Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

NVIDIA frenemy relation with OpenAI and Oracle

philippeoger.com

Post reply on HN
Print view View original

91–100 of 180 posts

Re: NVIDIA frenemy relation with OpenAI and Oracle

#91

Post by phailhaus »

Havoc wrote: post #81

Not a big fan of the circular observation. It's not the gotcha people seem to think. If the baker sells bread to the butcher, and the butcher sells meat to the baker then they can still both go to bed with a belly full of sandwich (aka actual utility & substance). Adding a third party to make it look more circle-y doesn't change that logic. Round trip financing is mostly an issue if it is artificial (e.g. a circle of…

That's the wrong analogy, because the butcher is giving money to the baker for the bread. If we we fix the analogy, then the baker gives money to the butcher so that he buys their bread with that money. The butcher cannot afford the bread without it!

Re: NVIDIA frenemy relation with OpenAI and Oracle

#92

Post by mattaw2001 »

wtallis wrote: post #54
zozbot234 wrote: post #33

Earlier quoted context omitted.

You can still scale SRAM by stacking it in 3D layers, similar to the common approach now used with NAND flash. I think HBM DRAM is also directly stacked on-die to begin with, apparently that's the best approach to scaling memory bandwidth too. It'll be interesting to see if we get any kind of non-NAND persistent memory in the near future, that might beat some performance metrics of both DRAM and NAND flash.

NAND is built with dozens of layers on one die. HBM DRAM is a dozen-ish dies stacked and interconnected with TSVs, but only one layer of memory cells per die. AMD's X3D CPUs have a single SRAM die stacked on top of the regular CPU+SRAM, with TSVs in the L3 cache to connect to the extra SRAM. I'm not aware of anyone shipping a product that stacks multiple SRAM dies; the tech definitely exists but it may not be economi…

I agree with your description and conclusion. Additionally the companies that can make chip stacks like HBM in volume are the HBM manufacturers. As they are bottlenecked by the packaging/stacking right now (while also furiously building new plant capacity) I can't see them diverting manufacturing to stacking a new SRAM tech.

Re: NVIDIA frenemy relation with OpenAI and Oracle

#93

Post by dboreham »

blindriver wrote: post #67

This is what happens when highly confident uneducated people read slop from other uneducated people with an agenda (Twitter, Burry et al) and then regurgitate more slop. There is no circular funding. There’s certainly circular speculation that is driving up the prices but the revenues are all accounted. The DSO change is meaningless if you understand accounting. The inventory building up is the cost of materials and…

Are partially processed wafers on Nvidia's balance sheet or TSMC's? Or are you saying Nvidia is holding sawn die in inventory? Confused as to what state the product might be in for it to make sense to hold for 4 months when supposedly you can sell everything you make.

Re: NVIDIA frenemy relation with OpenAI and Oracle

#96

Post by wmf »

mr_toad wrote: post #89
wmf wrote: post #65

Since SRAM is typically built in logic fabs (like TSMC) alongside the processors themselves, it theoretically shouldn't face the same supply chain crunch as HBM. DRAM and logic fabs are both sold out so replacing one with the other doesn't really help. And SRAM uses ~6x more silicon area than DRAM. Groq may be undervalued but not for supply chain reasons.

> Groq may be undervalued but not for supply chain reasons. Groq isn’t listed is it? Hard to know what its valuation might look like.

I see people routinely discussing valuations of private companies with certainty. I don't know how often those numbers are wrong.

Re: NVIDIA frenemy relation with OpenAI and Oracle

#97

Post by Magnets »

pinkmuffinere wrote: post #17
fwip wrote: post #13

- **The Cash Flow Mystery**: ... - **The Inventory Balloon**: ... - **The "Paper" Chase**: ... More AI slop.

I agree it's poorly written, but I'm _much_ more interested in whether it is correct, or incorrect. Do you believe it is incorrect?

75% margin means they have around $79.2bn of potential revenue sitting in inventory. Next quarter revenue is projected to be $65bn, so 110 days of stock.

Re: NVIDIA frenemy relation with OpenAI and Oracle

#98

Post by minimaxir »

mewpmewp2 wrote: post #41
khazhoux wrote: post #36

Earlier quoted context omitted.

So are we just disallowing multi-paragraph comments now?

If it's fully AI generated, perhaps add a disclaimer to it?

It is discouraged to post AI-generated comments to Hacker News, even if disclosed.

Re: NVIDIA frenemy relation with OpenAI and Oracle

#99

Post by danvayn »

mvkel wrote: post #16
kart23 wrote: post #3

The critiques of 'circular funding' don't really make sense to me. If you invest 20 billion and you get back 20 billion, your profit is the same. Sure your revenues look higher but investors have access to all that information and should be taking that into account, just like all the other financial data. Michael Burry is betting against AI growth translating into real profits as a whole, not the circular funding.

The problem is that stocks are often valued and traded on revenue growth, not profit[0] So circular funding generates stock price bumps when, as you said, there's no inherent value underneath. Creates a recipe for a crash. [0] consider pagerduty, incredibly profitable with little revenue growth. Trading at 1.5X revenue, where high revenue growth, unprofitable companies are trading at 10X revenue.

I feel like it's almost more of a Popular stock thing. Consider if pagerduty eked out an empty deal with any one of the "Pop stocks" that had little impact on their real profitability. Would stock trade differently or better? It feels like it really would in the modern market. Like even if the numbers weren't a big change, the buzz would be.

Re: NVIDIA frenemy relation with OpenAI and Oracle

#100

Post by danielmarkbruce »

next_xibalba wrote: post #86
tims33 wrote: post #79

The specifics of the article zero sense. Net Income and Operating Cash Flow are not the same thing so there is no mystery about whey they are different especially in a business with large Capex and long lead times. NVIDIA's historic DSO figures have also ranged from 41 to 57 over the last 5 years, so again not that crazy.

> Net Income and Operating Cash Flow are not the same thing This is the second article that gained a lot of traction on HN in the last few months with similar fundamental errors. The previous one was similarly written by someone with a dev background where they didn't understand the difference between gross margins, operating profit, and net income–concepts which were the basis for their analysis. I wonder with this…

A confident yet clueless swe? I don't believe you.
Post reply on HN