Outsized returns can be generated, just in due course.
Blowing $500M is great for building an Amazon or an Uber -- if you require a lot of hardware or logistics. It's useful in a 4-way rat race where it's first-past-the-goalpost.
I tend to prefer deep tech, where I build something hard (or relatively hard) that's never been built.
A company like Bose has $3.2B in annual revenues, 58 years after founding, and I believe they've never raised external capital. That's because they have a big pile of unique technology. Contrary to their image, they do a lot of work which is not speakers (and which you've likely never heard of), from architectural acoustics to government.
TinySeed looks to fund small, niche SaaS companies, mostly. Most of the stuff they fund, more or less anyone with SWE background could build.
For developing hard tech, I'd pick $10M and 10 years over $200M and 2 years, any day of the week.