Recently, Intel boasted that the EyeQ 5 chip from its Mobileye subsidiary has over twice the deep-learning performance efficiency than NVIDIA's Xavier SoC. But is that really the right metric to focus on? EE Times investigated the matter and notes that it's a misleading claim because comparing the specs of two SoCs is silly without taking into account the other chips that are needed to achieve a Level 4 or Level 5 autonomous driving platform.
The site notes that NVIDIA's Xavier has a much more powerful AI engine than the EyeQ5. Mike Demler, senior analyst at the Linley Group, points out that at the moment, efficiency doesn't matter if you don't have the performance level required for Level 4/5 autonomous vehicles.
EE Times writes analyst consensus is that the target performance of Mobileye's EyeQ 5 is too low compared to NVIDIA's Xavier. Demler sees the two Eye5Q's in Intel's system as the "eyes" of the self-driving car platform. Atom SoCs will form the "brain" to augment the Mobileye computer-vision cores.
Phil Magney, founder and principal advisor for Vision Systems Intelligence (VSI), told EE Times that both Intel and Nvidia “are making a bigger deal out of this than may be necessary.”
While all this specsmanship was originally about the power efficiency of the rival SoCs, Magney said, “I would expect most L4/L5 vehicles are going to have electric powertrains so the wattage may be a moot point.”
He observed that when Tesla switched from Mobileye to Nvidia with autopilot, “we saw no noticeable degradation in range. So electric powertrains will handle this without much impact.”
At the moment, NVIDIA has the lead but self-driving cars are a nascent market so it will be interesting to see who comes out ahead.