GPU Comparison
RADEON
AMD Radeon RX 640 OEM
CORE STATE
Polaris 23
VRAM
4 GB
CLOCK SPEED
—
TDP
50 W
BUS WIDTH
128 bit
ARCHITECTURE
GCN 4.0
PROCESS
14 nm
LAUNCH DATE
2020
VS
GEFORCE
Quadro FX 370M
CORE STATE
G98S
VRAM
256 MB
CLOCK SPEED
—
TDP
20 W
BUS WIDTH
64 bit
ARCHITECTURE
Tesla
PROCESS
65 nm
LAUNCH DATE
2008
DETAILED SPECIFICATIONS
SPECIFICATION
RX 640 OEM
Quadro FX 370M
Core Specs
Shading Units
640
8
-98.8%
Shaders
640
8
-98.8%
TMUs
40
4
-90.0%
ROPs
16
4
-75.0%
Compute Units
10
—
SM Count
—
1
Clocks
GPU Clock
1295 MHz
580 MHz
Shader Clock
—
1450 MHz
Memory Clock
1500 MHz
6 Gbps effective
400 MHz
800 Mbps effective
Memory
Memory Size
4 GB
256 MB
VRAM (MB)
4,096
256
-93.8%
Memory Type
GDDR5
GDDR3
Memory Bus
128 bit
64 bit
Bandwidth
96.00 GB/s
6.400 GB/s
Cache
L1 Cache
16 KB (per CU)
—
L2 Cache
512 KB
16 KB
Performance
Pixel Rate
20.72 GPixel/s
2.320 GPixel/s
Texture Rate
51.80 GTexel/s
2.320 GTexel/s
FP32 (TFLOPS)
1.658 TFLOPS
23.20 GFLOPS
FP64 (TFLOPS)
103.6 GFLOPS (1:16)
—
FP16 (TFLOPS)
1.658 TFLOPS (1:1)
—
Power
TDP
50 W
20 W
TDP (W)
50
20
-60.0%
Suggested PSU
250 W
—
Power Connectors
None
None
Architecture
Architecture
GCN 4.0
Tesla
GPU Name
Polaris 23
G98S
Generation
Polaris
(RX 600)
Quadro FX Mobile
(x700M)
Process Size
14 nm
65 nm
Transistors
2,200 million
210 million
Die Size
103 mm²
86 mm²
Foundry
GlobalFoundries
UMC
Density
21.4M / mm²
2.4M / mm²
API Support
DirectX
12 (12_0)
11.1 (10_0)
OpenGL
4.6
3.3
Vulkan
1.3
—
OpenCL
2.1
1.1
CUDA
—
1.1
Shader Model
6.7
4.0
Physical
Slot Width
Single-slot
MXM Module
Outputs
1x DisplayPort 1.4a2x mini-DisplayPort 1.4a
Portable Device Dependent
Bus Interface
PCIe 3.0 x8
MXM-II
Other
Production
End-of-life
End-of-life
Predecessor
Polaris
Quadro FX Go
Successor
Vega
Quadro Fermi-M