|
|
| Line 1: |
Line 1: |
| − | {{amd title|Vega|arch}}
| |
| − | {{microarchitecture
| |
| − | | atype = GPU
| |
| − | | name = Vega
| |
| − | | designer = AMD
| |
| − | | manufacturer = GlobalFoundries
| |
| − | | manufacturer 2 =
| |
| − | | introduction = 2017
| |
| − | | phase-out =
| |
| − | | process = 14 nm
| |
| | | | |
| − | | succession = yes
| |
| − | | predecessor = Arctic Islands
| |
| − | | predecessor link = amd/microarchitectures/arctic islands
| |
| − | | successor = Navi
| |
| − | | successor link = amd/microarchitectures/navi
| |
| − | }}
| |
| − | '''Vega''' (also known as '''Graphics core next 5th generation''') is a [[microarchitecture]] developed by [[AMD]] as a successor to {{\\|Arctic Islands}}.
| |
| − |
| |
| − | == Process Technology ==
| |
| − | {{see also|14 nm process}}
| |
| − | Vega is manufactured on [[Global Foundries]] [[14 nm process]].
| |
| − |
| |
| − | == Architecture ==
| |
| − | [[File:amd gpu roadmap.png|400px|right]]
| |
| − |
| |
| − | {{future information}}
| |
| − |
| |
| − | === Key changes from {{amd|Arctic Islands|l=arch}} ===
| |
| − | *New programmable geometry pipeline
| |
| − | **Up to 2x throughput
| |
| − | **Primitive shaders
| |
| − | **Improved workload balancing
| |
| − | *Tile based rendering
| |
| − | *New low precision instructions
| |
| − | **Half precision floating point with 2x performance
| |
| − | **8 bit integer with 4x performance
| |
| − | *15% higher clock speeds
| |
| − | *HBM 2
| |
| − | **2x bandwidth
| |
| − | **4x memory capacity
| |
| − | *New memory controller
| |
| − | *Render back-end is now a client of the L2 cache
| |
| − | *The instruction buffer has been enlarged
| |
| − |
| |
| − |
| |
| − | ====HBM 2====
| |
| − | Vega makes use of 2 stacks of HBM 2 (High bandwidth memory).
| |
| − |
| |
| − | {| class="wikitable"
| |
| − | |-
| |
| − | ! single stack !! HBM 1 !! HBM 2
| |
| − | |-
| |
| − | | dies || 4 + 1<info>4 dram die, 1 control die</info> || 2-8 + 1<info>2,4 or 8 dram die, 1 control die</info>
| |
| − | |-
| |
| − | | Gb/die || 2Gb || 8Gb
| |
| − | |-
| |
| − | | total Gb || 8Gb || 16-64Gb
| |
| − | |-
| |
| − | | bus width || 1024 || 1024
| |
| − | |-
| |
| − | | clock speed || 500 MHz || 945 MHz
| |
| − | |-
| |
| − | | bandwidth || 128 GB/s || 238 GB/s
| |
| − | |}
| |
| − |
| |
| − | == References ==
| |
| − | * AMD 2017 Financial Analyst Day, May 16, 2017
| |
| − |
| |
| − | == Documents ==
| |
| − | * [[:File:vega-whitepaper.pdf|White Paper: Radeon’s next-generation Vega architecture]]
| |
| − |
| |
| − | == See Also ==
| |
| − | * [[list of microarchitectures]]
| |
