Intel Core Ultra 9 285K vs Intel Core Ultra X7 368H Comparison

Intel
INTEL

Intel Core Ultra 9 285K

CORE STATE Arrow Lake-S
CORE SPECS 24 Cores / 24 Threads
CLOCK SPEED 3.7 Base / 5.7 GHz Turbo
CACHE 36 MB (shared)
MAX TDP 125W
ARCHITECTURE Arrow Lake
nm
PROCESS 3 nm
LAUNCH DATE 2024
VS
Intel
INTEL

Core Ultra X7 368H

CORE STATE Panther Lake
CORE SPECS 16 Cores / 16 Threads
CLOCK SPEED 2 Base / 5 GHz Turbo
CACHE 18 MB (shared)
MAX TDP 25W
ARCHITECTURE Panther Lake
nm
PROCESS 3 nm
LAUNCH DATE 2026

PERFORMANCE BENCHMARKS

cinebench_cinebench_r15_multicore
6,494
2,844
cinebench_cinebench_r15_singlecore
359
401
cinebench_cinebench_r20_multicore
24,003
11,852
cinebench_cinebench_r20_singlecore
3,388
1,673
cinebench_cinebench_r23_multicore
42,522
28,221
cinebench_cinebench_r23_singlecore
2,377
3,984
geekbench_multicore
26,702
N/A
geekbench_singlecore
2,870
N/A
passmark_data_compression
790,052
312,927
passmark_data_encryption
57,745
25,228
passmark_extended_instructions
62,277
25,669
passmark_find_prime_numbers
541
321
passmark_floating_point_math
224,324
105,681
passmark_integer_math
172,379
88,083
passmark_multithread
67,260
32,956
passmark_physics
3,938
2,857
passmark_random_string_sorting
94,927
38,093
passmark_single_thread
5,087
4,005
passmark_singlethread
5,087
4,005

Analysis: Intel Core Ultra 9 285K vs Intel Core Ultra X7 368H

Head-to-Head Benchmarks

The benchmark comparison between the Intel Core Ultra 9 285K and the Intel Core Ultra X7 368H is decisively lopsided. The desktop 285K wins 15 of the 17 recorded head-to-head tests, while the mobile 368H takes only 2. The average benchmark score tells the same story: the 285K posts 83807 against 40518 for the 368H, a gap that places the desktop part at the 96th percentile of all CPUs versus the 87th percentile for the mobile chip.

The largest single margin comes in Passmark data compression, where the 285K scores 790052 against 312927, a 152.5% advantage. Random string sorting shows a similar pattern at 149.2% (94927 versus 38093), and extended instructions deliver 142.6% (62277 versus 25669). Data encryption follows at 128.9% (57745 versus 25228). These are workload categories where the desktop chip's core count and memory bandwidth produce outsized gains.

Cinebench multicore results reinforce the pattern. In Cinebench R15 multicore, the 285K scores 6494 against 2844, a 128.3% lead. Cinebench R20 multicore shows 24003 versus 11852, a 102.5% margin, and Cinebench R23 multicore delivers 42522 versus 28221, a 50.7% advantage. Passmark multithread adds another data point: 67260 versus 32956, a 104.1% lead.

Integer math favors the 285K at 95.7% (172379 versus 88083), floating point math at 112.3% (224324 versus 105681), and prime number finding at 68.5% (541 versus 321). Physics simulation shows a more modest 37.8% edge (3938 versus 2857), and Passmark single thread still goes to the desktop chip at 27% (5087 versus 4005).

The 368H wins only two tests, and both are single-core oriented. In Cinebench R15 singlecore, the 368H scores 401 against 359, a 10.5% advantage. In Cinebench R23 singlecore, it scores 3984 against 2377, a 40.3% lead. These are the only recorded tests where the mobile part comes out ahead, and they point to a different architectural tuning rather than an overall performance advantage.

Where Each One Wins

The 285K dominates every multithreaded and throughput-oriented benchmark in the database. Data compression, encryption, extended instructions, integer and floating point math, multithread, physics, and random string sorting all favor the desktop chip by margins ranging from 37.8% to 152.5%. The Cinebench multicore tests confirm the same conclusion across three different rendering workloads. For any task that scales across cores, the 285K is the clear choice.

The 368H wins exclusively in single-core rendering workloads. Its Cinebench R15 singlecore score of 401 and Cinebench R23 singlecore score of 3984 indicate a design optimized for lightly threaded, short-burst tasks. The 40.3% lead in Cinebench R23 singlecore is substantial, suggesting that the mobile chip's newer architecture, Panther Lake, brings a meaningful single-thread efficiency advantage. However, this does not translate into a Passmark single-thread win, where the 285K still leads by 27%.

The split is straightforward: the 285K is the throughput champion, while the 368H offers superior single-core rendering in specific Cinebench tests. Users who prioritize multithreaded productivity will find the 285K vastly faster, while those who care about single-thread rendering scores in isolation would see the 368H as the winner in that narrow category.

The Verdict

The data indicates the Intel Core Ultra 9 285K is the stronger processor for almost every measured workload. It wins 15 of 17 head-to-head tests, including all multithreaded benchmarks, all Passmark tests, and the Cinebench R20 singlecore test. Its average benchmark score of 83807 is more than double the 368H's 40518, and its 96th percentile ranking places it among the top CPUs in the database, while the 368H sits at the 87th percentile.

The 368H should be selected only when single-core Cinebench rendering performance is the priority. Its Cinebench R23 singlecore score of 3984 is 40.3% higher than the 285K's 2377, and its Cinebench R15 singlecore score of 401 is 10.5% higher. These are specific, measurable wins. Outside of those two tests, the 285K delivers superior performance across the board.

For general computing, content creation, data processing, and any multithreaded workload, the 285K is the unambiguous choice. For a mobile platform where single-core rendering benchmarks are the sole concern, the 368H has a documented edge. The database further shows the 285K sits within 0.2% of the Intel Core Ultra 9 290K Plus and 0.9% ahead of the AMD EPYC 4584PX, while the 368H is effectively tied with the Intel Core 7 253PE at 0.1% and the Intel Core 5 223PE at 0.2%.

FAQ

Q: Which processor has the higher average benchmark score?

A: The Intel Core Ultra 9 285K posts an average benchmark score of 83807, compared to 40518 for the Intel Core Ultra X7 368H.

Q: In which benchmarks does the Intel Core Ultra X7 368H outperform the 285K?

A: The 368H wins Cinebench R15 singlecore (401 versus 359, a 10.5% advantage) and Cinebench R23 singlecore (3984 versus 2377, a 40.3% advantage).

Q: How large is the multicore performance gap in Cinebench R23?

A: The 285K scores 42522 against 28221 for the 368H, a 50.7% lead in the 285K's favor.

Q: What is the biggest recorded performance margin between the two?

A: Passmark data compression shows the largest gap, with the 285K scoring 790052 against 312927 for the 368H, a 152.5% advantage.

Q: How do the two processors rank against all CPUs in the database?

A: The 285K is at the 96th percentile, while the 368H sits at the 87th percentile.

Q: What is the nearest rival for each processor?

A: The 285K's closest rival is the Intel Core Ultra 9 290K Plus at a 0.2% difference, while the 368H's closest rival is the Intel Core 7 253PE at a 0.1% difference.

Architecture Differences

The two processors come from different Intel generations and market segments. The 285K uses the Arrow Lake architecture with the codename Arrow Lake-S, part of the Core Ultra Series 2 and the Ultra 9 generation. The 368H uses the Panther Lake architecture with the same codename, part of the Core Ultra Series 3 and the Ultra X7 generation. The 285K is a desktop part, while the 368H is a mobile part.

Both use a 3 nm process node, but the foundries differ. The 285K is fabricated by TSMC, while the 368H is fabricated by Intel. Transistor counts and die sizes are recorded only for the 285K: 17,800 million transistors on a 243 mm² die. No transistor or die size data is available for the 368H.

Cache configurations differ notably. The 285K has 24 cores and 24 threads, with 192 KB of L1 per core, 3 MB of L2 per core, and 36 MB of shared L3 cache. The 368H has 16 cores and 16 threads, with the same 192 KB of L1 per core, but 2.5 MB of L2 per core and 18 MB of shared L3 cache. Neither processor supports 3D V-Cache.

Memory support also diverges. The 285K uses DDR5 memory with dual-channel support and a memory bandwidth of 102.4 GB/s, plus ECC memory support. The 368H uses LPDDR5X memory with dual-channel support and a higher memory bandwidth of 153.6 GB/s, but no ECC support. PCIe lanes differ as well: the 285K offers Gen 5 with 20 lanes (CPU only), while the 368H offers Gen 5 with 4 lanes (CPU only).

Integrated graphics differ: the 285K uses Arc Xe-LPG Graphics with 64 execution units, while the 368H uses Arc B390. The 285K has an unlocked multiplier, while the 368H does not. The 285K's part number is SRQD5, and the 368H's part number is SA4R7Q9EJ.

Specification Differences

The core counts differ substantially. The 285K has 24 cores and 24 threads, while the 368H has 16 cores and 16 threads. Neither processor uses hyper-threading, so threads equal cores in both cases.

Clock speeds show a clear split. The 285K has a base clock of 3.70 GHz and a boost clock of 5.70 GHz. The 368H has a base clock of 2.00 GHz and a boost clock of 5.00 GHz. The desktop part runs at higher clocks across the board.

Thermal design power is a major differentiator. The 285K has a TDP of 125 watts, while the 368H has a TDP of 25 watts. This reflects the desktop versus mobile positioning.

The socket types are incompatible. The 285K uses Intel Socket 1851, while the 368H uses Intel BGA 2540. The 285K is a desktop socketed part, and the 368H is a mobile BGA part.

Release dates differ. The 285K was released on 2024-10-23, while the 368H was released on 2026-01-04. The 285K has a launch MSRP of $589; no launch MSRP is recorded for the 368H.

Production status is active for both. The 285K belongs to the Core Ultra Series 2, and the 368H belongs to the Core Ultra Series 3. The 285K's generation is recorded as Ultra 9 (Arrow Lake), and the 368H's generation is recorded as Ultra X7 (Panther Lake-H).

DETAILED SPECIFICATIONS

SPECIFICATION
Ultra 9 285K
Ultra X7 368H
Core Specs
Cores
24
16 -33.3%
Threads
24
16 -33.3%
Base Clock (GHz)
3.7
2 -45.9%
Boost Clock (GHz)
5.7
5 -12.3%
Frequency (GHz)
3.7
2 -45.9%
Turbo Clock (GHz)
5.7
5 -12.3%
Multiplier
37
20 -45.9%
SMP CPUs
1
1 0.0%
Cache
L1 Cache
192 KB (per core)
192 KB (per core)
L2 Cache
3 MB (per core)
2.5 MB (per core)
L3 Cache
36 MB (shared)
18 MB (shared)
Power
TDP (W)
125
25 -80.0%
PL1
250 W
—
PL2
250 W
—
Configurable TDP
—
45 W
Architecture
Architecture
Arrow Lake
Panther Lake
Codename
Arrow Lake-S
Panther Lake
Generation
Ultra 9 (Arrow Lake)
Ultra X7 (Panther Lake-H)
Process Size
3 nm
3 nm
Transistors
17,800 million
—
Die Size
243 mm²
—
Foundry
TSMC
Intel
Memory
Memory Support
DDR5
LPDDR5X
Memory Bus
Dual-channel
Dual-channel
Memory Bandwidth
102.4 GB/s
153.6 GB/s
ECC Memory
Yes
No
Platform
Socket
Intel Socket 1851
Intel BGA 2540
Chipsets
Z890, B860, W880, Q870, H810
—
PCIe
Gen 5, 20 Lanes(CPU only)
Gen 5, 4 Lanes(CPU only)
Intel Hybrid
Hybrid Cores
P-Cores: 8 E-Cores: 16
P-Cores: 4 E-Cores: 12
E-Core Frequency
3.2 GHz up to 4.6 GHz
1600 MHz up to 3.8 GHz
P-Core Turbo
5.5 GHz
—
LP E-Cores
—
4
AI/NPU
NPU
Yes / 13 TOPS
Yes / 50 TOPS
Graphics
Integrated Graphics
Arc Xe-LPG Graphics 64EU
Arc B390
Other
Market
Desktop
Mobile
Production Status
Active
Active
Launch Price
$589
—
Part Number
SRQD5
SA4R7Q9EJ
Package
FC-LGA18W
FC-BGA
Tj Max
105°C
100°C
View Core Ultra 9 285K Details View Core Ultra X7 368H Details