Intel Core Ultra 9 290K Plus vs Intel Xeon 6731P Comparison

The Intel Xeon 6731P and Intel Core Ultra 9 290K Plus represent two distinct poles of Intel’s current lineup: a 32-core Granite Rapids server processor built for bandwidth and sustained throughput, and a 24-core Arrow Lake Refresh desktop chip engineered for high clock speeds and responsiveness. The benchmark data shows a clear split, with the Core Ultra 9 dominating Cinebench and single-threaded workloads, while the Xeon 6731P takes the lead in several PassMark server-oriented tasks. This analysis walks through the head-to-head results, architecture differences, and the specific use cases where each processor’s design philosophy pays off.

Head-to-Head Benchmarks

The most striking pattern in the head-to-head results is the consistency of the Core Ultra 9 290K Plus’s victory across Cinebench. In every Cinebench test — R15, R20, and R23, both single-core and multi-core — the Core Ultra 9 wins by exactly 13.3%. The multi-core scores tell the story: the Xeon 6731P scores 44,871 in Cinebench R23 multi-core, while the Core Ultra 9 reaches 51,731. That 13.3% advantage is also present in Cinebench R20 (18,845 vs 21,727) and R15 (4,522 vs 5,214). The single-core margins are identical in percentage terms, with the Core Ultra 9 scoring 7,303 in R23 single-core against the Xeon’s 6,334.

The PassMark suite reveals a more nuanced picture. The biggest single win for the Core Ultra 9 is in floating point math, where it scores 214,760 against the Xeon’s 157,330 — a 26.7% advantage. It also leads in data encryption (52,563 vs 40,087, a 23.7% margin) and PassMark multi-thread (60,860 vs 52,790, again 13.3%). The single-thread PassMark test is the largest gap of all: the Core Ultra 9 scores 4,823, a 56.3% lead over the Xeon’s 2,107. This massive delta reflects the Core Ultra 9’s 5.80 GHz boost clock against the Xeon’s 4.10 GHz.

The Xeon 6731P, however, wins six of the seventeen head-to-head tests, and several of its victories are substantial. The most dramatic is PassMark physics, where the Xeon scores 7,105 against the Core Ultra 9’s 3,315 — a 114.3% lead. The Xeon also wins extended instructions by 25.4% (65,656 vs 52,338), integer math by 19.6% (198,761 vs 166,194), data compression by 14.5% (799,474 vs 698,346), and random string sorting by 10.4% (88,019 vs 79,744). The find prime numbers test is closer, with the Xeon ahead by 7.6% (541 vs 503). Overall, the Core Ultra 9 wins 11 tests to the Xeon’s 6, but the Xeon’s wins tend to be in areas that matter for server workloads.

FAQ

Q: Which processor has the higher average benchmark score?

A: The Xeon 6731P has an average benchmark score of 87,756, while the Core Ultra 9 290K Plus averages 84,003. This places the Xeon 3.2% ahead of the AMD EPYC 7F72 in its nearest rivals list, while the Core Ultra 9 sits 0.2% ahead of the Intel Core Ultra 9 285K.

Q: How do the two compare in single-threaded performance?

A: The Core Ultra 9 290K Plus is decisively faster in single-threaded tests. In PassMark single-thread, it scores 4,823 against the Xeon’s 2,107, a 56.3% lead. The Cinebench R23 single-core test shows a narrower 13.3% gap, with the Core Ultra 9 scoring 7,303 versus 6,334.

Q: Where does the Xeon 6731P show its biggest advantage?

A: The Xeon’s largest win is in PassMark physics, where it scores 7,105 — more than double the Core Ultra 9’s 3,315, a 114.3% difference. It also leads in extended instructions by 25.4% and integer math by 19.6%.

Q: Do both processors support ECC memory?

A: Yes, both the Intel Xeon 6731P and the Intel Core Ultra 9 290K Plus list ECC memory support as true in their specifications.

Q: What are the memory bandwidth figures for each?

A: The Xeon 6731P offers 409.6 GB/s of memory bandwidth across an eight-channel DDR5 bus. The Core Ultra 9 290K Plus provides 115.2 GB/s over a dual-channel DDR5 bus. This bandwidth difference is a key factor in the Xeon’s server positioning.

Q: Which processor has a higher percentile ranking among all CPUs?

A: Both processors rank at the 96th percentile versus all CPUs, indicating they are both in the top tier of performance according to the database’s aggregate scoring.

Architecture Differences

The two processors are built on fundamentally different designs. The Xeon 6731P uses the Granite Rapids architecture on a 5 nm process node, fabricated by Intel, with a die size of 598 mm². It packs 32 cores and 64 threads, with a base clock of 2.50 GHz and a boost clock of 4.10 GHz. The cache hierarchy is substantial: 112 KB of L1 per core, 2 MB of L2 per core, and a shared 144 MB L3 cache. The memory subsystem is designed for scale, with eight-channel DDR5 support delivering 409.6 GB/s of bandwidth. The Xeon supports PCIe Gen 5 with 136 lanes (CPU only) and has no integrated graphics. Its thermal design power is 245 W.

The Core Ultra 9 290K Plus takes a different approach. It uses the Arrow Lake Refresh codename on a 3 nm process node from TSMC, with a die size of 243 mm² and 17,800 million transistors. It has 24 cores and 24 threads (no hyperthreading), with a base clock of 3.70 GHz and a boost clock of 5.80 GHz. The cache is smaller: 192 KB L1 per core, 3 MB L2 per core, and 36 MB shared L3. Memory bandwidth is 115.2 GB/s over a dual-channel DDR5 bus. It includes integrated Arc Xe-LPG Graphics with 64EU, and the PCIe support is Gen 5, 20 lanes (CPU only). The Core Ultra 9 has a 125 W TDP and features an unlocked multiplier for overclocking.

The process node difference is notable — 3 nm for the Core Ultra 9 versus 5 nm for the Xeon — but the Xeon compensates with sheer core count and massive cache. The 144 MB L3 cache on the Xeon is four times the 36 MB on the Core Ultra 9, which explains its wins in data compression and extended instructions. The Core Ultra 9’s higher boost clock and newer process node drive its single-thread and floating-point advantages.

The Verdict

The data points to two distinct roles. The Intel Core Ultra 9 290K Plus is the clear winner for single-threaded and lightly threaded workloads, with a 56.3% lead in PassMark single-thread and consistent 13.3% wins across all Cinebench tests. Its floating-point math advantage of 26.7% and encryption lead of 23.7% make it suitable for desktop applications, development work, and tasks that rely on high clock speeds.

The Intel Xeon 6731P is the choice for server and workstation environments where memory bandwidth and parallel throughput across many threads matter. Its 114.3% lead in physics, 25.4% lead in extended instructions, and 19.6% lead in integer math indicate strength in scientific computing, data processing, and enterprise workloads. The eight-channel memory bus delivering 409.6 GB/s versus 115.2 GB/s is a decisive architectural advantage for memory-bound tasks.

Neither processor is universally superior. The Core Ultra 9 wins 11 of 17 head-to-head tests, but the Xeon wins the tests that align with server workloads. The Xeon’s average benchmark score of 87,756 is higher than the Core Ultra 9’s 84,003, but that aggregate figure is skewed by the Xeon’s server-oriented strengths. For a desktop user, the Core Ultra 9’s single-thread and Cinebench performance will be more relevant. For a data center operator, the Xeon’s bandwidth and multi-thread physics performance will matter more.

Specification Differences

The two processors differ across nearly every specification field. The Xeon 6731P has 32 cores and 64 threads, while the Core Ultra 9 has 24 cores and 24 threads. Base clocks are 2.50 GHz versus 3.70 GHz, and boost clocks are 4.10 GHz versus 5.80 GHz in favor of the Core Ultra 9. TDP is 245 W for the Xeon and 125 W for the Core Ultra 9.

The sockets are incompatible: the Xeon uses Intel Socket 4710, while the Core Ultra 9 uses Intel Socket 1851. The process node is 5 nm (Intel foundry) for the Xeon versus 3 nm (TSMC) for the Core Ultra 9. Die size is 598 mm² versus 243 mm². Cache differs substantially: L1 is 112 KB per core on the Xeon versus 192 KB per core on the Core Ultra 9; L2 is 2 MB per core versus 3 MB per core; L3 is 144 MB shared versus 36 MB shared.

Memory configuration is a major differentiator: the Xeon has an eight-channel bus with 409.6 GB/s bandwidth, while the Core Ultra 9 has a dual-channel bus with 115.2 GB/s. PCIe lanes are 136 versus 20, both Gen 5. The Xeon has no integrated graphics; the Core Ultra 9 includes Arc Xe-LPG Graphics with 64EU. The Xeon’s multiplier is locked, while the Core Ultra 9 is unlocked. The Xeon has a launch MSRP of $2700; the Core Ultra 9 has no listed launch MSRP.

Where Each One Wins

The Xeon 6731P wins in PassMark physics (114.3% ahead), extended instructions (25.4%), integer math (19.6%), data compression (14.5%), random string sorting (10.4%), and find prime numbers (7.6%). These are compute-heavy, parallel workloads that benefit from the 64 threads and 144 MB L3 cache. The physics result is particularly notable — the Xeon’s score of 7,105 versus 3,315 suggests simulation and collision-detection workloads will see dramatic gains.

The Core Ultra 9 290K Plus wins in all Cinebench tests (13.3% across the board), PassMark single-thread (56.3%), floating-point math (26.7%), data encryption (23.7%), and PassMark multi-thread (13.3%). The single-thread and floating-point wins point to desktop productivity, media encoding, and interactive applications. The encryption win suggests it handles cryptographic workloads well despite the Xeon’s server pedigree.

For a user choosing between them, the decision hinges on workload type. If the task involves large datasets, memory bandwidth saturation, or many parallel threads, the Xeon’s 409.6 GB/s bandwidth and 32 cores make it the logical pick. If the task is single-threaded, clock-sensitive, or involves floating-point arithmetic, the Core Ultra 9’s 5.80 GHz boost clock and 3 nm process deliver superior results. The data shows no single processor is best for everything — the Xeon excels in server-style throughput, while the Core Ultra 9 excels in desktop-style responsiveness and per-thread performance.