4 ms·
Ask HN: What's the Path for Innovation in Silicon?
If someone comes up with an idea, and pushes it through TinyTapeout or some other ASIC shuttle, and it actually meets design goals, such as lowering the cost of LLM computation by an order of magnitude. How does that lead to actually realized profit? Are there any VCs who specialize in this?
It seems to me that if Meta and others are going to spend a Trillion $USD on datacenter, saving even 50% out to be a good value add.
- Someone 1y agoI would think patenting the idea is the way to go, rather than verifying your design on an ASIC. Chances are that’s much more expensive, though, especially for a worldwide patent.
- mikewarot 1y agoPatenting it would have been a good idea a decade ago before I first started talking about it in public
- Someone 1y agoIf you have talked about an idea in public (and the public has noticed. If they didn’t, it can still effectively be a trade secret that’s worth something), the idea isn’t worth anything anymore. Money can be made in execution, but for this subject that requires deep pockets.
- bigyabai 1y ago> How does that lead to actually realized profit? It depends. How many billion dollars are you willing to pay TSMC for fab access? If it's not at least $5 bln then you might as well pack it up and go home, Nvidia and Apple can take things from here.
- mikewarot 1y agoSmall scale fab access is in my budget, enough to prove the point. (Less than $500)
- mikewarot 1y agoIt occurs to me that I assumed, from my observations here on HN, that YCombinator doesn't do hardware startups, and never questioned that assumption.