Intel is quitting on its adorable, powerful, and upgradable mini NUC computers

tuxprint@lemmy.tuxprint.com to Selfhosted@lemmy.world – 755 points –
Intel is quitting on its adorable, powerful, and upgradable mini NUC computers
theverge.com

Damn, this is a sad day for the homelab.

The article says Intel is working with partners to "continue NUC innovation and growth", so we will see what that manifests as.

235

Jesus Christ. Why does it feel like tech industry is just getting shittier and more expensive, while all the cool consumer options are being axed. Intel Nucs were a relatively cheap way to get a cute little desktop machine or a home server. I am sad that they're going away. I guess there's always Minisforum, but still...

Because infinite growth of profits on a finite planet.

Yeah this part bothers me. To these companies a solid profit stream is not viable. It has to be iPhone level growth year after year or they think it’s failing and axe it. It’s quite annoying. Eventually you will hit a plateau. That just means it’s a mature market, not failing. Grrrr…

You see the same shit on streaming services. "Oh this show has been out for two days and hasn't reached Game of Thrones level of popularity already? Let's remove it from existence forever."

Just throwing shit at a wall to see what sticks.

Capitalism is unsustainable. We're seeing what happens in late capitalism. The belts tighten, the workers get left in the dust, the products consumers actually want get the axe.

We don't even have capitalism yet, what late stage are you talking about?

What are you defining capitalism as, and what word would you use to describe our current system?

You can read about capitalism in Wikipedia.

Most countries today move towards economical fascism, where governments exercise control over private property but do not nationalize it. Lobbying, donor interest protection, cronyism, rise of oligarchy - you can see it in many countries. And then inevitable radicalisation of the public and scapegoating everything else as the core issue. Capitalism, migrants, ecology - everything is a problem but the government.

Contemporary capitalist societies developed in the West from 1950 to the present and this type of system continues to expand throughout different regions of the world—relevant examples started in the United States after the 1950s

This Wikipedia article says that the US is a capitalist system.

Lobbying, donor interest protection, cronyism, rise of oligarchy

Where are these things listed in the article as being incompatible with capitalism, and their presence meaning it's some other system?

I guess that really depends on where you live. I can only speak on behalf of the US.

Where do you have capitalism in US? US is probably one of the most anti capitalist countries in the world right now.

That's not really true though and it's anecdotal. The anti-capitalist mindset might be growing due to awareness and people suffering at the hands of capitalism (continued layoffs, increased cost of groceries and rent, union busting, worker exploitation), but that's because of the ever-tightening squeeze of late capitalism. When you have a structure that requires infinite growth to exist, in a world with finite resources, you end up with the current state of the US.

I think it would be more accurate to say that the anti-capitalist mindset among the working class has definitely grown in the US, but at its core, the US is pro-capitalist.

Where's US pro capitalist? It's one of a few countries with legal corruption called lobbying, which helps big corps to shield themselves from competition. US today has a plethora of laws and regulations which create and sustain monopolies. US has whole industries created by lawmakers and completely stonewalled from anyone entering them. Capitalism my ass...

Also capitalism doesn't require infinite growth. I don't know where you people are getting that lunacy from.

Infinite growth is not a core part of capitalism. You're right there. But do you know what is? Pursuit of profit. And do you know what leaves dollar signs in companies eyes? Pursuing infinite growth. Infinite growth results in infinite capital, in theory. Such growth is not a requirement of capitalism, but it is the logical conclusion when you throw sustainability out of the window. And boy, do we know that corps love doing that!

You summarized the infinite growth aspect better than I did. This is exactly what I was referring to. Thank you!

I’m not sure where they are getting their info or how the US isn’t a capitalist hellscape. The US in its current state is exactly what happens when capitalism reaches a boiling point because all of the people driving it pursue infinite growth with zero accountability.

1 more...
1 more...

Laws and regulations that allow capitalists to continue their pursuit of infinite growth. One of the definitions of capitalism is simply:

The concentration or massing of capital in the hands of a few

This is like a 1:1 definition of what we have in the US today, and our government enables, protects, and benefits from it. It’s “late” capitalism because it’s grown into a completely unsustainable system.

Late capitalism is the acceleration of growth and concentration of wealth in the hands of a few, with various crises being the result (layoffs, inflated prices, union busting, cuts in safety—e.g train derailments, etc).

1 more...
1 more...
2 more...
2 more...

Bruh, I don't believe in late stage capitalism either but we are definitely living in capitalist economies in most of the world.

Capitalism isn't just laissez-faire, completely free market type stuff. It's a spectrum.

2 more...
2 more...

While companies today are certainly overzealous in their drive for growth, it is a myth that infinite economic growth is impossible. It is not only possible but necessary: https://medium.com/@oliverwaters\_76079/the-strange-necessity-of-infinite-economic-growth-ebc2e505cdf1

Only two kinds of people believe in infinite growth; economists and psychopaths.

Only two kinds of people believe in infinite growth; economists and psychopaths.

But you repeat yourself :)

When the money supply grows infinitely then everything priced in it has to grow infinitely

Maybe capitalists instead of economists? 😂

Capitalists are behind the most prelavent economic school (neoliberalism) today—just look at the history of the "Chicago school". I doubt the capitalists themselves believe that BS, but it's profitable for them to make the rest of the world to believe it.

I highly recommend evonomics.com, some rally good essays on there about the cult-like economic beliefs of today. Written by economists who've seen through the BS.

So we're going with an ad hominem attack instead of engaging in good faith?

Pretending like capitalism is this new concept that needs to be fully explored and debated before we understand that it's bad is a pretty bad faith framing of the issue. Infinite economic growth is literally impossible because Earth has finite resources and there is a finite number of humans. There is no necessity or imperative behind infinite economic growth other than to make the ruling class richer at everyone else's expense.

I would say just generalizing capitalism as 'bad' is also not in good faith. It is not without issues, and letting it be completely unrestrained would probably be disastrous. But no other economic system has lifted more people out of abject poverty or driven technological innovation as hard. There are benefits.

There's the old "more people were in poverty before capitalism" argument.

Did capitalism bring people out of poverty? Or did access to education, healthcare, social safety nets, and proper food bring people out of poverty? Where I live, capitalism is what's driving people into tent cities.

How does one person controlling the capital in an area, help other people if they're gatekeeping the economic prosperity from by forcing them to perform labour, at a disproportionately low rate of recompense, to help them (the capital owner) increase their net worth? Don't even say trickle down economics or I'll deck you.

This has nothing to do with capitalism. And my source explains how infinite growth is possible. Consuming the resources of a finite system is not the only factor that goes into economic growth.

That article is utterly unconvincing. It just handwaves the finite nature of our material reality with a very weak appeal to "infinite" human creativity. And then the conclusion is that infinite growth is necessary because there's no way to change the status quo of wealth hoarding. It's just apologism for the very worst aspects of capitalism without a single iota of serious thought.

I don't think there is any hand waving. Consuming a resource is not the only factor that goes into economic growth. Can you address that point specifically?

No I won't because it's irrelevant if it is the only factor or not. It's the limiting factor. Please don't engage in red herrings.

You won't because you don't understand what you're talking about.

Seems that you're the one doing the hand waving.

In a finite system, infinite anything is an impossibility.

This sounds true on its face, but if you had read my source, you would see how that argument is refuted. The problem is that you are assuming the resources of the system must be used up for growth, but that is not true.

If the last 300 years are anything to go by, we clearly do need resources if we are to maintain growth at a rate high enough to barely keep pace with the needs of the market. Coal, steal, oil, cement, water, food, etc.

The reality is, we can't replace the current demand on renewable energy sources alone. You seem to believe the system can pivot and adapt fast enough to fix itself. While I'm of the mindset the system will follow the path of least resistance even if that means killing itself.

People used to say this about energy as well, yet in the past 5-10 years, I’ve read several articles demonstrating that we appear to have decoupled energy growth from economic growth

We need resources, yes, of course! However, consuming those resources is not the only way to generate growth. My linked post lays it out fairly clearly, I think.

Whether or not I think we, currently, can pivot quickly enough to a model that doesn't kill us all, I don't know. I think it's possible, but like you, I'm also pessimistic about it happening. In any case, that is not at all what I was suggesting. My only point was that infinite economic growth is feasible in general.

Do you have the text of that article you linked? I'll confess I hit a login wall nearly immediately into the discussion and I never log in to any of that stuff. But I am curious to read more.

Apologies! I didn't realize it was walled.


The Strange Necessity of Infinite Economic Growth
Nov 13, 2018

Despite intuitive claims to the contrary, infinite economic growth is not just possible — it’s essential for human flourishing.

In an article for Foreign Policy, the anthropologist Jason Hickel has recently called for an end to global economic growth, as the only way to avoid total ecological destruction. He argues that so called ‘green growth’ — the idea that we can decouple intensive resource use from economic growth — is a myth. George Monbiot has shared a similar concern, echoing previous warnings by influential writers Naomi Klein and Paul Ehrlich.

I’m going to focus on Hickel’s work, since it contains several related misconceptions that are nonetheless highly intuitive. In his article and across his other writings, Hickel makes three key basic claims:

  1. There are hard physical limits to how efficiently we can use resources. It follows that infinite economic growth is impossible on our physically finite planet.

  2. Empirical studies are showing that we cannot ‘decouple’ resource use from economic growth, meaning we will run out of resources if economic growth continues.

  3. We can maintain high living standards even with zero global economic growth.

Let’s look at each claim in turn.

  1. Is infinite growth impossible?

In his article, Hickel claims that there are strict physical limits to how efficiently we can use resources, which means pursuing economic growth in the form of rising GDP will eventually exhaust all resources on the planet. Hickel relies on conclusions from three empirical studies that purport to demonstrate this, which I will unpack in my response to his second claim. But here I just want to clarify what exactly economic growth is. By doing so we can understand why economic growth does not in principle depend on unsustainable resource depletion, before examining if such future growth is an empirical plausibility.

‘Economic growth’ refers to the increase in, and improvement of, goods and services that are useful to human beings. This is often conflated with what is called ‘extractivism’: which is the obviously finite process of extracting material resources from the Earth to sell on the market. The former is a necessary process for the continuous flourishing of humanity, the latter erroneously conceives of nature as, in Naomi Klein’s words, a ‘bottomless vending machine’.

Since the Earth consists only of a finite amount of matter and energy, we clearly cannot indefinitely consume its contents without eventually drowning in waste or starving to death. But even this obvious insight contains a misconception. We never actually ‘consume’ matter and energy in the first place. We are forever bound by the first law of thermodynamics, which dictates that matter and energy are never created or destroyed, only transformed.

This brings us back to economic growth. Fundamentally, it is the process of human beings transforming matter and energy into more valuable forms via the development of theoretical and practical knowledge. It is ‘economic value’ which increases exponentially, measured (very)roughly by GDP. The crucial point is that while the extraction of finite physical resources cannot increase indefinitely on a finite planet, economic growth actually can.

Steven Pinker describes the infinite potential of economic growth eloquently in his book Enlightenment Now:

‘…it’s a fallacy to think that people “need resources” in the first place. They need ways of growing food, moving around, lighting their homes, displaying information, and other sources of well-being. They satisfy these needs with ideas: with recipes, formulas, techniques, blueprints, and algorithms for manipulating the physical world to give them what they want. The human mind, with its recursive combinatorial power, can explore an infinite space of ideas, and is not limited by the quantity of any particular kind of stuff in the ground.’

Of course, it’s abundantly clear from history that most civilisations have failed miserably to generate and implement the requisite knowledge to provide for the needs and desires of their citizens. This brings us to Hickel’s empirical claim: that our own current trajectory is leading us to disaster.

  1. Prediction or prophesy?

One might reasonably suspect that the promise of infinite economic growth would only be plausible for a sufficiently advanced civilisation — particularly one that has managed to wean itself off fossil fuels. We should therefore ask if decoupling economic growth from resource use is likely to be feasible given our relatively primitive level of technological development.

On Hickel’s account, things are not looking good. He cites several studies which suggest there is no way to avoid running out of resources if we continue with our current rate of economic growth.

The main problem with these studies is that they tend to presume a certain fixed ‘biocapacity’ of the Earth. This concept and the corresponding notion of humanity’s ‘ecological footprint’ have many problems, the most fundamental of which is that they depend arbitrarily on our current level of scientific and technological development. This is because the maximum level of human consumption that our planet can support is not fixed by some natural law — it depends entirely on the sophistication of our technology to convert raw materials efficiently into life-supporting forms.

The very same lump of matter and energy has vastly different properties to us humans depending on the level of and quality of our knowledge. Major scientific breakthroughs therefore allow us to do dramatically more with less. One kilogram of uranium contains two to three million times more energy than the same amount of coal or oil, but this fact went completely unnoticed by everyone up until fundamental breakthroughs in physics in the 20th century.

Similarly, the invention of desalination techniques unlocked the effectively boundless supply of seawater. While the rest of the Middle East suffers water shortages, Israel has a surplus — thanks to advances that have reduced the costs of desalination by two-thirds since 1990.

One study mentioned by Hickel calculated that we’ll be using 95 billion metric tons of resources globally in 2050. But this figure is meaningless without a corresponding estimate of what fraction that is of the Earth’s total resources. And we can’t know what Earth’s total resources are because we cannot predict future fundamental technological advances.

At one point Hickel declares that the sustainable level of global resource use is about 50 billion metric tons a year, without citing any source or justification. The authors of the 2012 study he references cited this same figure as a possible upper limit on global resource extraction, being roughly the level of extraction there was in 1992. But why choose this year? Because that was the year of the first United Nations Conference on Environment and Development. In other words, it was picked out of a hat.

Even where these studies make allowances for future improvements in how efficiently we use resources, they only allow for greater efficiency in the use of resources we currently know about. If the future is anything like the past, we will discover whole new ways to build and power things, and replace whole classes of raw materials we currently depend upon.

Then there’s the fact that these studies’ predicted improvements in resource efficiency are themselves woefully pessimistic. The most optimistic prediction in the studies Hickel mentions is that resource efficiency will double by 2050. But almost any consumer product we use today requires far fewer resources to build and run than their equivalent 32 years ago. While it’s a somewhat trite example, take the iPhone. This one product has replaced landline push-button phones, pagers, cameras and camcorders, calendars, alarm clocks, audio-recorders, flashlights, maps, GPS, credit cards, and more. We should only expect this kind of dematerialization to accelerate over the next 32 years, given the rapid advances being made in nanotechnology and materials science.

This fundamental fallacy driving the pessimism of these studies was eloquently captured by David Deutsh in his book The Beginning of Infinity. In the chapter ‘Unsustainable’, Deutsch reflects on Paul Erlich warning his high school class in 1971 of the impending global ecological collapse, a tragedy which never came to pass:

‘Ehrlich thought that he was investigating a planet’s physical resources and predicting their rate of decline. In fact he was prophesying the content of future knowledge. And, by envisaging a future in which only the best knowledge of 1971 was deployed, he was implicitly assuming that only a small and rapidly dwindling set of problems would ever be solved again.’

It’s perfectly acceptable and prudent to make predictions about future resource scarcity, so long as you remember that such predictions are based on our current level of scientific knowledge. Such analyses play a crucial role in motivating us to invest more in advancing our knowledge, but Hickel mistakenly interprets their findings as fatalistic signs that we must cease economic growth altogether. And this, as I’ll show below, is simply not an option.

  1. A world with no growth?

How might the world look if Hickel has his way, and global economic growth is brought to a halt? He thinks we’ll be just fine, claiming that ‘our planet provides more than enough for all of us; the problem is that its resources are not equally distributed.’

It’s a claim often made: that the world is plenty rich enough for everyone, if only those wealthy fat-cats had the empathy to share their billions with the rest of humanity. Recently the distinguished British physicist Sir Martin Rees similarly exclaimed:

‘We have a billion people in the world in abject poverty, which could be alleviated by the wealth of the thousand richest people on the planet. That we allow that to continue surely says something significant about how much — or little — moral progress we’ve made since medieval times.’

In his critique of Hickel’s argument, Noah Smith dismissed the panacea of global wealth redistribution as a political and logistical impossibility. In his response, Hickel reassured Smith that we can indeed distribute global wealth with policies like introducing a global minimum wage, democratizing global economic governance institutions, implementing subsidies and tariffs in developing countries, cancelling national debts, or rolling out a global basic income.

Hickel appreciates the political difficulties of implementing these policies, but the more important question is whether such measures can actually achieve their intended aims in principle.

The notion that redistributing world income is the solution to global poverty is both intuitively plausible and spectacularly false. It makes sense to us on an emotional level when we are confronted with images of billionaires swanning around on mega-yachts on the Mediterranean, juxtaposed with children starving in a war-induced famine in Yemen. But we need to put this visceral imagery aside for a moment and look at some stark numerical realities.

How much wealth is there in the whole world? Credit Suisse puts it currently at around $317 trillion USD. How much would each person in the world have if that were divided equally among all 7.6 billion of us? $41,710.

To be clear, this is the amount of cash each human being would have if everything of economic value (houses, cars, furniture, spectacles, intellectual property, everything) was traded to aliens for cash in return. (Try to ignore the logical and inflationary issues of aliens forging vast amounts of US currency — it’s just a thought experiment).

As is comically depicted by Tim Urban on his blog Wait But Why, we would then all just be huddled naked together next to our equal piles of cash, which we would then presumably try to eat or burn for warmth. But suppose we could actually buy goods and services back from the aliens, at their current day prices. What could each of us afford? Perhaps a new car and some tanks of fuel, enough basic food supplies for a few years, a decent tent, warm clothes, and other survival equipment. That’s about it.

No houses. No electricity. No internet. No schools, no hospitals, no government.

In other words, we would all be living in what is generally considered today to be dire poverty. And this is presuming that all wealth is somehow convertible into cash, which it isn’t. Thus the aliens. In reality the vast majority of economic wealth does not exist in fungible, divisible, monetary form.

Take the approximately one trillion dollars of value that the company Apple represents. You can’t divide up ownership of such a company equally among the world population and expect it to still exist in any recognisable form. And a company is one of the more abstract kinds of things one can own. Forget about divvying up something more physically tangible like a toy-making factory or a uranium mine.

Economic wealth is something that needs to be created, by people, wherever they are situated. People in impoverished countries need better knowledge, skills, and institutions to create more economic wealth, if they are going to reach the living standards enjoyed in OECD countries. Those in advanced economies need to keep increasing their wealth in order to be able to tackle the increasingly complex challenges they’ll be facing in the future.

This doesn’t mean that we shouldn’t give generously to people in need. But we need to keep in mind the end goal: assisting them to achieve sustainable economic growth. Indeed merely redistributing wealth can end up crowding out the development of sustainable local economies, as the philosopher Larry Temkin has discussed.

Going forward, there is no alternative to huge increases in global economic growth. Contrary to our strong intuitions, infinite economic growth is both logically and practically possible on our finite planet, and utterly necessary if we want to provide every person on Earth with a decent lifestyle. We should not be taken in by prophesies to the contrary, which have come and gone since the beginning of human progress itself.

Who is Oliver Waters and why should I listen to them regarding economic theory? I read the post, and it reads more like a philosophical thought experiment than any applicable economics theory.

While I don’t believe someone needs a higher education degree to speak on complex topics, I’m not going to take a Medium blog post from someone who lists no demonstrable experience in theoretical or practical economics as a central source for discussions, sorry.

It's not philosophical at all; it's rather straightforward in its arguments, IMO. Not sure why nobody wants to discuss the points directly, and they are cogent points regardless of whose keyboard they originated. If the points made are incorrect, they should be relatively easy to refute.

2 more...

Relatively cheap? Huh? At $500-$1000 they were exactly the opposite of a relatively cheap desktop machine.

There was a great resale market for them. I got an i7 8th gen for about $200-300 new when the 10th gen came out. It was clearly never used overstock that a reseller picked up cheap. Its a champ of a machine, still going strong.

They also made cheap celeron models that sold in the $100-200 range that were 5x as powerful as the raspi that would normally fill the niche.

Yeah the celeron and pentium models are amazing low power machines to run Home Assistant on. Mine is running half a dozen other docker addons including frigate to do ai object detection (offloading most of the heavy lifting to a Google coral chip plugged into usb)

Being the default industry standard meant drivers were never a hassle

IKR? For what they wanted I could get a faster full size machine with better expandability. I get the value in a small box, but unless you had some commercial application or wanted some special architectural aesthetic in your home that required that size, it was a waste of money.

Intel NUCs were very good machines but honestly they were completely overpriced compared to Chuwi/Minisforum/etc.

My guess is they were just not enough sales, that's all.

What’s the Chuwi Equivalent to a Nuc? Not being snarky, im genuinely looking for a small server.

Yeah, mini-computers are one thing, but the NUCs were more than that. Having a PCI-E card that you can slot into your computer to literally run a PC inside your PC is super unique and not something anyone else offers.

Sad to see them drop this. I can understand that it's not an in-demand market segment, but it was cool none-the-less

Having a PCI-E card that you can slot into your computer to literally run a PC inside your PC is super unique and not something anyone else offers

My hope has been from the start that that product line would lead to some compute module-style clustering motherboards for really clean & compact x86 clusters. It would especially make sense for dedicated server/VPS providers which already rely on similar dense blade systems from Supermicro.

Imagine a box that would take 3 of them, give each a PCIe slot and an NVMe slot, and an then give you 3 power buttons, 3 sets of IO and maybe an integrated network switch so you only need 1 Ethernet cable to connect the swarm to your network. That would be useful not just for clustering in homelabs and SOHO but also for offices and such if they want to reduce the physical footprint of their PCs while maintaining pretty good serviceability for "go swap this PC out" scenerios

Oh lort. You just gave me flashbacks. One of my kids bought one of those $200 Chuwi laptops and it would barf all over itself about once a month, so badly it would require a reinstall.

Chip shortage. Since COVID, chip companies have been having a really hard time getting properly restocked. This impacts all electronics industries. Cars, computers, even Apple had to redesign some of their products to accommodate the shortages, so has many other companies big and small. The Raspberry Pi prices have soared. So products that take a chip away from a more mainstream or lucrative market are being axed.

There’s a chip shortage. Most people just use web based apps, so stay on their phones / cheap laptops Enthusiasts usually just build their own machines. Everything is more expensive. The list goes on

Probably because people aren't spending their money on it.

For most people, why get a nuc when you can get a laptop? Nuc fills a niche.

Or why get a nuc when you can get a decommissioned Enterprise sff PC like a thinkcentre tiny for a quarter of the price

These are amazing. Dell, Lenovo and I think HP made these tiny things and they were so much easier to get than Pi's during the shortage. Plus they're incredibly fast in comparison.

Better cooing which means it would last longer.

No display, battery, camera, etc should be cheaper.

2 more...

They were too pricy for me. I ended up with Bee-link machines (SER4/5/5Pro) and am happy with them.

Yeah I always coveted one but couldn't justify the cost over second hand dell or lenovo SFF PCs.

Sad really, but the issue, as someone as mentioned already is they were too expensive.

Yeah. Not sure why people would be proud of paying more for less.

It's not like the size difference is prohibitive compared to a normal workstation.

Every time I've had a use for these either a business PC (or ex-business referb for home) has always been a better, cheaper answer.

where do you find refurbs? I'd love to get my hands on an ex-business refurb!

@damnthefilibuster @Kushia

You can get a hold of them via #eBay. Every now and then a business upgrades and floods the market.

One of these days I want to buy a whole bunch of them and build a #Beowulf cluster.

You could use eBay but that's usually the option of last resort.

Your local city probably has referb shops that sell them or if you're keen you can pick them up directly from auction for peanuts.

If you're in a major city theres likely a recycling centre just for old office machines. You can snag them dirt cheap, but with no Harddrive. Theyre a bit dated, but will work great as a server.

In a similar vein is to look for government auctions in town. I've got a major public university in my city, and it maintains a permanent auction warehouse. Like once a month they sell all kinds of stuff, from mini fridges to laptops by the pallet.

+1 I'm curious too

My city has a couple mom-and-pop type businesses doing it, I'd hazard a guess it's similar elsewhere - never heard of any 'big name' outfits doing it on any real scale.

What's your mom-and-pop businesses called? They have similar names or similar ways of finding them... Would make it easier to find those around me.

I think there's a niche for a computer slightly more powerful than a raspberry pi, with no need for active cooling, capable of running as a basic always-on server.

The Intel NUCs were always a bit too expensive for that, and the Raspberry Pis are slightly underpowered (plus the SD-card as the primary storage is limiting). But, there are increasingly ways that people who aren't massive computer geeks would want an always-on computer. Things like a home security system, a media downloader, a home automation machine, etc. The power consumption, noise and size of a desktop computer is just overkill for that. A Raspberry Pi could be, but the default versions are not designed as servers. They're more robotics sandboxes.

Each generation brought incremental improvements and I feel like they were just starting to hit their stride and get somewhere, but your comment does allude to the issues NUCs have in their current state.

For me it's not a comparison with a Raspberry pi, NUC is far too expensive for that. It's that I'm paying top dollar for a less capable system than I can build in a small form factor from standard parts.

They made some decent leaps forward in recent years, but they've been passed as if they were standing still by the likes of the Beelink GTR6. Better price, better thermals, better for gaming, better by every metric you could throw at it.

Again I think it would be a real shame for intel to give up right now because it seems as if the gap between a low-spec traditional gaming PC and what can be achieved in one of these little boxes is all but closed with AMD hardware, and the NUC wasn't really that far off either: they just needed another couple of little boosts and a reality check on their pricing.

The GTR6 sells for $619USD and will play games at 1080p or 2560x1080 with performance far better than anything I can build myself for anywhere close to that price. In traditional computing workloads, it's even better! It will handily beat my Jan 2021 balls-to the wall $6000 PC in most CPU tasks.

Say for example I was looking to build a PC for my dad to game on at the above resolutions. By the time I've bought a decently rated PSU, Motherboard and a modest CPU: the GTR6 has already beaten me. My build can't go any further because I can't beat it without spending dumb money.

I'm not personally in the market for one of these things, but the moment they provide an easy means to mate a high-spec GPU to the crazy hardware already inside a NUC or GTR6 style box for a competitive price...it's going to be a pretty difficult decision to justify another monster desktop PC build.

The stupid thing is, Intel were already so close to being there! The NUC 11 Extreme Kit was exactly this, it was just priced in the most noncompetitive manner and for that stupid money, it only came barebones - still requiring you to buy further components as well as add a GPU. https://www.pcmag.com/reviews/intel-nuc-11-extreme-kit-beast-canyon

I've rambled enough. I really wish intel hadn't given up on this space, but I have a bit of faith that smaller operators are going to continue to leverage the power of AMD's mobile offerings and fairly soon, land on a formula that near enough eliminates the appeal of my beloved custom PC.

https://youtu.be/iaYHtfa1-pY

https://youtu.be/Ye7BmiPsqiA

Thanks for the info. I haven't paid much attention to the NUCs lately, because the Raspberry Pis, despite their limitations, were closer to the specs I needed, and you can't beat their price to performance ratio.

I didn't realize quite how good the NUCs and the NUC-likes had become. Way overkill for what I wanted though.

There's a few boards that bridge the gap between pi and a pc for media servers and small NAS uses. Look at Asus Tinker board, Odroid, Udoo Bolt, Orange Pi, Rockpro64, BeagleBone

I've only recently been thinking of setting up a media server or NAS. Currently have a RaspberryPi running a 3D print server, but like you say RaspberryPi's are a bit weak hardware wise and limited by the SD card. But I never wanted to spend the money on a NUC. I'll have to check out these other options you mentioned, thanks for listing them.

I just bought a used Lenovo ThinkCentre M710Q Mini Tiny Desktop PC Computer i5 6400T 1TB SSD Win 10 Pro from Ebay for $289 AUD and plugged in some oldish external SSDs and HDDs and now have 10TB of storage. I'm really pleased with it, it took about half an hour to install Proxmox and I've now got 5 VMs up and running.

the Raspberry Pis are slightly underpowered (plus the SD-card as the primary storage is limiting).

OrangePi has been my go-to since these got expensive. More features, including a 8gb emmc module built in, and just as cheap.

I'll have to look into that, it may be more what I need.

They are out there but not in large quantities.

i.e. my new home server runs on an odroid H3

I'm fine with it. Their competitors passed them by a few years ago anyway. The only thing the Intel branded stuff was better at now- was being more expensive.

Agree, love my NUC but it seems the last few years they haven't been the best option. It seems like they lost touch with what people wanted from them around the time they started releasing models that supported a full size GPU.

started releasing models that supported a full size GPU.

Exactly what nobody on earth wants from a mini pc.

Real shame. Best purchase I've made for running Proxmox with Plex, Radarr, Sonarr, Home Assistant

Between Minisforum and Beelink putting out NUC-likes with AMD, Intel just can't compete. I'm biased in favor of team red to begin with, but you just cannot tell me an Intel NUC provides better per dollar value than the above's offerings. I've used NUCs, I like NUCs, but why pay more for less when there exist alternatives?

Exactly, for a home lab I would pick an Amd over Intel just to have the extra cores on top of costing less.

The reason for wanting intel is the iGPU to get quicksync.

Exactly. The only reason why I'm waiting for NUC12 right now. As far as my limited knowledge goes, AMD is behind here. Correct me, if I'm wrong.

I’m not too knowledgeable on the topic but I thought the amd iGPU had vce, which is a their version of quicksync?

For me it's the hardware transcoding capabilities of the Nuc is what makes it stand out.

Quick sync is so good and well supported that Intel is a no brainier for me.

I mean, they're the OEM, they could easily have lowered their own prices. It's not like they were taking a loss on each unit.

Funny timing on this since the mini pc market is picking up steam from what I can tell. Then again, these are overpriced compared to the competition.

That depends. I don't think Intel actually wants to be in the market for whole (or barebones) systems. they probably would much rather just sell the processors and leave the rest to others. The NUCs were just a tool to kickstart the market, which seems to have worked quite nicely. The only issue being that now both AMD and Apple are strong competion.

So under that assumption this withdrawal makes a lot of sense, especially now that they need to focus all of their resources to catch up in their main business segment.


Didn't Valve make similar comments for the steam deck? That they see it as a tool to create a new market and hope that others follow.

Even if someone else were to make a much better handheld. As long as it runs Proton/Steam Valve would still win.

I own a bunch of them, generations five through ten, and have always had a love/hate relationship with them. None has ever died on me. My main workstation at home, as well as two "homelab" servers are NUCs. They Just Work<tm> under both Ubuntu and Proxmox.

The love is for them just working. The hate is for Intel :-)

What they got wrong:

  • cooling. CPU cooling is finely tuned and controllable through the BIOS, no qualms there. The disk and the NVME SSD have no cooling whatsoever. Sticking an small 40mm fan to the side and running it at the minimum RPM drops the case temperature from 60°C to 40°C and avoids the NVME SSD burning out. Needless to say, a glued on fan looks fugly.
  • opening. By refusing to let their firmware be accessible to the fwupdmgr mechanism, Intel forces its Linux users to physically go to the machine, stick in a USB thumbdrive, keyboard and a monitor, and click their way through the BIOS update. In contrast, my Dell gear gets updated online through fwupdmgr, and I just have to suffer a reboot with a few minutes of downtime. I don't even have to be at the keyboard.
  • remote monitoring. I bought two NUC's with vPRO support, to allow for remote management. But the remote console sucks eggs even from a Windows management station, so I wound up disabling it on all of them. Both Dell's iDRAC and HP's ILO run circles around vPRO based remote management.

That's not a lot to go wrong for such a big endeavour, which is why I will keep hating Intel and sorely missing the upgrade opportunity. Just hoping Dell will step into the void.

What do you recommend for desktops that aren't the big ass tower?

Look at minisforum and beelink.

I can second Beelink here. I bought a Beelink SER5 for US$380 as a gaming computer for my kids. It's an AMD Ryzen 7 5800H with a Vega GPU, 16G RAM and a 500GB SSD. It probably won't work well with the latest graphics-intensive games, but it's been great so far with a bunch of games my kids like.

That one worked so well that when I needed a new desktop computer for their schoolwork and similar, I got another Beelink, this time a Mini S12 for US$200. It's an Intel N95 with 8G RAM and a 256G SSD. Works absolutely fantastically for its purpose.

Both are tiny and silent.

14 more...

I got one for my mother when she needed a new PC and it died within a month. Not intel's fault though, chip on the SSD died, first time I've seen an m.2 SSD die like that. Replacement going strong.

14 more...

The article makes it sound they cost over $1,000 (USD?) and were impossible to find but here in Australia I never had any issues finding and unless you were going for the extreme versions, there closer to $5-600AUD which made them a great fit. All we can hope is that there’s a few other brands who are willing to fill the space with equal quality products.

there closer to $5-600AUD

New or used?

That was for new entry level specs, you could obviously spend a lot more on the highest specs but often the NUC fit a segment that didn’t need to be bleeding edge of performance.

5 more...

I got an i7-6700 skull canyon? for free through work many years ago, absolutely love it, it now serves as a Linux box and hosts server stuff on it. Only issue is a ram port died and seemed a common problem!?

Still enjoying using it and it's form factor is fantastic, not sure if I would replace my own desktop with it but would have been an easy consideration for the kids first PC although it may benefit them actually building a tower and learning.

Shame to hear they are stopping

Go used. Lots of people get rid of their hardware when just a bit of care and repairs will make it as useful as brand new.

Damn. I may need to buy a couple

That sucks, I hope this isn't a statement about the "Mini-Pc" market in general. I've been thinking about getting one as a "Steam machine/ emulation station" for a long time but the stars never really lined up.

I've got a full sized PC in the front room getting long in the tooth and looking ridiculous that could easily be replaced. But while the 970 still plays Dave the Diver, well there's other shit money can be spent on.

Wasn't meant as a reply, pressed the wrong thing, my bad

Right there with you. Full size ATX machine circa 2010ish, can still play GTA V fine enough. The only reason it isn’t my media server is because my Mac mini does that for less power.

The big guy keeps chugging along when I need him, so the funds go elsewhere.

I think this has more to do with the refurbished small form factor business PCs eating up their market share as they flooded the market. I can get a decent i5 unit for $100and throw a $100 into it in upgrades and hit the same performance as their $300-400+ price range.

I found an HP SFF for like $60 at the thrift store with a 4th gen i5 and it was kitted out with more ram and a 250gb SDD. Perfect HTPC for what I do. I was shopping NUCs too.

Good find! I live in the San Francisco Bay Area and all the thrift stores near me are overpriced, so I never find good deals like that.

I got this at my local overpriced thrift store and was surprised they didn't want a shit ton for it. This place will put ebay listing (not even sold) prices on their electronics. I think it came out of their office or something.

That’s a bummer. Maybe framework will fill the void lol

There are plenty of alternatives to the NUC. MinisForum, StarLabs and System76 out of the top of my head

Oh man i was thinking of getting one of these to replace my raspberry pi

Maybe ironically with the prices dropping on these people will actually buy them..

Lenovo or HP mini PC would be a much better bang for your buck.

They're also a lot bigger and don't really fall under the same miniPC classification.

They're not a lot bigger than a NUC. My HP mini PC's footprint is like 8"x8"

And NUCs are usually 4x4. That's literally half the footprint.

Edit: a quarter of the size. This is why I don't do math before coffee.

Okay, sure, but we're talking about inches. 8x8 isn't a large footprint. Don't be obtuse. Also 4x4 is 1/4 the footprint of 8x8.

Unless space is the absolute unchangeable primary concern than the size difference doesn't matter.

I kind of get it. MinisForum and companies like it have sort of carried the torch of what the NUC started. I loved the NUCs, but this was kind of inevitable.

I have two MinisForum miniPCs and I absolutely love them, I've had them on for months at the time without any issues. Before I got them I was looking into the Intel NUCs and they were way too expensive for the specs. Sure, their top of the line NUCs are absolute beasts in a tiny form factor, but their basic entry level stuff is for burning money

100% but its a lot easier for a business to go "we need to purchase X number this intel product" vs "We need to spend X on product from some company your non-technical ass has never heard of"

In the consumer/small business space I think we will be fine for options but the intel NUC was great for a lot of business applications and I will miss it!

I hope Valve release a home console with SteamOS like this.

They did, and it was a market failure, so they dumped all their inventory for cheap.

They weren't distributed directly by Valve though, there wasn't a standard hardware configuration, and SteamOS 3 and Proton didn't exist then.

I think with the strength of the Steam Deck now it'd really help to solidify the Valve ecosystem. Why buy a PlayStation and re-buy your games when you can just use Steam?

EDIT: That reminds me I really want a Steam Controller 2.0 too!

The frustrating thing about the steam link is how locked down it is. I'm not mad that they discontinued it or that they made the software available for raspberry pi. That last part is actually really cool.

The thing is, you can't do shit with it other than steam link. I want to hack this thing man! I want to install other shit on it and add it to the lab lol.

Damn, we are using them at my work and they have been very good as remotely updateable media kiosks. I just started to learn how to use them. Ofcourse well keep using them for some time still, but at some point we'll need to find another solution.

I was also thinking getting one to work as a streaming computer. Currently I use one computer setup, which causes performance issues with some games. Would a nuc work as a computer to encode the video live or would it make more sense to use a machine with s proper GPU? Any thoughts?

Encoding uses the iGPU. The iGPU should usually support 4k 60fps if it's a recent CPU.

Yeah, I'm not watching Tek Syndicate content.

Point is not who made it, but the PC. Here's the pure link since clicking on video description was too hard: https://www.bee-link.com/catalog/product/index?id=493

Maybe it doesn't matter to you but it does to me. I'm subscribed and watch Level1 videos so I still remember the mess Logan caused. They had a nice channel that could be even better today but he fucked it up.

(thanks for the link, though)

I do remember there was some drama, but to be honest I never followed them nor do I follow now. Saw the video some days ago, found hardware presented interesting and shared that. That about sums it up.

Minisforum is taking the torch from them. I just bought one from them which is essentially a NUC, it has a Core i7 and RTX 3070 mobile in it. It's pretty much a laptop without a screen. They make tons of smaller ones if you forgo the integrated high-end GPU.

Sad to see these go. I use one for my Nextcloud home server and am happy with it.

Sad, I have one right now and it's great. Sleek small form factor with the power of a regular PC for not really that much more money is a great idea. I haven't been the kind of guy to want to build a big rainbow LED PC in a long time, I've been appreciating I can get a great machine the size of a large hard drive

I have been using a Beelink mini PC in my home entertainment setup for about a year. It has been very reliable and solid. No issues with 4k content.

What's the opinion about System76's mini PCs? I've just ran across them and thinking of getting one.

I bought one several years ago (at least 6 years) and I.find it still works great. Though i'm not very demanding in how I use it

I don't own one, but want to as well. Commenting here to return and see what anyone replies to you.

Commenting here to return and see what anyone replies to you.

Not sure about Kbin, but Lemmy has a bookmarks feature for this.

I've bought a few dozen of these things, shame to see them go.

My wife just asked me about a backup solution for pictures. Is a small pc like this onnected to network with some drives in raid the best option? Should I use to also replace our Amazon fire stick?

I bought a synology for this. I still need to add a backup plan to it though.

That's sort of how I do mine. I put all my data onto dropbox/onedrive. I've got a $100 HP USFF hooked up in my office that is a 100% online mirror for those cloud accounts, and it backs up to an 8TB external each week. I rotate that drive with a spare each month (give or take), putting the "offline" one in a firesafe. It means I have a live copy (my pc), a cloud copy (OD/DB), a second hot copy (USFF PC), a near-line backup no more than 7 days old that isn't "live" and a cold storage copy that is no more than a month old (aka less than Apple's deleted-pictures and Dropbox's previous version storage time). It cost me two external drives and the mini-pc. And if all those fail I'll probably be roaming the radioactive wasteland looking for food and losing that data won't matter.

Oh, and that little box also runs a small FTP server and my Torrents for my Linux distro collection.

I use OpenMediaVault for my NAS

But if you don't want to be the IT of your family, I'd just go with an easy solution like WDs my cloud or one drive

I wouldn't recommend a WD My Cloud Home - it's not a NAS as such, it's a bit limited; I'd go for a Synology. or One Drive as you suggest - a 1TB plan is quite reasonable with regards to cost.

Looks like they're trying to get 3rd parties to make them. Oh well, pour one out for the quirky little machine.

Great machines, I use an NUC8i7 as our HTPC. Supports 4K 60fps. Got it hooked up to a Denon amp for Dolby Atmos. At some point i hope I'll find time to look into Home Assistant, I'd use another NUC for running that.

This is unfortunate, these NUC are inexpensive and reliable for the conference room.

AMD seems to be eating their lunch in small computers for consumers with their APUs in the Steamdeck and the more than a half dozen like handhelds, mini-pcs, etc. I'm sure intel will hang onto small embedded devices for industrial applications for some time but it's puzzling that they would just drop RISCV which seems poised to proliferate in this sector as well. It could just be that intel seeing that manufacture in China is and will continue to be very tricky has to narrow focus while they move their manufacture closer to home.

Lame. I was just thinking about possibly picking up a NUC to run a Jellyfin home media server and such. Seemed like a perfect use case. Oh well, guess we'll see where intel goes with it...

Plenty of alternatives to a NUC still out there. I like the MSI Cubi personally.

I was just looking at buying one second hand yesterday... Better buy one before everyone ramps up their prices!

Ah this sucks. They're such a great size and very capable. I'm currently using one as my all in one home server - it's been flawless.

It’s so easy to hate that company

I mean, they're just doing whatever they believe will make them the most amount of money for the least amount of effort.

All publicly-traded corporations do the same. Intel has just been very good at it because they used to have a product that was better than the competition.

Intel would better focus on the things they master instead of building shitty gpus