Interesting caveat:
> Our agreement with Meta is valued at ~$3B, with a five-year term. The size of the contract was limited to the amount of capacity that we had available.
He did an interview and said that because of it’s size, Brk will no longer out perform the general market. He also said that the stock will drop 50% like it has 3 times before. Sounds like he’s setting a 🌈 🐻 trap
Openai tells me "NVIDIA’s next-gen Rubin R100 AI GPU … on 3 nm process node in second half of 2025." I don't think it's only about the node size either, the platform and consistent releases/execution are part of the NVDA story.
I was a bit annoyed. My broker wouldn't let me short it, describing the stock as unborrowable. I had originally been planning to attempt a sort of spread bet long-straddle\* because I honestly wasn't sure what it was going to do. Anyway, in the end I took a punt and went long on a couple of £0.01/point bets, which left me up £1.19 due to some judicious scalping.
No way I was going to put even remotely serious money on that if I'm only allowed to go long.
*\*Normally a long straddle is done with options, but I'm in the UK and don't have an account I can trade options directly with. However, I do have a spread bet account. There are downsides to attempting a long straddle trade this way. With options, IIRC, your maximum loss for a long straddle is limited to whatever you paid for the options. With spread betting your maximum loss is the distance between your stop losses multiplied by your stake size - which isn't ideal - although if you were paying attention you'd be pretty unlucky to do that badly, but it's quite really easy for volatility to ding both your stop losses so you have to put them out quite a long way - maybe 5% pre-earnings to avoid that (which means you need a big move after the announcement to make any money). If the price doesn't move that much then you'll lose the spread multiplied by your stake when you close out your positions.*