DanteErodov
Galvanizer
Intel has now launched its first DG1 discrete GPU in three laptops from Acer, ASUS, and Dell. It is mostly identical to the inbuilt GPU present in Tiger Lake iGPU with some differentiations, notably dedicated 4GB LPDDR4X-4266 memory on 128 bit bus as well as a slight overlock of 1650MHz compared to 1350MHz on Tiger Lake iGPU.
Specifications:
The ace up the sleeve with this GPU is in content creation applications where Intel's workload sharing feature, called Deep Link, kicks in. This essentially allows the inbuilt GPU to work with the discrete GPU providing addition performance benefits. According to Intel, this allows DG1 to perform 7 times faster than the Nvidia MX350. It can also run Hyper Encode 78% faster than an Nvidia RTX 2080 Super Max-Q.
This looks interesting if it can provide the productivity benefits in other applications. Intel is planning to provide support to software developers to take advantage of this so support should improve in the future. What are your thoughts on this?
Sources:
Intel’s Discrete GPU Era Begins: Intel Launches Iris Xe MAX For Entry-Level Laptops
Intel introduces the Iris Xe Max: a discrete GPU that merges with integrated GPUs
Specifications:
- 96 EUs (768 ALUs)
- 1650MHz
- 2.46 TFLOPs
The ace up the sleeve with this GPU is in content creation applications where Intel's workload sharing feature, called Deep Link, kicks in. This essentially allows the inbuilt GPU to work with the discrete GPU providing addition performance benefits. According to Intel, this allows DG1 to perform 7 times faster than the Nvidia MX350. It can also run Hyper Encode 78% faster than an Nvidia RTX 2080 Super Max-Q.
This looks interesting if it can provide the productivity benefits in other applications. Intel is planning to provide support to software developers to take advantage of this so support should improve in the future. What are your thoughts on this?
Sources:
Intel’s Discrete GPU Era Begins: Intel Launches Iris Xe MAX For Entry-Level Laptops
Intel introduces the Iris Xe Max: a discrete GPU that merges with integrated GPUs