Hardware

Google optimizes new Tensor G6 chip for efficiency, not performance

Google optimizes new Tensor G6 chip for efficiency, not performance

Google continues to develop its own Tensor chips for the Pixel line of smartphones, starting with the Pixel 6. While those chips have previously lagged behind the competition in performance and power efficiency, leaked data indicates the company plans to change that with the release of the new Tensor G6, codenamed “Malibu.”

The company’s new Tensor G6, which is codenamed “Malibu,” is the first of its kind in the Pixel lineup.

Main changes to the Tensor G6:

  • Focus on thermal efficiency and energy savings. The focus is on reducing heat dissipation and improving efficiency rather than increasing performance.
  • .

  • Limited performance improvements. Even the predecessor Tensor G5 is expected to bring only modest performance gains, and Google doesn’t seem to be looking to force development in that direction.

.

Google optimizes new Tensor G6 chip for efficiency, not performance (tensor g6 specs)

Potential changes and reductions:

  • Use of old GPU design. Tensor G6 may use a GPU originally designed for Tensor G4, resulting in the lack of ray tracing support expected in Tensor G5.
  • Reduction in crystal area. For optimization purposes, Google intends to reduce the crystal area of the Tensor G6 to 105 mm², whereas the Tensor G5 occupies 121 mm². This will be achieved by, among other things, using TSMC’s 3nm N3P technology and optimizing the crystal layout.
  • .

  • SLC and DSP reduction. The Tensor G6 is expected to have a reduced top-level cache to 4MB and one less core in the DSP.

.

This way, Google is tackling the challenge of making smaller, more efficient chips without putting much emphasis on power gains.

Subscribe
Notify of
guest
0 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments

You may also like

More in:Hardware