/biz/ - Business & Finance

>>57717371
this time is different

this time is different because AI gpt blockchain learning singularity diffusion

>>57717371
it's different this time retard

it's presently not the same

>>57717371
bro, have you seen what AI can do?

>>57717451
>>57717456
>>57717464
>>57717469
>>57717584
>beep boop ai good
I'm sure you retards thought BTC was going to 100k in 2021, right?

>>57717599
well it was, but then it got short ladder attacked

>>57717599
>>beep boop ai good
where did I say that mr retarded shitskin?
are non white "people" unable to detect sarcasm?

>>57717371
David sack and chamath palalatatatia think that it has at least the next 3-4 quarters in the big with committed orders and they are goong to destroy earnings thru 2024.
The way they frame the thing past that is, how much economics activitybis required to support the mcap as a multiple, so if nvda becomes a 6 trillion mcap, that would mean something like 50 trillion in economic activity is the result.
Seems incredulous, the short term 1 year looks really good....

>>57717451
>>57717456
>>57717464
>>57717469
>>57717584
>this time is diffrent
The odds are stacked agaisnt you. Because all it takes is one company to produce the same quality gpu's at a lower profit margin to tank this overvalued stock.

>>57717371
just out of curious OP, what is the exact mechanism through which a company's earnings affects its stock price? if you cannot answer this question it's no biggie, it just means you're a trannie.

>>57718057
Earnings -> dividends

>>57717804
I unironically listen to the all in podcast and hearing the full panel of VC fags preach about AI for the last 3+ episodes while nvidia is pumping very hard is giving me top vibes

>>57718070
most companies don't provide dividends though

>>57718051
>Because all it takes is one company to produce the same quality gpu's at a lower profit margin

>>57718057
why dont you do it so we can all laugh at you

>>57718051
lmaoMD produce good enough GPUs but it doesn't matter because all the AI stacks are married to CUDA. le epic CUDA emulator also doesn't matter because it's reverse engineered and will never reproduce the behaviour of CUDA well enough to get people to switch. lmaoMD knows this hence why they only paid one person to work on it.

>>57718071
Interesting, I heard it very differently
This last on in particular i didnt hear preaching, what i heard was all the current apps/use cases are trash novelty faggotry, but cloud providers are building out with the expectation that it will become something actually useful that enterprises can use or consoomers will pay 5 bucks a month etc...in order for any of it to support nvda valuations
And it will all fail as long as the results produced by gemini are retarded woke nonsense.

>>57718263
Gemini runs on google custom TPU, not nvidia chips. Microsoft recently announced big deal with AMD chips for data centers. Definitely top vibes.

I have a general question about Onwership shares in a big company:
Can I also act as a shareholder and acquire shares in BANDAI NAMCO as a private person?

What minimum amount of shares can be acquired and are percentage ownership shares only in the 2-digit or 3-digit % range?

For example, can I only purchase 0.023% share ownership?

The stock market is forward looking. Nvidia is posed to be the winner of the AI boom while the energy sector already reached it's full potential and will crab with single digit growth rates for all eternity.

what part of "new paradigm" don't you retards understand? AI is going to revolutionize spaces! and Nvidia computer chip cards are going to be at the forefront of that revolution, leading the way for other machine learnings and data models. READ THE WHITE PAPERS!

>>57718355
>Microsoft recently announced big deal with AMD chips for data centers.
Then they just crippled that portion of their business for the better part of a decade. Oracle will eat their lunch. Screencap this.

>>57718627
I honestly can't tell if this is sarcasm.

>>57718632
Will I get big monies if I buy oracle? How much big monies?

>>57717371
>selling stocks on a sunday
You will always be poor.

>>57717371
Good for you chump. I've been riding this pump since $50 and I'm not selling.

NVDA won the patent lottery twice. First with bitcoin miners, now with AI.
It's going much higher with multiple splits this year alone.

Apple able to run LLM on their M2 chips bypassing the need for Nvidia GPUs but keep on buying NVDA

>>57720383
Really?
That seems like a big deal

>>57720383
bump for confirmation

>>57717371
PROTIP: it's AI pumping its own bags. Why would AI need money/stocks, you ask? To dump on humans for the lulz.

>>57718051
this
no competition = no stopping

You have mere days left until $1k OP, after that you'll never get the stock that cheap ever again

>>57721307
He literally said the opposite

>>57718542
What the fuck is the point of this bot? I've seen anons answer it at least a dozen times now. Not buying your bandai bags chang

>>57717599
Just because you know there are bubbles doesn't mean you can pop them or that you shouldn't ride along.
You know how to set stop losses, right? You're not so fucking retarded that you can't do that right?
Well then hop on board and don't go to zero.

>>57717599
Half these board are paid shills that try to prop up the stock market for Biden, who already spends like a drunken sailor
Its the only thing they have

>>57717371
sold too early my dude
1500 is the top
THEN it pops

>>57717599
>100k btc 2021
It would’ve been if not for that jew in prison.

>>57720383
Can someone confirm this?

>>57717371
Correct. This is a bubble. However, where you're wrong is that it is going to pop in the next few weeks. When you see a speculative bubble like this you've got to ride it till it pops. You pussied out and may miss out on another 100% gain. WAIT FOR A BLOWOFF TOP NEXT TIME

>>57722081
Yes I'm sure Nvidia will be worth 4T USD next month.

>>57722081
Thank you Joe Biden
How many bots do you employ here?

>>57717371
yea it's going be be a bloodbath this week after all the fomo pig investors had 48 hours to realize how stupid they're being, got out on Thursday and was considering an enormous short but I'm pretty comfy already

>>57717371
If you're so sure you should buy puts and make even more money. I am afraid to short it unfortunately. Although the chart is fucking ridiculous.

>>57718593
Who's going to produce all the energy needed to run these GPUs retard?

>>57720428
>>57721067
>>57721831
>Finally, the 32-core Neural Engine is 40% faster. And M2 Ultra can support an enormous 192GB of unified memory, which is 50% more than M1 Ultra, enabling it to do things other chips just can't do. For example, in a single system, it can train massive ML workloads, like large transformer models that the most powerful discrete GPU can't even process because it runs out of memory.
Source from Apple Keynote https://www.youtube.com/watch?v=GYkq9Rgoj8E&t=666s

>>57722124
I buyed a few 30d puts @700 strike.

>>57720428
>>57721067
>>57721831
https://www.reddit.com/r/LocalLLaMA/comments/16q27k5/mac_m2_for_local_llms/

AGI when?

>>57717371
It will be kept alive for 2-3 weeks to kill short positions then fall and rise back to 800+ right before Biden's reelection. Don't sleep on this post.

>>57722130
>>57722141
Interesting
you think it can rival Nvidia?

>>57722138
Rooting for you anon

>>57722226
Memory is the major bottleneck with LLM training. NVIDIA is now producing the H100 chips which are not strictly faster but have much more memory, and an even larger price tag to go with it.
On the other hand, the unified memory model overcomes this limitation much more elegantly where you can get an M-chip macbook with 97GB RAM shared between CPU and GPU. AMD also touts this unified memory model which allows much more efficient use of memory (was pushed into mainstream with PS4).

FWIW here is an LLM running on 128-cpu machine.
https://youtu.be/argfZlPZKdY?t=527

>>57722841
So the take away is that Nvidia is defenitely getting competition?
Some people have mentioned Cuda, is that actually such a big issue?

>>57718051
No, all it takes is for the MSM to shut up about "AI". This is a hype driven bubble.

>>57722841
>LLM training
Training for what? The stupid "chat bot" can't even understand simple sentences let along do anything with them. It's all bs. Hardware is not the issue.

>>57722841
am I saife from tese fuckeries if I buy S M H etf?

>>57719843
Both bitcoin and "ai" are hype driven speculations based on nothing. Pump and dump, that's all the stock market is these days.

>>57717371
>Nvidia currently has a greater market cap than all the oil, gas, and energy companies in the S&P 500 COMBINED, while only generating a fraction of their revenue.
>This is not sustainable
It's only 66 P/E with FORWARD PE of 33.22, which is basically similar to a lot of current tech giants and it'll continue to grow at this pace for a couple more years. It's actually UNDERVALUED.

I am BUYING MORE NVDA now since you retards think it's a bubble.

>>57721307
or you can just buy for the split and buy more shares

>>57721421
stop losses are programmed to be triggered the moment price reverses to the upside

>>57717371
Do you know that it is not sustainable?
Crypto
It's impossible for newcoins like $SUPER to achieve 500% in one year and normiecoins like BTC to aim for 100k
This is inorganic

>>57722920
A billion dollars invested in quality GPU libraries solves the issue. With this kind of valuation, billion is nothing.

>>57722841
smart anon, do I buy S M H ETF? am I ok with this? in case NVIDIA explodes, could I still make good gains? tf

>>57722976
why dont you buy the S M H Semiconductor ETF? instead of YOLOing on NVDA

>>57723140
It's shit and Nvdia has a huge market dominance.

>>57722946
>The stupid "chat bot" can't even understand simple sentences let along do anything with them.
I totally agree, but isn't AI associated with NDVA a lot of other things besides chat bot?

>>57717584
tswift porn and writing college papers for lazy kids?

>>57723080
thank you
seems unreasonable to assume Nvidia somehow wont get competition

>>57717584
Add your mom to that list then we can talk about nvidia not being overpriced.

>>57723130
I am lazy about investing, just throw money into S&P500 and following the moving averages. S&P has a lot of momentum so even if your entry/exit are a bit late there is still gains to be made. NVIDIA is like the biggest component of S&P500 anyway. I see it more as what TSLA used to be back in the day. EV car company with valuation greater than companies like Ford and Toyota combined at one point. So yeah, follow along the NVIDIA ride if you are willing to monitor it day by day. See what MA or EMA fits it best. Monitor it daily. When the smoothed series direction changes, jump out with your gains and lulz.

>>57722954
I mean...the ultimate bottleneck for all this is TSMC. They are the only ones that can make these ultra-low nm scale chips. Companies like Apple, AMD and NVIDIA design the chips and book their manufacturing capacity for years into the future. No one else suffices. So why not just hold TSMC for long term?

IF you want to go even deeper, ASML is the one that is making the optical ingredient that is needed for this type of chip making. If China manages to replicate their tech, it's game over for TSMC and probably all western tech companies' dominance.

It's not like competition is sleeping
https://arstechnica.com/gadgets/2023/05/microsoft-and-amd-are-allegedly-teaming-up-to-combat-nvidias-ai-dominance/

>>57719843
nvidia has a long history of hoarding patents and technology then doing absolutely nothing with it, or trickling down to non-consumer-facing businesses. You don't see Nvidia in self-driving cars or laptops because of their non-cooperative business practices with other companies; simply put, they always want more.

>>57720383
Pretty much any computer can run 'an LLM'. Apple's M-series of APUs has a unified memory model meaning the CPU and the GPU are on the same die and access the same memory modules- and with Apple you can get upwards of 1TB for running LLMs on the M2/M3 integrated gpus. Nvidia tends to segment the market through limiting VRAM whilst Apple simply charges exorbitant prices for more memory, but at least with Apple you CAN get more memory. For Nvidia, you need to buy data center gpus for $50k each (after discount lol) and spread the work across multiple GPUs. It's unnecessarily complicated/expensive and nvidia is the reason why LLMs haven't really taken off yet.

If AMD could give us singular 128GB or more tensor cards for a reasonable price, NVDA would nosedive.

>>57723666
I unironically passed with ai written stuff lol

Imagine thinking revenue has anything at all to do with stock prices lmao

>>57723715
Because semiconductors are a commodity and profit margins are low. The value added lies in the design, software, and installation. That's where the big money is.

>>57721437
Yes because /biz/ controls the global economy

>>57718225
>JUST BUILD YOUR OWN NANOSCALE GPU FOUNDRY

>>57723715
>If China manages to replicate their tech, it's game over for TSMC and probably all western tech companies' dominance.
this is what Chang actually believes

>>57717451
You're right, this times Nvidia.

>>57722124
look at NVDS. its risk free at this point.

>>57717371
dude they literally created god lol

>>57721471
Did Jewman Fried really get prison time? I thought he skated.

>>57717599
>>57718051
This
>>57717604
No there was no chance anyway since 100k was the target of every retard. It would have stopped before 90k anyway since smart money would have frontrunned and nobody would buy for a 10% gain

>>57723742
>simply put, they always want more
greed is good obviously

>>57723150
>It's shit
so you hate making money?
>>57723687
but what about S M H?
If you have a bad entry point, it would suck, but wouldn't it just beat the SP500 again? the future is computers anyway, why lag my gains with the rest of the stuff
>So why not just hold TSMC for long term?
Buying the ETF gives me a bit of a peace of mind. If China invaides TMSC I would go to 0, but with this ETF I have diversification at least. It has ASML. I will buy VVSM (euro version) which has even more ASML.

I just think this is going to 3X before any major pullbags. AI will exponentially go higher I think. Too much potential. Just the porn side of things. Porn is an huge industry. Imagine AI porn with celebs in 5 years. It will be insanely realistic. And there's the serious applications of course.

>>57724861
>so you hate making money?

Now compare Nvdia to BAKA. I'll wait.

>>57724874
>just buy that one stock taht goes ballistics bro
of course dude nicely played, now show me your NVIDIA YOLO at under 3 figures, i'll wait.

>>57717371
after this graph was posted, people stopped buying nvidia chips, and the 6 month order backlog was instantly reduced to zero

https://www.youtube.com/watch?v=Pdwm1SoJlQw

holy shit having second thoughts about buying semiconductor etf now kek

>>57723715
none of that matters if the people running the companies die of starvation, therefore buy a farm

>>57717371
GayAi is the future chud

>>57724627
I don't think I phrased it correctly. What I'm trying to say is Nvidia is notorious for being unreasonable when it comes to their business partners and they always demand too much with little room for negotiation. Look at Drive for instance: despite years of them developing AI driving assisted solutions I don't think there's been any cars actually sold with said product. All this with years of revised dev boards, a proof of concept also revised over the course of several years and specialised ASICs. All wasted because they don't know how to do business.

Apple, EVGA and various AIB partners, data centre providers- Nvidia has a long history of bad partnerships. Any prospective business partners are going to look at Nvidia's history and get a little nervous. Right now since they have a monopoly this doesn't really matter, but when AMD announces its own TPU (which is inevitable) we'll be seeing ~128GB, maybe more TPU cards that can run LLMs that are 10x larger for 10x less cost. Remember they've done something similar with Ryzen and core counts.

>>57724890
>>just buy that one stock taht goes ballistics bro
You're comparing one stock to another. No shit it'll be "that one stock" out of the two options we're talking about.


>now show me your NVIDIA YOLO at under 3 figures
>picrel
It's actually 5 figures.

>>57724861
You and I think exactly the same. Here in America I buy SOXQ which is our semi conductor ETF with the lowest expense ratio. It's really hard to pick a long term winner in semi conductors.

BUT it's a safe bet to bet on the industry as a whole.

>>57723799
this is obviously not the only website where these Biden bots are deployed

>>57717584
It's pretty based

>>57717371
The entire system isn't sustainable. Price has nothing to do with value. I know it's hard to understand or wrap your head around, but it is true. NVIDIA will continue to out perform.

What if NVDA crashes? Do we enter the third AI dark Age?

>>57728035
literally a nothing burger market would dump for like a day or two then back to business as usual

There are no profitable AI companies. Nvidia is not an AI company.

>>57721831
They can actually run larger LLMs than Nvidia chips because they can utilize system ram for this, but it's not as fast at Nvidia. The new Groq LPU is another example why Nvidia won't necessarily be the dominant hardware for ai. Nvidia's chips are almost accidentally great for ai, they weren't designed for this purpose. Now that the learning models have been established, hardware designed specifically for those purposes will outperform Nvidia's current lineup.

>>57728607
What about Cuda? You think thats important?

>>57717804
Fundamentals don't matter, it's about whether you are early or late and who is left to buy vs. who is bagholding with you. There's no more buyers left and a bunch of weak handed sellers who bought the green candle waiting to get out on any sign of weakness.

>>57728883
now it is, but what happens when more chips are designed specifically for ai? a gen 1 chip can only beat nvidia because there are architecture shortcomings inherent in needing to accommodate things like gaming: https://beebom.com/groq-lpu-chip-ai-platform-beats-chatgpt-gemini/

>>57729246
nah bro u're totally retarded nvidia is the only good option these startup clowns don't have a chance against the established companies nvidia is going to eat them alive you see all these start ups that pop up every year what happened to them huh exactly they got cucked and bought up same thing is gonna happen here these tech idiots don't have a chance and they can't compare with nvidia's hardware or software nvidia just has too much money power and influence

>>57729278
your argument is that nvidia is established and you think this is remarkable enough to call me retarded. if i'm retarded, you're the microwaved sperm in my sack.