Intel Core Ultra 5 235T vs Intel Core Ultra 9 285 Comparison

Head-to-Head Benchmarks

The benchmark data paints a very one-sided picture. Across all 17 recorded tests, the Intel Core Ultra 9 285 takes the win with no exceptions. The margin varies significantly by workload, ranging from a modest 11.1% advantage in single-threaded tests to a commanding 52% lead in random string sorting.

Starting with the Cinebench suite, the Ultra 9 285 delivers consistent dominance. In Cinebench R15 multicore, it scores 4933 against the Ultra 5 235T's 2644, a 46.4% gap. The single-core result follows the same pattern: 696 versus 373, also 46.4% apart. Cinebench R20 shows identical relative positioning, with the Ultra 9 scoring 20556 multicore and 2901 single-core, compared to 11017 and 1555 for the Ultra 5. Cinebench R23 continues the trend, with the Ultra 9 reaching 48945 multicore and 6909 single-core, versus 26232 and 3703.

The Passmark suite reveals where the gap widens most. The largest discrepancy appears in random string sorting, where the Ultra 9 scores 73651 versus 35377, a 52% advantage. Data compression shows a 51% gap, with scores of 602121 and 295100. Data encryption follows closely at 50%, with 46949 against 23457. Extended instructions tests show a 48.8% difference, with 45357 versus 23212, while integer math sits at 48.9%, scoring 164869 against 84244.

Floating-point math and multithread performance both show 45.4% gaps. The Ultra 9 posts 194988 in floating-point math versus 106546, and 56602 in multithread versus 30918. Physics tests show a 40.1% gap, with 3598 against 2157. Prime number finding shows the narrowest multicore margin at 30.7%, with 459 versus 318.

The only test where the Ultra 5 comes reasonably close is single-thread performance. Passmark single-thread shows 4881 for the Ultra 9 versus 4339 for the Ultra 5, an 11.1% gap. This suggests the architectural design of the Ultra 5 235T does not suffer as severely in lightly threaded workloads, but the Ultra 9 still wins outright.

Where Each One Wins

The recorded data shows no benchmark category where the Intel Core Ultra 5 235T comes out ahead. The Ultra 9 285 wins every single workload category. However, the magnitude of the wins varies enough to identify what each processor does relatively better.

The Ultra 5 235T performs least badly relative to the Ultra 9 in single-threaded tasks. The 11.1% deficit in Passmark single-thread is its closest result. This indicates that for basic responsiveness, everyday application launches, and lightly threaded productivity tasks, the gap narrows considerably. The 30.7% deficit in prime number finding also shows that some computational workloads that do not scale perfectly with core count remain closer.

The Ultra 9 285 shows its largest advantages in memory-intensive and highly parallel workloads. Random string sorting at 52% ahead, data compression at 51%, and data encryption at 50% all require substantial memory bandwidth and many cores operating simultaneously. These are the workloads where the extra 10 cores and the larger shared cache make the biggest measurable difference.

The Cinebench results, all at 46.4% gaps, represent the middle ground. These render workloads scale well with cores but do not saturate memory as aggressively as the Passmark compression and sorting tests. The Ultra 9 consistently delivers roughly 86% more multicore performance in these tests, while the single-core Cinebench results show the same percentage gap, which is notable given the much smaller single-thread Passmark gap.

Architecture Differences

Both processors share the same Arrow Lake-S architecture, manufactured by TSMC on a 3 nm process node. The die size is identical at 243 mm², and the transistor count matches at 17,800 million. Both use the Intel Socket 1851 platform with PCIe Gen 5 support at 20 CPU lanes, dual-channel DDR5 memory, and identical memory bandwidth of 102.4 GB/s.

The core counts differ substantially. The Ultra 5 235T has 14 cores and 14 threads, while the Ultra 9 285 has 24 cores and 24 threads. Neither processor supports Hyper-Threading, so threads equal cores in both cases. The base clock on the Ultra 9 is 2.50 GHz versus 2.20 GHz on the Ultra 5, and the boost clock reaches 5.60 GHz versus 5.00 GHz.

The cache hierarchy shows both shared and distinct elements. L1 cache is 192 KB per core on both, and L2 cache is 3 MB per core on both. The L3 cache differs: the Ultra 9 has 36 MB shared, while the Ultra 5 has 24 MB shared. This 12 MB difference in last-level cache contributes to the larger performance deltas in memory-heavy workloads.

The integrated graphics also differ. The Ultra 9 285 uses Arc Xe-LPG Graphics with 64 execution units, while the Ultra 5 235T uses the same architecture but with only 24 execution units. This represents a significant difference in integrated graphics capability, though the benchmark data focuses on CPU compute.

ECC memory support is another differentiator. The Ultra 9 285 supports ECC memory, while the Ultra 5 235T does not. Both processors ship with a 65 W TDP, indicating that the performance gap comes from core count, cache, and clock speed rather than power budget. The Ultra 9 was released on 2024-12-31, while the Ultra 5 followed on 2025-01-06.

The average benchmark score reinforces the overall positioning. The Ultra 5 235T averages 38561 points, while the Ultra 9 285 averages 75488 points. The Ultra 5 sits at the 86th percentile of all CPUs in the database, while the Ultra 9 reaches the 95th percentile.

The Verdict

The data directs different buyers toward each processor. The Intel Core Ultra 9 285 is the clear choice for workloads that use all available cores and memory bandwidth. Its 52% lead in random string sorting and 51% lead in data compression show that data-intensive batch operations benefit enormously. The 50% advantage in data encryption and 48.9% in integer math further reinforce this positioning. For rendering, the Cinebench R23 multicore score of 48945 versus 26232, a 46.4% gap, makes the Ultra 9 the stronger option for content creation.

The Ultra 9 also wins in single-threaded performance, but by a much smaller margin. The 11.1% gap in Passmark single-thread means that for general desktop use, web browsing, and office applications, the Ultra 5 235T will feel nearly as responsive. The Ultra 5's 86th percentile ranking places it among capable desktop processors, and its nearest rivals include the Intel Core i5-14500 with an average score of 38531, only 0.1% behind.

The Intel Core Ultra 5 235T makes sense for systems where the 14-core configuration meets the workload requirements and the lower core count is acceptable. Its 86th percentile placement and average score of 38561 put it in a similar class to the Intel Xeon w3-2525 at 38392 and the Intel Core 9 270H at 38335. These are meaningful performance levels for mainstream desktop computing.

The Ultra 9 285 sits in a different performance class entirely. Its average score of 75488 puts it alongside AMD EPYC 8224P at 75582, AMD Ryzen 7 PRO 9755X3D at 75716, and AMD Ryzen 7 PRO 9755 at 75738, all within 0.3% of each other. The 95th percentile ranking reflects its position among the top desktop processors in the database.

For users running heavily threaded rendering, compression, encryption, or sorting workloads, the Ultra 9 285 delivers roughly double the performance of the Ultra 5 235T in most multicore tests. For users with lighter workloads, the Ultra 5 235T offers a closer single-thread experience, trailing by only 11.1%, while providing 14 cores for occasional parallel tasks.

FAQ

Q: Which processor has more cores and threads?

A: The Intel Core Ultra 9 285 has 24 cores and 24 threads, while the Intel Core Ultra 5 235T has 14 cores and 14 threads. Neither processor uses Hyper-Threading, so thread counts equal core counts for both.

Q: What is the performance difference in single-threaded workloads?

A: In Passmark single-thread tests, the Ultra 9 285 scores 4881 versus 4339 for the Ultra 5 235T, an 11.1% advantage. This is the smallest performance gap recorded between the two processors across all benchmark tests.

Q: Do both processors use the same manufacturing process?

A: Yes, both are built on the 3 nm process node by TSMC. They share the same die size of 243 mm² and the same transistor count of 17,800 million. Both use the Arrow Lake-S architecture.

Q: What is the L3 cache difference between the two?

A: The Ultra 9 285 has 36 MB of shared L3 cache, while the Ultra 5 235T has 24 MB. Both have 192 KB of L1 cache per core and 3 MB of L2 cache per core.

Q: Which processor supports ECC memory?

A: The Intel Core Ultra 9 285 supports ECC memory. The Intel Core Ultra 5 235T does not support ECC memory. Both use DDR5 with dual-channel memory and 102.4 GB/s memory bandwidth.

Q: How do the integrated graphics compare?

A: The Ultra 9 285 uses Arc Xe-LPG Graphics with 64 execution units, while the Ultra 5 235T uses Arc Xe-LPG Graphics with 24 execution units. Both processors have a 65 W TDP.