That's from April, but in short there's a huge order backlog that they're still working through apparently, despite producing half a million units per month. With the lockdown gap in production and the Pi 4s getting increasingly integrated into various 3rd party products I suppose that's no surprise.
It's quite apparent that there's little demand for the Pico, since it's always in stock.
> It's quite apparent that there's little demand for the Pico, since it's always in stock.
Without additional context (which perhaps you have and used subconsciously) that's not evidence that there's less demand - just evidence that the ratio of demand to stock is lowest. It could be that it has 2x the demand but they prioritised it and produced 3x as much stock, or it could be that a specific component makes one product easier/less delayed than the other to make (in which case equal demand could still lead to only one being regularly in stock).
If for example 100 people a year want to buy a product $A, and 10 a year want to buy product $B, and the company manufactures 200 $A's a year but only 5 $B's, then $B will be out of stock more despite being far less popular.
Or course this partly relates to how well a company predicts future demand when deciding how much of each product to create. But in many cases (though I would guess not when it comes to The Raspberry Pi Foundation) marketing therefore also becomes a factor - in that companies may see value in either creating slightly less than they expect there to be demand for, or artificially limiting / lying about stock levels, in order to get people thinking "wow it's out of stock so it must be popular!"
Well it's either an overestimation in production or an underestimation in demand. Or likely both to some extent in this case.
I bought a few of them a while back and have only recently managed to integrate one of them into a really basic project. They tried to make some kind of middle ground between an ESP and an Arduino, while providing an incredibly buggy MicroPython build and no Arudino IDE integration. Some of that's been corrected, but it still remains this all rounder thing that's never the best choice for the application.
> Well it's either an overestimation in production or an underestimation in demand.
Or they correctly estimated, planned not to go out of stock
and were able to succeed. Jumping from "it never shows as out of stock" to "therefore they must have badly estimated one or both of supply or demand" is even stranger a leap of thinking than the initial misconception of thinking that not going out of stock proves low demand.
In both this comment and the previous one, you're guessing at a possible explanation while writing as if you know it to be the correct explanation.
(Sorry for coming across all critical, hopefully learning what can and can't be construed from a product being in stock is worth my negativity!)
You can probably just ignore them as a concern at this point. Nobody is going to want to use them in new products. They might keep going as a educational novelty but the magic is gone as far as using them in commercial/industrial products.
Because you can't purchase them at scale. And haven't been able to for a year. If I'm making a product I'm not going to pick something I can't get. And I'm not going to switch back once I've found something else.
Eben says to embrace the Pico and buy the rPi400 because it does not compete with orders from industrial customers (that spent bazillions on testing/certification of their rPi4/rPi3-derived products and therefore receive some priority above the poor huddled masses yearning to breath free air). If you aren't down with Pico yet, I recommend googling Limor Fried and searching her company's site.
As far as Intel goes, they've been on oxygen for decades with the technological advancements they appropriated from DEC, while at the same time selling off the DEC-designed StrongARM technology and exploring new ways to generate heat and waste power. At this point, they have a formal relationship with TSMC and a government mandate to turn the Rust Belt into the Silicon Belt, so don't count them out (unless you are 75 or something, because it will be 10-15 years for all of that to happen), but to guys like me (and I would imagine most people on a site like this), they're about as relevant now as IBM (/s) In the meantime, we need to convince Apple to sell its consumer chipsets, maybe with an incentive from USA (either money or an agreement not to prosecute them for investing so much training and capital in China that they feel comfortable announcing their plan on TV yesterday to murder as many people as necessary to return Taiwan to 1895 legal structure).
That's the problem. He's not even wrong. That whole comment is just incoherent rambling.
What does the Pico have to do with RPi shortages? Why would anyone buy a rPi400 as a RPi replacement? WTF does 30 year old DEC tech have to do with today's Intel? Why force Apple to sell consumer chips and how does he think that's gonna work out in the long run when all the software required to use them is single-source?
Still no sign of stability in supply.