Samsung, finally the 4 nm production process works properly | Rumor

0
14
samsung, finally the 4 nm production process works properly |
samsung, finally the 4 nm production process works properly |

 

If TSMC trudges with the 3nm process, apparently Samsung is making major progress in the 4 one: according to reports Pulse News, the foundries of the South Korean giant would have managed to achieve a rate of return almost equal to what was seen in the 5 nm. This means that in all likelihood Google’s next top-of-the-range SoC, the Tensor G3, will finally “drop” to 4nm, after two generations stuck on 5nm.

As we know, Samsung started producing 4nm chips even at the end of 2021, but things didn’t go very well. In addition to having a very low performance, Qualcomm’s Snapdragon 8 Gen 1 were struggling a bit from a thermal and energy point of view, so much so that the US chipmaker decided to switch to TSMC for the mid-cycle refresh, the Snapdragon 8 Plus Gen 1, with very tangible differences.

The 4nm node improvement might seem almost irrelevant by now, calculating that the time is now ripe for 3 nm (in fact Samsung started mass production already last month), but it is not necessarily said. The most refined process of all is generally dedicated to smartphones or at least more generally to mobile chips, but there are many other manufacturers impatiently waiting their turn. Like AMD, for example, which again according to the source would have signed a new supply contract, even if for the moment it is not clear for what purpose.

It’s worth remembering that, as far as Google is concerned, only the Pixel 8 and Pixel 8 Pro, which are likely to arrive this fall, should feature the new generation chip; the other upcoming devices, including the first foldable, the budget Pixel 7a and the tablet, are all expected to be based on the Tensor G2.

Previous articleOne UI 5 Watch, Samsung will focus on improving three features
Next articleHonor Magic VS review: first view in the Foldable world
Abraham
Expert tech and gaming writer, blending computer science expertise