AMD Ryzen AI Embedded P164 vs Intel Arc G3 EXTREME Comparison

FAQ

Q: Which processor has the higher multithreaded PassMark score?

A: The Intel Arc G3 EXTREME records a PassMark multithread score of 30659, which is 15.6% higher than the AMD Ryzen AI Embedded P164's 25889.

Q: How much faster is the AMD part in integer math workloads?

A: The AMD Ryzen AI Embedded P164 scores 87940 in PassMark integer math, a 23.6% advantage over the Intel Arc G3 EXTREME's 71164.

Q: Which processor leads in single-thread performance?

A: The Intel Arc G3 EXTREME scores 4293 in PassMark single-thread tests, 6.1% ahead of the AMD Ryzen AI Embedded P164's 4029.

Q: What is the difference in average benchmark scores between the two?

A: The AMD Ryzen AI Embedded P164 has an average benchmark score of 52901 and sits in the 91st percentile, while the Intel Arc G3 EXTREME averages 36699 and sits in the 85th percentile.

Q: Which processor has more cores and threads?

A: The Intel Arc G3 EXTREME has 14 cores and 14 threads, while the AMD Ryzen AI Embedded P164 has 8 cores and 16 threads.

Q: What are the closest rivals for each processor based on average score?

A: The AMD Ryzen AI Embedded P164's nearest rival is the AMD Ryzen 5 9500F with an average score of 52873 and a delta of 0.1%. The Intel Arc G3 EXTREME's nearest rival is the Intel Core Ultra 5 225 with an average score of 36938 and a delta of -0.6%.

Architecture Differences

The AMD Ryzen AI Embedded P164 uses the Gorgon Point codename with a Zen 5 / Zen 5c generation, built on a 4 nm process at TSMC. The Intel Arc G3 EXTREME uses the Panther Lake codename with an Arc G3 (Panther Lake) generation, built on a 3 nm process at Intel's own foundry. This process difference gives Intel a smaller transistor geometry, though the die size for the Intel part is not recorded in the database, while the AMD chip measures 233 mm².

Core counts differ significantly. Intel fields 14 cores and 14 threads, while AMD uses 8 cores and 16 threads. The thread count on the AMD part indicates simultaneous multithreading, which Intel's 14-core configuration does not employ. Cache hierarchies also diverge: AMD allocates 80 KB of L1 per core and 1 MB of L2 per core, with 8 MB of L3. Intel provides 192 KB of L1 per core and 2.5 MB of L2 per core, with a larger 18 MB shared L3 pool.

Memory support separates the two as well. Both support dual-channel memory, but AMD accepts DDR5 and LPDDR5X, while Intel only lists LPDDR5X. Memory bandwidth favors Intel at 136.5 GB/s versus AMD's 89.6 GB/s. ECC memory is supported on the AMD part but not on the Intel part. PCIe connectivity also differs: AMD offers Gen 4 with 16 CPU lanes, while Intel provides Gen 5 with 4 CPU lanes. Integrated graphics are distinct, with AMD using the Radeon 880M and Intel using the Arc B390.

Where Each One Wins

The AMD Ryzen AI Embedded P164 wins 3 of the 11 head-to-head benchmark comparisons. Its victories come in data compression, extended instructions, and integer math. The data compression score of 327891 is 6.8% higher than Intel's 307094, which indicates a meaningful edge in workloads that rely on repetitive data manipulation. The integer math result of 87940 beats Intel by 23.6%, a substantial gap that suggests AMD's cores handle general arithmetic operations with greater efficiency. The extended instructions score of 24193 is only 0.7% ahead, a marginal win.

The Intel Arc G3 EXTREME wins 8 of the 11 comparisons. Its largest wins come in find prime numbers, where it scores 248 versus AMD's 71, a 71.4% advantage. Physics simulation also shows a strong Intel lead: 2515 versus 1210, a 51.9% difference. Floating point math favors Intel by 35.8% with scores of 86929 versus 55799. Data encryption shows a 32.8% Intel advantage (23881 versus 16055). The multithread score is 15.6% higher on Intel, and random string sorting is 6.8% higher. Single-thread performance is 6.1% higher on Intel.

The split is clear: AMD dominates integer-heavy and compression tasks, while Intel leads in floating point, encryption, prime number generation, physics, and overall multithread throughput. The Intel part also holds the single-thread edge.

Specification Differences

The two processors differ across every major specification field recorded in the database.

• Cores: AMD 8, Intel 14
• Threads: AMD 16, Intel 14
• Base clock: AMD 2.00 GHz, Intel 1.90 GHz
• Boost clock: AMD 5.00 GHz, Intel 4.70 GHz
• TDP: AMD 28 W, Intel 25 W
• Socket: AMD Socket FP8, Intel BGA 2540
• Codename: Gorgon Point, Panther Lake
• Generation: Ryzen AI Embedded (Zen 5 / Zen 5c), Arc G3 (Panther Lake)
• Process node: 4 nm, 3 nm
• Foundry: TSMC, Intel
• Die size: 233 mm², not recorded
• L1 cache: 80 KB per core, 192 KB per core
• L2 cache: 1 MB per core, 2.5 MB per core
• L3 cache: 8 MB, 18 MB shared
• Memory support: DDR5, LPDDR5X, LPDDR5X only
• Memory bandwidth: 89.6 GB/s, 136.5 GB/s
• ECC memory: Supported, not supported
• PCIe: Gen 4, 16 lanes, Gen 5, 4 lanes
• Integrated graphics: Radeon 880M, Arc B390
• Part number: unknown, SA4R3
• Release date: 2026-03-08, 2026-05-27
• Multiplier unlocked: False for both

The AMD part has a higher base and boost clock, a higher TDP, and a larger L3 cache is absent (it has 8 MB total L3). The Intel part has more cores, more L1 and L2 per core, a much larger shared L3, higher memory bandwidth, and a newer process node. The release dates are separated by about two and a half months, with AMD launching first.

Head-to-Head Benchmarks

The benchmark data shows a lopsided result overall, but the magnitude of each win varies considerably. Starting with AMD's strongest performance, the integer math score of 87940 represents a 23.6% lead over Intel's 71164. This is the largest margin in AMD's favor across all tests. Data compression also favors AMD, with 327891 versus 307094, a 6.8% difference. Extended instructions are nearly tied, with AMD ahead by just 0.7% (24193 versus 24017).

For Intel, the most dramatic win is find prime numbers. The Intel part scores 248, which is 71.4% higher than AMD's 71. This is the single largest relative difference in the entire comparison. Physics simulation shows a 51.9% Intel advantage (2515 versus 1210), and floating point math follows at 35.8% (86929 versus 55799). Data encryption gives Intel a 32.8% edge (23881 versus 16055). The multithread score of 30659 is 15.6% ahead of AMD's 25889. Random string sorting is a closer contest, with Intel at 37331 versus 34801, a 6.8% gap. Single-thread performance is also an Intel win at 4293 versus 4029, a 6.1% margin.

The average benchmark scores confirm the overall positioning. AMD's average of 52901 places it in the 91st percentile of all CPUs, while Intel's average of 36699 places it in the 85th percentile. The nearest rival data reinforces this: AMD's closest competitor is the AMD Ryzen 5 9500F at 52873, a 0.1% difference, while Intel's closest is the Intel Core Ultra 5 225 at 36938, a -0.6% difference. The percentile gap of 6 points and the average score gap of 16202 indicate that the AMD part occupies a higher performance tier overall, despite losing 8 of the 11 individual head-to-head comparisons. The wins Intel does secure are often by large margins, but AMD's wins in data compression and integer math, plus its strong showing in extended instructions, keep its average score substantially higher.