While recent reports had indicated that the physical size of the die that makes up the CPU and GPU on Microsoft’s Project Helix might be considerably larger than the one in Sony’s PS6, technical analysts believe that this ultimately won’t lead to too wide of a gap in horsepower between the two next-gen consoles. During a recent episode of its podcast, Digital Foundry noted that the difference in power won’t “get you a whole lot,” and that it is “basically not that meaningful.”
According to Digital Foundry, the difference between the two next-gen consoles is likely to come down to the fact that Project Helix has around 26 percent more compute units than the PS6. For the sake of comparison, the Xbox Series X has 44 percent more compute units than the PS5, with the latter having faster CUs. They made a note of the fact that the current generation of consoles didn’t really see a meaningful difference when it comes to quality or frame rate across most titles, largely thanks to more modern software features like dynamic resolution and image upscaling through FSR.
Industry insider KeplerL2 took to the NeoGAF forums to add their own thoughts on the matter, noting that the gulf of power between the two consoles will be wider than what we saw between PS5 and Xbox Series X. While the latter had 20 percent higher TFLOPS (trillion floating point operations per second), along with 20 percent more LLC (last level cache) and memory bandwidth paired with 18 percent lower front-end bandwidth, geometry rate and pixel fill rate when compared to the PS5, the Magnus APU powering Project Helix is slated to have around 25 percent higher TFLOPS than the PS6’s Orion APU, along with 33 percent higher front-end bandwidth, geometry rate, and pixel fill rate, as well as 140 percent more LLC and 20 percent more memory bandwidth.
Despite this, even KeplerL2 believes that this won’t amount to too much of a difference when it comes to real-world use and performance. It will come down to Project Helix being capable of higher frame rates than the PS6 in some cases, for example, or being able to handle path tracing where Sony’s system can only do regular raytracing.
“It’s definitely a bigger difference than this gen,” wrote KeplerL2. “XSX has 20% higher TFlops and Texture Fill rate plus 20% more LLC and memory bandwidth but 18% lower Front-End bandwidth, Geometry rate and Pixel Fill rate. For Magnus it’s ~25% higher TFlops/Tex rate, ~33% higher Front-end BW, Geom rate, Pixel rate plus 140% more LLC and 20% more memory bandwidth.”
“That said, I agree that it’s not enough to make a huge difference, like Magnus running something at 60 FPS while the PS6 can only handle 30 FPS, or running Path Tracing in a game where the PS6 can only handle RT.”
The insider went on to note that, while the hardware differences are big on paper, in practice, it will likely come down to Project Helix being able to run games at higher internal resolutions before the image is upscaled through FSR than the PS6. “I mean that the [hardware] difference is not big enough to allow that to happen,” they wrote. “In most cases I expect Magnus to run higher internal res (e.g PS6 1080p upscaled to 4K, Magnus 1440P upscaled to 4K) or using slightly higher quality settings.”
However, the major difference between Project Helix and the PS6 is expected to be in pricing. While Sony will likely stick to its current console pricing strategies to help grow its user base, Microsoft’s console is expected to be much more expensive. Digital Foundry’s Oliver Mackenzie noted that, “Because the new Xbox Magnus die is also over 400mm squared, even though I think it is a dual die design, that is a considerably large die for a console. Whereas PS6 seems to be a die around PS5 Pro size, as it’s a very svelte die that’s monolithic, so that should be cheaper to produce just by its nature as well.” The bill of materials for Project Helix due to its larger die size is expected to be considerably higher.
