Not saying anything about the valuation but they just provided a guidance of $11B revenue for the next quarter which is about 50% more compared to the quarter they just reported for...
Yeah, but let's more than double that - and say they make $100B of revenue per year. That's still a Price / Sales ratio of 10 ! Price to sales, not price to earnings !!!
You never know, maybe this AI thing is not just hype (I think it is for now), and maybe NVidia will basically own the future of AI computing chips - if that's the case, then okay - this valuation will be OK - but this valuation is here, now !
And a huge company stock price is moving by 20-30%. This is absolutely bonkers. Know, a lot of financial reporting happens on periodical basis (be it end of day or month) - it kind of assumes that things are not moving too wildly and you can put some trust in it. Which is usually the case for big companies. But lately, we see those top N companies in US, with prices having huge swings. I wonder, what are the implications of that.
This makes no sense. Already trading at 50x EBITDA on forecasted datacenter demand. How does this actually pan out? AWS, Microsoft, other large internet companies buy a bunch of GPUs to train models _one time_, or at best at an annual cadence. Others rent once a month to train.
Model inference will be done on the edge. Apple makes their own silicon and will never return to NVDA, so count out all iDevice revenue. Same with Tesla.
> This makes no sense. Already trading at 50x EBITDA on forecasted datacenter demand. How does this actually pan out? AWS, Microsoft, other large internet companies buy a bunch of GPUs to train models _one time_, or at best at an annual cadence. Others rent once a month to train.
OpenAI trained on Nvidia cards, but doesn't OpenAI's ChatGPT run on Nvidia cards as well? So it's not just the training, but the execution as well, right?
I train Dreambooth models for Stable Diffusion, which takes up a lot of GPU time, but then I generate images with those Dreambooth models, which takes less time (but still uses my Nvidia card).
> This makes no sense. Already trading at 50x EBITDA on forecasted datacenter demand. How does this actually pan out?
Easy: Within 20 years, we'll see the first publicly-traded company > $100B trading at 1000x EBITDA, before crashing down to 200x EBITDA.
As shown in the last few years, money can be loaned out easily by the Fed & banks in times of extreme economic stress. With more crises inevitably coming up in some form or another, more loans will eventually be given out to incentivize the continued existence of the market.
This kind of demand has to spur competitors into action. I know it will take many quarters, but I assume AMD/Intel/others will catch up? So why do investors believe the current growth will continue many years, to justify this valuation?
They have had a 60% plus margins for years, I think they're just out competing. Or the others have stagnated. The valuation is ridiculous but they have built a product/sales pitch that is extremely difficult to disrupt/replicate.
This is now a 1T$ valued business, based on ... 20B$ annual revenues, say (generously) 10B$ annual profit, and A LOT of hype.
(I would never recommend trying to short it though - the market can stay irrational longer than ... )