bitzu101 wrote on 2026-07-08, 10:10:

europe done nothing

If I'm not mistaken, AMD had a chip factory in Dresden, Germany for quite a while. Last year, GlobalFoundries announced a big investment there.

GlobalFoundries wrote:

DRESDEN, Germany, 28 October 2025 – GlobalFoundries (NASDAQ: GFS) (GF) today announced plans to invest €1.1 billion to expand its manufacturing capabilities at its Dresden, Germany site. The investment will enable a production capacity increase to more than one million wafers per year by the end of 2028, making it the largest site of its kind in Europe.

The expansion, known as project SPRINT, is expected to be supported by the German federal government and the State of Saxony under the framework of the European Chips Act, with EU approval for the full program expected later this year. This investment underscores Saxony’s role as a critical hub for semiconductor manufacturing and innovation and reinforces Europe’s strategic goal of supply chain resilience.

bitzu101 wrote on 2026-07-09, 15:01:

the costs would be way to high. the level of competence of the people is almost 0. […]
Show full quote

MagefromAntares wrote on 2026-07-09, 14:55:

bitzu101 wrote on 2026-07-09, 14:40:

there is no distributed capacity already. capitalism works in the way that all companies went where it is cheapest. hence the is […]
Show full quote

there is no distributed capacity already. capitalism works in the way that all companies went where it is cheapest. hence the issues today.

hungary , romania (my country) , pretty much any country in easter europe is incapable of producing any chips.

think in europe , the only countries that manufacture some chips are Germany, France, Austria, Italy, Ireland, and the Netherlands. and these are not big producers. but they can satisfy the needs of basic chips if need be , because they have supply chains already in place. it is not smart to start from scratch in hungary or any other country that does not have the supply chain.

also , because the EU , you don t need to really.

that being said , the 6 counties that do make microchips in Europe are not producing high end chips on the latest nodes , etc...

I would argue against Hungary not having the capability to produce chips, we don't have it right now, but we had it in the past(Mikroelektronikai Vállalat(MEV)), we lost it on 1986-May-26(It is also a good example why centralizing it even in a single country is a bad idea) and the MSZMP(The then ruling party) decided to not rebuild it, I'm almost 100% sure that Romania also had something similar, if a nation were capable of producing something in the past, then I think it should be able to produce it now, I don't think that the resource allocation around the world changed so much that it would be impossible to rebuild the supply chains and the facilities.

the costs would be way to high. the level of competence of the people is almost 0.

not saying it s impossible , but for sure it s improbable and if started would cost staggering amounts of money and for countries like hungary and romania , just not feasable.

probably the only logical thing to do is to invite a company that does chips already and give them huge benefits to set up production in your country. but even the, lots of money spent , requires supply chains , requires talent , requires people from abroad to be hired , requires local training , etc... it s so complex... we in the east struggle with idiotic/stupid people in power , huge corruption , we are so far away from such things... have u seen certain villages in romania? with no asphalt on the street and no rusnning water? and u want us to make chips? 🤣

Don't worry Hungary has a lot of the same problems, but in this same topic it was also argued that microchip production is not purely about business but strategic capability and also part of technological independence, prioritizing it would be more important than simply the profit it would make. Yes there are more pressing issues, but this should also get more resource allocation than 0 percent.

Joseph_Joestar wrote on 2026-07-09, 15:05:

bitzu101 wrote on 2026-07-08, 10:10:

europe done nothing

If I'm not mistaken, AMD had a chip factory in Dresden, Germany for quite a while. Last year, GlobalFoundries announced a big investment there.

GlobalFoundries wrote:

DRESDEN, Germany, 28 October 2025 – GlobalFoundries (NASDAQ: GFS) (GF) today announced plans to invest €1.1 billion to expand its manufacturing capabilities at its Dresden, Germany site. The investment will enable a production capacity increase to more than one million wafers per year by the end of 2028, making it the largest site of its kind in Europe.

The expansion, known as project SPRINT, is expected to be supported by the German federal government and the State of Saxony under the framework of the European Chips Act, with EU approval for the full program expected later this year. This investment underscores Saxony’s role as a critical hub for semiconductor manufacturing and innovation and reinforces Europe’s strategic goal of supply chain resilience.

need more and more and more. i want to buy my new cpu and instead of malasya or china to say EU or usa ...

eu produces some chips , but still need way more to compare to asia ... think israel also produces some intel chips...

MagefromAntares wrote on 2026-07-09, 15:05:

bitzu101 wrote on 2026-07-09, 15:01:

the costs would be way to high. the level of competence of the people is almost 0. […]
Show full quote

MagefromAntares wrote on 2026-07-09, 14:55:

I would argue against Hungary not having the capability to produce chips, we don't have it right now, but we had it in the past(Mikroelektronikai Vállalat(MEV)), we lost it on 1986-May-26(It is also a good example why centralizing it even in a single country is a bad idea) and the MSZMP(The then ruling party) decided to not rebuild it, I'm almost 100% sure that Romania also had something similar, if a nation were capable of producing something in the past, then I think it should be able to produce it now, I don't think that the resource allocation around the world changed so much that it would be impossible to rebuild the supply chains and the facilities.

the costs would be way to high. the level of competence of the people is almost 0.

not saying it s impossible , but for sure it s improbable and if started would cost staggering amounts of money and for countries like hungary and romania , just not feasable.

probably the only logical thing to do is to invite a company that does chips already and give them huge benefits to set up production in your country. but even the, lots of money spent , requires supply chains , requires talent , requires people from abroad to be hired , requires local training , etc... it s so complex... we in the east struggle with idiotic/stupid people in power , huge corruption , we are so far away from such things... have u seen certain villages in romania? with no asphalt on the street and no rusnning water? and u want us to make chips? 🤣

Don't worry Hungary has a lot of the same problems, but in this same topic it was also argued that microchip production is not purely about business but strategic capability and also part of technological independence, prioritizing it would be more important than simply the profit it would make. Yes there are more pressing issues, but this should also get more resource allocation than 0 percent.

i agree it should get more than 0% resorce allocation. I also agree about the strategic capability and tech independence. but i am talking about reality. and the reality is that all eastern europe is poor compared to western europe , we are also way more corrupt. and no one in power even thinks about fab productions let alone about how they would even go about something like this...

i do like your enthusiams though 😀

bitzu101 wrote on 2026-07-09, 15:15:

MagefromAntares wrote on 2026-07-09, 15:05:

bitzu101 wrote on 2026-07-09, 15:01:

the costs would be way to high. the level of competence of the people is almost 0.

not saying it s impossible , but for sure it s improbable and if started would cost staggering amounts of money and for countries like hungary and romania , just not feasable.

probably the only logical thing to do is to invite a company that does chips already and give them huge benefits to set up production in your country. but even the, lots of money spent , requires supply chains , requires talent , requires people from abroad to be hired , requires local training , etc... it s so complex... we in the east struggle with idiotic/stupid people in power , huge corruption , we are so far away from such things... have u seen certain villages in romania? with no asphalt on the street and no rusnning water? and u want us to make chips? 🤣

Don't worry Hungary has a lot of the same problems, but in this same topic it was also argued that microchip production is not purely about business but strategic capability and also part of technological independence, prioritizing it would be more important than simply the profit it would make. Yes there are more pressing issues, but this should also get more resource allocation than 0 percent.

i agree it should get more than 0% resorce allocation. I also agree about the strategic capability and tech independence. but i am talking about reality. and the reality is that all eastern europe is poor compared to western europe , we are also way more corrupt. and no one in power even thinks about fab productions let alone about how they would even go about something like this...

i do like your enthusiams though 😀

While this isn't really on topic I want to leave this conversation thread on a positive note, so I say regarding this:
I was not yet alive to witness the events of 1956, but after the Soviets crushed the Hungarian uprising for a lot of people it seemed that stopping being a Soviet satellite state was impossible, however both my grandparents and parents said that the Soviets will eventually leave even if we were unsuccessful at that time, and yes it took 33 years, but the Soviets did leave after 1989 and I actually witnessed that part. Maybe a situation seems hopeless, maybe it seems like that it is the darkest before pitch black, but in my whole life I had the following way of thinking: "It will not be quick, it will not be easy, but eventually it will happen, might be not my generation, might not even be the next one, but it will happen, and I should work in the direction of being a positive influence on the matters even if the results are not immediate and I personally might never see them or only see them after I became a pensioner walking with a stick."

Sorry to derail this thread a bit, but I wanted to explain why I keep my enthusiasm 😀

Always good to have the enthusiasm.

Hungary doesnt have the money required to build a tier 1 or tier 2 fab. The situation on the ground would make investors nervous about dropping that much into it.

Joseph_Joestar wrote on 2026-07-09, 15:05:

If I'm not mistaken, AMD had a chip factory in Dresden, Germany for quite a while.

Please refrain from introducing facts to these discussions 😅

badmojo wrote on 2026-07-09, 22:18:

Joseph_Joestar wrote on 2026-07-09, 15:05:

If I'm not mistaken, AMD had a chip factory in Dresden, Germany for quite a while.

Please refrain from introducing facts to these discussions 😅

when people say europe needs to produce chips is mean we need to be self sufficient or to produce a large procent of chips to matter worldwide. and those chips need to be ultra high end.

we also need to quadruple the chip manufacturing. maybe spend a few bucks into creating a new company , a EU company to rival TSMC or Samsung... put some money together to build the future.

I mean we have ASML in europe but no world wide company with fabs to produce...

Europe must:

Choose one host country (NL, DE, or BE).
Invest €300–500 billion over 20 years.
Centralise semiconductor strategy at the EU level.
Build a workforce pipeline of 50,000+ engineers.
Cluster suppliers around the fab.
Guarantee access to ASML’s High‑NA EUV.
Create a pure‑play foundry (no design).
Secure global customers.
Maintain political stability for decades.

If Europe does all of this, it can build a TSMC‑equivalent. If it doesnt, it never will.

Good luck getting French and Germans to work together on something like this. You'll end up with two rival programs with all the associated duplication of effort and expenditure.

Errius wrote on 2026-07-11, 09:29:

Good luck getting French and Germans to work together on something like this. You'll end up with two rival programs with all the associated duplication of effort and expenditure.

Exactly.
TSMC happened in Taiwan because it could happen in Taiwan. So many of the right things came together in the right place at the right time, thats why its in a place that is so vulnerable.

It should be noted that Airbus only works because aircraft are highly modular machines, so there's not too much inefficiency involved with making the wings in one country and the engines in another and the fuselages in another and assembling everything in yet another.

You could save a lot of money by doing everything in just one country, but national rivalries don't allow that.

wierd_w wrote on 2026-07-07, 21:27:

Be careful. 'Programmer' is a loaded word. […]
Show full quote

Be careful. 'Programmer' is a loaded word.

It's true that *good* programmers care about how long a function takes to complete, and that it does not engage in anomalous behaviors, or fails to free memory after its done with it.

Then there are programmers that 'dont care about that', because 'It does what I want it to, and that's good enough. CPUs are fast enough that it's not important anyway.'

And, outside of those boxes, there's more circles on the diagram, where 'But the host OS's function call takes too long to complete! So we reimplemented it inside our own codebase!' (Which is a major reason why wdb browsers are mini-operating systems these days, and 'need' to be, because the afore mentioned bad programmers are writing webapps, and their turds need to get polished), and other sins 'of necessity' live.

'If I write it as a webapp, I dont have to think about where and how it runs! I save so much time!'

Is just an abstraction of 'CPU, RAM, and diskspace are plentiful!'

It leads to things that should not be web apps, being web apps, and the resultant turd polishing to make those performant.

Sorry, this a pile of romanticized nonsense about “good programmers,”

The whole “good programmers care about performance, bad ones don’t” thing is a fairy tale. Modern software isn’t typically slow. It is very large however because teams are huge, deadlines are tight, features pile up, and hardware is cheap enough that companies prioritize shipping over hand‑tuning every line. Pretending it’s all about personal virtue is just gatekeeping.

But last time I experienced a slow program was... in the 90s.

Your whole argument is basically “things were better when I was younger,”. The idea that people write webapps because they “don’t have to think about where it runs” fails to see the point: The Web brings one runtime, one deployment, one update pipeline. To want to go back to a platform that is not the Web is not justified.

Software is bigger because the problems are bigger, the systems are bigger, and the expectations are bigger than memory or performance: integration, UX, security take a precedence. Programmers did not suddenly become idiots, they have runtimes, SDKs, and hardware that allow them to bring a better experience.

I prefer a programmer that focus on shipping a feature that works, rather than focusing on deailing with 640KB of memory...

No, no sir, no.

Today, people throw huge libraries at small problems, and ignore the costs this incurs.

They do this because they want just the small bit of premade code (and the outside support team that maintains it, to maintain it for them), and believe 'cpu, disk, and memory are cheap and plentiful', and feel their own time is vastly more important than any other concern.

All you've done is make a big show of why you agree with that paradigm, and why you dont want to make quality software. You believe 'getting it done in a hurry, so quarterlies look good, and investors stay happy', is the single most important thing in the universe.

This is why 'hello world' can be a multimegabyte consuming behemoth, and why people like yourself think this is OK.

In the actual world, having shittons of code that your program does not actually need means your attack surface is many many times larger, you burn the user's resources for nothing that benefits them, get fussy when called out on it, and just reiterate how your boss wants it yesterday.

You can take your 'you insulted my mother' attitude elsewhere.

The enemy of a job well done, is a job done 'good enough.'

Jobs done 'good enough' are why the girders in the old pfizer building buckled, why the rhinoliner in the reflecting pool peeled off, and why the doors blew off boeing aircraft.

It's also why software is rife with attack surfaces, consume many times the memory and cpu needed to process them, and why people get less and less capable of doing things correctly as time marches on.

xcomcmdr wrote on Today, 07:04:

Sorry, this a pile of romanticized nonsense about “good programmers,” […]
Show full quote

wierd_w wrote on 2026-07-07, 21:27:

Be careful. 'Programmer' is a loaded word. […]
Show full quote

Be careful. 'Programmer' is a loaded word.

It's true that *good* programmers care about how long a function takes to complete, and that it does not engage in anomalous behaviors, or fails to free memory after its done with it.

Then there are programmers that 'dont care about that', because 'It does what I want it to, and that's good enough. CPUs are fast enough that it's not important anyway.'

And, outside of those boxes, there's more circles on the diagram, where 'But the host OS's function call takes too long to complete! So we reimplemented it inside our own codebase!' (Which is a major reason why wdb browsers are mini-operating systems these days, and 'need' to be, because the afore mentioned bad programmers are writing webapps, and their turds need to get polished), and other sins 'of necessity' live.

'If I write it as a webapp, I dont have to think about where and how it runs! I save so much time!'

Is just an abstraction of 'CPU, RAM, and diskspace are plentiful!'

It leads to things that should not be web apps, being web apps, and the resultant turd polishing to make those performant.

Sorry, this a pile of romanticized nonsense about “good programmers,”

The whole “good programmers care about performance, bad ones don’t” thing is a fairy tale. Real software isn’t slow. It can be slow because teams are huge, deadlines are tight, features pile up, and hardware is cheap enough that companies prioritize shipping over hand‑tuning every line. Pretending it’s all about personal virtue is just gatekeeping dressed up as wisdom.

But last time I experienced a slow program was... in the 90s.

Your whole argument is basically “things were better when I was younger,” wrapped in tech jargon. The idea that people write webapps because they “don’t have to think about where it runs” is just wrong. They think about it constantly: one runtime, one deployment, one update pipeline. To want to go back to a platform that is not the Web is not justified.

Software is bigger because the problems are bigger, the systems are bigger, and the expectations are bigger. Not because programmers suddenly became idiots.

No, sorry but programmers are just like everyone else, if there is an easy lazy path to get something done they will take it even if its slower and requires more resources to execute. (Java Script was a perfect example of this in action, same for Adobe Flash)

Seen it time and time again across a dozen different industries . .they will ALWAYS take the path of least resistance even if it costs more.

In fact in one position they would give the worst job to the dumbest employee because idiots will always find an easy lazy way to complete a task. They would then evaluate the method and decide if it was worth allowing it to keep being done that way, 90% of the time only a few changes were needed to make it acceptable to HR and OSHA.

Programmers are no different here, do I write this library myself or do I just use the massive library that isn't really fit for purpose but saves me writing 20 - 30 lines of code ...or perhaps Ill just go to Github and use someone else's code.

Nahhh Ill just ask Claude Code to vibe code this for me. I mean Who's gonna know right...

xcomcmdr wrote on Today, 07:04:

The whole “good programmers care about performance, bad ones don’t” thing is a fairy tale. ... Modern software isn’t typically s […]
Show full quote

The whole “good programmers care about performance, bad ones don’t” thing is a fairy tale.
...
Modern software isn’t typically slow.
...
But last time I experienced a slow program was... in the 90s.

Presented without comment

Law212 wrote on 2026-07-08, 15:12:

Agree completely. Canada needs to become way more self sufficient but the government is hellbent on selling every bit of the country to foreign governments.

Definitely. You guys should never have allowed ATI to be sold to AMD. I know, hindsight and all that, and over here in Europe, we've made our fair share of similar mistakes as well.

There is software that is the output of a passionate programmer in his spare time and software that is a company's product, do not confuse the two.

I've been wanting to write more or less what xcomcmdr said on multiple occasions but I just don't have the energy for these things anymore.
Everyone who has actually done this job outside of a bedroom knows how it goes, and it's not "I was told to ship this feature by the end of the week but let me take 3 months just to rewrite this library from scratch and introduce 10 bugs to an already solved problem, so that it uses a few MB less RAM and completes an edge case 0.5 sec faster so that a stranger on the internet who is not even a colleague doesn't call me lazy".

Sorry if I struck a nerve with some people but this is a recurring statement implied as a fact.

More, the baseline truth of this xkcd;

as it relates to libraries, and library use.

The use of a well maintained library is fine, as long as that library is sparse, and does not *itself* demand many more layers of dependency, that are also not sparse.

openssl only does one thing. It will implement secure sessions better than a diy session will, and it does not have extraneous parts. It's a fine thing to use. It's also written to compile everywhere, and does not lug around a virtual machine or high level interpreter.

Then you have stuff like microsoft's .net framework, which is the exact opposite of that approach. A sprawling swiss army knife, that has lots of rube-goldbergian parts that your project most certainly wont need, that relies on an interpreter and post-hoc optimization service to attain any kind of performance.

One can very much make good software without wheel reinvention.