GPU Comparison
RADEON
AMD Radeon Instinct MI200
CORE STATE
Aldebaran
VRAM
64 GB
CLOCK SPEED
1700 MHz
TDP
300 W
BUS WIDTH
4096 bit
ARCHITECTURE
CDNA 2.0
PROCESS
6 nm
LAUNCH DATE
—
VS
GEFORCE
GeForce RTX 3050 6 GB
CORE STATE
GA107
VRAM
6 GB
CLOCK SPEED
1470 MHz
TDP
70 W
BUS WIDTH
96 bit
ARCHITECTURE
Ampere
PROCESS
8 nm
LAUNCH DATE
2024
PERFORMANCE BENCHMARKS
3dmark_3dmark_steel_nomad_dx12
DETAILED SPECIFICATIONS
SPECIFICATION
Instinct MI200
RTX 3050 6 GB
Core Specs
Shading Units
6,656
2,304
-65.4%
Shaders
6,656
2,304
-65.4%
TMUs
416
72
-82.7%
ROPs
0
32
+∞%
Compute Units
104
—
SM Count
—
18
Clocks
Base Clock
1000 MHz
1042 MHz
Boost Clock
1700 MHz
1470 MHz
Memory Clock
1600 MHz
3.2 Gbps effective
1750 MHz
14 Gbps effective
Memory
Memory Size
64 GB
6 GB
VRAM (MB)
65,536
6,144
-90.6%
Memory Type
HBM2e
GDDR6
Memory Bus
4096 bit
96 bit
Bandwidth
1.64 TB/s
168.0 GB/s
Cache
L1 Cache
16 KB (per CU)
128 KB (per SM)
L2 Cache
16 MB
2 MB
Performance
Pixel Rate
0 MPixel/s
47.04 GPixel/s
Texture Rate
707.2 GTexel/s
105.8 GTexel/s
FP32 (TFLOPS)
22.63 TFLOPS
6.774 TFLOPS
FP64 (TFLOPS)
22.63 TFLOPS (1:1)
105.8 GFLOPS (1:64)
FP16 (TFLOPS)
181.0 TFLOPS (8:1)
6.774 TFLOPS (1:1)
AI/RT
RT Cores
—
18
Tensor Cores
—
72
Power
TDP
300 W
70 W
TDP (W)
300
70
-76.7%
Suggested PSU
700 W
250 W
Power Connectors
1x 8-pin
None
Architecture
Architecture
CDNA 2.0
Ampere
GPU Name
Aldebaran
GA107
Generation
Radeon Instinct
(MIx)
GeForce 30
Process Size
6 nm
8 nm
Transistors
58,200 million
8,700 million
Die Size
724 mm²
200 mm²
Foundry
TSMC
Samsung
Density
80.4M / mm²
43.5M / mm²
AMD MCM
MCM
1
—
API Support
DirectX
—
12 Ultimate (12_2)
OpenGL
—
4.6
Vulkan
—
1.4
OpenCL
3.0
3.0
CUDA
—
8.6
Shader Model
—
6.8
Physical
Slot Width
OAM Module
Dual-slot
Length
—
242 mm
9.5 inches
Height
—
112 mm
4.4 inches
Outputs
No outputs
1x HDMI 2.13x DisplayPort 1.4a
Bus Interface
PCIe 4.0 x16
PCIe 4.0 x8
Other
Launch Price
—
179 USD
Production
End-of-life
End-of-life
Predecessor
FirePro Data Center
GeForce 20
Successor
—
GeForce 40