-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Project Dojo - the SaaS Product?
- Thread starter Buckminster
- Start date
Knightshade
Well-Known Member
huh?
This story is about wanting to build an actual chip fab (multiple ones from the sounds of it)
Nvidia has 0 of those. Tesla also has 0 of those.
This would be competition for folks like TSMC and Intel.
Elon predicted this. And why he said it would come down to FLOPS/$.
Knightshade
Well-Known Member
Yup, the "DOJO AS A SERVICE" folks gonna be mad
As I suggested in other thread rumors of the head of Dojo no longer being at Tesla due to Dojo 1.0 not performing well look more and more likely to have been accurate
Last edited:
Exactly, if Dojo costs 25% less than unobtainium Nvidia and is half the speed they still have a winner.
jabloomf1230
Minister of Silly Walks
What is the difference between a bad idea, and idea that takes a lot of hard work over a long timeframe to achieve...?
The difference is most people give up far too early, and far too easily, that is why they never achieve anything..
It takes tenacity to make things work...
Elon does a few things that are hard to achieve, or which might never be achieved, which if they can be done have a large pay-off. Someone has to tackle the difficult problems.
jabloomf1230
Minister of Silly Walks
Having an inhouse supercomputer is a great idea. Having it based on your own proprietary chip isn't. Elon gambled and failed.
I think the project is proceeding on the basis that it may work out, and may come in handy, that was always the plan..
It was revealed at AI day just to help attract talent,,
The problem with the early reveals is the media and investors expect quick and certain results...
Doing their own chip worked out well for the FSD computer, at worst they are batting 1 out of 2.
jabloomf1230
Minister of Silly Walks
Post in thread 'Tesla, TSLA & the Investment World: the Perpetual Investors' Roundtable' Tesla, TSLA & the Investment World: the Perpetual Investors' Roundtable
Knightshade
Well-Known Member
Other thread suggests this is 500M over 5 years, so 100M/yr... which in AI compute spending terms is very very little... sounds like indeed V1 turned out quite underwhelming and this is more like a larger scale testbed situation to see if they can get a better V2 or V3 product that's genuinely worthwhile compared to giving Nvidia/AMD money further in the future.
Tesla and NVDA have clear line of sight of major long term volume plus the cash to make big orders. This gives them the advantage of being able to hoover up existing capacity:
Similar threads
