A featured concept image of Intel Xe GPU. Image Credits: Cristiano Siqueira (@ CSiqueira97)
Intel will launch next year its first gaming discrete graphics cards based on its next-generation Xe-HPG GPUs. The gaming setup will have a range of products, but unlike the Xe-LP & Xe-HP GPU line manufactured on Intel’s own 10nm SuperFin process node, the Xe-HPG GPUs will be produced on an external foundry in which TSMC becomes the most likely candidate.
TSMC to win Intel’s Xe-HPG GPU orders, first gaming discrete setup to be produced on 6nm process node
Reports coming from Taiwanese-based stores seem to indicate that TSMC is confident in winning orders for Intel’s next-generation Xe-HPG GPUs. As reported by IThome (via @ harukaze5719), Intel’s Xe-HPG GPUs will be manufactured at TSMC’s factory, utilizing its thriving manufacturing technologies. Taiwan-based sources claim that Intel will likely use TSMC’s 6nm process node for its gaming GPUs, which will compete with AMD’s RDNA 2 and NVIDIA’s Amper GPU lineup by the time it hits the store shelves.
The TSMC 6nm process node was unveiled last year in the company’s latest roadmap. Codenamed as ‘N6’, the TSMC 6nm process will use an advanced version of EUV lithography technology. It will offer an 18% increased logic density over TSMC’s 7nm (N7) process and is compatible with 7nm design, which is said to shorten the time it takes to market consumer products.
The Xe-HPG-class gaming GPUs will also use a standard packaging design that should be relatively easy to produce, while datacenter and HPC chips will use a mix of advanced packaging technologies that Intel is running. of the years has updated, such as EMIB, CO-EMIB, and FOVEROS.
Here’s all we know about Intel’s discrete gaming graphics cards powered by Xe-HPG GPUs
Intel’s gaming lineup for graphics cards will be powered by the Xe-HPG GPU. This specific GPU is another category within the Xe micro-architecture family. It falls between the Xe-LP and Xe-HP and is primarily aimed at the gaming audience. The Xe-HPG GPU is expected to use a single tile and assuming a single tile consists of 512 EUs, we look at 4096 cores on the flagship.
A featured concept image of Intel Xe GPU. Image Credits: Cristiano Siqueira (@ CSiqueira97)
Based on our exclusive report and from what Intel has talked about regarding its Ponte Vecchio chip, it looks like Intel is all on board the MCM train, with each chip consisting of several Xe GPU tiles that together they will be connected to a sample of a GPU. Here are the actual EU counts of Intel’s various MCM-based Xe HP GPUs along with estimated core counts and TFLOPs (You can read more about these impressive performance figures presented here by Intel):
- HP (12.5) 1-Tile GPU: 512 EU [Est: 4096 Cores, 10.6 TFLOPs 1.3 GHz, 150W]
- HP (12.5) 2-Tile GPU: 1024 EU [Est: 8192 Cores, 21.2 1.3 GHz, TFLOPs, 300W]
- HP (12.5) 4-Tile GPU: 2048 EU [Est: 16,384 Cores, 42.3 TFLOPs 1.3 GHz, 400W/500W]
Of course, Intel may opt for a higher EU count on its Xe-HPG gaming-specific GPUs, but that remains to be seen until the definitive specifications are revealed. In terms of game-specific features, the Intel Xe-HPG-powered graphics cards will have hardware-accelerated ray-tracing and GDDR6 memory to optimize performance and value, while the Xe-HP setup, which focuses on ‘ the data center market, will opt for HBM memory.
Intel recently gave us the first demo of their Xe LP GPU within their upcoming Tiger Lake CPUs, which proves to be a big leap in integrated graphics performance for Intel as seen in the demonstration. Expect more information about the Xe HP & Xe HPC GPUs in the coming months.
