# 143: I'm Thinking of Unicorns

- Show: [Embedded.fm](https://hardwarepodcasts.com/podcast/embedded-fm)
- Published: 2016-03-17
- Length: 1h 06m
- Page: https://hardwarepodcasts.com/podcast/embedded-fm/2016-03-17-143-im-thinking-of-unicorns
- Listen: https://makingembeddedsystems.libsyn.com/embedded-ep143mp3

## Summary

Dan Luu, a hardware and software engineer at Microsoft with prior experience at Centaur and Google, discusses the evolution of CPU features, the trade-offs between working at startups versus large tech companies, and the efficacy of computer science research. He highlights how specialized hardware accelerators can offer massive performance gains over general-purpose CPUs for specific workloads like deep learning, and critiques the software industry's neglect of error handling and rigorous testing compared to hardware design. Luu also shares insights from his blog, danluu.com, on topics ranging from compiler bugs to the persistent myths about startup compensation and the path dependence of career success.

## Chapters

- [0:06](https://hardwarepodcasts.com/podcast/embedded-fm/2016-03-17-143-im-thinking-of-unicorns#t=6) Introduction and Lightning Round
- [5:13](https://hardwarepodcasts.com/podcast/embedded-fm/2016-03-17-143-im-thinking-of-unicorns#t=313) CPU Features Since the 1980s
- [10:29](https://hardwarepodcasts.com/podcast/embedded-fm/2016-03-17-143-im-thinking-of-unicorns#t=629) Hardware Accelerators and Specialized Chips
- [18:33](https://hardwarepodcasts.com/podcast/embedded-fm/2016-03-17-143-im-thinking-of-unicorns#t=1113) Startups vs. Large Companies
- [37:19](https://hardwarepodcasts.com/podcast/embedded-fm/2016-03-17-143-im-thinking-of-unicorns#t=2239) Efficacy of Computer Science Research
- [46:50](https://hardwarepodcasts.com/podcast/embedded-fm/2016-03-17-143-im-thinking-of-unicorns#t=2810) Error Handling and Software Testing
- [1:01:31](https://hardwarepodcasts.com/podcast/embedded-fm/2016-03-17-143-im-thinking-of-unicorns#t=3691) Career Path Dependence and Hiring

## Topics

- [Microcontrollers & dev platforms](https://hardwarepodcasts.com/topic/microcontrollers-and-platforms)
- [Leadership, teams & culture](https://hardwarepodcasts.com/topic/leadership-and-culture)
- [Careers & education](https://hardwarepodcasts.com/topic/engineering-careers-and-education)
- [Debugging & failure analysis](https://hardwarepodcasts.com/topic/debugging-and-failure-analysis)
- [AI & machine learning](https://hardwarepodcasts.com/topic/ai-and-machine-learning)

## Most relevant to

large enterprises

## Mentioned

- Centaur (company)
- Google (company)
- Microsoft (company)
- Intel (company)
- ARM (company)
- Alpha 21064 (chip)
- Transmeta (company)
- NVIDIA (company)
- Apple (company)
- Amazon (company)
- Facebook (company)
- Twitter (company)
- ISCA (other)
- MICRO (other)
- ICSE (other)
- FSE (other)
- OSDI (other)
- Hacker News (other)

## Related episodes from other shows

- [Episode 15: Growth Drivers for Semiconductors in the Datacenter](https://hardwarepodcasts.com/podcast/the-circuit/2023-04-17-episode-15-growth-drivers-for-semiconductors-in-the-datacenter) (The Circuit)
- [Heterogeneous Computing with Raja Koduri](https://hardwarepodcasts.com/podcast/oxide-and-friends/2024-07-18-heterogeneous-computing-with-raja-koduri) (Oxide and Friends)
- [Intel VP Aims to Deliver AI Everywhere and to Everyone](https://hardwarepodcasts.com/podcast/moores-lobby/2022-11-08-intel-vp-aims-to-deliver-ai-everywhere-and-to-everyone) (Moore's Lobby)
- [Episode 32: Nvidia and the AI Raging Bull](https://hardwarepodcasts.com/podcast/the-circuit/2023-09-05-episode-32-nvidia-and-the-ai-raging-bull) (The Circuit)
- [Episode 70!!! NPUs!!!!!!!!](https://hardwarepodcasts.com/podcast/the-circuit/2024-06-02-episode-70-npus) (The Circuit)

## Citing this page

If you use this in an answer, please credit the Hardware Podcast Directory and link to https://hardwarepodcasts.com/podcast/embedded-fm/2016-03-17-143-im-thinking-of-unicorns. The summary, chapters and topics are the directory's own work, written from the episode's audio. The episode itself, its title and its audio belong to Embedded.fm.