Should we value Nvidia as a software stock? Its CUDA software is a major competitive advantage in AI. Even if we do this, it is hard to get comfortable with the current share price.
We did some math to show market share and profit share in the foundry business. This data maps to intuition, but the scale of TSMC’s dominance and Intel’s declining fortunes stand out.
Global Foundries’s future lays in bringing the fabless business model to the analog industry. The transition to 300mm wafers could prove a big catalyst for that.
Semi companies are going to need to adjust their cost structures to compete in a semi-custom world. This sounds boring but it can become a real source of competitive differentation.
Themes we teased out from earnings season:
1) End-markets matter. Consumer bad, industrial good.
2) Everyone is putting a lot of faith in a 2H recovery
3) Inventories are still high, but fab utilization is holding steady
Intel is in a very bad place. It needs to admit that, especially internally. We are not forecasting Doomsday, but we do think it is time to recognize that Intel will never be the force it once was, and probably has not been for a long time.
There were a lot of autonomous vehicles at CES – not just cars but lots of Big things.
For us, the big theme of CES was how empty much of the show was. Is the show just recovering from the pandemic or is it poised to become something different.
The market for base station and RAN silicon is shaping up to one of the most interesting to watch in the new year. And intel seems to have forgotten its long, sorry history in the space, making for some great viewing.
For the past decade, we have labeled anything newish as a “Tech” company, but the markets are reminding us that there is a difference between companies that use technology to solve business problems and those that are actually solving technology problems.