It was undoubtedly the most important news of yesterday, Raja Koduri confirmed his departure from Intel and ended it an almost six-year adventure in which he held very important positions in the chip giant, in fact he became the executive vice president and head of Intel’s graphics division.

Raji Koduri’s arrival at Intel was not accidental. His departure from AMD to the ranks of the chip giant came at a time when GPUs grew in importanceand in which Intel was fully aware that it had to put in the work to get back into a market it had never succeeded in.

It can be said that the chip giant was not wrong in its new commitment to the graphics core. The importance of GPUs is greater today than everboth in the mainstream consumer sector and in the professional sector, where it is used to accelerate many workloads and to support the development and evolution of artificial intelligence, inference and deep learning, and this reality will only grow in the coming years.

Raja Koduri and his goals at Intel

The truth is that it wasn’t easy at all from the start and it wasn’t even a matter of resources or human talent, rather the opposite. Intel is a company that has the necessary resources to face any project with guaranteesbut the truth is that it was particularly complicated.

To understand this better, we need to analyze the key objective that Raja Koduri had to accomplish at Intel. This was it develop a GPU architecture, As many of our readers will know, but it didn’t work with any architecture, it had to meet a number of very specific keys to really compete with NVIDIA and AMD in both the professional sector and the mainstream consumer market. I’ll summarize the keys:

• High performance, at the level of the latest generations of NVIDIA and AMD, which at the time were Ampere and RDNA2.
• Easily scalable to design more or less powerful GPUs depending on current needs and to create a diverse product catalog more easily.
• Efficient and optimized, with a good power-to-watt ratio and adequate software and driver support.
• Compatible with the latest generation technologies and equipped with hardware specialized in ray tracing and artificial intelligence.

design from scratch A GPU capable of meeting these four goals wasn’t easy, that’s for sure, and with this simple breakdown of the keys, I’m sure you’ll be able to better understand the delays and issues Intel has faced and overcome since launch. .. embarked on the adventure of returning to the field of accelerators and GPU-based graphics cards.

Arc Alchemist: goal achieved, but half done

Development Intel Xe architecturewhich underpins both professional GPUs and those aimed at the general consumer market of the chip giant, was very complicated and they ended up suffering significant delays. These delays meant that the first commercial products were launched late, in a limited way, and that driver-level support was not really optimal.

I think it’s obvious these delays also played a role in the marketing of Intel Arc Alchemist graphics cards and had a negative impact on its sales. This is completely normal, because they were released when NVIDIA and AMD already had their new graphics generations, GeForce RTX 40 and Radeon RX 7000 ready, and this reduced the interest and value of the Arc Alchemist series.

Raja Koduri he was very late, there’s no denying that, and when you work for a company at Intel’s level, it’s clear that there comes a point where it can’t be tolerated. All in all, and despite the delays suffered by Arc Alchemist, it has to be recognized that the former VP of Intel’s graphics unit was able to lead his team on a path that allowed for an architecture that met many of these goals. we saw.

Intel Xe is an architecture highly scalable, as it is possible to combine Xe blocks to create higher performance graphics cores. It also proved to be a strong architecture and able to compete with Ampere and RDNA2both for raw performance and to support advanced technologies as it has specialized hardware for them speed up ray tracing and with specialized units in the field of artificial intelligence, the so-called XMX fieldwhich is exactly what Intel XeSS intelligent rebuilding and scaling accelerates.

Important goals were accomplished with Arc Alchemist, but Koduri left things behind along the way. Intel Xe architecture the power per watt consumed didn’t scratch the surfacecame to the market with a strong limitations in games based on outdated APIs (before DirectX 12 and Vulkan), lost a lot of performance with processors that didn’t allow ResizableBAR to be turned on, and Intel’s XeSS implementation wasn’t as good as expected.

Intel knows how to play cards and has taken the necessary precautions improve the value of Arc Alchemist A750 and A770, focusing mainly on improving performance with outdated APIs and reducing the price to accommodate the latest moves of NVIDIA and AMD in the mid-range, and finally we can say that it managed to save a release that clearly reeked of disaster for some. No, it wasn’t a success, that’s clear, especially since this generation of graphics cards came very late, but it’s also not fair to call it a failure, and I think it’s a decent first step, if possible, in Intel’s return to the graphics sector.

Realignment of graphic division, departure of Raja Koduri and legacy

Intel’s displeasure with the performance of Raja Koduri’s graphics division was made clear when the giant decided to reorganize the division, a move accompanied by a demotion, with Koduri stepping down as executive vice president. From that moment there was a very significant silence that led us to announce his departure from Intel, all this in a period of only three months.

The official version provided by Intel and Raja Koduri is this friendly trip, and that the latter will embark on a new adventure at the head of his own startup, which will aim to develop AI solutions for games, multimedia and entertainment. However, it is clear that despite everything the main trigger for this departure was this goal being half-fulfilled which we told you in the previous section.

In case anyone doubts this, I remind you that the Intel Xe architecture is also used in high-performance graphics accelerators for the professional sector, known as Ponte Vecchio, and that at this level Koduri was not able to meet all its goals, or the expectations it raised. You can’t even talk about an absolute failure, but that’s clear didn’t take the opportunity nor the resources and time that Intel has made available to you.

Intel’s new graphics division will continue, and so far it appears that cHe will continue his work based on the legacy left by Raja Koduri. This means that, in principle, there should be no major changes in the short and medium term future of the chip giant’s GPU strategy, and that the release of Arc Battlemage and Arc Celestial should be assured.

It is important to note that both of these generations of graphics cards target the general consumer market and They will use the Intel Xe architecture as a basis. Based on this, they will apply revisions and changes that should improve performance and efficiency, but there will be no architecture change, so in the end the differences in brute force could be relatively modest.

Arc Battlemage should hit the market between the end of 2023 and the first half of 2024and Arc Celestial will be available at some point 2025, if there are no delays and if Intel can meet its plan. In principle, we can expect the same hardware distribution at the GPU level as well (shaders, geometry and texturing engines, cores for ray tracing and XMX matrices for AI).

These two new generations of graphics cards will be the legacy of Raj Koduri at Intel, because as I said, they will use the Xe architecture with some modifications and improvements. The big change at the architectural level will come after the launch of Arc Druid, planned for 2026 first. Details are still very scarce as we’re talking about a graphics generation that’s still a few years away, so I’m sorry I can’t share more information with you at this time.

Will Intel succeed in the graphics sector?

Focusing on the general consumer market, which is ultimately what MuyComputer readers are interested in, I think he has achieved very positive things with Arc Alchemist and that this generation showed that it could indeed create GPUs capable of running current games and supporting next-gen technologies such as ray tracing and intelligent upscaling.

However, deep down it’s not easy at allAnd that’s what I think it all depends on what you do with the Arc Battlemage. This second generation of graphics cards will be crucial for Intel, and its success or failure will determine whether the company should continue to work to improve its position in the general consumer graphics space, or whether it should throw in the towel.

Both professionals and users can understand that the first step is always complicated and that not everything can be perfect. That’s why I get along so well with Arc Alchemist, But now that Intel has experienced that first release, it can’t afford certain things and we can’t accept certain bugs either.

Arc Battlemage must come to market in a very good state of maturity both at the level of hardware and drivers and support, and it is necessary for the chip giant to cover a wider spectrum of the market and close new deals to bring more titles to XeSS.

It is also necessary that this new graphics generation be cable to compete with what NVIDIA and AMD have on the market, either by matching or surpassing them in terms of raw performance and technology, or by offering attractive value for money in performance if it falls short of both (selling high-performance products at a lower price to compensate for their lower performance) .

If you are asking about a possible exit from Intel, I can assure you that it will not happen in the short or medium term, although some rumors pointed to it at the time. There’s a good reason for that, and that’s the Santa Clara Giant invested a lot of moneyresources and time in your adventure within the graphics sector and throwing in the towel would be a waste of all those resources.

On the other hand, you should keep in mind that the GPU sector is extremely important, namely Intel needs this component at different levels, from graphics cards for ordinary consumers to integrated graphics cores and graphics accelerators for the professional sector. Ultimately, there are many fronts and a company like Intel cannot afford to be left out of the race in the GPU sector. But of course, that doesn’t mean you can’t take a break and face some challenges or new development cycles in the long run.