Difference between revisions of "Template:AMD Radeon VII"
Jump to navigation
Jump to search
blackwiki>Elk Salmon m (ditch XHTML) |
blackwiki>Wikiinger (Undid revision 902530132 by Elk Salmon (talk) - that's not XHTML) |
||
| Line 1: | Line 1: | ||
<!-- Changes to the table *layout* should be proposed and discussed first on Talk:List_of_AMD_graphics_processing_units --> | <!-- Changes to the table *layout* should be proposed and discussed first on Talk:List_of_AMD_graphics_processing_units --> | ||
{| class="wikitable center" style="font-size: 85%;" | {| class="wikitable center" style="font-size: 85%;" | ||
| − | ! rowspan=2 | Model<br>([[codename]]) | + | ! rowspan=2 | Model<br />([[codename]]) |
| − | ! rowspan=2 | Release Date<br>& Price | + | ! rowspan=2 | Release Date<br />& Price |
| − | ! rowspan=2 | [[Microarchitecture|Architecture]]<br>& [[Semiconductor device fabrication|Fab]] | + | ! rowspan=2 | [[Microarchitecture|Architecture]]<br />& [[Semiconductor device fabrication|Fab]] |
| − | ! rowspan=2 | Transistors<br>& Die Size | + | ! rowspan=2 | Transistors<br />& Die Size |
! colspan=2 | Core | ! colspan=2 | Core | ||
! colspan=2 | [[Fillrate]]{{efn|name="boost"}}{{efn|name="texture fill"}}{{efn|name="pixel fill"}} | ! colspan=2 | [[Fillrate]]{{efn|name="boost"}}{{efn|name="texture fill"}}{{efn|name="pixel fill"}} | ||
| − | ! colspan=3 | Processing power{{efn|name="boost"}}{{efn|name="FLOPS"}}<br>([[GFLOPS]]) | + | ! colspan=3 | Processing power{{efn|name="boost"}}{{efn|name="FLOPS"}}<br />([[GFLOPS]]) |
! colspan=4 | Memory | ! colspan=4 | Memory | ||
! rowspan=2 | {{abbr|TBP|Typical board power}} | ! rowspan=2 | {{abbr|TBP|Typical board power}} | ||
| − | ! rowspan=2 | Bus<br>interface | + | ! rowspan=2 | Bus<br />interface |
|- | |- | ||
! Config{{efn|name="cconfig"}} | ! Config{{efn|name="cconfig"}} | ||
! Clock{{efn|name="boost"}} ([[Hertz|MHz]]) | ! Clock{{efn|name="boost"}} ([[Hertz|MHz]]) | ||
| − | ! Texture <br>([[texel (graphics)|GT]]/s) | + | ! Texture <br/>([[texel (graphics)|GT]]/s) |
| − | ! Pixel <br>([[pixel|GP]]/s) | + | ! Pixel <br/>([[pixel|GP]]/s) |
! [[Half-precision floating-point format|Half]] | ! [[Half-precision floating-point format|Half]] | ||
! [[Single precision floating-point format|Single]] | ! [[Single precision floating-point format|Single]] | ||
! [[Double precision floating-point format|Double]] | ! [[Double precision floating-point format|Double]] | ||
| − | ! Bus type<br>& width | + | ! Bus type<br />& width |
| − | ! Size <br>([[gibibyte|GiB]]) | + | ! Size <br/>([[gibibyte|GiB]]) |
| − | ! Clock <br>([[transfer (computing)|MT/s]]) | + | ! Clock <br/>([[transfer (computing)|MT/s]]) |
| − | ! Bandwidth <br>([[Gigabyte|GB]]/s) | + | ! Bandwidth <br/>([[Gigabyte|GB]]/s) |
|- | |- | ||
| − | ! style="text-align:left;" | Radeon VII<br>(Vega 20)<br><ref name="anand_radeon_vii">{{cite news|last1=Smith|first1=Ryan|title=AMD Reveals Radeon VII|url=https://www.anandtech.com/show/13832/amd-radeon-vii-high-end-7nm-february-7th-for-699|accessdate=10 January 2019|publisher=anandtech|date=9 January 2019}}</ref><ref name="extreme_radeon_vii">{{cite news|last1=Hruska|first1=Joel|title=The AMD Radeon VII’s Core Configuration Has Been Misreported|url=https://www.extremetech.com/gaming/283649-the-amd-radeon-viis-core-configuration-has-been-misreported|accessdate=11 January 2019|publisher=etremetech|date=1 January 2019}}</ref><ref name="techgage_radeon_vii_fp64">{{cite news|last1=Williams|first1=Rob|title=AMD’s Radeon VII GPU Will Not Support Uncapped Double-Precision (FP64)|url=https://techgage.com/news/radeon-vii-caps-fp64-performance/|accessdate=14 January 2019|publisher=techgage|date=13 January 2019}}</ref><ref name="tomshardware_radeon_vii">{{cite news|last1=Liu|first1=Zhiye|title=AMD Radeon VII Will Ship Without Double-Precision|url=https://www.tomshardware.com/news/amd-radeon-vii-double-precision-disabled,38437.html|accessdate=16 January 2019|publisher=Tom's Hardware|date=14 January 2019}}</ref><ref name="techpowerup_radeon_vii">{{cite news|title=AMD Radeon VII|url=https://www.techpowerup.com/gpu-specs/radeon-vii.c3358|accessdate=16 January 2019|publisher=techgage}}</ref><ref name="ryan_smith_radeon_vii_fp64">{{cite tweet |user=RyanSmithAT |number=1085680805802733568 |author=Ryan Smith |date = 16 January 2019|title = Alright, I finally have the Radeon VII FP64 performance matter sorted out with AMD. Contrary to earlier statements, it is being throttled. Radeon VII's rate will be 1:8, versus Vega 20's native 1:2 rate. Notably, this is still twice the native FP64 rate of all other Vegas.}}</ref> | + | ! style="text-align:left;" | Radeon VII<br />(Vega 20)<br /><ref name="anand_radeon_vii">{{cite news|last1=Smith|first1=Ryan|title=AMD Reveals Radeon VII|url=https://www.anandtech.com/show/13832/amd-radeon-vii-high-end-7nm-february-7th-for-699|accessdate=10 January 2019|publisher=anandtech|date=9 January 2019}}</ref><ref name="extreme_radeon_vii">{{cite news|last1=Hruska|first1=Joel|title=The AMD Radeon VII’s Core Configuration Has Been Misreported|url=https://www.extremetech.com/gaming/283649-the-amd-radeon-viis-core-configuration-has-been-misreported|accessdate=11 January 2019|publisher=etremetech|date=1 January 2019}}</ref><ref name="techgage_radeon_vii_fp64">{{cite news|last1=Williams|first1=Rob|title=AMD’s Radeon VII GPU Will Not Support Uncapped Double-Precision (FP64)|url=https://techgage.com/news/radeon-vii-caps-fp64-performance/|accessdate=14 January 2019|publisher=techgage|date=13 January 2019}}</ref><ref name="tomshardware_radeon_vii">{{cite news|last1=Liu|first1=Zhiye|title=AMD Radeon VII Will Ship Without Double-Precision|url=https://www.tomshardware.com/news/amd-radeon-vii-double-precision-disabled,38437.html|accessdate=16 January 2019|publisher=Tom's Hardware|date=14 January 2019}}</ref><ref name="techpowerup_radeon_vii">{{cite news|title=AMD Radeon VII|url=https://www.techpowerup.com/gpu-specs/radeon-vii.c3358|accessdate=16 January 2019|publisher=techgage}}</ref><ref name="ryan_smith_radeon_vii_fp64">{{cite tweet |user=RyanSmithAT |number=1085680805802733568 |author=Ryan Smith |date = 16 January 2019|title = Alright, I finally have the Radeon VII FP64 performance matter sorted out with AMD. Contrary to earlier statements, it is being throttled. Radeon VII's rate will be 1:8, versus Vega 20's native 1:2 rate. Notably, this is still twice the native FP64 rate of all other Vegas.}}</ref> |
| − | | rowspan= 1 |{{dts|2019|02|07|format=mdy}}<br>$699 USD | + | | rowspan= 1 |{{dts|2019|02|07|format=mdy}}<br />$699 USD |
| [[Graphics Core Next#fifth|GCN 5<sup>th</sup> gen]]<br>[[TSMC]] 7FF | | [[Graphics Core Next#fifth|GCN 5<sup>th</sup> gen]]<br>[[TSMC]] 7FF | ||
| − | | {{val|13.2|e=9}}<br>331 mm<sup>2</sup> | + | | {{val|13.2|e=9}}<br />331 mm<sup>2</sup> |
| − | | 3840:240:64<br>60 CU | + | | 3840:240:64<br />60 CU |
| − | | 1400<br>''1750'' | + | | 1400<br />''1750'' |
| − | | 336<br>''420'' | + | | 336<br />''420'' |
| − | | 89.6<br>''112'' | + | | 89.6<br />''112'' |
| − | | 22,272<br>''27,648'' | + | | 22,272<br />''27,648'' |
| − | | 11,136<br>''13,824'' | + | | 11,136<br />''13,824'' |
| − | | 2,784<br>''3,458.5'' | + | | 2,784<br />''3,458.5'' |
| − | | HBM2<br>4096-bit | + | | HBM2<br />4096-bit |
| 16 | | 16 | ||
| 2000 | | 2000 | ||
Revision as of 20:57, 19 June 2019
| Model (codename) |
Release Date & Price |
Architecture & Fab |
Transistors & Die Size |
Core | Fillrate[lower-alpha 1][lower-alpha 2][lower-alpha 3] | Processing power[lower-alpha 1][lower-alpha 4] (GFLOPS) |
Memory | TBP | Bus interface | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Config[lower-alpha 5] | Clock[lower-alpha 1] (MHz) | Texture (GT/s) |
Pixel (GP/s) |
Half | Single | Double | Bus type & width |
Size (GiB) |
Clock (MT/s) |
Bandwidth (GB/s) | ||||||
| Radeon VII (Vega 20) [1][2][3][4][5][6] |
February 7, 2019 $699 USD |
GCN 5th gen TSMC 7FF |
Template:Val 331 mm2 |
3840:240:64 60 CU |
1400 1750 |
336 420 |
89.6 112 |
22,272 27,648 |
11,136 13,824 |
2,784 3,458.5 |
HBM2 4096-bit |
16 | 2000 | 1028 | 300 W | PCIe 3.0 x16 |
- ↑ 1.0 1.1 1.2 Boost values (if available) are stated below the base value in italic.
- ↑ Texture fillrate is calculated as the number of Texture Mapping Units multiplied by the base (or boost) core clock speed.
- ↑ Pixel fillrate is calculated as the number of Render Output Units multiplied by the base (or boost) core clock speed.
- ↑ Precision performance is calculated from the base (or boost) core clock speed based on a FMA operation.
- ↑ Unified Shaders : Texture Mapping Units : Render Output Units and Compute Units (CU)
References
- ↑ Smith, Ryan (9 January 2019). "AMD Reveals Radeon VII". anandtech. Retrieved 10 January 2019.
- ↑ Hruska, Joel (1 January 2019). "The AMD Radeon VII's Core Configuration Has Been Misreported". etremetech. Retrieved 11 January 2019.
- ↑ Williams, Rob (13 January 2019). "AMD's Radeon VII GPU Will Not Support Uncapped Double-Precision (FP64)". techgage. Retrieved 14 January 2019.
- ↑ Liu, Zhiye (14 January 2019). "AMD Radeon VII Will Ship Without Double-Precision". Tom's Hardware. Retrieved 16 January 2019.
- ↑ "AMD Radeon VII". techgage. Retrieved 16 January 2019.
- ↑ Ryan Smith [@RyanSmithAT] (16 January 2019). "Alright, I finally have the Radeon VII FP64 performance matter sorted out with AMD. Contrary to earlier statements, it is being throttled. Radeon VII's rate will be 1:8, versus Vega 20's native 1:2 rate. Notably, this is still twice the native FP64 rate of all other Vegas" (Tweet) – via Twitter.