Linux Driver patch notes reveal AMD’s planned dual Vega GPU
Linux Driver patch notes reveal AMD’s planned dual Vega GPU
- table->Tliquid1Limit = cpu_to_le16(tdp_table->usTemperatureLimitLiquid1);
- table->Tliquid2Limit = cpu_to_le16(tdp_table->usTemperatureLimitLiquid2);
Another line also suggests the use of a PLX chip in a GPU, which again hints at a dual-GPU configuration. A PLX chip is used to connect two GPUs dies in a single graphics card and allow them both to operate on a single PCIe connection. The below line refers to the cooling of said chip, allowing fans to respond to the temperature of the PLX chip.
- table->FanGainPlx = hwmgr->thermal_controller. advanceFanControlParameters.usFanGainPlx;table->TplxLimit = cpu_to_le16(tdp_table->usTemperatureLimitPlx);
(AMD’s R9 295X2, the companies last consumer-oriented dual GPU)
AMD is expected to reveal more details on their RX Vega GPU on May 16th as part of their Financial Analyst Day, though up until now the company has been strangely quiet about this new hardware launch.
Vega will release by the end of Q2 2017, which means that Vega will release before the end of June.
You can join the discussion on AMD’s potential Dual-Vega GPU on the OC3D Forums.