In a recent interview with PCGamer, the chip giant said that the Intel Arc Alchemist will not only be “competitively” priced, but will also offer very good performance in ray tracingthanks to the configuration of the RTUs (Ray Tracing Units) that these graphics cards use.

Before I get down to business, I want to remind you that the Intel Xe-HPG architecture used by the Santa Clara giant in Intel Arc Alchemist has one RTU for each Xe block, which means that a graphics card like the Intel Arc 3 A380, which has 8 Xe blocks, has 8 ray tracing units. These units take care of accelerating workloads that represent ray tracing, including calculation of collisions and intersectionsand it is very likely that the XMX matrices will take care of the subsequent noise reduction in the end.

Well, according to Tom Petersen, RTU is very ready to work with real-time ray tracing and offers very good performance compared to GeForce RTX 3060. He didn’t give specifics, but said they should offer “similar or better” performance compared to NVIDIA, suggesting that their architecture is generally on par with NVIDIA, at least as far as ray tracing is concerned.

In this sense, we have to highlight two important keys, the first is that Intel used a BVH cache on GPUwhich makes it possible to speed up BVH intersections, and the so-called “yarn sorting unit”, which should simplify the organization and resolution of different work threads. It is clear that Intel Arc Alchemist not only offers full ray tracing acceleration, but is also a highly specialized architecture for the task.

Why is Intel comparing Arc Alchemist with GeForce RTX 30 and not Radeon RX 6000?

Well, it’s very simple, because when we talk about ray tracing, the GeForce RTX 30 is clearly above the Radeon RX 6000. The architecture that NVIDIA uses in its second generation RT cores plays in a different league, so much so that these offer more than double the performance in ray traced gamesas we saw when we analyzed the ray-tracing performance of the GeForce RTX 3090 Ti against the Radeon RX 6950 XT.

Intel has been directly compared to NVIDIA because it is the gold standard when it comes to gaming performance with spokes, nothing else. AMD managed to introduce acceleration units of this technology in the Radeon RX 6000, but they have important shortcomings due to which they lose a lot of performance, and one of the most important is related to the BVH intersections, which in the Radeon RX 6000 they are not accelerated by specific hardware, but are calculated directly in the shaders.

To be fair, we have to acknowledge that Intel made a huge effort with Arc Alchemist and that creating a competitive GPU architecture from scratch It is not easy. In addition to all the challenges it presents at the hardware level, we have to consider all that this means at the software, driver and optimization levels. With all this in mind, I think the result that Intel achieved with its first generation of graphics cards is reasonably good, at least with the data we have on the table right now.

Can someone tell me that their drivers are still very green and that they are making serious problems, and if that’s truein fact, they also need to improve performance with outdated APIs, but I remind you that a veteran like AMD released some drivers a year ago that disabled some Radeon RX 5700 and RX 5700 XT, and that it also had performance problems with DirectX 11 for years Let’s be a little more fair and impartial.