Home > GPU Comparison > RTX 4090 vs M4 Max GPU (40-core) – what's better?

GeForce RTX 4090 vs Apple M4 Max GPU (40-core)

GeForce RTX 4090
VS
Apple M4 Max GPU (40-core)
Nvidia GeForce RTX 4090
AD102
Apple M4 Max GPU (40-core)
Custom

We performed a head-to-head comparison of the GeForce RTX 4090 24 GB with 128 pipelines and 16384 shaders against the 2 years and 2 months newer Apple M4 Max GPU (40-core) that utilizes 640 pipelines and 5120 shaders. Here you will find complete details about specs, efficiency, performance tests, and more.

  1. Review
  2. Differences
  3. Gaming performance
  4. Benchmarks
  5. AI
  6. Specs
  7. Comments (6)

Review

General comparison of performance in games, applications, power efficiency, and other metrics
Gaming
Performance in DirectX, OpenCL, and Vulkan games
Workstation
Perf. in 3D modeling, video editing and rendering apps
AI/ML
Capabilities for machine learning and AI-related tasks
Energy Efficiency
Power consumption efficiency in different scenarios
NanoReview Final Score
Overall video card score
The "Energy Efficiency" metric has less impact on the NanoReview Score for desktop GPU.

Key differences

Key distinctions and advantages of M4 Max GPU (40-core) over RTX 4090
Reasons to consider the GeForce RTX 4090
  • Performs significantly better (up to 2.9x) in 3DMark Steel Nomad Lite
  • 5.1x higher maximum theoretical performance (82.6 vs 16.2 TFLOPS)
  • Supports Nvidia DLSS 3 technology
  • Achieves 2.7x more points in the GeekBench 6 Compute test (316K vs 116K)
  • Has 85% higher memory bandwidth: 1010 vs 546 GB/s
  • Has 3.2x more shading units (16384 vs 5120)
Reasons to consider the Apple M4 Max GPU (40-core)
  • Manufactured using a more efficient 3 nm process technology

Gaming Performance

Frame rate comparison across popular AAA titles at different resolutions
GPU
RTX 4090
vs
M4 Max GPU (40-core)

Games

FPS Table
Forza Horizon 5
275
199
193
155
 -
The Witcher 3
292
268
191
101
 -
Counter-Strike 2
339
238
188
90
 -
Far Cry 6
203
179
170
104
 -
Hogwarts Legacy
200
163
132
74
 -
CoD: Modern Warfare III
266
263
219
151
 -
Ghost of Tsushima
188
152
137
91
 -
Cyberpunk 2077
226
208
143
66
 -
Shadow of the Tomb Raider
302
284
222
118
 -
Average FPS by Resolution
1080p High 255 -
1080p Ultra 217 -
1440p Ultra 177 -
4K Ultra 106 -
Margin of Error Low Low

Benchmarks

Graphics cards’ performance in recent benchmarking apps

3D Mark

Multiplatform graphics benchmark suite that directly correlates with performance in modern games
Steel Nomad Lite Score
RTX 4090 +190%
42169
M4 Max GPU (40-core)
14542
Time Spy 36299 -
Solar Bay 187525 61312
Port Royal 26127 -
Fire Strike 72432 -
Wild Life Extreme 85218 36641
Night Raid 195813 -
Sources: 3DMark [1], [2]

GeekBench 6 OpenCL

GPU test for computational tasks (image processing, photography, computer vision, and ML)
GB6 Compute Score
RTX 4090 +171%
316301
M4 Max GPU (40-core)
116754
Background Blur 318.9 img/sec 195 img/sec
Face Detection 222.8 img/sec 125.9 img/sec
Horizon Detection 14.2 Gpixels/sec 4.82 Gpixels/sec
Edge Detection 21 Gpixels/sec 7.47 Gpixels/sec
Gaussian Blur 23.9 Gpixels/sec 5.8 Gpixels/sec
Feature Matching 2.38 Gpixels/sec 1.06 Gpixels/sec
Stereo Matching 1950 Gpixels/sec 417.4 Gpixels/sec
Particle Physics 48090.4 FPS 17079.9 FPS
API OpenCL OpenCL
Sources: Geekbench [3], [4]

Cinebench 2024 GPU

Hardware benchmark using Maxon's Cinema 4D rendering engine
Cinebench 2024 GPU
RTX 4090 +94%
32578
M4 Max GPU (40-core)
16783

Passmark Graphics

Videocard test that focuses on compute shaders, multi-texturing, tessellation, and other features
G3D Mark Score
G2D Mark 1303 -
DirectX 11 323 FPS -
DirectX 12 151 FPS -
GPU Compute 25991 Ops/s -
Sources: PassMark [5] – 22563 samples

Blender

Rendering performance test for 3D modeling
Blender GPU
RTX 4090 +124%
11681.51
M4 Max GPU (40-core)
5208.05
Sources: Blender [9], [10] – 8966 & 460 samples
Add your GPU benchmark result

Recent User Tests

The latest benchmark tests that have been submitted by users
GeForce RTX 4090
DateBenchmarkResult
📘 2026-04-03 (afroman420IU)Cinebench 202432578
Add View all (1)
Apple M4 Max GPU (40-core)
DateBenchmarkResult
📘 2025-07-10 (thomas)Geekbench 6 OpenCL116478
Add View all (1)

Artificial Intelligence Tests

Performance in machine learning and artificial intelligence tasks

GeekBench 6 ML

Tests throughput of AI operations in single, half, and quantized precision
GB6 ML Single Precision
RTX 4090 +82%
42962
M4 Max GPU (40-core)
23632
GB6 ML Half Precision
RTX 4090 +129%
59636
M4 Max GPU (40-core)
26019
GB6 ML Quantized
RTX 4090 +29%
31670
M4 Max GPU (40-core)
24606
Image Classification (SP) 15600 10302
Image Segmentation (HP) 36657 23763
Image Super Resolution (Q) 48114 29314
Face Detection (HP) 78111 42016
Pose Estimation (Q) 289639 105972
Text Classification (SP) 4074 2963
Machine Translation (HP) 6349 6594
Object Detection (SP) 21186 8935
Depth Estimation (Q) 75599 43143
Style Transfer (SP) 601428 221563
Framework ONNX Core ML
Backend DirectML GPU
Sources: Geekbench [9], [10]
Results were compared using different ML frameworks

Specifications

Technical specifications of GeForce RTX 4090 and Apple M4 Max GPU (40-core)

General

Vendor Nvidia Apple
Build Discrete Integrated
Released September 20, 2022 October 30, 2024
Launch price (MSRP) $1599 -
Case Desktop Laptop
Purpose Gaming Professional
Segment High-end Mid-range
Architecture Ada Lovelace Apple M GPU
GPU Codename AD102 Custom
Rival Equivalent - - GeForce RTX 4070 Laptop
Successor - GeForce RTX 5090 - Apple M5 Max GPU (40-core)
Recommended CPU - Intel Core i9 14900K or above - Apple M4 Max (16-Core) or above
Used in CPUs - - Apple M4 Max (16-Core)
Desktop GPU rating (#3rd place)
Laptop GPU ranking (#14th place)

Graphics Processing Unit

Base Clock 2235 MHz 500 MHz
Boost Clock 2520 MHz 1578 MHz
Shading Units 16384 5120
Texture Mapping Units (TMUs) 512 320
Render Output Units (ROPs) 176 160
Compute Units (Pipelines) 128 640
Tensor Cores 512 -
Ray-tracing Cores 128 -
L1 Cache 128KB per cluster -
L2 Cache 72MB shared -
Instructions Per Cycle 2 IPC 2 IPC

Raw Performance

Pixel Fill Rate 444 GPixel/s 252 GPixel/s
Texture Fill Rate 1290 GTexel/s 505 GTexel/s
FLOPS (FP32)
RTX 4090 +410%
82.6 TFLOPS
M4 Max GPU (40-core)
16.2 TFLOPS

Physical

Interface PCIe 4.0 x16 Custom
TGP 450 W 50 W
Manufacturing TSMC TSMC
Fabrication Process 5 nm 3 nm
Die Size 609 mm² -
Transistor Count 76.3 billion -
Transistor Density 125.29 MTr/mm² -
Max. Temperature - 100°C

Memory

Memory Type GDDR6X System Shared
Memory Size 24 GB -
Memory Clock 2625 MHz 8533 MHz
Effective Memory Speed 21000 Mbps -
Bus 384-bit 512-bit
ECC No No
Memory Bandwidth
RTX 4090 +85%
1010 GB/s
M4 Max GPU (40-core)
546 GB/s

API

DirectX 12 -
Vulkan 1.3 -
OpenGL 4.6 -
OpenCL 3.0 -
CUDA 8.9 -
Ray Tracing Yes Yes
DLSS DLSS 3 No
DisplayPort 1.4a -

Cast your vote

Choose between two graphics cards
183 (65.1%)
98 (34.9%)
Total votes: 281

Related comparisons

1. Apple M4 Max GPU (40-core) vs GeForce RTX 5090 Laptop (3)
2. Apple M4 Max GPU (40-core) vs Apple M1 Ultra GPU (64-core) (2)
3. Apple M4 Max GPU (40-core) vs Apple M4 Max GPU (32-core)
4. Apple M4 Max GPU (40-core) vs GeForce RTX 5090 (2)
5. Apple M4 Max GPU (40-core) vs Apple M4 Pro GPU (20-core)
6. GeForce RTX 4090 vs Apple M3 Max GPU (40-core) (2)
7. GeForce RTX 4090 vs Nvidia RTX 5000 Ada (2)
8. GeForce RTX 4090 vs GeForce RTX 5070
9. GeForce RTX 4090 vs GeForce RTX 5090 (18)
10. GeForce RTX 4090 vs Radeon RX 9070 XT

User opinions

You can share your opinion or ask a question in the comments below
🌐 Register your profile and become part of NanoReview community!
Avatar
ComputerNerd November 1, 2024 at 10:05 PM
The M4 Max is amazing from a CPU standpoint, but it's still efficient minor league baseball in the GPU world. Nvidia and AMD high end GPUs are still major league.
+24 Reply
Avatar
GPU November 3, 2024 at 05:36 PM
Dude, don't forget this performance coming from a soc (apu). Of course, RTX 4090 is better than M4 Max but M4 Max is good enough for gaming. Only if devs brings more games to Mac.
+8 Reply
Avatar
ehhhh November 4, 2024 at 07:56 AM
the whole m4 max package including cpu, gpu and RAM runs in under 100 watts. its nuts, its just not aimed at burning 400 watts or more.
+8 Reply
Avatar
Sayaf Alwagdany November 12, 2024 at 06:05 PM
GPU, Guys there is point we need to take a close look on it while comparing. RTX-4090 needs 82.6 TFLOPS to the flowing: Pixel Fill Rate 444 GPixel/s Texture Fill Rate 1290 GTexel/s M4 Max needs 18.4 TFLOPS to the following: Pixel Fill Rate 288 GPixel/s Texture Fill Rate 576 GTexel/s Not to mention that the M4 Max has half of memory bandwidth. So, if M4 Ultra comes with close to double performance (2 x M Max) as this is their target, with 36.8 TFLOPs and the same memory bandwidth the M4 Ultra can do: Pixel Fill Rate 576 GPixel/s Texture Fill Rate 1152 GTexel/s The M4 Ultra will exceed the RTX-4090 or at least match with it while needing half of the TFLOPS. Or in another word half of the work! And if the rumors are true and there is an M4 Extreme which will be double of M4 Ultra, Nvidia has nothing to compete. This shows how Apple Architecture is good and shows the advantage of TBDR. Not to mention that no one is measuring using a Metal API for Apple Silicon which will give more performance and it is the way the GPU works while they accept to use CUDA for Nvidia!
+19 Reply
Avatar
Anonymous November 18, 2024 at 05:04 PM
Sayaf Alwagdany, Texture fill rate is a factor in GPU performance but there are many other important ones. You can't just say better texture fill rate means better performance. So an M4 Ultra will still not match a 4090 in 3D rendering, although I'm sure it will destroy it in video encode/decode (if the M4 Max doesn't already). The M Extremes have always been rumored and never one has come out, so that means nothing until we actually see it. Metal API is meaningless because no one uses it, I doubt the games industry will ever try to adopt it so even if theoretical performance is good with Metal, it's irrelevant. Apple's architecture is certainly better but that doesn't mean they are willing to let the power draw go through the roof like Nvidia, a disadvantage for them. Because in the end, 99% of people will take performance over power efficiency any day.
+9 Reply
Avatar
Also anonymous August 7, 2025 at 09:29 PM
Anonymous, The Metal API is certain used for for games on Apple's platforms. And it seems at least a bunch of so called AAA titles have been released for the Mac (with Metal support) since you post less than a year ago, just recently Cyberpunk 2077.
0 Reply
EnglishРусский