I Finally Dug Into NVIDIA. Here’s What I Found.

[@newhampshirelifestyle4233]

I have owned Nvidia for over 5 years. As a self-directed investor since 1998 and a professional Software Engineering since 1991 - I can share some insights. Your analysis is correct, and the incredible thing about Nvidia - is they have not taken a proprietary approach to their product. Not only are they developing unique, cutting edge hardware, Nvidia has developed an entire Software Ecosystem that runs on their product line and they have made it freely available. CUDA, Tensor Core, Jet Pack SDK, DDS, Ada LoveLace Ray Tracing, etc. - the list goes on. They have supported OpenGL, Cuda, DirectX graphics libraries instead of forcing their own proprietary API/Applications on the world. I own a Jetson Nano, and RTX 4070, joined the Nvidia Developer form and LOVE their SDK. Everything JUST WORKS! On Windows and Linux all their drivers and API's just run seamlessly. This is why Nvidia has and will continue to grow.

[@mattderron]

This is awesome thanks for sharing

[@mindcache5650]

Yes. Last tech downturn, I bought in. Then when the market corrects I will buy more. You only need 1 great stock to make a fortune.

[@tomasondra6472]

It's never too late; I bought 300 shares for $200 in July 2021, and I am $250K+ as of today :)

[@mattderron]

that's amazing!

[@mikedowell8465]

I only own 10 shares but plan on buying more eventually.

[@ola_thripsala]

Wow good for you! :)

[@baalzhamon8491]

also a funny thing worth noting... Lisa Su, the CEO of AMD is the cousin of Jensen Huang, CEO of Nvidia

[@mattderron]

lol what?? Did not know that 🤯

[@uTube00789]

Lol, it's a rumor started on blind or Reddit.

[@jeffrose5622]

@@uTube00789 It's NOT a rumor. They are first cousins once removed.

[@HeroDai2448]

not the cousin. Lisa is jensens cousin daughter

[@jaxx4040]

As we so far know it’s possible and been said by some, yes lol Taiwanese

[@IsaacGonzalez-ku2pj]

Welcome to the NVDA team Matt! I was skittish when I bought 50 shares in early Dec at 475! People were saying it was overvalued then too 😂 I didn't care because I loved the growth numbers and story. So glad I pulled the trigger. I don't think the big rug pull comes this year because they are so far ahead in the space. Plus, all their customers are mag 7 companies with nothing but free cash flow to spend on NVDA chips! Also, be aware this thing is prone to huge intra-day swings so get ready for the ride! People sell it heading into earnings too - great opp to buy more imo!

[@offthetrail5675]

I had some Nvidia at 60. Sold out at 190 thinking I was smart lol.😢

[@mattderron]

Honestly if you have something that does over 3x it’s hard to argue that taking profits is a bad thing. Only easy in hindsight to know that it was going to go this far

[@Wakeywhodat]

I sold LLY at $350 ::takes a bow: You win, but everyone misfires!

[@jeffrose5622]

NVIDIA will end 2024 somewhere between $1,200 and $1,500. They are NOT overpriced at all. They are actually one of the least expensive M7 companies with forward PE in the 30's. They are THE COMPANY OF THE FUTURE!

[@Positively46]

I think it will hit $1200 Thanks for digging in to share in this video.

[@mattderron]

No worries!

[@kurniawanms2]

CSCO stock went almost 10x from early 98 to Feb 2000. It was the internet boom. All the telco companies were buying Cisco routers to build internet infrastructure. The demand for Cisco routers was so high that some companies were doubling their orders just trying to get the routers. Basically 4 years of demand was being pulled forward into 6 quarters. Then the demand slowed rapidly and Cisco stock crashed and never recovers till now. We have similar situation now with NVDA. Now substitute CSCO for NVDA, routers for H100 GPUs, internet for AI and telco companies for big tech companies like META, GOOGL. We might have several years of AI GPU demand being pulled forward into 4-6 quarters. Time will tell. Good luck

[@mattderron]

We shall see 🙂

[@TrumpxDaily]

Fair points but there weren’t miners back then, gaming industry has exploded, AI, so much more you are not accounting for in your comparison.

[@crawfish91]

the forward PE for cisco was in the 100s, foward PE for nvidia now is in the 30s. Cisco is only a hardware company. nvidia is also a software company. and also a platform. the big issue is whether or not nvidia can monopolize also the "inferencing". Nvidia has the machine learning in the bag. But all in all, if the big companies can't monitize AI, then they'll lower their cap ex.

[@mattderron]

@crawfish91 this is actually a great point that I didn’t discuss. Right now everyone is building to build and planning to figure out monetization later. I agree that if there’s no clear path to revenue for some of these things it could reduce capex spend

[@RS-lw9cd]

I bought 200 shares of CSCO in the mid 1990s. It split multiple times (2:1, 3:2, 3:2, 2:1) after I bought the stock. So after the multiple splits, I held 1800 shares. The stock price started to decline in price. I sold half my shares to buy other stocks. I still hold the 900 remaining shares. I probably should have sold them all, because it dropped so much in price. At that time, I figured I made a big mistake by not selling all of it. However, CSCO started paying dividends in 2011. It was my first real dividend growth stock at that time. My cost basis is approximately $6.12 per share. The dividend has risen to $1.60 per share. I am receiving $1,440 in annual dividends (900 shrs x $1.60 = $1,440). My dividend yield on cost (DYOC) is slightly more than 26% ($1.60/$6.12). As long as CSCO continues to pay its annually increasing dividend, I plan to hold the stock for the rest of my life. Selling CSCO would be like killing the goose that lays the golden eggs. How many investments can you get a 26% return (and increasing) every year? So, CSCO was great, then bad, and now, hopefully, will be good for years to come...only time will tell.

[@arigutman]

You're videos always are some of the most insightful out there... This channel should have over 100K subs. Easily.

[@mattderron]

Thanks Ari, much appreciated

[@cthornton99109]

I liked your comments. I have owned the stock since January 2020. I've got some great gains. I anticipate at least another 2 years' worth of gains.

[@mattderron]

Sorry missed this comment. Thanks!

[@supercruchynoodle]

I did the same. I ignored it for the same reason but then changed my mind a month ago. I dollar cost average a small daily amount and have taken advantage of a little of the growth.

[@HeroDai2448]

18:15 its ok. I bought 4 shares at 665 and thought that was high and it still went up more and it will likely go higher in the next months

[@toshirojira]

I just recently started investing at 42 (!!!) and have always felt I was way, way behind. Turns out, I've had a 401k (w/ mostly maxed contributions) for a decade, which (due to index fund tracking ETFs) has had me invested in Apple, Nvidia, Tesla, etc. So turns out I bought early! Need to keep reminding myself, it's never too late. Love your work, sir!

[@mattderron]

Thank you!

[@jsolo890]

​@mattderron the time stamp says Amazon and not Nvidia fyi

[@mattderron]

@@jsolo890 oh good call! I’ll update it

[@sk.n.9302]

I was really needing this analysis. Very helpful!

[@mattderron]

Glad it was helpful!

[@vicgill1980]

Good work! Nice independent analysis.

[@mattderron]

Thanks!

[@wonderz3217]

Been a holder since 2019…when no one knew anything about NVDA. I saw the most diversified futuristic company that will touch ALL markets… proud holder here. But no one is talking about that this is among the biggest Bitcoin player out there OMG!!! I actually bought it because that year was BTC’s bear market….and it was 50% on sale! As I dig in deeper, I knew this was a gold mine… I loaded the bought and told my co workers about it… only a few listened but most have sold. Now they all think I’m a genius!

[@rossmccabe7671]

Matt:: good presentation to the details of Nvidia business particulars. I have been all in on Nvidia for the past 6 months and have been most impressed-with Jensen Huang presentations during these past 6 months. I have no experience in business but I do have an interest and some understanding of the dynamics of leadership. Jensen is an exquisite example of the ultimate leader-that leadership sifts down to each employee and provides an expample of work ethic and willing to experiment and explore. Thanks again for your presentaton.

[@mattderron]

That’s awesome! My video for today that just got released actually includes Jensen’s comments about pain and suffering, which I loved!

[@kristentheologus-KTechnogal]

Great video! Congrats on taking the leap into Nvidia 😳 I bought 10 shares below $500 at the end of 2023, mostly feeling it was entirely speculative. But have been educating myself on their business, as you have, but it doesn't ease the “how far can it climb” worries 😉

[@mattderron]

Thanks! That’s awesome you’ve almost doubled your money already

[@OGBhyve]

You did a great job on the research! We will only see continued growth for NPUs dedicated to common inference tasks.

[@mattderron]

Thanks!

[@money.hungry]

Thanks again for your transparency 👍🏽

[@indersingh1212]

It's not about the Chips Matt! Nvidia is about both the hardware and software stack that provides an ecosystem for accelerated compute. Once an AI has trained their models, it's extremely difficult to leave the NVIDIA platform. Investors need to understand this and you haven't explained it. Growth and longevity comes from the software. Not the chips!

[@mattderron]

Thanks, I specially mentioned CUDA and their software being a big part of their moat in the video multiple times though?

[@SOAP-jf7ue]

I'm just waiting for them to announce net navis and battle chips.

[@JAMGAR369]

I also own 20 shares that I bought last year at an average cost of 488 so it’s almost doubled The way I see it is AI is just getting started and has a long way to go,while there is some competition I believe Nvidia will be dominant for two or three years for now and they could be dominant for many years after that I think it’s still a good buy now and will probably continue its rapid growth

[@Santanacarlos1016]

I bought Soxq (it’s a semi ETF, copy of Soxx with a way lower expense) soxx 8x the last 10 years. Way beating out Nasdaq and S&P. When I looked into the sector, there’s so much competition but a lot of the companies are great. Wanted ASML, looked into TSMC, and Broadcom, Nvidia… applied materials… a lot of these have excellent margins, FCF, top and bottom line growth. So I said you know what, I’m glad I’m an individual investor and not institutional, I can do a whatever I want. So I bought Soxq with a max 5% weighting, how I treat all my stocks. I expect to hold this very long term and expect Soxq to beat out the S&P the next 10. Which is my overall portfolio goal.

[@mattderron]

That's awesome, love the thought process. You're absolutely right that as individual investors we have flexibility that institutions don't have. A fact that is lost on many with the "90% of fund managers can't beat the market" stuff

[@dominiquetheeasyminimalist]

I agree that there's soooo much to discover in that industry! Exactly as you said "for the time being". I was very late to the party, but I'm up over 40%, and I intend to sell at the end of the year no matter what. NVIDIA had a a sharp pullback in the past, so I'll enjoy the ride for a while then move on. It is a cyclical industry and I'm not willing to risk it in the long run.

[@navegandolejanooriente6268]

There quarterly expectation are getting ridiculous, one miss short by even a dollar and that priced in future P/E ration are going right out the window with a ride down in the stock price. A correction is not the same as a crash in the stock but I could see it's price returning into the highs 400s to low 500s after a missed quarterly EPS on May 22.

[@ksrengaram]

Very good analysis on overall semiconductor industry and its way of doing business, hope your purchase moves the needle to become a multibagger Matt

[@mattderron]

Thank you! I don’t know if it has the same runway at this size that it did before, but hopefully still some solid returns to be had

[@yuriizbenko6115]

super, Matt, as always :) thanks a lot for your efforts, it helps us. I'm curious about your strategy regarding VLO

[@mattderron]

Thanks! Glad it’s helpful. In terms of VLO, I trimmed 25% of my position (that’s part of what I used to buy NVDA). I still have 150 shares of it and holding but it’s run up a lot and wanted to take some profits. It’s my only energy stock though so I’m not planning to get out of it completely unless something drastically changes.

[@tonyscinemascope]

my exposure to nvidia is indirect through my etfs (jqua, vong, soxq, nvdy) but maybe it's time i pick up a couple of shares, can't hurt

[@nfo2527]

Well-done Matt! Keep up the good work! 🥂I'm long & strong 700 shares and writing wkly calls. It's been crazy!

[@mattderron]

Wow!!! That’s amazing, congrats!!

[@TheDanielKahl]

Great analysis. Keep up the good work.

[@mattderron]

Thank you!

[@suriyar8135]

Good video with lot of information!

[@mattderron]

Thanks!

[@aymanhindam1123]

Good summary, just my 2 cents: Inference is processed on GPU not CPU for better performance (Neural Network Model (for AI) is much faster on GPU for both Training and Inference). Since 2020, TSMC is constructing 2 fabs in Phoenix , Arizona the first one will start production in 2025

[@mattderron]

Yeah this a good point, I was trying to make the point that in general it's viewed as a "less intensive" task where you didn't need the top of the line chips to be successful, thus giving others an easier path to take market share from Nvidia. "More like" CPUs but likely not

[@johnmcquaid7524]

Bought 3 shares around the same time as your original purchase (I paid $895/share). Of course; share price has dipped since my purchase. Thanks for sharing your thoughts!

[@mattderron]

I'll be adding to this position this week

[@KTrain82]

I added NVDA in my portfolio in January and already have very nice gain.

[@mattderron]

That's awesome! You definitely do

[@JonMcDough]

I'll solve the Nvidia price problem by buying one share. The demons that chase my every trade will drop it significantly and fast hoping that I'll be scared off. Then I'll buy the next share and have them repeat. It takes patience but eventually the investors jump in aggressively and drive it back up. I've done this slow walk numerous times on numerous positions. It's the only way I've managed to make money. Of course, I stick with quality stocks only.

[@RS-lw9cd]

You provided a nice analysis and explanation about the chip companies. It seems like people think it is easy for companies to replicate (without infringing on patents) A.I./GPU chips. The simple truth, it is NOT easy. It takes years and years of research and development, which is very costly. NVDA has been the dominant GPU designer (for producing PC video cards) for many, many years. NVDA's only real competition is AMD. There were many other GPU designers, but they could not compete with NVDA and they either went broke, or moved their efforts into designing other types of chips. Serious PC gamers will spend thousands of dollars to build their own PC, and a large cost of that is the GPU card. Anyway, NVDA realized early that GPUs would be the brains for A.I. and has a big lead in its development. AMD is second, but still behind in development. Yes, other companies are trying to produce their own GPU chips, but that will take some time... probably a lot of time. As they say, time will tell (in this battle to produce competitive GPUs). Anyway, there are lots of opportunities to invest in A.I. companies, including a plethora of chip companies, software companies, service companies, etc. There are so many. I just hope that the ones I have invested in will be the most profitable ones!!!

[@mattderron]

Yes we all are wishing that 🙂

[@paulschaaf8880]

High growth companies almost always look terrible if you analyze it like a value stock. What made me decide to buy was my outlook on AI. If you think AI is a fad, or a bubble then IMO you shouldn't touch Nvidia with a ten foot pole because without AI you have to focus mostly on their gaming business which is MUCH less lucrative. But if you think AI is here to stay and is going to develop into a huge market then Nvidia is one of the only companies around that is pretty much guaranteed to be a big winner. Lots of companies are going to make money on the software side, but which ones? There will be plenty of losers on the software side as well. On the hardware side though, who is everyone using? Nvidia. I do think more competition will eventually emerge but they have a huge lead right now. I bought around $150 and Nvidia is by far my largest holding now.

[@heyflo1]

Bought some shares last year and it's incredible. They're leading the industry by such a huge margin that if you want THE BEST you have to buy NVIDIA so... they're fine for a long time.

[@user-qh2kw3ln9h]

SOXX is another way to get into the semiconductor industry but to take a broader stance by investing into the ETF that has exposure to many of the leading players but yet if a pullback happens you are insulated

[@WanderingExistence]

I like SMH more, higher weighting on better stocks.

[@mattderron]

I almost went this route actually, but decided to buy NVDA directly

[@BrankoDimitrijevic021]

The problem is that NVIDIA’s growth depends on TSMC’s production capacity. They could try to tap Samsung (as they did in the past) or even Intel (if their IDM 2.0 strategy comes to fruition), but the fact remains that the number of advanced fabs in the world is limited. All manufacturers are currently in an investment spree to increase their capacity, but that’s only an incremental increase that will come on-line over a number of years.

[@kyukyu90002]

What software do you use to valuate companies? Do you have any tips on valuating? If you already have a video about it please do share! I found your channel a week ago and I subbed alarmed right away!!

[@mattderron]

I don't really have specific software, I use the spreadsheets you see in my videos (I provide them free with a sign up for my newsletter). If you want to see me go through - just watch videos on my "Stock Analysis" playlist on my channel. I go through my method for analyzing in each of those videos pretty much

[@evadeanu1]

Thanks for this video. You made all this AI a bit clearer in my head.

[@mattderron]

Awesome, glad it helped!

[@angelost1467]

Rumour is nvidia is planning to get their Grace CPU successor out early next year, that will give keep them way ahead of the competition in the inferencing business. I don’t see Jensen letting up.

[@AmericanPirate]

Good vid. I think a lot of us are now looking for the next Nvidia. I'm not saying it's not still a buy but profit wise it's at that point where it's not as attractive due to the price point

[@wendao8647]

I like your style - smooth, critical elements, well-rounded, no waste of seconds.

[@mattderron]

Thank you!

[@rssharma9]

I have NVDA in my portfolio and I have no idea whether to buy more or sell. I am just holding on to them for now.

[@UNDERDOG18UNDERDOG18]

I would hold until the Fed stops easing. Fed will probably cut rates incrementally starting in summer and stocks will rally in anticipation of QE.

[@Gode1967]

I have significantly invested into the SOXX etf …. So Nvida is biggest position and I do not care who wins the chip race in the end

[@Coyotehello]

Hi Matt, I just noticed (late on the game again) ... (but I am not a trader, I am an investor). META just started paying dividends. .39%. A game changer to me (I just put a stop on my "sell" for 20 shares of META at 330% profit). Any Co. paying dividends in my portfolios totally change category. Studently a stock dive in a year or two or more, does ot seems so scary anymore becase I am making money every quarter. Great move. Zuki got himself a nice boat. Sorry to dive on your episode here. Cheers, a.

[@mattderron]

META is a beast and I agree having the dividend makes it easy to just hold since they’re a cash flow beast

[@MukeshSharmabhatt9]

I invested in QCOM and NVDA at the same time when both were trading at 120. x amount in NVDA and 2x in QCOM. I wish i would have done other way around.

[@zebbie09]

One thing you are missing is that NVDA is anticipating markets and creating them…. They are also creating software and have solutions for specific verticals. They are also developing humanoid robots. It will be a wild ride, but hang on. This stock reminds me of Apple in 2010. The one thing that scares me is the dependency on Taiwan Semiconductor and the DOJ coming after tech companies….

[@mattderron]

Yes, I didn't bring up government / DOJ in this video but was going to make a video about it soon with all the activity (which IMO is annoying)

[@bjgould83]

Great video!

[@mattderron]

Thanks!

[@nataraja3030]

i also recently added nvda to my portfolio at a similar price after doing some research and reaching similar conclusions as yourself. great video, thanks for sharing.

[@mattderron]

That's awesome, thank you!

[@Coda1850]

Excellent vid, thx!

[@mattderron]

Thank you!

[@jaxx4040]

I got in originally at like $580 a share, thankfully a background in comp Sci and some business has helped me predict this.

[@mattderron]

That's awesome. My undergrad was in Computer Science as well...but from a while ago lol

[@jeffcargo4287]

Nvidia is a transformative company like Ford with the assembly line. AI is going to affect our lives more than the internet. One example: training robots. Nvidia is training robots using digital worlds much faster and cheaper than if they were trained in the real world. That has the potential to make businesses more profitable by replacing human labor with robots on a scale not seen before. Another Industrial Revolution with leaps in productivity. Also the way Nvidia chips will help transform medical research is just beginning. I could go on. And they are way ahead of their competitors. Can’t assign value to Nvidia like traditional companies.

[@loakland2773]

GREAT video Matt.... Very well done and presented....Hold NVDA and look forward to getting the split.... This company is very well run and the CEO is awesome... (unlike SBUX CEO).... This industry is all about the future which you identified... Robotics, gaming, cloud, AI etc plus throw in the fact chips will needed for everything we do as that world will continue to grow..... NVDA could in fact grow into other avenues and eliminate or lessen competitors. All the best to you and your family. Appreciate your hard work on making your videos so educational, informing and thought provoking.... Very much appreciated....

[@mattderron]

Thanks, totally agree about Jensen. Easily my favorite CEO right now

[@eversunnyguy]

Just curious..Why is the TSM stock not so hight...is it because of split or something ? Looks like TSML is kind of monopoly ?

[@davidwan685]

Excellent video !!! Learn so much. Thank You. I do not own NVDA and am still concern that all of those wonderful future sales and profits are already in the current stock price. Also, TSMC can only successfully make so many of these advanced chips (I used to work for a company that designed a chip to be manufactured by TSMC. The production time is long and fail rate is like 30%), thus foundry bottleneck. Even if the other big tech companies designed a "better" chip, they still need a foundry, and TSMC is the best. I see a similarity between NVDA and a stock that I own - NVO. I bought NVO AFTER everyone and their grandmothers around the world had already bought the stock. The stock actually falls after the "wonderful future sales and profits" excitement started to fade. Going to stay on the side line and cheer for you.

[@mattderron]

Thanks, I appreciate it. I think it will be a wild ride. If they have any weakness in earnings to come, I know it will correct and probably hard. I think it's got a way to go first though, so as always only time will tell

[@DJaquithFL]

What people failed to understand besides what you said .. 90%+ of the underlying AI is powered by Nvidia. By 2030 Nvidia is set to have a Market Cap of 10 Trillion and nearly 100% growth YoY.

[@ashishpatel621]

Hi Matt, Great video. Where to you get all those data and numbers, please? Thank you.

[@mattderron]

Thanks! It's mostly Seeking Alpha. What I showed is available for free on the site. I did a review of their premium service that talks about what's free and what's not, so if you want to see that it's here: kzbin.info/www/bejne/famraGSIfpV4itU

[@MukeshSharmabhatt9]

Where does Applied material, KLA and ASML stand in the food chain?

[@jonathanguzman8584]

Great video Matt,question, do you use a certain tool to get all those numbers for the P/E ? I'm trying to become a better investor/trader and would like to know how you calculated all those metrics.

[@mattderron]

Thanks! I think what you're referencing is Seeking Alpha. I use that to do the bulk of my research. The metrics you see in the video are available for free on the site. I did a review of their premium service and what is free vs not free a while back. It's here if you want to check it out: kzbin.info/www/bejne/famraGSIfpV4itU

[@jonathanguzman8584]

@@mattderronThank you so much, keep up the good work.

[@lucass.1295]

I’ve held NVDA for a while but I’m out at $1k.

[@mindfulJK]

Thanks Matt for the great video. I am luckily holding NVDA at appox. $210 range, but I was wondering why ASML didn't come up to you when researching? Maybe, it did and you make another video at a another time?! TSMC, Intel, and Samsung are basically begging ASML to sell the machines that make chips. It also pays better dividend than NVDA.

[@mattderron]

That's a really good point, ASML came up initially but because they aren't a designer or manufacturer I left them out. Same with SMCI for similar reasons, even though ASML is more relevant. However I put them on my research list to dig into more. Thanks!

[@WanderingExistence]

​@@mattderron There's a good CNBC video about ASMLs market strategy and business partnerships.

[@minhvukimnhat1323]

Thank you Matt for this nice vid!

[@mattderron]

No worries!

[@eversunnyguy]

Great overview of the life cycle of AI.

[@mattderron]

Thanks!

[@nick_g]

In since $423, plan to get out next year around $2300

[@mattderron]

Nice!

[@mathclass1357]

Bought and sold NVDA a few times. Always had to buy back at a higher price. Final bought at 420 last year and would not sell in a few years.😂

[@GaneshD123]

Great work - thank you!

[@mattderron]

Thanks!

[@gianmarco8583]

This is a great video to understand the market thank :)

[@mattderron]

Thank you!

[@anzatzi]

Great precis of Nvidia and the semi market. Imo-the biggest risk for Nvidia is that AI tech proves to be largely non-monetizable, ie, failure of MS Co-pilot or Siri 2.0.

[@mattderron]

Yes this is an excellent point that I didn’t put in the video. Totally agree that this is a risk, even if it may take companies a while to give up on it

[@kurniawanms2]

Another risk is the GPU power consumption. 1 x NVDA H100 GPU has peak power consumption of 700W. At 50% utilization, that is about 8.4kWh per day or 252 kWh per month. The average power consumption for a US home in 2022 was 886 kWH / month. So 3.5 units (886/252) of NVDA GPU running at 50% capacity will consume the same amount of power as 1 typical American household. META will install 350K of H100 GPUs in their data centers. That will consume the power for about 100,000 US households (350k/3.5). That is only for META. We still have AMZN, GOOGL, MSFT, ORCL etc. NVDA is projected to sell 3.5 million GPUs to big tech companies. 3.5 million GPUs at 50% usage will consume the same amount of power equivalent to 1 million US household. US energy infrastructure will not be able to cope with that rapidly increasing power demand from AI data centers.

[@mattderron]

This is a really good point that I didn’t touch on at all. I know NVDA’s new chips are supposed to be even more efficient but ultimately this problem will be there no matter what

[@MatthewCleere]

Don't forget the threat of China attacking Taiwan. TMSC in Taiwan produces a vast majority of the AI capable chips in the world right now, including almost all of Nvidia's chips.

[@davidhastings2603]

Great Detailed analysis ❤

[@mattderron]

Thank you!

[@soylentgreenb]

Nvidia is not s semiconductor company. They only design the chips; TSMC fabs them. They are mainly a software company. Their lead in AI is not about having the best chips for AI, it’s making software. CUDA has been used for a long time by researchers and this has created a large base of existing software and familiarity. That’s the moat; if someone like cerebras makes a faster more power efficient chip for training LLMs (which has been the case for years) this doesn’t allow them to work like a drop in replacement because all the existing stuff is CUDA and all the research is CUDA. The biggest threat is that Google and Microsoft etc are big enough to make their own chips. Graphics cards are a poor fit for AI and nvidia is playing catch up to make spinoff products that are significantly more efficient at AI related work. The moat is software and will remain so unless they manage to also pull another rabbit out of the hat and get best in class AI performance.

[@kevinlee3137]

I heard one analyst put a price target on Nvidia for $5500. I am a stockholder. Therefore, I would be OK with that $5500 price.

[@mattderron]

Wow, yeah totally fine by me 🙂

[@jakedamis]

I bought in at $106 and sold when it hit $430. Boy do I feel dumb lol.

[@mattderron]

I mean, 4x your money isn't dumb at all

[@ddubsr5886]

There’s always another opportunity around the corner

[@shaneomack5018]

How is that dumb? You did great man

[@jakedamis]

@@mattderron Haha yeah no doubt. It’s hard to know when to take profits. Mainly why I’m going with more of a dividend approach going forward.

[@mattderron]

100% agree, I talked about that in my newsletter last week. Knowing what to do with a winner is harder than knowing what to do with a loser IMO

[@Coyotehello]

Hi Matt, good o you for jumping on the Nvidia band wagon. At 800$ it is too rich for me, I feel like I missed the boat. Putting aside their design capabilities, the achillea heel is TSMC. In my experience the semi conductor manufacturers tend to leapfrog each other, in part because of the time and cost to build foundries. An older foundry might be able to produce 128 core chip or a 6 layers instead of the newer 8 layer technology. etc. So all that being said there are a few reasons you see the likes of Microsoft signing deals with Intel: - Brand new foundry opened just a few months ago. A foundry capable of building chips that TSMC cannot. - Located in the USA. This is important because the Biden admin has been pushing for more chip manufacturers in the USA and the USA military has said that they favour purchasing from USA manufacturers. - Finally I have full confidence in CEO Gelsinger. I bought Intel at 29$ and it now trades at 43$ I fully (hope) expect it to return to the 60-70$ range from the early 2020 years and more! But for now... Apple stock... $10B here, a lawsuit there... Cheers, a.

[@BrianHuynh95]

Very informative video Matt. I really like your analysis on this and respect your decision to add. To me, the “expected” growth has to be perfect for the valuation and I believe Nvidia could achieve it in a short term, say 2-3 years. I’d rather get exposed to it with my SCHG with 8% of Nvidia. Again, thanks for your research.

[@mattderron]

Thank you! For sure getting exposure to NVDA through an ETF is a great way to go with significantly less risk. The valuation is definitely stretched and they'll need to continue growing the business faster than the stock price.

[@aaa4181]

8% exposure isn’t going to make any significant gains for your portfolio.

[@albertlergier663]

Got to be in to win...

[@dlai70]

No company comes close. They are competing against themselves. I ignored NVIDIA because of the crazy valuation. I'm just riding the wave and look for an exit out of the position.

[@soylentgreenb]

Another threat to nVidia is that AI is too overhyped and won’t be seriously profitable and world changing for another decade or two. If that’s the case earnings are going to collapse; sentiment on AI is going to become too negative before slowly starting to climb back up. This is usually how these things work. E.g. Internet bubble; big spike; crash; slow climb to world changing and profitable. VR; big hype; hype dies and what remains is real; slow climb. I suggest AI hype will die; leaving what is real about AI and slowly climbing for another decade or two into something really useful. Another challenge with AI is that now may have been the only time you could train it. You could scrape the internet for images, texts etc and train an AI and this is rapidly being locked down. The ethics and legality of training an AI on someone elses copyrighted work is being challenged. Language models are being tweaked with woke nonsense (see google particularly). In 5 years there may have been such a flood of AI generated nonsense that AI will eat its own output from inferior versions of itself and it won’t get access to better training data. This is a technical challenge. That’s much slower to solve than just making the language model bigger. It required getting better at e.g. making a better discriminatory LLM to judge the output on writing style, factualness etc. It also involves purging HR of blue haired gremlins with San fransisco intersectional ethics that a vanishingly small part of the world actually agrees with.

[@mikerusso7095]

Welcome aboard the NVDA train. Enjoy the next 10 years of innovation and growth. I also worry about supply chain risk, but we are all connected …

[@mattderron]

That would be amazing

[@BanksBandz]

Nice add. I got in at $785

[@mattderron]

Nice!

[@mario3454]

Nice shirt!

[@mattderron]

Thanks! Go Spurs Go!

[@barrysharer3401]

Years ago I looked at buying Apple but I decided it was too expensive and didn’t. Later I bought Apple after it 7x’d from the previous time I looked at it. Bottom line it’s been a good stock. Lesson learned.

[@robertwilliam9558]

NVDA to the Moon.

[@quangyung6751]

to the sun.

[@ikyiAlter]

I love how this is a great analysis but it doesn't add any value to a 3 mins Google search on most people's impression on Nvidia. That said, I did learn a lot more. I mean there are a ton of people selling Nvidia and saying it is a bubble and it'll pop. So they're selling out...20%. lol yeah, that's just migitating risk but they still believe Nvidia will continue to do well for the near future. Thus, they still have majority of their original money in it. 🤦🏻‍♂️ Simply put, still a great stock to invest but one you should monitor closely.

[@87vortex87]

I'm not sure if your goal was to give a full supply chain run-down, but you missed ASML which develops the machines that makes these chips even possible. They have a monopoly on EUV and High N/A EUV. Also, BESI the market leader in hybrid bonding making chiplet design possible. Please include those two company, by default, when talking about the chip industry. It's such a gigantic blindside of US investors, it's ridiculous. Just look up both stocks and you'll see my point.

[@mattderron]

It wasn't, but it's a good point. Someone else asked about ASML in the comments, I didn't include it since it's different than the others, but definitely critical to the supply chain

[@lisagayhart2482]

Yep not a value company. Growth for sure . I am still in and buying on the small dips

[@rahhjur]

Google and Amazon have both been "building their own chips" for 7+ years (google with the tensor TPU and amazon with Graviton) and they're still both highly dependent on Nvidia (buying up billions each year). Nvidia has been providing datacenter chips to all the large cloud providers for more than a decade. Outside of the large cloud providers, government, research facilities, and entire nation states are looking to buy Nvidia's chips. AMD will have a chip competitive in specs this year with their MI300 chip (with much lower volume production), but Nvidia still has a much stronger moat around their software stack, developer base and overall CUDA libraries just offering so much more than AMD. Intel has Gaudi 2, but that's not a real competitor at all (low volume, overall lower performance, and they're new the game so software and developer support is extremely subpar, Intel's Gaudi 3 is expected to launch at the very end of 2025. AMD MI300 and Intel's Gaudi 3 is constantly benchmarked against Nvidia's current gen H100, however Nvidia just announced their new B100, which is 2.5-6X more powerful. So once that's launched AMD and intel will only fight for the much lower margin $5-10K price point GPU business. For full disclosure, 80%+ of my portfolio consist of semiconductor companies. Nvidia makes up 40%, AMD makes up about 10%, and Broadcom 13%.

[@mattderron]

Yup great point, I didn’t talk about it in the video but they’ve been making their own chips for a while and still depending on NVDA

[@JazevoAudiosurf]

see i can do this whole analysis or just see that AI is taking off and there is no end to acceleration

[@travistarr9433]

My boy Jimmy Cramer been on Nvidia since $40/share.

[@mattderron]

That's crazy, because normally inverse Cramer is undefeated 😂

[@hobsonschoice8649]

Have you looked at super micro computing

[@mattderron]

Only briefly while researching this video, but not in-depth

[@EyaEya2024]

wow big money🎉

[@wilmarkjohnatty4924]

Nvidia is incredibly underpriced. Ive been buying NVDA since 2017, and back then its future was more risky and over valued based on the information we had back then. My advice is to wait for a pullback which happens often because the stock is being manipulated by options and short traders and buy on the Dip. Overall its going up and up. But because you arent knowledgeable about the stock buy and hold. There is going to be excitement and volatility every quarter. AMD is very similar because this tech came from Gaming GPU and at one time AMD made GPU's almost as good as Nvidia's - (ATI). Also, Jensen Huang once worked for AMD and they have collaborated many times in the past especially against Intel which INTC was screwing up with Memory standards etc. AMD also was in talks to buy Nvdia back in the early 2000's and they ended up buying ATI. Also you are missing out a big piece - the infrastructure companies like SMCI, Dell etc., ARM and the related companies like Sound Hound AI etc.

[@Klausi666]

AI Bubble, i would still recommend it to buy.

[@Thailand-iz7bx]

I WANT TO BUY MORE

[@rubidiumcratchet3420]

nvda will hit 1k next week