|
gptkbp:instance_of
|
gptkb:Graphics_Processing_Unit
|
|
gptkbp:aienhancements
|
gptkb:Yes
|
|
gptkbp:application
|
gptkb:simulation
gptkb:Deep_Learning
gptkb:Data_Analytics
Rendering
Scientific Computing
|
|
gptkbp:architecture
|
gptkb:Vega
|
|
gptkbp:availability
|
Widely Available
|
|
gptkbp:colorway
|
3840
|
|
gptkbp:connects
|
PCIe 4.0
|
|
gptkbp:cooling_system
|
Active Cooling
|
|
gptkbp:direct_xsupport
|
gptkb:Direct_X_12
|
|
gptkbp:form_factor
|
Dual-slot
PCIe Card
|
|
gptkbp:gpu
|
gptkb:Yes
|
|
gptkbp:has_ability
|
6.1
|
|
gptkbp:has_units
|
240
60
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
AMD Radeon Instinct MI50
|
|
gptkbp:manufacturer
|
gptkb:AMD
|
|
gptkbp:manufacturing_process
|
7 nm
|
|
gptkbp:market_segment
|
Professional
|
|
gptkbp:memory_type
|
gptkb:HBM2
2048-bit
|
|
gptkbp:monitors
|
gptkb:None
|
|
gptkbp:number_of_cores
|
Not Supported
|
|
gptkbp:open_glsupport
|
gptkb:Open_GL_4.5
|
|
gptkbp:opened
|
gptkb:Open_CL_2.0
|
|
gptkbp:peak_fp32_performance
|
13.3 TFLOPS
|
|
gptkbp:peak_fp64_performance
|
6.6 TFLOPS
|
|
gptkbp:performance
|
13.3 TFLOPS
26.6 TFLOPS
6.6 TFLOPS
|
|
gptkbp:power_connector
|
2 x 8-pin
|
|
gptkbp:power_consumption
|
300 W
|
|
gptkbp:price
|
Varies by retailer
|
|
gptkbp:ram
|
32 GB
1 TB/s
|
|
gptkbp:ray_tracing_support
|
Not Supported
|
|
gptkbp:release_date
|
gptkb:2018
|
|
gptkbp:release_notes
|
Optimized for AI and HPC workloads.
|
|
gptkbp:released
|
gptkb:Radeon_Instinct
|
|
gptkbp:successor
|
gptkb:AMD_Radeon_Instinct_MI100
|
|
gptkbp:target_market
|
gptkb:research_center
|
|
gptkbp:tdp
|
300 W
|
|
gptkbp:thermal_design_power
|
300 W
|
|
gptkbp:use_case
|
gptkb:machine_learning
High-Performance Computing
|
|
gptkbp:vrready
|
No
|
|
gptkbp:bfsParent
|
gptkb:AMD_Radeon_Instinct_MI25
|
|
gptkbp:bfsLayer
|
6
|