I thought George was going to save AMD. Now he's saving tenstorrent? Busy guy!
I wish there was a thousand more geohots than all the mediocre middle-managers at AMD or tenstorrent; or people who have never done anything beyond posting snarky comments in online forums.
Sadly, I think geohot is an example of someone who earned some cred for impressive accomplishments in the past and then tried to cash in that cred over and over again in unrelated future domains.
His brief and very public flame out at Twitter after mysteriously abandoning another project and the bold claims about his AMD work that never really translated to anything have really detracted from whatever past “cred” he built up. I really hope he can find a new niche and succeed, but until then it might be time to lie low on social media and avoid throwing more mud.
>I think geohot is an example of someone who earned some cred for impressive accomplishments in the past [...]
Huh? Yeah, that's what I wrote.
You cut out the part of my post where I made my point.
Earning “cred” for past accomplishments doesn’t give someone a free pass forever to be a loud mouth.
>Earning “cred” for past accomplishments doesn’t give someone a free pass forever to be a loud mouth.
What the ... ? Sorry, but you don't have to earn that. I suggest you familiarize yourself with human rights, particularly if you won the lottery and were born in a country like the US; it is very clearly written, in unambiguous terms, the rights their citizens enjoy.
And still, I'd rather listen to geohot rambling about whatever he wants to say, than some randos on the internet who have accomplished nothing but "arguing". Respect is not a right, it is earned, he's earned mine.
This is obtuse. I obviously wasn’t talking about human rights or the 1st amendment.
I was saying it’s illogical to suggest that because someone did one impressive thing in another domain a long time ago, we should therefore continue to value their input on every topic forever.
Let's see your code.
comma.ai is an impressive achievement on its own!
One of the lessons I wish I'd learned earlier is that indulging bombastic behavior neither benefits them nor you.
Strong opinions, weakly held are good. Well-reasoned opinions to the contrary are even encouraged. Provided, of course, that they're smart enough to know when to persist and when to disagree and commit.
If you just indulge it, they end up with the engineering equivalent of Nobelitis. And if they're on your team, you end up with more burden than asset no matter how brilliant they are.
With AMD the experience is so poor that you have to save the company from itself if you want to make progress.
pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/rocm6.3
today on stream and it just worked. No external ROCm install, and just the amdgpu driver that's in Arch.
We also have our own complete driver/runtime now in tinygrad; it's so much nicer to build on a foundation when you can blame yourself for the bugs.
Maybe SA might set their sights on you next?
The short-term payoff of that drama created a long-term problem where the only way they could look good was by outrunning the progress of the engineers who were inside the company, well funded, and already familiar with everything. It was an impossible goal from the start but he made it even more impossible by attacking AMD. AMD was smart to basically ignore them and wait for him to give up, as opposed to inviting that drama to crossover in-house or split their user base.
Like it's strange people think I give up on things, I think they listen to the media too much. This is a 2+ year long project that I've worked on almost every day. https://geohot.github.io/blog/jekyll/update/2023/05/24/the-t...
There was a big public episode where he appealed to Elon Musk for a job at Twitter, was given an “internship”, tried soliciting public submissions for the code he was tasked with, left a lot of people shocked that he was struggled with basic FE work, and then resigned 4 weeks in: https://news.ycombinator.com/item?id=34074344