Big picture: Intel has been fighting a long battle with main rival AMD in the desktop space, which has recently been popular in the Team Red market. The Ryzen series has pushed Intel out of the spotlight, but as the company gets closer to launching its Alder Lake processors to level the playing field and reclaim some market share in a business it once dominated, they are also gearing up for an acquisition. in another highly competitive industry.
V recent interview with CRN, which covers multiple topics with Intel CEO Pat Gelsinger, one of the topics highlighted was the upcoming battle with Nvidia. Moving forward, Intel plans to differentiate itself from the GPU giant by being “much more ecosystem-friendly” in terms of the software that powers its components.
“Nvidia has become too proprietary, and this is widely seen in the industry, so we are going to populate this stack with oneAPI, but in a much more favorable and open way for the industry and their innovation,” he said. …
oneAPI is an open programming model covering various types of architectures, including Intel silicon products such as processors, GPUs, and FPGAs.
Addressing Intel’s “underdogged” graphics business, Gelsinger emphasized that the company’s efforts in the GPU market must go beyond features, performance and power, all at an attractive price point to the consumer.
Gelsinger attributed Nvidia’s success in large part to the way its executives have run the company over the past decade, but also pointed out how “they really got lucky” in the field of artificial intelligence.
“Nvidia has had 10x or more performance leadership in ten years. If you have this, 10x leadership for 10 years, people will take advantage of it. And they were really lucky: AI appeared. 30 years overnight. success and they put it together really well. So they worked hard, they made money, and then they got lucky in that regard, ”explained Gelsinger.
In detailing how Intel specifically intends to deliver compelling products to the GPU market, Gelsinger touched on how they are going to make a smooth transition from integrated to discrete on the Intel platform.
“So what do we need to do? To create great products in these segments (…) and the market is hungry for us to provide them with an alternative. Then we need to supply them with unique differentiated value added.
And in GPUs, we go to the customer and say, “Well, guess what, we’re just the undisputed leader in integrated graphics. will ship, and we’re going to make it a smooth transition from integrated to discrete on the Intel platform.
And even better than that, we’re going to do an integrated and discrete work together. So if you have three [execution units] stands in an integrated [GPU]then you have 10 EUs on the discrete level, we are going to give you 13 EUs and you are only going to buy 10 EUs on the discrete GPU and you are going to qualify one product that [works] smoothly between the two. “Well, it’s pretty differentiated. And that’s just one example.
Attaching importance to his words, Intel poaching former AMD Lead GPU Architect, Vinit Goel, who oversaw the Xe GPU architecture on which future graphics products will be based. He will lead a team of architects and design engineers to “architect, design and validate the Intel Xe IP roadmap”.
Related reading: Last time Intel tried to build a graphics card
Intel is certainly in no rush if it wants to leave its mark on the graphics market. Nvidia is firmly entrenched in the industry, and AMD plays a strong minor player – not to mention its dominance among chipmakers – and is set to become an even bigger force to be reckoned with if the Arm acquisition materializes.
Nonetheless, Intel is heading off to a good start judging by the expectations of future Xe-HPG graphics cards. As for the battle with AMD, Gelsinger believes that the success his rival has made in recent years will not last due to the upcoming launch of Alder Lake and Sapphire Rapids.