Home GPU comparison Apple M3 Ultra GPU (80-core) and Nvidia GeForce RTX 4090 - Which is better?

Apple M3 Ultra GPU (80-core) vs Nvidia GeForce RTX 4090

53 out 100
Apple M3 Ultra GPU (80-core)
VS
92 out 100
Nvidia GeForce RTX 4090
Apple M3 Ultra GPU (80-core)
Nvidia GeForce RTX 4090

Compared 2 processors: Apple M3 Ultra GPU (80-core), vs Nvidia GeForce RTX 4090. Below are specs, tests, pros and cons for each device

  1. Differences
  2. Review
  3. General
  4. Benchmarks
  5. Processor
  6. Memory
  7. Comments (0)

Review

Key Specs Apple M3 Ultra GPU (80-core) and Nvidia GeForce RTX 4090
Games
Performance in DirectX, OpenCL, Vulkan
AI/ML
AI workload testing
Professional Apps
3D modeling, video editing, rendering
Energy Efficiency
Power usage in various scenarios
NanoReview Score
Overall score by NanoReview

Tests & specs

Specs & test comparison
Processors:
M3 Ultra GPU (80-core)
vs
GeForce RTX 4090

Games

FPS Table
Game title
1080p (High)
1080p (Ultra)
1440p (Ultra)
4K (Ultra)
Forza Horizon 5
0
0
0
0
277
199
191
154
The Witcher 3
37
25
0
0
288
265
187
100
Counter-Strike 2
0
0
0
0
329
244
185
91
Far Cry 6
0
0
0
0
203
182
167
104
Hogwarts Legacy
0
0
0
0
201
164
132
76
COD Modern Warfare 3
0
0
0
0
274
258
221
149
Ghost of Tsushima
0
0
0
0
183
154
137
91
Cyberpunk 2077
0
0
0
0
227
208
143
68
Shadow of the Tomb Raider
0
0
0
0
304
277
215
119
Avg FPS
1080p Highn/a254
1080p Ultran/a217
1440p Ultran/a175
4K Ultran/a106
Deviationn/aLow

Benchmarks

CPU performance comparison across tasks
3DMark
Steel Nomad Lite Score
Apple M3 Ultra GPU (80-core)
17182
Nvidia GeForce RTX 4090
42191
Time Spyn/a36322
Solar Bayn/a186930
Port Royaln/a26115
Fire Striken/a72253
Night Raidn/a195300
Sources: 3DMark [Nvidia GeForce RTX 4090]
GeekBench v6
GB6 Compute Score
Apple M3 Ultra GPU (80-core)
147719
Nvidia GeForce RTX 4090
316301
Background Blur204.5 frames/s318.9 frames/s
Face Detection126 frames/s222.8 frames/s
Horizon Detection6.53 Gpix/sec14.2 Gpix/sec
Edge Detection10.7 Gpix/sec21 Gpix/sec
Gaussian Blur9.5 Gpix/sec23.9 Gpix/sec
Feature Matching1.05 Gpix/sec2.38 Gpix/sec
Stereo Matching560.3 Gpix/sec1950 Gpix/sec
Particle Physics25267.8 fps48090.4 fps
APIOpenCLOpenCL
Sources: GeekBench v6 [Apple M3 Ultra GPU (80-core)], [Nvidia GeForce RTX 4090]
PassMark
PassMark Graphics
Apple M3 Ultra GPU (80-core)
n/a
Nvidia GeForce RTX 4090
38303
G2D Markn/a1297
DirectX 11n/a329 fps
DirectX 12n/a150 fps
GPU Computen/a27111 Ops/s
Sources: PassMark [Nvidia GeForce RTX 4090]
GFXBench 5
Aztec Ruins (High 4K)
Apple M3 Ultra GPU (80-core)
375.1 fps
Nvidia GeForce RTX 4090
615.9 fps
Aztec Ruins (Normal)831.5 fps1617.6 fps
Car Chase517.6 fps1228.6 fps
Manhattan 3.1.1 (1440p)1083.9 fps1234.6 fps
T-Rex3144.1 fps6286 fps
ALU 22765.4 fps15073.4 fps
Driver Overhead 2570.2 fps303.7 fps
APIMetalOpenGL
OSOS XWindows
Tessellationn/a12119.1
Sources: GFXBench 5 [Apple M3 Ultra GPU (80-core)], [Nvidia GeForce RTX 4090]
Geekbench 6 ML
GB6 ML Single Precision
Apple M3 Ultra GPU (80-core)
21356
Nvidia GeForce RTX 4090
n/a
GB6 ML Half Precision
Apple M3 Ultra GPU (80-core)
23530
Nvidia GeForce RTX 4090
n/a
GB6 ML Quantized
Apple M3 Ultra GPU (80-core)
21706
Nvidia GeForce RTX 4090
n/a
Image Classification (SP)8094n/a
Image Segmentation (HP)24595n/a
Horizon Detection25190n/a
Face Detection (HP)38337n/a
Pose Estimation (Q)111021n/a
Text Classification (SP)2786n/a
Machine Translation (HP)4339n/a
Object Detection (SP)7668n/a
Depth Estimation (Q)37488n/a
Style Transfer (SP)244483n/a
FrameworkCore MLn/a
BackendGPUn/a
Sources: GeekBench v6 [Apple M3 Ultra GPU (80-core)]

Blender

Real-world 3D modeling test
Blender GPU
Apple M3 Ultra GPU (80-core)
7492.44
Nvidia GeForce RTX 4090
n/a

Tech Specs

Specs & test comparison

General Info

Manufacturer Apple Nvidia
Build Integrated Discrete
Announced March, 5, 2025 September, 20, 2022
Form Factor PC use PC use
Use case Professional Gaming
Segment High-end High-end
Architecture Apple M GPU Ada Lovelace
Codename Custom AD102
Price n/a 1599

GPU

GPU Frequency n/a 2235 MHz
Boost GPU Frequency n/a 2520 MHz
Shader Units 10240 16384
TMUs 640 512
ROPs 320 176
Execution Units 1280 128
Ray Tracing Cores n/a 128
Tensor Cores n/a 512
L1 cache n/a 128 KB/core
L2 cache n/a 72 KB/core
Instructions per Clock 2 IPC 2 IPC

Performance

Pixel Fill Rate n/a 444 Gpix/sec
Texture Fill Rate n/a 1290 GTex/s
FLOPS n/a 82.6 TFLOPS

Process

Interface Custom PCIe 4.0 x16
TGP 140 W 450 W
Process TSMC TSMC
Process tech. 3 nm 5 nm
Max temperature 100 °C n/a
Transistor Size n/a 609 mm²
Transistor count 184 billion 76.3 billion
Transistor density n/a 125.29 MTr/mm²

Memory

Interface System Shared GDDR6X
Memory Frequency 6400 MHz 2625 MHz
Bus 1024 bit 384 bit
ECC support No No
Effective Memory Speed n/a 21000 MB/s
Memory Size n/a 24 GB
Max Throughput
Apple M3 Ultra GPU (80-core)
819.3 GB/s
Nvidia GeForce RTX 4090
1010 GB/s

API

DirectX n/a 12
Vulkan n/a 1.3
OpenGL n/a 4.6
CUDA n/a 8.9
DisplayPort n/a 1.4a
Ray Tracing Yes Yes
DLSS No DLSS 3

Results

After analyzing the data, we consider Nvidia GeForce RTX 4090 the clear winner in this comparison.

Survey

Which GPU would you pick?
0 (0%)
1 (100%)
Total Votes: 1

User comments

Share your opinion or ask questions below