Intel’s high-end Arc A7 graphics cards will be available “very soon,” and Team Blue has been forthcoming with the everything we need to know, including how they’ll perform in comparison to Nvidia’s offerings.
Intel revealed some intriguing information regarding the state of Arc desktop GPUs, including the confirmation that the next Alchemist graphics cards to come, the A770 and A750 high-end models, will join the cheap A380 that is currently out there “very soon,” so potentially later this month.
At the same time as the release of bespoke boards from third-party card makers, reference cards from Intel will be made available for purchase on Team Blue’s website. Germany, along with other “important” nations, will be a part of the rollout (so this time it won’t be limited to Asia).
The A770, when it launches, will be positioned as a midrange option between Nvidia’s RTX 3060 and 3060 Ti graphics cards, with the A750 roughly matching the performance of the base RTX 3060.
The sort of performance you receive will naturally vary widely from game to game based on a wide range of circumstances, as Petersen notes in his interview with Digital Foundry (with one notable element being wonkier performance with DX11 titles, a much talked about point of late, compared to DX12 or Vulkan). Nonetheless, those are the ballpark figures for performance (on the AMD front, we’re talking about somewhere in the middle between the Radeon RX 6600 and RX 6650 XT for the A770).
The performance-dependent dependency of Intel’s Arc GPUs on ReBAR was also discussed (Resizable BAR, a PCIe feature that allows the CPU to get full access to GPU memory for considerable frame rate boosts). Intel is hard at work optimising drivers to enable older PCs that don’t have ReBAR support, which is a concern because those machines see a significant performance hit when using Arc GPUs.
Intel’s positive outlook is a breath of fresh air
What’s nice about Petersen’s Intel insights is that he’s being forthright. We just brought up the tricky subject of ReBAR compatibility; while Intel is attempting to optimise for older PCs, Petersen recommends that people choose Intel’s GPU competitors, AMD and Nvidia, instead. It’s encouraging to see that type of forthrightness and honesty throughout the interviews here.
Petersen readily admits that while supply constraints slowed down the desktop Arc GPU deployment, the main challenges focused around making sure games worked well and customers had a positive purchasing experience. The problem was certain to arise, what with the thousands of already-popular games available.
Intel undoubtedly was aware of this fact; nevertheless, a closer examination of the situation suggests that perhaps it proved to be a far greater source of stress than expected, which would explain the lengthy delays.
In addition, it is made clear that future tweaking of graphics drivers will not result in a significant increase in the performance levels of the Arc graphics cards; nonetheless, we can expect much improved compatibility across a wide range of games, and a better overall experience.
Given that Intel has no plans to increase performance, and that both AMD and Nvidia are working on next-generation graphics processing units, the rivalry in the graphics card market is set to get significantly heated up. We’re anticipating some reasonably priced desktop GPUs from the A7 series to compete with existing options, as the implication is that here is where Intel can best hold its own.
Petersen also emphasises that Intel isn’t in this for a year, a few years, or even a decade; it wants to be a player in the graphics card industry for a long, long time, and as the months pass, we should anticipate those drivers to grow more and more polished, and in better form overall.
While the Arc Alchemist launch has been less than inspiring so far, Intel’s approach in this regard has been anything but, and it gives us reason to believe that the future of GPUs will be brighter and more competitive with three competitors rather than a duopoly.