Report an Error

AMD Durango 2

Default
Durango 2
M1004145-001
M1004145-001
AMD's Durango 2 GPU uses the GCN 1.0 architecture and is made using a 16 nm production process at TSMC. With a die size of 240 mm² and a transistor count of 5,000 million it is a medium-sized chip. Durango 2 supports DirectX 12 (Feature Level 11_1). For GPU compute applications, OpenCL version 1.2 can be used. It features 768 shading units, 48 texture mapping units and 16 ROPs.

Graphics Processor

GPU Name
Durango 2
Architecture
GCN 1.0
Foundry
TSMC
Process Size
16 nm
Transistors
5,000 million
Density
20.8M / mm²
Die Size
240 mm²
Released
Aug 2nd, 2016

Graphics Features

DirectX
12 (11_1)
OpenGL
N/A
OpenCL
1.2
Vulkan
1.1
Shader Model
5.1

Render Config

Shading Units
768
TMUs
48
ROPs
16
Compute Units
12
Max. TDP
95 W

All GCN 1.0 GPUs

AMD GPU Architecture History

Graphics cards using the AMD Durango 2 GPU

Name Chip Memory Shaders TMUs ROPs GPU Clock Memory Clock
AMD Xbox One S GPU M1004145-001 8 GB 768 48 16 914 MHz 1066 MHz

Durango 2 GPU Notes

Architecture Codename: Sea Islands
Codename: Gladius / Bishop
CLRX Version: GCN 1.0

Instruction issue rate:
12 SCs * 4 SIMDs * 16 threads/clock = 768 ops/clock

FLOPs:
768 ops/clock * (1 mul + 1 add) * 914 MHz = 1.404 TFLOPS

Interpolation:
(768 ops/clock / 2 ops) * 914 MHz = 350.97 Gfloat/sec

Triangle rate:
2 tri/clock * 914 MHz = 1.828 Gtri/sec

Vertex rate:
2 vert/clock * 914 MHz = 1.828 Gvert/sec

Vertex/buffer fetch rate: (4 bytes)
4 elements/clock * 12 SCs * 914 MHz = 43.872 Gelement/sec

Vertex/Buffer data rate from cache:
43.872 Gelements/sec * 4 bytes = 175.488 GB/sec

Peak throughput from main RAM:
68.2 GB/sec

Peak throughput from ESRAM:
128 bytes/clock * 914 MHz = 116.992 GB/sec

ESRAM size:
32 MB

GSM size:
64 KB

LSM size:
12 SCs * 64 KB = 768 KB

L2 cache size:
4 x 128 KB = 512 KB (shared)

Bilinear fetch rate: (4 bytes)
4 fetches/clock * 12 SCs * 914 MHz = 43.872 Gtexels/sec

Bilinear data rate from cache:
43.872 Gtexels/sec * 4 bytes = 175.488 GB/sec

L1 cache size:
16 KB/SC * 12 SCs = 192 KB (nonshared)

Pixel clear rate:
1 8×8 tile/clock * 4 DBs * 914 MHz = 233.984 Gpixel/sec

Pixel hierarchical Z Rate:
1 8×8 tile/clock * 4 DBs * 914 MHz = 233.984 Gpixel/sec

Sample Z rate:
16 /clock * 4 DBs * 914 MHz = 58.496 Gsample/sec

Pixel rate:
4 /clock * 4 DBs * 914 MHz = 14.624 Gpixel/sec