# data center revenue — X 热门讨论 (2026-10-01 13:10 UTC)

## @Normal_2610 (Normal Guy) · 10-01 09:57 · ♥38 ↻8 💬4 AI will make software much cheaper and easier to build, but that does not mean Indian IT companies will make more money. Earlier, companies like TCS and Infosys grew because more software work meant more engineers, more billable hours and more revenue

But AI may change the whole process. Instead of 10 people doing a project, AI agents may do most of the coding, testing and documentation, while only 2–3 experienced people supervise.

So even if the world builds much more software, Indian IT companies may not capture the same value.

The big question is More software → yes But who makes the money?

It could be TCS/Infosys, AI companies, cloud companies like Microsoft/AWS/Google, new AI-native firms, or even the customers themselves.

Indian IT still has strengths like deep client knowledge, old complex systems, regulation and cybersecurity expertise. But the old model of more people = more revenue could become much weaker

At least the clear win is for Data Center & Energy

The bulb example explains this. Early machines simply helped glassblowers work faster. Then the Ribbon machine completely changed the manufacturing process, jumping production from roughly hundreds of bulbs per hour to tens of thousands.

The glassblower didn't become 200× more productiv - the glassblower stopped being the centre of production

https://t.co/36hZTMk3hV https://x.com/Normal_2610/status/2105597907038871746

## @MilkRoadAI (Milk Road AI) · 09-30 14:17 · ♥30 ↻7 💬10 The Super Bowl of semiconductor earnings is here, $MU reports today.

Wall Street is looking for roughly $51.2 billion in revenue and $31.50- $31.60 in adjusted EPS for fiscal Q4 but I am slightly more bullish, projecting around $52.2 billion in revenue, 87.5% gross margins and $33 in EPS.

The quarter matters, but guidance matters much more. I want to see whether Micron can guide next quarter toward roughly $58 billion in revenue, margins near 88% and EPS in the high $30s. That would be another strong signal that the memory shortage and pricing cycle remain strong heading into 2027.

These are the the biggest questions I want answered.

Micron had signed 16 strategic agreements last quarter. How many more have been signed, are they getting larger or longer, and how much 2027 DRAM and HBM capacity is already committed?

How are HBM4 shipments and customer qualifications progressing? How much 2027 capacity is already sold, and what is the timeline for HBM4E?

Can demand continue growing faster than new capacity from Micron, Samsung, SK Hynix and Chinese manufacturers?

Pricing: Are customers still willing to lock in memory years in advance, and does Micron expect pricing to remain strong throughout 2027?

If free cash flow keeps growing, how much goes toward new fabs and advanced packaging versus eventually being returned through buybacks?

The bigger reason this report matters is the scale of memory demand ahead.

Nomura estimates memory demand tied to global data-center capex could increase from $107 billion in 2025 to $454 billion in 2026 and nearly $1.4 trillion by 2030 and the AI data center portion alone is forecast to grow from $106 billion in 2026 to $517 billion by 2030.

These are forecasts but they show how much memory could be required as models get larger and AI inference, agents and context windows continue scaling.

For the stock, simply beating this quarter is not enough because expectations are already extremely high.

The real test is whether Micron confirms that pricing, margins, customer commitments and AI memory demand remain strong throughout 2027. > 引用 @MilkRoadAI: If you own Micron stock, You NEED to read this!

Wall Street is currently expecting roughly $51 billion in revenue and around $31.50 to $31.60 in adjusted EPS. I think Micron comes in higher. My estimate is around $52.2 billion in revenue, 87.5% gross margins and $33.00 in adjusted EPS. That would put my revenue estimate roughly $1.2 billion above the Street and EPS around 4% to 5% above consensus.

The reason I am comfortable being slightly above the Street is pretty simple. Memory pricing continues to remain extremely strong while supply is still struggling to keep up with demand. Micron originally guided for around $50 billion in revenue, 86% gross margins and $31.00 in EPS, so expectations have already moved meaningfully above management's original outlook. I think pricing has remained strong enough for Micron to come in above those expectations again. Gross margin is going to be one of the most important numbers for me. The Street is already looking for margins around 87%, which is insane when you think about where this business was just a few years ago. I am looking for around 87.5%. If Micron can continue growing revenue while keeping margins anywhere near these levels, the amount of earnings and free cash flow this company can produce becomes ridiculous.

DRAM is another big reason I am staying above consensus. AI infrastructure continues to become more memory intensive with every generation. Larger models, longer context windows, inference and eventually AI agents all require more memory alongside the compute. At the same time, customers are trying to lock up supply years in advance because they are worried future capacity will not be available when they need it.

This is also why the long term agreements are probably one of the most important things I will be listening for on the call. Last quarter Micron had signed 16 strategic customer agreements and I want to know how much that number has increased since then. Are we talking about 18 agreements now? 20? Even more? More importantly, I want to know whether the size and duration of these agreements are getting larger as customers become increasingly worried about securing enough memory. These agreements matter because they give Micron something the memory industry historically has not had much of, which is long term demand visibility. Memory has always been extremely cyclical because manufacturers usually do not know exactly what demand or pricing will look like several years into the future. If customers are now willing to commit to supply years in advance, provide deposits, agree to minimum purchases or accept pricing protections, Micron suddenly has much better visibility into future revenue and utilization.

I also want more detail around HBM because this is becoming an increasingly important part of the Micron thesis. The company has already started high volume HBM4 shipments for its lead customer and has sent qualification samples to additional customers. I want to know whether Micron is maintaining or gaining HBM market share as the industry moves from HBM3E into HBM4 and eventually HBM4E. If Micron can hold around 20% share or move even higher, it becomes an even bigger beneficiary of the AI infrastructure buildout.

HBM4E is another area I want management to talk about. I want to know when additional customers are expected to qualify the product, how much of 2027 HBM capacity is already committed and whether customization allows Micron to capture even better pricing. The more customized HBM becomes for specific AI accelerators, the harder it becomes to look at Micron as just another commodity memory company.

The most important part of the entire earnings report, though, is probably going to be next quarter guidance. The market already expects a monster fiscal Q4, so whether Micron reports $51.8 billion, $52 billion or $52.5 billion might not matter nearly as much as what management says comes next. I would personally like to see Micron guide toward roughly $58 billion in revenue next quarter, gross margins around 88% and EPS somewhere in the high $30s. If management gets anywhere near those numbers, that would tell me the memory pricing environment is still extremely strong heading into 2027.

The biggest risk I want addressed is supply. Samsung, SK Hynix, Micron and Chinese memory manufacturers are all investing heavily because the economics are so attractive right now. Eventually that capacity will come online (I personally believe it has no meaningful effect but I want the management to address the Chinese memory makers) .The real question is whether AI memory demand can continue growing faster than the industry can add supply. I want management to give us more clarity on 2027 DRAM supply growth, how quickly new fabs can actually contribute meaningful capacity and whether Chinese memory companies are starting to change the supply picture. If supply suddenly starts growing much faster than demand, pricing could turn very quickly, but right now I still think demand is winning.

I also want to hear more about capital returns. Micron is spending aggressively because it needs more capacity but if revenue and margins remain anywhere near current levels, the company should also generate an enormous amount of cash. At some point investors are going to start asking how much of that cash gets reinvested versus eventually being returned through buybacks or other capital returns. Overall, I am going into Wednesday expecting another beat. My numbers are around $52.2 billion in revenue, 87.5% gross margins and $33.00 in adjusted EPS, compared with the Street at roughly $51 billion and around $31.50 to $31.60 in EPS. But I am not going to make a prediction on whether the stock goes up or down after earnings because honestly that is basically a coin toss at this point. Micron has already had a massive run and expectations are extremely high. The company could beat the Street and still sell off if guidance does not clear the bar investors have built into the stock. It could also report numbers close to expectations and move higher if management gives an extremely bullish outlook for pricing, long term agreements and 2027 demand. That is why I care much more about what Micron actually tells us about the business than trying to guess how the stock trades the next morning.

If you enjoyed reading this and you want to see exactly how I’m positioned in Micron ahead of earnings and the rest of the memory names I hold, check out my portfolio below.

https://t.co/WtE7ibqvoU https://x.com/MilkRoadAI/status/2105300968455983357

## @rickyho_1989 (Ricky Ho) · 10-01 08:20 · ♥36 ↻4 💬1 The more I look at the economics of AI infrastructure, the more convinced I become that one of the most popular bear arguments against the AI capex cycle is beginning to weaken, because the market has spent the last two years assuming that GPUs are effectively brutally depreciating assets with extremely short economic lives, where every new NVIDIA generation supposedly renders the previous one obsolete and leaves hyperscalers and neoclouds holding stranded hardware, yet the data in these charts suggest something much more interesting is happening, with older GPUs maintaining or even improving rental economics, inferred residual values remaining surprisingly resilient, and this occurring at exactly the same time that the end-user cost of AI, measured through token prices, is falling sharply.

That combination matters enormously because if the bear case were correct and every new generation of GPUs immediately destroyed the value of the previous generation, then we should already be seeing H100, A100 and H200 rental rates collapse as customers migrate toward newer hardware, secondary-market values deteriorate and utilization falls, yet the opposite appears to be happening, with H100 residual values recovering toward roughly US$30,000, H200 values around US$38,000, A100 values still holding near US$18,000, while B200 rental rates remain extremely strong around US$5.69 per GPU-hour, H200 around US$3.29, H100 around US$2.63 and A100 around US$1.59, which is not what an oversupplied or rapidly obsolescing market is supposed to look like.

What makes the second chart even more interesting is the overlay with the token price index, which has fallen dramatically from above US$2 per million tokens toward roughly US$1, meaning that the cost of consuming intelligence is declining at the same time that the productive value of the hardware producing that intelligence remains resilient, and I think this is the point that many investors are still missing because falling token prices are often interpreted as evidence that AI economics are deteriorating, when in reality they may instead be evidence that AI is becoming dramatically more efficient while usage grows even faster.

This is the same pattern we have seen repeatedly throughout the history of technology, where storage became cheaper and the world responded by storing exponentially more data, bandwidth became cheaper and people responded by streaming more video, compute became cheaper and companies responded by building more software, and now intelligence itself is becoming cheaper, which means the natural outcome may not be that the world spends less on AI but rather that the world consumes far more intelligence because the number of economically viable applications expands every time the cost of inference falls.

That is why I think investors need to be careful when they look at AI hardware through the lens of traditional server depreciation, because older GPUs do not necessarily become useless when a new generation arrives, particularly when the underlying demand for compute is expanding much faster than new supply can be installed, which means the market can evolve into a hierarchy where the newest GPUs are allocated to frontier training and the most compute-intensive reasoning workloads, the previous generation migrates into mainstream inference and fine-tuning, and older generations continue serving enterprise workloads, batch processing, smaller models and latency-insensitive tasks.

The best analogy may actually be aviation rather than smartphones, because a new Boeing 787 does not suddenly make every 777 worthless, and while the newer aircraft may be more fuel efficient and better suited to certain routes, the older aircraft can still generate attractive returns if demand for air travel remains strong enough, which is increasingly how AI compute appears to be behaving, with Blackwell not necessarily killing Hopper, Hopper not necessarily killing Ampere, and each new generation instead creating a new upper tier while the older hardware cascades downward into other economically useful workloads.

That is not instant obsolescence. It is segmentation. And segmentation matters because it expands the total market by allowing different classes of hardware to serve different levels of workload intensity, which means that if aggregate AI demand continues to grow faster than aggregate compute supply, then every layer of that hierarchy can remain valuable even as the frontier keeps moving forward.

This is exactly why the rental-rate data are so important, because H100 pricing stopped declining and began recovering, A100 pricing stabilized, H200 economics strengthened and B200 pricing remained strong, all of which suggest that the market is not experiencing widespread compute oversupply but rather continued absorption across generations, and if that interpretation is correct, then one of the key arguments behind the AI capex bear case, namely that hyperscalers and neoclouds are overbuying GPUs that will become economically obsolete before earning back their capital cost, begins to look much less persuasive.

The useful economic life of these assets may simply be longer than the pessimists assume, and even more importantly, the payback period may be materially shorter than the depreciation period, which is the combination that actually matters for capital returns because if a GPU costs US$30,000 but can generate thousands of dollars of monthly rental revenue at high utilization, the owner can recover a substantial portion of the original investment long before the hardware becomes economically irrelevant, meaning that whatever residual value remains after payback becomes optionality rather than a prerequisite for achieving an acceptable return.

That is a fundamentally different economic setup from buying an asset that loses most of its value before it has generated sufficient cash flow, and this distinction becomes particularly important when evaluating the hyperscalers, because Microsoft $MSFT, Amazon $AMZN, Alphabet $GOOG and Meta $META are all spending extraordinary amounts on AI infrastructure while investors continue to ask whether this capex eventually becomes stranded, yet if older GPUs retain economic usefulness while newer hardware commands premium workloads, then the effective return on invested capital could turn out to be much better than simplistic depreciation assumptions suggest.

This is also why I think investors should stop treating falling token prices as automatically bearish for AI infrastructure, because the relationship between price and revenue is not linear, and a 50% decline in token price does not mean AI revenue necessarily falls 50% if the lower price causes usage to increase fivefold, particularly as AI usage moves away from simple chatbot interactions toward agentic workflows that can consume orders of magnitude more compute because they reason across multiple steps, search, call tools, retry, verify outputs, coordinate with other agents and maintain memory.

A simple chatbot interaction may require only a few thousand tokens, while an autonomous agent performing a complicated workflow could eventually consume hundreds of thousands or millions of tokens as it iterates through a task, which means the move from chatbots to agents can massively increase aggregate token consumption even while the cost per token continues falling, and this is why I think the right metric to watch is not price per token but total useful intelligence consumed.

That is the more important economic variable because enterprises do not ultimately care about tokens in isolation, they care about whether the output creates more value than it costs, and if an AI agent can perform a task for US$2 that previously required US$50 of human labor, then the economics are still extraordinary even if token prices fall by another 50%, while that lower cost actually unlocks new workflows that would previously have been too expensive to automate and expands the addressable market further.

This is why deflation at the unit level can coexist perfectly well with a massive boom at the aggregate level, because semiconductor cost per unit of compute fell for decades while semiconductor demand exploded, storage cost per gigabyte collapsed while global data generation became enormous, telecommunications cost per bit fell while internet traffic multiplied, and AI could follow the same path except that what is becoming cheaper is not simply storage or bandwidth but intelligence itself, which could make this one of the most economically important cost curves of our generation.

The GPU residual-value data are also important from a financing perspective because if lenders, infrastructure investors and private-credit providers become more confident that GPUs retain meaningful residual value rather than becoming near-worthless after a short period, then the cost of capital for AI infrastructure can decline, since a bank financing a GPU cluster cares enormously about what the underlying hardware is worth in a downside scenario, and there is a huge difference between assuming that a US$30,000 GPU will be worth US$5,000 after two years and assuming that it can still be rented profitably or sold for US$15,000.

The second scenario produces a much stronger collateral profile, which can support lower financing spreads, higher advance rates and larger pools of capital, which in turn supports more infrastructure buildout, more model development, more inference capacity and ultimately more AI applications, creating a reinforcing flywheel where stronger residual values make the financing of compute easier while greater availability of compute helps expand the market for intelligence.

This is particularly relevant for neoclouds and GPU infrastructure providers, where much of the bearish thesis depends on the idea that they are funding short-lived assets with long-duration liabilities, because if the assets retain economic value for longer than expected, the asset-liability mismatch becomes less severe, even though utilization, customer concentration, financing costs and power availability still remain critical risks.

The same dynamic strengthens NVIDIA’s strategic position because if customers believe its GPUs retain stronger residual values, then the effective total cost of ownership falls, which means the willingness to pay for new generations can actually remain high even as NVIDIA continues releasing better hardware, and this is where the annual product cadence can become bullish rather than cannibalistic, because the conventional bear argument says a new architecture every year accelerates obsolescence while the alternative view is that the newest architecture takes the highest-value frontier workloads, the previous generation migrates into mainstream inference and older hardware finds lower-cost applications where it still generates cash.

Again, that is not obsolescence. It is a cascading compute stack. And if demand is expanding fast enough, the stack can keep growing even while every individual generation becomes less cutting-edge over time.

That said, I would not take the charts at face value without caveats because the residual values here are inferred from rental rates rather than actual realized secondary-market transactions across a deep and liquid resale market, while rental rates themselves can be influenced by temporary shortages, regional constraints, specific configurations, cluster topology, power costs, networking quality and utilization, which means a GPU sitting idle in the wrong data center is not economically equivalent to the same GPU operating inside a highly utilized, well-networked cluster.

Compute is not merely the chip. It is the chip plus networking, power, cooling, orchestration, software and customers.

That is why utilization remains the key risk, because if AI demand were to disappoint materially and large portions of the installed base became underutilized, rental rates and residual values could fall very quickly, which means the bullish interpretation depends fundamentally on continued growth in economically useful workloads rather than on the hardware itself possessing some magical immunity to depreciation.

But that is also why the current trend matters so much, because if we were already in a severe AI infrastructure bubble with widespread overcapacity, I would expect the oldest hardware to be collapsing first, yet instead H100, H200 and even A100 economics remain resilient while token prices continue declining, which suggests the industry is becoming more efficient without simultaneously destroying demand for compute.

That is an unusually bullish combination because it means cheaper intelligence is not translating into weaker hardware economics but into broader adoption, higher utilization and stronger aggregate consumption, which is exactly what you would expect to see if the demand curve were shifting outward faster than the cost curve were moving downward.

Cheaper intelligence. More applications. More agents. More inference. Higher utilization. Resilient rental rates. Stable or rising residual values. New generations commanding premium pricing. Older generations remaining economically useful. That is not what the end of a capex cycle looks like. It looks much more like an industry moving down the cost curve while the market for its output expands even faster.

This is why I remain bullish on AI stocks, because I do not believe every company associated with AI will win, I do not believe every data center project will earn attractive returns, I do not believe every neocloud will survive, and I certainly do not think valuation can be ignored, but I increasingly think the core infrastructure thesis remains intact because the market continues to underestimate how much compute the next generation of agents, reasoning systems and autonomous workflows could consume.

The bear case assumes rapid technological obsolescence, while the data increasingly point toward persistent economic usefulness.

The bear case assumes falling token prices destroy AI economics, while I think falling token prices expand the market.

The bear case assumes hyperscalers are overbuilding capacity, while I see older GPUs retaining value even as newer hardware is being installed at record speed.

The bear case assumes saturation is approaching, while I think we are moving from a world where humans occasionally chat with models toward one where billions of software agents consume intelligence continuously in the background.

That is a completely different demand curve.

So my takeaway from these charts is relatively simple: AI compute is getting cheaper at the unit level while becoming more valuable at the system level, and that is not a contradiction but exactly what happens when a technology becomes economically transformative, because lower costs unlock more use cases, more use cases create more consumption, more consumption supports higher infrastructure utilization and higher utilization preserves the value of the hardware.

If intelligence follows the same path as bandwidth, storage and compute before it, where unit costs collapse while aggregate consumption explodes, then the biggest risk for investors may not be that the world has already built too much AI infrastructure.

It may be that we are still underestimating how much intelligence the world will eventually consume.

Long AI. https://x.com/rickyho_1989/status/2105573677114999294