Scored using Processors Methodology v3.0
Processors Methodology v3.0
Mathematical & Unbiased Scoring
Products in this category are scored under Frontum Processors Methodology v3.0: the technical score from the specifications is the base, and independent expert and video reviews (up to ±0.5 points), owner opinions (up to ±0.3) and best-of lists (up to +0.3) move it up or down.
Explore Processors Methodology v3.0

AMD Ryzen AI Max 390 review: 74 specs compared

Avg. price:~£1,870PassMark multi-core score:41552N. of physical cores:12CPU boost clock speed:5.0 GHzCPU socket:FP11TDP (Thermal design power):55 W

AMD Ryzen AI Max 390 Review: Compare 74 technical specifications to see how this product ranks in the Processors category and whether it is worth buying.

AMD Ryzen AI Max 390
7.7/10

Overall score

Score
7.7/ 10
✕
What it is
Our main score for this model, 1–10. It combines specs, expert reviews, best-of lists and owners.
How we count it
Parts: Technical score 40%, Reviews score 25%, List score 25%, Sentiment score 10%
Marks: Each part gets a mark out of 10. For reviews, lists and owners, a typical model in this category gets about 8 and the best one gets 10.
No data yet: A part with no data shows "?". It counts as the mark of a weaker model in this category (4 in 5 models with data score higher). This way, models with reviews, lists and owners usually come out ahead.
Score distribution
â–¼
<55-66-77-88-99-10
Score components

40%

8.4

Technical score

25%

?

Reviews scoreNot enough data yet, so we use 7.2, the mark of a weaker model in this category

25%

?

List scoreNot enough data yet, so we use 7.5, the mark of a weaker model in this category

10%

?

Sentiment scoreNot enough data yet, so we use 7.0, the mark of a weaker model in this category

â–²Better than 87% of processors
5.7

Value for money

Value for money
5.7/ 10
✕
What it is
An indicator that combines the processor's overall rating with its cost.
When it matters
When you are looking for a processor with a good balance between performance, efficiency, and price.
Score distribution
â–¼
<55-66-77-88-99-10
Score components

70%

7.7

Overall score

30%

1.0

Price

â–¼Worse than 99%
8.4

Technical score

Technical score
8.4/ 10
✕
What it is
How good the specs are compared with the other models in this category, 1–10.
How we count it
Areas: We rate the specs in each area against the other models in this category.
Weighting: The areas that matter most to buyers count more. The shares are shown under Score components.
Share: The technical score is 40% of the overall score.
Score distribution
â–¼
<55-66-77-88-99-10
Score components

60%

8.7

Performance

18%

8.3

Architecture

10%

7.5

Memory & PCIe

7%

7.0

Power & Thermal

4%

8.9

Platform

1%

9.3

Integrated graphics

★Better than 93%
?

Reviews score

Reviews score
✕
What it is
Average grade from independent test sites, magazines and video reviewers, 1–10.
How we count it
Grades: Each review's own score is converted to 1–10; without one, we grade the review ourselves. Big test sites count more, short reviews half as much as full tests. A video review counts half.
Score: The average grade is 80% of the review mark; the number of reviews is the other 20% (1 review = 5, the typical number in this category = 8, the most-reviewed model = 10).
Enough data: With few reviews, the average grade is pulled toward the typical grade of this category. No reviews yet: 7.2 for now.
Share: Reviews count 25% of the overall score.
Score distribution
â–¼
<55-66-77-88-99-10
Grades by review (0)
We have not found an independent test of this model yet. Until we do, its review part counts as 7.2.
User ratings collected online are not part of this score.
No expert tests yet
?

List score

List score
7.5/ 10
✕
What it is
How often independent test sites, magazines and video reviewers put this model in their best-of lists for this category.
How we count it
Points: The model in the most lists of this category gets 10; the others get 8 to 10 in proportion. In no list yet: 7.5. Lists count 25% of the overall score.
Enough data: Every list counts, even a single one. With fewer than 5 best-of lists in the category, this part is left out.
Lists with this model (0)
Not in any best-of list yet, so this part counts as 7.5. Models in a list get 8 to 10.
Not in any best-of list yet
  • amd-ryzen-ai-max-390
amd-ryzen-ai-max-390
Where to buy
    N/A~ £1,870
Best rankings
Available: ranking among products currently available (including other versions of this product).All: ranking among all products in the database.
Loading...
Loading category ranking...

Technical Specifications of processor AMD Ryzen AI Max 390

Technical score
Technical score
8.4/ 10
✕
What it is
How good the specs are compared with the other models in this category, 1–10.
How we count it
Areas: We rate the specs in each area against the other models in this category.
Weighting: The areas that matter most to buyers count more. The shares are shown under Score components.
Share: The technical score is 40% of the overall score.
Score distribution
â–¼
<55-66-77-88-99-10
Score components:

60%

8.7

Performance

18%

8.3

Architecture

10%

7.5

Memory & PCIe

7%

7.0

Power & Thermal

4%

8.9

Platform

1%

9.3

Integrated graphics

8.4
Reviews score
7.2
List score
?
Value for money
Value for money
5.7/ 10
✕
What it is
An indicator that combines the processor's overall rating with its cost.
When it matters
When you are looking for a processor with a good balance between performance, efficiency, and price.
Score distribution
â–¼
<55-66-77-88-99-10
Score components:

70%

7.7

Overall score

30%

1.0

Price

5.7
Processor type
mobile
This processor belongs to the mobile processor class, which is more advanced than that of 6.9% of processors and equal to that of 47.7% of processors.
CPU socket
FP11
This processor uses the FP11 CPU socket, which is newer than that of 52.2% of processors and equal to that of 0.6% of processors.
Chipset
Z790, B660, X670E
This processor supports Z790, B660, X670E chipsets, which is broader compatibility than 89.9% of processors.
CPU architecture
x86-64
This processor uses the x86-64 architecture, which is more advanced than that of 1.8% of processors and equal to that of 98.2% of processors.
Instruction sets
MMX, SSE, SSE2, SSE3, SSSE3, SSE4A, SSE4.1, SSE4.2, AVX, AVX2, AVX-512, FMA3, AES, SHA, AMD-V
This processor supports MMX, SSE, SSE2, SSE3, SSSE3, SSE4A, SSE4.1, SSE4.2, AVX, AVX2, AVX-512, FMA3, AES, SHA, AMD-V instruction sets, which is broader support than 69.3% of processors and equal to that of 7.1% of processors.
Show more
N. of physical cores
12
This processor has 12 CPU cores, which is more than 79.9% of processors and equal to 5.7% of processors.
CPU threads
24
This processor offers 24 CPU threads, which is more than 89.7% of processors and equal to 4.4% of processors.
CPU boost clock speed
5 GHz
This processor reaches a boost clock of 5 GHz which is higher than that of 78.5% of processors and equal to that of 5.9% of processors.
CPU base clock speed
12x 3.2 GHz
This processor has a base clock of 12x3.2 GHz which is higher than that of 90.6% of processors and equal to that of 0.1% of processors.
PassMark multi-core score
41,552
This processor scores 41,552 in PassMark, which is higher than 93.2% of processors.
Show more
Semiconductor size
4 nm
This processor uses a 4 nm process node, which is more advanced than that of 86.8% of processors and equal to that of 7.7% of processors.
Foundry
TSMC 4 nm
This processor is built on the TSMC 4 nm foundry process, which is more advanced than that of 86.4% of processors and equal to that of 7.8% of processors.
L3 cache
64 MB
This processor has an L3 cache of 64 MB which is larger than that of 94.5% of processors and equal to that of 2.8% of processors.
L2 cache
12 MB
This processor has an L2 cache of 12 MB which is larger than that of 83.1% of processors and equal to that of 2.9% of processors.
L1 cache
960 KB
This processor has an L1 cache of 960 KB which is larger than that of 84.6% of processors and equal to that of 1% of processors.
Show more
DDR memory version
DDR5
This processor supports DDR DDR5, which is newer than that of 64.2% of processors and equal to that of 12.8% of processors.
Maximum memory speed
8,000 MHz
This processor supports memory speeds up to 8,000 MHz, which is higher than that of 95.9% of processors and equal to 1.4% of processors.
Max memory speed (JEDEC)
LPDDR5X-8000 MHz
This processor supports JEDEC memory speeds up to LPDDR5X-8000 MHz, which is higher than that of 96.4% of processors and equal to 1.5% of processors.
Max memory speed (XMP / EXPO)
N/A
Maximum memory capacity
128 GB
This processor supports up to 128 GB of memory, which is more than 59.3% of processors and equal to 22.8% of processors.
Show more
Integrated graphics
yes
This processor includes integrated graphics. 14.7% of processors offer this.
Integrated GPU model
Radeon 8050S
This processor uses the Radeon 8050S integrated GPU, which is more advanced than that in 98.2% of processors and equal to that in 0.2% of processors.
Integrated GPU execution units
32
This processor has 32 GPU execution units, which is more than 70.1% of processors and equal to 10.1% of processors.
Integrated GPU base frequency
1,295 MHz
This processor has an integrated GPU clock of 1,295 MHz which is higher than that of 95.9% of processors and equal to that of 0.6% of processors.
Integrated media encoders/decoders
H.264 (HW decode/encode), H.265 (HW decode/encode), VP9 (HW decode), AV1 (HW decode/encode)
This processor supports H.264 (HW decode/encode), H.265 (HW decode/encode), VP9 (HW decode), AV1 (HW decode/encode) media codecs, which is broader support than 90.8% of processors and equal to 3.4% of processors.
Show more
TDP (Thermal design power)
55 W
This processor has a TDP of 55 W which is higher than that of 61.7% of processors and equal to that of 4.2% of processors.
Base power (PL1)
55 W
This processor has a base power of 55 W which is higher than that of 61.8% of processors and equal to that of 4.3% of processors.
Boost power (PL2)
?
Tau (power duration limit)
N/A
Configurable TDP
yes
This processor supports configurable TDP. 46.4% of processors offer this.
Show more

AMD Ryzen AI Max 390 vs the average processor

246810TotalPlatformPerformanceArchitectureMemory & PCIeIntegrated graphicsPower & Thermal
  • 3.37x higher multi-core score
    This processor has a higher Geekbench 6 multi-core score than the average processor (17,557 vs 5,209). The average processor scores 5,209 in Geekbench 6 multi-core.
    17,557 vs 5,209
  • 3.62x higher PassMark score
    This processor has a higher PassMark benchmark score than the average processor (41,552 vs 11,469). The average processor scores 11,469 in PassMark benchmark.
    41,552 vs 11,469
  • More advanced microarchitecture
    This processor uses a more advanced microarchitecture than the average processor (Strix Halo vs Kaby Lake).
    Strix Halo vs Kaby Lake
  • 8 more performance cores
    This processor has more performance cores than the average processor (12 vs 4). The average processor has 4 performance cores.
    12 vs 4
  • 57.2% better single-core performance
    This processor has a higher PassMark single-core score than the average processor (4,020 vs 2,558). The average processor scores 2,558 in PassMark single-core.
    4,020 vs 2,558
  • 16 more CPU threads
    This processor has more CPU threads than the average processor (24 vs 8). The average processor has 8 CPU threads.
    24 vs 8
  • 77.8% higher single-core score
    This processor has a higher Geekbench 6 single-core score than the average processor (2,729 vs 1,535). The average processor scores 1,535 in Geekbench 6 single-core.
    2,729 vs 1,535
  • 6 more CPU cores
    This processor has more CPU cores than the average processor (12 vs 6). The average processor has 6 CPU cores.
    12 vs 6
  • 5 years newer release date
    This processor has a newer release date than the average processor (2025 vs 2020).
    January 2025
    2025 vs 2020
  • 3.37x higher multi-core score
    This processor has a higher Geekbench 6 multi-core score than the average processor (17,557 vs 5,209). The average processor scores 5,209 in Geekbench 6 multi-core.
    17,557 vs 5,209
  • 3.62x higher PassMark score
    This processor has a higher PassMark benchmark score than the average processor (41,552 vs 11,469). The average processor scores 11,469 in PassMark benchmark.
    41,552 vs 11,469
  • 8 more performance cores
    This processor has more performance cores than the average processor (12 vs 4). The average processor has 4 performance cores.
    12 vs 4
  • 57.2% better single-core performance
    This processor has a higher PassMark single-core score than the average processor (4,020 vs 2,558). The average processor scores 2,558 in PassMark single-core.
    4,020 vs 2,558
  • 16 more CPU threads
    This processor has more CPU threads than the average processor (24 vs 8). The average processor has 8 CPU threads.
    24 vs 8
  • 77.8% higher single-core score
    This processor has a higher Geekbench 6 single-core score than the average processor (2,729 vs 1,535). The average processor scores 1,535 in Geekbench 6 single-core.
    2,729 vs 1,535
  • 6 more CPU cores
    This processor has more CPU cores than the average processor (12 vs 6). The average processor has 6 CPU cores.
    12 vs 6
  • 63.1% higher Cinebench R20 single-core score
    This processor has a higher Cinebench R20 single-core score than the average processor (799 vs 490). The average processor scores 490 in Cinebench R20 single-core.
    799 vs 490
  • 13.6% higher boost clock
    This processor has a higher boost clock speed than the average processor (5 GHz vs 4.4 GHz). The average processor reaches boost clock speed of 4.4 GHz.
    5 GHz vs 4.4 GHz
  • Higher base clock
    This processor has a higher base clock than the average processor (12x3.2 GHz vs 4x3.4 GHz).
    12x3.2 GHz vs 4x3.4 GHz
  • 4 wider front-end design
    This processor has a higher front-end width than the average processor (8 vs 4). The average processor uses front-end width of 4.
    8 vs 4
  • More advanced microarchitecture
    This processor uses a more advanced microarchitecture than the average processor (Strix Halo vs Kaby Lake).
    Strix Halo vs Kaby Lake
  • 7.11x larger L3 cache
    This processor has a higher L3 cache than the average processor (64 MB vs 9 MB). The average processor has L3 cache of 9 MB.
    64 MB vs 9 MB
  • 4x larger L2 cache
    This processor has a higher L2 cache than the average processor (12 MB vs 3 MB). The average processor has L2 cache of 3 MB.
    12 MB vs 3 MB
  • 2.88x more L3 per core
    This processor has more L3 cache per core than the average processor (5.3 MB/core vs 1.85 MB/core). The average processor provides 1.85 MB/core of L3 cache per core.
    5.33 MB/core vs 1.85 MB/core
  • 66.7% smaller process node
    This processor has a lower process node than the average processor (4 nm vs 12 nm). The average processor uses a process node of 12 nm.
    4 nm vs 12 nm
  • More advanced foundry
    This processor uses a more advanced foundry process than the average processor (TSMC 4 nm vs Intel 14 nm).
    TSMC 4 nm vs Intel 14 nm
  • 2.5x larger L1 cache
    This processor has a higher L1 cache than the average processor (960 KB vs 384 KB). The average processor has L1 cache of 384 KB.
    960 KB vs 384 KB
  • 2x more L2 per core
    This processor has more L2 cache per core than the average processor (1 MB/core vs 0.5 MB/core). The average processor provides 0.5 MB/core of L2 cache per core.
    1 MB/core vs 0.5 MB/core
  • 5.46x higher memory bandwidth
    This processor has a higher memory bandwidth than the average processor (256 GB/s vs 46.9 GB/s). The average processor offers memory bandwidth of 46.9 GB/s.
    256 GB/s vs 46.9 GB/s
  • 2.5x higher memory speed
    This processor has a higher maximum memory speed than the average processor (8,000 MHz vs 3,200 MHz). The average processor supports memory speed of 3,200 MHz.
    8,000 MHz vs 3,200 MHz
  • Newer PCIe version
    This processor supports a newer PCIe version than the average processor (4 vs 3).
    4 vs 3
  • Newer DDR support
    This processor supports a newer DDR generation than the average processor (DDR5 vs DDR4).
    DDR5 vs DDR4
  • Higher JEDEC memory speed
    This processor has a higher JEDEC memory speed than the average processor (LPDDR5X-8000 MHz vs DDR4-3200 MHz). The average processor supports JEDEC memory speed of DDR4-3200 MHz.
    LPDDR5X-8000 MHz vs DDR4-3,200 MHz
  • 2x more memory capacity
    This processor has more maximum memory capacity than the average processor (128 GB vs 64 GB). The average processor supports 64 GB of memory.
    128 GB vs 64 GB
  • Supports ECC memory
    This processor supports ECC memory, while the average processor does not. 32% of processors support it.
  • Better integrated GPU
    This processor uses a better integrated GPU than the average processor (Radeon 8050S vs Intel UHD Graphics 630).
    Radeon 8050S vs Intel UHD Graphics 630
  • 3.7x higher GPU clock speed
    This processor has a higher integrated GPU frequency than the average processor (1,295 MHz vs 350 MHz). The average processor has integrated GPU frequency of 350 MHz.
    1,295 MHz vs 350 MHz
  • 1 more supported displays
    This processor has more supported displays than the average processor (4 vs 3). The average processor supports 3 displays.
    4 vs 3
  • 2.49x larger die size
    This processor has a higher die size than the average processor (441 mm² vs 177 mm²). The average processor has a die size of 177 mm².
    441 mm² vs 177 mm²
  • 7.48x more expensive
    AMD Ryzen AI Max 390 is more expensive than the average processor (£1,870 vs £250).
    £1,870 vs £250
  • 27.4% worse value for money
    AMD Ryzen AI Max 390 has worse value for money than the average processor (5.72 vs 7.3).
    5.72 vs 7.29
  • 22.2% higher base power
    This processor has a higher base power draw than the average processor (55 W vs 45 W). The average processor has a base power draw of 45 W.
    55 W vs 45 W
  • 16% slower Blender rendering
    This processor has a higher Blender BMW27 render time than the average processor (334.5 vs 398). The average processor needs 398.015 for the Blender BMW27 test.
    334.47 vs 398.015
  • Limited PCIe bifurcation
    This processor supports less flexible PCIe bifurcation than the average processor (x16, x8/x8, x8/x4/x4 vs x16, x8/x8, x4/x4/x4/x4).
    x16, x8/x8, x8/x4/x4 vs x16, x8/x8, x4/x4/x4/x4
  • 2 fewer memory channels
    This processor has fewer memory channels than the average processor (4 vs 2). The average processor supports 2 memory channels.
    4 vs 2
  • Older TPM support
    This processor supports an older TPM version than the average processor (fTPM 2 vs PTT 2).
    fTPM 2.0 vs PTT 2.0
  • 7.48x more expensive
    AMD Ryzen AI Max 390 is more expensive than the average processor (£1,870 vs £250).
    £1,870 vs £250
  • 27.4% worse value for money
    AMD Ryzen AI Max 390 has worse value for money than the average processor (5.72 vs 7.3).
    5.72 vs 7.29
  • Older TPM support
    This processor supports an older TPM version than the average processor (fTPM 2 vs PTT 2).
    fTPM 2.0 vs PTT 2.0
  • 16% slower Blender rendering
    This processor has a higher Blender BMW27 render time than the average processor (334.5 vs 398). The average processor needs 398.015 for the Blender BMW27 test.
    334.47 vs 398.015
  • 2.49x larger die size
    This processor has a higher die size than the average processor (441 mm² vs 177 mm²). The average processor has a die size of 177 mm².
    441 mm² vs 177 mm²
  • Limited PCIe bifurcation
    This processor supports less flexible PCIe bifurcation than the average processor (x16, x8/x8, x8/x4/x4 vs x16, x8/x8, x4/x4/x4/x4).
    x16, x8/x8, x8/x4/x4 vs x16, x8/x8, x4/x4/x4/x4
  • 2 fewer memory channels
    This processor has fewer memory channels than the average processor (4 vs 2). The average processor supports 2 memory channels.
    4 vs 2
  • 22.2% higher base power
    This processor has a higher base power draw than the average processor (55 W vs 45 W). The average processor has a base power draw of 45 W.
    55 W vs 45 W
  • 22.2% higher TDP
    This processor has a higher TDP than the average processor (55 W vs 45 W). The average processor has a TDP of 45 W.
    55 W vs 45 W

Video reviews

Hands-on testing, unboxing, and real-world video evaluations

Overview video

Third-party reviews

What customers like about AMD Ryzen AI Max 390

  • Exceptional workstation performance with 12 full Zen 5 cores, excelling in multi-threaded tasks like code compilation and Linux kernel builds.
  • Integrated Radeon 8050S graphics offer 'discrete-class' performance, rivaling entry-to-mid-range dedicated GPUs (like the RTX 4060) in some scenarios.
  • Massive AI capabilities with a 50 TOPS NPU and support for up to 96GB of unified memory for large language models.
  • Higher base clock (3.2GHz) compared to the flagship 395 model, providing a slight edge in bursty single-core workloads like web browsing.
  • Impressive performance-per-watt in low-power modes, making it a viable option for high-end thin-and-light workstations.

What customers don't like about AMD Ryzen AI Max 390

  • High platform costs; devices like the ASUS ROG Flow Z13 or HP ZBook Ultra G1a carrying these chips occupy a premium, often expensive price tier.
  • Power consumption can peak up to 120W TDP, which may lead to significant fan noise and heat in smaller laptop chassis.
  • Lower performance-per-watt efficiency compared to the 16-core flagship (395) at the same default 55W TDP.
  • Soldered LPDDR5X memory in most implementations limits future upgradability.
  • Niche market appeal; the heavy focus on 'Halo' performance makes it overkill for average users and potentially overpriced for budget gamers.

Verdict

The AMD Ryzen AI Max 390 is a high-performance 12-core, 24-thread mobile processor built on the 4nm TSMC Zen 5 'Strix Halo' architecture, featuring a 3.2 GHz base and up to 5.0 GHz boost clock. Key characteristics include an integrated Radeon 8050S GPU with 32 RDNA 3.5 compute units, a dedicated XDNA 2 NPU delivering 50 TOPS (110 TOPS total chip performance), and support for up to 128GB of LPDDR5x-8000 unified memory. Its main pros are workstation-class CPU and iGPU performance in portable form factors, high-speed quad-channel memory bandwidth, and the ability to run large language models (LLMs) locally. Cons include a high configurable TDP up to 120W that requires robust cooling, potential value competition from the 16-core flagship Max+ 395, and a lack of support for the VVC video codec.

Compare AMD Ryzen AI Max 390 with other processors

1Product to compare no. 1
VS
2Product to compare no. 2
VS
3Product to compare no. 3 (optional)

Graphic comparison of AMD Ryzen AI Max 390 and other processors

No results found