the higher boost clock?
A: The AMD Ryzen 9 7900 boosts to 5.40 GHz, slightly ahead of the Intel Core Ultra 7 265T at 5.30 GHz. The Intel part has a much lower base clock at 1.50 GHz versus 3.70 GHz for AMD.
Q: How many benchmark wins does each processor claim in the head-to-head tests?
A: The AMD Ryzen 9 7900 wins 9 of the 15 recorded tests, while the Intel Core Ultra 7 265T wins 6.
Architecture Differences
The AMD Ryzen 9 7900 is built on the Zen 4 architecture (Raphael) using a 5 nm TSMC process, while the Intel Core Ultra 7 265T uses the Arrow Lake architecture (Arrow Lake-S) on a 3 nm TSMC node. The Intel part integrates more transistors, 17,800 million versus 13,140 million, but its die is significantly larger at 243 mm² compared to 2x 71 mm² for the AMD chip.
Core configuration differs fundamentally. The AMD part uses 12 full cores with 24 threads, enabling simultaneous multithreading. The Intel part has 20 cores but only 20 threads, meaning each core is single-threaded. The AMD chip carries 64 MB of shared L3 cache, more than double the 30 MB shared L3 on the Intel part. Per-core caches also differ: AMD provides 64 KB L1 and 1 MB L2 per core, while Intel provides 192 KB L1 and 3 MB L2 per core.
Memory support is DDR5 for both, with dual-channel buses, but the Intel part has higher theoretical bandwidth at 102.4 GB/s versus 83.2 GB/s. AMD supports ECC memory, while Intel does not. PCIe connectivity differs as well: AMD offers Gen 5 with 24 lanes (CPU only), Intel offers Gen 5 with 20 lanes (CPU only). Integrated graphics differ, with AMD using Radeon Graphics and Intel using Arc Xe-LPG Graphics 64EU.
The AMD chip has an unlocked multiplier for overclocking, while the Intel part is locked. Production status is active for both, with the AMD released on 2023-01-13 and the Intel released on 2025-01-06. The AMD part carries a launch MSRP of $429, the Intel part a launch MSRP of $384.
Head-to-Head Benchmarks
The data reveals a split personality between these two processors. The AMD Ryzen 9 7900 leads in 9 tests, mostly in integer-heavy and compression workloads, while the Intel Core Ultra 7 265T leads in 6 tests, with its biggest wins in single-threaded and floating-point performance.
The largest AMD victory comes in PassMark integer math, where the Ryzen 9 7900 scores 164075 versus 104943, a 56.3% advantage. Data compression shows a nearly identical pattern: 577847 against 370158, a 56.1% lead. The AMD part also excels in extended instructions (42253 versus 28609, 47.7% ahead) and random string sorting (68474 versus 44400, 54.2% ahead). Multithreaded PassMark gives AMD a 30.4% win (48347 versus 37084), and physics simulation goes to AMD by 27.9% (3059 versus 2391). Data encryption favors AMD by 16.9% (34708 versus 29687), and prime number finding favors AMD by 18% (380 versus 322).
The Intel part counters with a commanding single-thread lead. Cinebench R23 single-core shows 4455 versus 1966, a 55.9% margin. Cinebench R15 single-core follows with 449 versus 315, a 29.8% lead. In floating-point math, Intel scores 129817 against 97943, a 24.6% advantage. Cinebench R23 multicore goes to Intel at 31558 versus 24776, a 21.5% win, while Cinebench R15 multicore goes to AMD at 4020 versus 3180, a 26.4% margin. PassMark single-thread slightly favors Intel at 4338 versus 4130, a 4.8% edge.
The Cinebench results are particularly telling. The Intel chip wins the newer and more demanding R23 multicore test, but AMD wins the older R15 multicore test. This suggests the two respond differently to workload scaling. Intel's 20 cores without hyperthreading excel in the highly parallel R23 test, while AMD's 12 cores with 24 threads handle R15 more efficiently.
The Verdict
The benchmark data supports a clear division of roles. The AMD Ryzen 9 7900 is the stronger choice for integer math, data compression, encryption, and general multithreaded throughput as measured by PassMark. Its 30.4% lead in PassMark multithread and 56.3% lead in integer math indicate a processor built for sustained computational work across many threads.
The Intel Core Ultra 7 265T is the better option for single-threaded responsiveness and floating-point workloads. Its 55.9% lead in Cinebench R23 single-core is decisive, and its 24.6% advantage in floating-point math shows strength in scientific and graphics-related calculations. The Intel part also wins the Cinebench R23 multicore test by 21.5%, despite losing PassMark multithread, meaning its performance profile depends heavily on the specific workload.
For users prioritizing integer-heavy tasks like data processing, compression, or encryption, the AMD Ryzen 9 7900 is the data-backed pick. For users who need maximum single-thread speed or floating-point throughput, the Intel Core Ultra 7 265T wins in those measured categories. The AMD part also offers ECC memory support and an unlocked multiplier, while the Intel part offers higher memory bandwidth and a smaller process node.
Specification Differences
| Specification | AMD Ryzen 9 7900 | Intel Core Ultra 7 265T |
|---|---|---|
| Cores | 12 | 20 |
| Threads | 24 | 20 |
| Base clock | 3.70 GHz | 1.50 GHz |
| Boost clock | 5.40 GHz | 5.30 GHz |
| TDP | 65 W | 35 W |
| Architecture | Zen 4 | Arrow Lake |
| Process node | 5 nm | 3 nm |
| Transistors | 13,140 million | 17,800 million |
| Die size | 2x 71 mm² | 243 mm² |
| L1 cache (per core) | 64 KB | 192 KB |
| L2 cache (per core) | 1 MB | 3 MB |
| L3 cache (shared) | 64 MB | 30 MB |
| Memory bandwidth | 83.2 GB/s | 102.4 GB/s |
| ECC memory | Yes | No |
| PCIe lanes (CPU) | Gen 5, 24 | Gen 5, 20 |
| Integrated graphics | Radeon Graphics | Arc Xe-LPG Graphics 64EU |
| Multiplier | Unlocked | Locked |
| Launch MSRP | $429 | $384 |
Where Each One Wins
The AMD Ryzen 9 7900 wins in data compression (56.1% ahead), integer math (56.3% ahead), extended instructions (47.7% ahead), random string sorting (54.2% ahead), data encryption (16.9% ahead), prime number finding (18% ahead), PassMark multithread (30.4% ahead), physics simulation (27.9% ahead), and Cinebench R15 multicore (26.4% ahead). This makes it the choice for compression-heavy workflows, encryption tasks, and general multithreaded integer processing.
The Intel Core Ultra 7 265T wins in Cinebench R23 single-core (55.9% ahead), Cinebench R15 single-core (29.8% ahead), Cinebench R23 multicore (21.5% ahead), floating-point math (24.6% ahead), and PassMark single-thread (4.8% ahead). This makes it the choice for single-threaded applications like legacy software, floating-point calculations, and the specific parallel scaling pattern used in Cinebench R23.
Users on constrained power budgets should note the Intel part's 35 W TDP versus 65 W for AMD, though the AMD part delivers its multithreaded wins at that higher envelope. The AMD chip also provides ECC support for reliability-focused builds, while the Intel chip offers higher memory bandwidth for memory-sensitive tasks. Neither processor is uniformly faster; the recorded data shows they excel in different domains.