Report an Error

AMD Xbox One S GPU

Graphics Processor
Durango 2
Cores
768
TMUs
48
ROPs
16
Memory Size
8 GB
Memory Type
DDR3
Bus Width
256 bit
GPU Chip
GPU
Front 2
Front 2
Back
Back
The Xbox One S GPU was a high-end gaming console graphics solution by AMD, launched on August 2nd, 2016. Built on the 16 nm process, and based on the Durango 2 graphics processor, in its M1004145-001 variant, the device supports DirectX 12. The Durango 2 graphics processor is an average sized chip with a die area of 240 mm² and 5,000 million transistors. It features 768 shading units, 48 texture mapping units, and 16 ROPs. AMD includes 8 GB DDR3 memory, which are connected using a 256-bit memory interface. The GPU is operating at a frequency of 914 MHz, memory is running at 1066 MHz.
Its power draw is rated at 95 W maximum. The console's dimensions are 295 mm x 230 mm x 64 mm, and it features a igp cooling solution. Its price at launch was 399 US Dollars.

Graphics Processor

GPU Name
Durango 2
GPU Variant
M1004145-001
Architecture
GCN 1.0
Foundry
TSMC
Process Size
16 nm
Transistors
5,000 million
Density
20.8M / mm²
Die Size
240 mm²
Chip Package
BGA-515

Graphics Card

Release Date
Aug 2nd, 2016
Generation
Console GPU
(Microsoft)
Production
End-of-life
Launch Price
399 USD

Clock Speeds

GPU Clock
914 MHz
Memory Clock
1066 MHz
2.1 Gbps effective

Memory

Memory Size
8 GB
Memory Type
DDR3
Memory Bus
256 bit
Bandwidth
68.22 GB/s

Render Config

Shading Units
768
TMUs
48
ROPs
16
Compute Units
12

Theoretical Performance

Pixel Rate
14.62 GPixel/s
Texture Rate
43.87 GTexel/s
FP32 (float)
1,404 GFLOPS

Board Design

Length
295 mm
11.6 inches
Width
230 mm
9.1 inches
Height
64 mm
2.5 inches
Weight
2.9 kg (6.4 lbs)
Storage
1 TB HDD
TDP
95 W
Outputs
1x HDMI 2.0a
Inputs
1x HDMI 1.4b

Graphics Features

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

Durango 2 GPU Notes

Generation: Sea Islands
Codename: Gladius / Bishop
CLRX: 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
May 21st, 2024 10:08 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts