Template:AMD Radeon Instinct

From blackwiki
Revision as of 06:52, 9 December 2018 by blackwiki>RaptaGzus (Added MI50 and MI60, merged rows that were next to each other and had the same data, added a boost clock clarification for the double precision GOPS figure, and did a minor notelist formatting fix on CU display.)
Jump to navigation Jump to search
Model Launch Code Name Archi-
tecture
Fab (nm) Transistors (Billion) Die Size (mm2) Bus interface Clock rate Core config[lower-alpha 1] Fillrate Memory Processing Power
(GFLOPS)
TDP (Watts) API support (version)
Core (MHz) Boost (MHz) Memory (MT/s) Pixel (GP/s)
(Boost)[lower-alpha 2]
Texture (GT/s)
(Boost)[lower-alpha 3]
Size (GiB) Bus width (bit) Bus type Bandwidth (GB/s) Half Precision
(Boost)
Single Precision
(Boost)[lower-alpha 4]
Double Precision
(Boost)
OpenCL
Template:Rh | Radeon Instinct MI6 [1][2][3][4][5] 2016-12-12 Polaris 10 GCN 4th gen 14 5.7 232 PCIe 3.0 x16 1120 1233 1750 7000 eff 2304:144:32:36 39.46 177.6 16 256 GDDR5 224 5800 5800 358 150 2.0
Template:Rh | Radeon Instinct MI8 [1][2][3][6][7] Fiji XT GCN 3rd gen 28 8.9 596 1000 1000 1000 4096:256:64:64 64.0 256.0 4 4096 HBM 512 8200 8200 512 175
Template:Rh | Radeon Instinct MI25 [1][2][3][8][9][10] Vega 10 XT GCN 5th gen 14 12.5 510 1400 1500 1704 4096:256:64:64 96.0 384 16 2048 HBM2 436.2 24600 12300 768 300
Template:Rh | Radeon Instinct MI25 mxgpu (Prototype, near equal Radeon Pro V340 mxgpu)[11] 2017-06-17 2x 12.5 2x 510 1400 ? 1500 ? 2x 4096:256:64:64 2x 96.0 ? 2x 384 ? 2x 436.2 2x 24600 ? 2x 12300 ? 2x 768 ?
Template:Rh | Radeon Instinct MI50 [12][13][14] 2018-11-18 Vega 20 GL 7 13.2 331 PCIe 4.0 x16 1746 2000 3840:240:64:60 111.744 419.04 4096 1024 26800 13400 6700
Template:Rh | Radeon Instinct MI60 [12][15][16] 1800 4096:256:64:64 115.2 460.8 32 29450 14725 7362.5
  1. Single-precision shader processors : Texture Mapping Units : Render Output Units : Compute Units
  2. Pixel fillrate is calculated as the number of ROPs multiplied by the base (or boost) core clock speed.
  3. Texture fillrate is calculated as the number of TMUs multiplied by the base (or boost) core clock speed.
  4. Single precision performance is calculated from the base (or boost) core clock speed based on a FMA operation.

References

  1. 1.0 1.1 1.2 Smith, Ryan (12 December 2016). "AMD Announces Radeon Instinct: GPU Accelerators for Deep Learning, Coming in 2017". Anandtech. Retrieved 12 December 2016.
  2. 2.0 2.1 2.2 Shrout, Ryan (12 December 2016). "Radeon Instinct Machine Learning GPUs include Vega, Preview Performance". PC Per. Retrieved 12 December 2016.
  3. 3.0 3.1 3.2 Kampman, Jeff (12 December 2016). "AMD opens up machine learning with Radeon Instinct". TechReport. Retrieved 12 December 2016.
  4. "Radeon Instinct MI6". Radeon Instinct. AMD. Retrieved 22 June 2017.
  5. https://www.techpowerup.com/gpudb/2927/radeon-instinct-mi6
  6. "Radeon Instinct MI8". Radeon Instinct. AMD. Retrieved 22 June 2017.
  7. https://www.techpowerup.com/gpudb/2928/radeon-instinct-mi8
  8. Smith, Ryan (5 January 2017). "The AMD Vega Architecture Teaser: Higher IPC, Tiling, & More, coming in H1'2017". Anandtech.com. Retrieved 10 January 2017.
  9. "Radeon Instinct MI25". Radeon Instinct. AMD. Retrieved 22 June 2017.
  10. https://www.techpowerup.com/gpudb/2983/radeon-instinct-mi25
  11. https://www.techpowerup.com/gpudb/3269/radeon-instinct-mi25-mxgpu
  12. 12.0 12.1 "Next Horizon – David Wang Presentation" (PDF). AMD. AMD.
  13. "Radeon Instinct MI50". AMD. AMD.
  14. "Radeon Instinct MI50 Datasheet" (PDF). AMD. AMD.
  15. "Radeon Instinct MI60". AMD. AMD.
  16. "Radeon Instinct MI60 Datasheet" (PDF). AMD. AMD.