I don't understand what you mean. These chips are available in bulk from the manufacturer and from all major distributors worldwide. Why would you want to go through China? These chips aren't even manufactured there. The fab is in Thailand.
If you mean a PCBA shop, Mouser or DigiKey can ship your order directly to your favorite PCBA place.
The wisdom of the crowd can be substantial in supply chain. That is to say, for non-trivial projects, selecting any kind of component which may not be immediately available in arbitrary quantities represents a huge project risk. The numbers currently available in China show that the chips' position in the market is too early days / small volume to be considered a viable development target given the fact that the world center of electronics production is yet to see enough aggregate demand to stock more than a handful.
Now if you want to develop against a chip on a 10 year outlook, go for it. But if you want to get work done this week, or this month, or even this year, I'd argue that the distribution is too premature to make that a smart decision.
I don't follow. You can source the chips, in large quantities, right now, and have them delivered to any non-embargoed country in a matter of days.
That they're in limited stock at a particular outlet in China means very little, no? And conversely, that STM32 chips happened to be in stock in 2018 meant very little in the following four years.
If you want the chips right now, you can have them. If you want long-term availability, either buy ahead of the time or sign a contract with the manufacturer (and hope it doesn't fall apart due to a geopolitical crisis or whatnot).
I think perhaps you miss the difference between ordering from the vendor and having stock in the marketplace. The first is just a promise, which everyone learned had little value during chipaggedon. It turns out vendors have all sorts of interests which affect which customers they prioritize stock for in the event that a shortage is anticipated, and most of us aren't on the list. Stock in the distribution marketplace means that not only does the quantity really exist, but it has been in demand long enough that the distribution chain has decided to buffer additional volume. This means there's volume moving and you can to a much greater extent than a vendor promise actually count on it being available and the price representative (rather than promotional, temporary, or highly subject to fluctuation). In general, you also get your stock quicker. I hope that clarifies.
China is full of counterfeits though. I don't actually expect those chips in China to be legitimate. Its public information that these chips are made in Taiwan or other locations, so Chinese supply is pretty equestionable.
I can order 3000+ from Microchip's website right now for any of the AVR Dx series. And Microchip has decades-long reputation of a solid supply chain, even in the 2020 COVID19 period when STM32 got issues.
--------
The older chips are fully obsolete compared to these modern designs. The newer AVR DD is cheaper, faster, loses the Voltage-slowdown issue (1.8V is only qualified for like 4MHz on ATMega328. 1.8V on AVR DD can go full 24MHz), etc. etc.
I don't see any reason why anyone would ever choose the decades-old ATMega328 when AVR DD is right here, cheaper and better in every conceivable way. One exception: if you really need the ATMega328 pinout _exactly_ because you can't change the PCB or update the electronics, I can understand legacy issues. But new projects should pretty much switch to AVR Dx or AVR Ex (depending on your project's needs, since the different chips specialize in different purposes). Ditto for legacy software because ATMega328 peripherals are older. (Modern TimerA and TimerB are much better than ATMega328's legacy Timer0 and Timer1)
China-bashing is naive - most of the world's electronics are produced there for a reason. Here's how it works. Projects aren't open shut, and they don't exist in a vacuum. The cost of targeting a new hardware platform is nontrivial. All hardware costs are trivial vs. overall project costs (time, people, shipping, testing, production, packaging, distribution, etc.). Even then, outside of launch-time near-shore marketing-land, the chips currently cost double the old ones. No prior software will run without modification. Even the programming interface is new, which means retooling production code, jigs and fixtures in addition to firmware and schematics. Nowhere did I say "the hardware is worse", rather I said "the hardware is a premature choice for commercial projects with rapid delivery requirements at this time". Understand the difference. I stand by that assessment.
> China-bashing is naive - most of the world's electronics are produced there for a reason.
I didn't say that China didn't have most of the world's production. What I said is that a large number of Chinese chips are counterfeit. So when we're talking about discussion points like:
> Even then, outside of launch-time near-shore marketing-land, the chips currently cost double the old ones.
Well, have you done the QA to assure that these "old chip lots" are actually legitimate AVR ATMega328pb or are they some king of counterfeit chip? I know the STM32 chips in a lot of Chinese shops are just clever replicas, and that's key to their lower prices.
I don't fully trust prices, especially of old stock in China.
---------
China has enormous potential in terms of electronics manufacturing. But the supply chain problem / counterfeit problem certainly exists. No matter how clever their replicas get, there are minor concerns about power-delivery differences or minor differences to the ADC (or whatever). Maybe the counterfeit chips are good enough for your projects, or maybe not. But its still a concern that floats in the back of my mind, especially if the prices are much cheaper than the legitimate sources.
Like: maybe the chips don't sleep quite as low power as a legitimate chip, or the ADC is slightly less linear than a legitimate chip. Etc. etc. Minor differences that with good testing you could actually work with the counterfeit and get a usable product, but a risk nonetheless if you have an old design that depends on the specifications of the original.
> China has enormous potential in terms of electronics manufacturing.
this is like saying 'south africa has enormous potential in terms of diamond mining' or 'the us has enormous potential in terms of mass shootings'. china had enormous potential in terms of electronic manufacturing 30 years ago. today, electronics manufacturing outside of china is a footnote. an increasing number of manufacturers don't bother to produce non-chinese-language datasheets. in this context, if a chip's availability in china is sketchy, that's reason to question its viability
the hackaday article does start with talking about stm32 counterfeits but is mostly about legitimate clones, most of which are improvements over st's chips
I dunno why you're so focused on trying to make China look better. It's well acknowledged by pretty much everyone that supply chains in China have a counterfeit problem.
There are pretty much two solutions. Either you more closely handle supply chain issues yourself by buying from the OEM or reputable sources (which increases costs). Or you accept the counterfeits as good enough. Based on this discussion, you've gone with the second option which is fine. But it's something people should know about your argument.
You CANNOT compare prices in the way you've been doing in your argument. That's all I'm saying.
When we compare OEM ATMega328pb vs AVR64DD32, the OEM prices for AVR DD is cheaper. That's the ground truth.
The methodology you've chosen as the basis of your argument is not accounting for the huge, well known risks of counterfeit ATMega328.
you have me confused with someone else; i haven't compared any prices in this thread
i agree that there's a counterfeit problem, and it's worse in china, but that's irrelevant to my argument
i don't think it makes sense to say that i'm trying to make china look better. you're painting a pants-on-head absurd picture of china's role in electronics manufacturing, apparently taken from some kind of incredible propaganda. i'm just giving you the basic facts that everyone agrees on but you don't know yet
The issue here is ATmega328 pricing vs AVR DD pricing. Which is the crux of the discussion.
The pricing estimates discussed earlier have to be factored against the well known, widespread counterfeits. Especially because ATMega328 is a well known counterfeited chip, and the Chinese market well known for being full of not only counterfeits in general, but even counterfeits of this specific chip under discussion.
You cannot compare chip prices like the poster did earlier. And all this Chinese vs whatever discussion you're trying to distract me with aside doesn't change the core fact.
i think the pricing is a mostly irrelevant detail, actually, which is why i didn't mention it. what's relevant is availability
you're right that availability of counterfeits isn't real availability, but i think you're vastly exaggerating the seriousness of that problem. a more likely reason for chinese distributors lowering their prices for the atmega328 is that it's not appealing for new designs, so they need to offload their leftover inventory
Very few hobby designs that use more capable chips actually need more capable chips, and the added complexity - hardware and software - often bites hard.
That said, wifi is a major selling point, and you need a beefy 32-bit computer to run the protocol. This is the brilliance of ESP32: you can actually use it just as a wifi dongle for another microcontroller, but since it needs so much computing power, might as well give you some RAM and CPU to do your thing directly on the wifi chipset...
The ESP32 is very overpowered for random home projects. But the cost is low, and they can be used with the Arduino ecosystem which is very beginner friendly.
Often the ESP32 is far cheaper than whatever sensor you're making it monitor. WiFi isn't the perfect technology to upload sensor data in the home, but anything proper Zigbee is more expensive and more complex to use. So the ESP32 ends up in all kinds of projects.
The ESP32 is not overpowered! Either for the cost or the power consumed. This idea that if an MCU is too easy to program for or has "resource left over" after completing the task that it is ill suited for the task has to go.
I haven't tried the ESP32-H2 or ESP32-C6 yet, but they have 802.15.4 radios so they can be used with Zigbee or Threads (plus WiFi 6 on the C6). I definitely am eager to play around with one, most of my current projects have been on the ESP32-C3.
What's interesting to me is why this story leaked. I bet it's just Morgan Stanley (the owner of E-Trade) trying to control they narrative. If they just cut him off, he gets to tell his story, MS can't officially discuss the particulars, and the company ends up getting a lot of hate mail.
But if they leak the official justification beforehand, then Roaring Kitty will have a harder time spinning it.
New construction is rapidly getting more expensive in many of the markets where it's needed the most. While it'd be cool to point to a single factor, it's a combination of several things. People are expecting more (more square footage, higher-end finish), the government is demanding more (environmental and energy regulations), labor is getting costlier, and yes - there's NIMBYism manifesting through zoning laws and other "conservation" rules.
Not entirely, but either way, land value is tied to all of that. If the city can't grow due to zoning and conservation rules, and if you can't subdivide existing parcels, you basically have more and more money chasing the same lots.
By "not entirely", I mean that while land value often dominates, lowering it doesn't solve the problem, because even if you somehow acquired a city lot for $50k in the Bay Area, you wouldn't be able to build a house on that lot for $100k. So, solving one problem doesn't make it affordable.
It is a lossy compressed index. It has an approximate knowledge of law, and that approximation can be pretty good - but it doesn't know when it's outputting plausible but made-up claims. As with GitHub Copilot, it's probably going to be a mixed bag until we can overcome that, because spotting subtle but grave errors can be harder than writing something from scratch.
There's already a fair number of stories of LLMs used by an attorney messing up court filings - e.g., inventing fake case law.
I am not suggesting that the generative aspects would be useful in drafting motions and such. I am suggesting that their tendency towards false results is harmless if you just use them as a complex index. For example, you could ask it to list appellate cases where one party argued such-and-such and prevailed. Then you would go read the cases.
This, along with several other "meta" objections, is a significant portion of the discussion in the paper.
They basically say two things. First, although the measurement is repeatable at face value, there are several factors that make it less impressive than assumed, and the model performs fairly poorly compared to likely prospective lawyers. Second, there is a number of reasons why the percentile on the test doesn't measure lawyering skills.
One of the other interesting points they bring up is that there is no incentive for humans to seek scores much above passing on the test, because your career outlook doesn't depend on it in any way. This is different from many other placement exams.
They by and large don't. Most creators on YouTube aren't even eligible to monetize, especially if they're making niche educational content.
Google just pockets all the revenue if you have fewer than 4,000 "public watch hours". If you put together an a concise 4-minute DIY video, an average view time will probably hover around 1 minute, and you will need 240k views to qualify. Most videos on YouTube get under 1k views.
There is a negligibly small percentage of content creators / content farmers who live in a symbiotic relationship with the platform and actually make good money, but let's not pretend you're patronizing the creators when you're paying for YouTube Premium.
> Google just pockets all the revenue if you have fewer than 4,000 "public watch hours". If you put together an a concise 4-minute DIY video, an average view time will probably hover around 1 minute, and you will need 240k views to qualify. Most videos on YouTube get under 1k views.
With less than 4k public watch hours your channel has barely made any money. It is accumulated for your videos for the entire channel, that isn't per video.
And no, youtube isn't making money on those videos, it costs them more to store etc and most of them aren't getting any ads anyway. Instead you get free video upload on youtube, normally you pay for that.
Youtube makes their money from the big creators that draws in a lot of views per video, there efficiency of scale lets them get over the profitability hump.
YouTube runs ads on all my videos, including the ones with fewer than 200 views. I'm not sure what you're getting at otherwise. The point I was making is that paying for YouTube does absolutely nothing to support the vast majority of content creators on the platform. All you're doing is paying Google, and a tiny slice of that supports a handful of celebrity content creators like Mr. Beast. Maybe it's a bargain that makes sense to some, but the parent was deriding the grandparent for not willing to pay.
How many times in history have we regretted over-building this kind of infrastructure? We have no shortage of good ways to use electric energy. If AI turns out to be a fad, others will benefit.
Pollution is a concern, but I think don't think that anti-growth or anti-consumption sentiments are useful. All this money pouring in is an opportunity to invest in clean energy.
