Slashdot Mirror
Microsoft Quietly Cuts Off Windows 7 Support For Older Intel Computers (computerworld.com)
An anonymous reader shares a report: If your PC doesn't run Streaming Single Instructions Multiple Data (SIMD) Extensions 2, you apparently won't be getting any more Win7 patches. At least, that's what I infer from some clandestine Knowledge Base documentation changes made in the past few days. Even though Microsoft says it's supporting Win7 until January 14, 2020, if you have an older machine -- including any Pentium III -- you've been blocked, and there's nothing you can do about it.
...if you're not an idiot and don't go to random sites/click "run" on downloaded files, you're reasonably safe...
Excellent, using your idea, we don't need to patch anything ever again. The answer has been right in front of us all this time, we would have never figured it out if you hadn't come along. I won't even need to help older relatives when they fuck up their PCs, I'll just print out your post and stick it below their screen
Windows 7 is the Best Windows. , better than all versions which came before it and after it. The only thing it lacks is out of the box USB 3.0 support. The drivers not on the install disk but you ca add it.. That's all. It's everything you need.
Windows 8 was a stupid movie. "Let's change the UI, because, fuck it, let's change the UI." Nothing else.
And Windows 10 with its intrusive spying and adverts truly sucks ass. It didn't add anything of value either.
Microsoft is pushing out new versions because no one has gone for their subscriptions so new versions is how they make money. That is all.
> b0s0z0ku : Also, if you're not an idiot and don't go to random sites/click "run" on downloaded files, you're reasonably safe.
Wow. You really are an idiot. Precisely the sort of idiot who needs to be protected with security patches.
Well said. You responded far more politely than I would have to a person is a living, breathing example of the kind of fuckwit who for years helped to keep Linux from being widely accepted. Newbies would drop by a Linux site hoping for help. They would ask questions that were very elementary but not stupid, often after having invested hours trying to fix a problem themselves. Far too often the "advice" they got from twatwaffles like this was that they were morons who shouldn't be allowed to own a computer.
Some day, perhaps creatures like this will have their noses rubbed in the fact that for most people, a computer is a means to an end, not the end itself.
I've calculated my velocity with such exquisite precision that I have no idea where I am.
I enjoy how that's fit in there as if this is 2008 and there are still a good number of P3s in use.
So, if a company just decides they don't want to honor the promises they made, they should be able to walk away whenever they want? That's incredibly moronic and you know it. Microsoft led customers to believe they would receive support through the EOL date. Prematurely ending support is going back on their word, and it shouldn't be legal for them to do this.
Doesn't change the fact that most malware is actually installed by its users, voluntarily. And that most Pentium III systems are likely running in specific, non-Internet-connected applications.
The last Pentium III chip ever created was released in 2001.
2001. That is 17 years ago. Last computer built with this ancient technology is probably from 2005, over ten years ago. Much of the hardware machinery, such as mechanical drives and fans, should've stopped working by now. If they miraculously still work, Linux is a prime candidate to run on this super-old system for that last mileage.
Face it, Windows 7 is on it's death bed, and if you do not like it, go Linux or go home. :)
systemd is not an init system. It's a GNU replacement.
I'm not arguing that what MS is doing here is right. It's just LESS shitty than cramming Windows 10 down users' gullets, with its built-in spyware cr@p and ever-changing interface.
"And that most Pentium III systems are likely running in specific, non-Internet-connected applications." = You made that up from pure spun hubris. You're an idiot sometimes when you do this.
Last P-III was 2001 or 2002. People/companies generally nurse such old hardware along because they have to (it controls factory hardware, etc), not because they want to.
The last PC I had that wasn't capable of SSE2 was an Athlon T-Bird, which had a quite beefy for its time 768MB of RAM. I wouldn't want to even try running Windows 7 on that.
Doesn't change the fact that Windows 7 officially supports Pentium IIIs (32-bit x86 1GHz or higher) and has stated support to 2020. And that this inherently means if they're running in specific applications and are exploited because of malware because of a lack of updates, Microsoft can be sued through the ass not to mention be potentially brought up on fraud charges*. But, yea, let's try to spin this into being a minor issue and wholly the users fault.
PS - Even using your out of the ass numbers of 0.5% of computers (really using just Windows 7 sales (300 million), which is a smaller number), and you're talking 1.5 million computers. You don't think 1.5 million computers as part of a botnet would be a big thing? Or replacing 1.5 million computers early would be a big thing? Seriously, the sheer scale of expected support should have shut down Microsoft considering such changes.
Of course, the real lesson here is to cut Windows products out of your environment as much as possible. Any company that believes it can just mandate substantial changes to your business so you can keep accepting patches to fix ITS bugs is a trainwreck.
* Large corporations quite specifically are maximizing the life of bought hardware and scheduling software purchased based upon a support cycle. Anything that upsets that can be a massive loss.
Sure, most people do not run Win7 on computers that old. But there are embedded systems like displays, measurement equipment, medical equipment, etc. that will be affected by this and MS was fine doing this deceptively and without warning and without giving people time do make arrangements. They also did it _while_ these systems are officially compliant with the Win7 minimum requirements. That is just completely unacceptable, but so very much like MS. No honor, no care for the customer, just always after the biggest profit they can get for cheap.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Embedded systems like medical equipment, displays, measurement equipment, etc.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Not importance-wise. Likely applications that will be networked is medical imaging, display boards, some measurement equipment, SCADA system front-ends, etc. This is a real fail on their part and, if I were up to me, they would be liable for any and all damage they cause.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Why would anyone be running Windows 7 on a P3??? There are two use cases where anyone aside from a collector would still be actively using such old equipment. #1, you need it to control or interface with specific and valuable hardware from that time period. In this case, you would also want to be running the original OS from that time period too, to ensure maximum compatibility. On the other hand, if you're using it for retro gaming, you also want to be running at least a relatively time accurate OS, since you're using an old computer over an emulator precisely because it's more authentic. If you're using a P3 as a home computer, you can literally buy a machine with a Core 2 Duo or a Sandy bridge processor for $30 or less, and you should have done so a long time ago anyway, since a P3 hasn't been able to handle anything more complex then static html for like 10 years now, making it largely useless as a general purpose home computer. And even if for some reason you did, Linux would be the obvious choice as it's both free and more capable then Win 7. I simply can't see any situation for why you'd ever want to run Windows 7 on a computer like that...
"Set a man a fire, he'll be warm for the rest of the night. Set a man afire, he'll be warm for the rest of his life."
Also, if you're not an idiot and don't go to random sites/click "run" on downloaded files, you're reasonably safe.
Code Red would like to have a word with you.
Il n'y a pas de Planet B.
Less than that. I had a computer without SSE2 until 2011, and by the end nothing worked. The MSVC compiler automatically enabled SSE2 and used it, which meant that all Windows application compiled after 2009 depended on SSE2. So support has been dead for at least 7 years already
The last PC I had that wasn't capable of SSE2 was an Athlon T-Bird, which had a quite beefy for its time 768MB of RAM. I wouldn't want to even try running Windows 7 on that.
I had a dual Athlon 1800MP for years with 2GB of RAM, it survived surprisingly long by being dual processor, and having the GPU upgraded 3 times. It mostly died 7 years ago because everything compiled with a microsoft compiler depended on SSE2, so practical support ended 7 years.
Even using your out of the ass numbers of 0.5% of computers (really using just Windows 7 sales (300 million), which is a smaller number), and you're talking 1.5 million computers. You don't think 1.5 million computers as part of a botnet would be a big thing? Or replacing 1.5 million computers early would be a big thing?
Most of those have already been replaced and sent off to the metal recyclers. Congratulations, you managed to offer even dumber numbers than the GP.
Of course, the real lesson here is to cut Windows products out of your environment as much as possible. Any company that believes it can just mandate substantial changes to your business so you can keep accepting patches to fix ITS bugs is a trainwreck.
"Linux Set To Shed Nearly 500k Lines Of Code By Dropping Old CPUs"
https://www.phoronix.com/scan....
it's not that it is older architecture that they're dumping support for, it's HOW THEY'RE DOING IT.
windows 7 has a published EOL. they aren't adhering to it.
they release buggy patches. they said they'd fix them. they are not.
__
similarly, and even more fucked up, microsoft killed support for windows 7 and 8.x for *NEW* processors (amd am4 bristol ridge/ryzen, intel kaby lake and newer).. and tried to nix it for skylake as well. also violating their own published lifecycle.
THIS IS NOT OK. regardless of the age of the affected systems. they have policies, we expect them to fucking follow them.. just as they expect us to abide by their draconian license agreements and bless their invasive data gathering and 'privacy' policy.
I keep saying - This is just the thin end of the wedge. Or in this case, now a slightly thicker part of the thin end of the wedge; It'll keep getting worse and worse until 'no support in 2020 but you can still use it on existing systems' slowly turns into 'deliberately crippled on all systems so you have to use windows 10'
Don't listen to all these astroturfers putting down old systems just because they're old.
If they could run Win7, there's no good reason why the goal posts should suddenly be moved.
Why do American nutjobs keep flooding the World Wide Web with their private political opinions? Nobody gives a shit about your baboon of a President or your fruitcake political opinions.
They also failed to update their minimum requirements. On https://support.microsoft.com/en-us/help/10737/windows-7-system-requirements, there is no reference to SSE2 or SIMD at all.
C - the footgun of programming languages
That was my feeling too, I mean a Pentium III? That'd be roughly a 20-year-old machine, contemporaneous with Windows 98 and Windows 2000 and for very late models, XP. Who's even running Windows 7 on a PIII?
Based on the timeline of the CPU, these would have been supplied with Windows XP, and probably still run XP behind a firewall or offline.
I'm not a member of the ReactOS project, just shilling as an alpha user who's impressed with the progress so far.
Yes, the project has been around for two decades, but it's made remarkable strides in the last couple of years. Give it two more—coincidentally when all support for Win7 ceases—and I think people will be pleasantly surprised by its usability. My only concern is that the Kremlin has dumped a bunch of money into the project, and I'd feel better if someone did an independent security audit of the code to see that Vlad didn't have some backdoor put in.
'He who has to break a thing to find out what it is, has left the path of wisdom.' -- Gandalf to Saruman
Doesn't change the fact that most malware is unknowingly installed by its users, involuntarily, likely because they are logged in with local admin rights that make it rather easy.
FTFY, because you know this. Yes, user error certainly does play a part, but these days it's rather easy to find a browser or plugin exploit, and that was before the browser market started growing and creating multiple vectors. Deception does not validate the use of the term "voluntarily", as no one would happily install malware if they knew what it was.
And that most Pentium III systems are likely running in specific, non-Internet-connected applications.
And probably haven't received patches in quite a long time anyway. A 20-year old PIII-powered system isn't running Windows 7, it's walking it.
I'll just leave this here.
* Large corporations quite specifically are maximizing the life of bought hardware and scheduling software purchased based upon a support cycle. Anything that upsets that can be a massive loss.
Large corporations that are still demanding support for 20-year old systems are not relying on a reasonable support cycle. They're acting like a bunch of whiny cheap-asses who would be willing to spend FAR more money fighting fine-print support bullshit in court than they would simply upgrading their hardware.
Actually, I was using the GP's numbers. Total Windows PCs is over a billion (of which according to MS as of April 2017 400 million were on Windows 10). According to some Steam Hardware Survey stats, ~0.44% of Intel CPUs are under 1.4GHz of which I'm not sure if that's a lot of Pentium 3s, misconfigured CPUs/incorrect answers, or what. Point is, it's not very hard to imagine a bunch of old PCs doing some PoS or other job for which they get monthly or quarterly patch updates and for which there are millions that exist.
Likely to be used as a botnet? No. Likely to be infected by malware by a user? No. Still officially supported until 2020? Apparently irrelevant to Microsoft. The rest is just irrelevant, really. As a company, you don't pull out of supporting something you promised to support without a good reason. Even if there is only one user on a Pentium 3s (which is highly unlikely), they're still obligated to continue support. Why is that so hard to understand? If Microsoft didn't want to offer support until 2020 under the listed required specs for Windows 7, they simply shouldn't have promised such. It wouldn't have been hard to have required SSE2 back when Windows 7 was released.
Sunk cost fallacy.
"It does not do to leave a live dragon out of your calculations, if you live near him." - Tolkien
What other OSes have been supported since 2009?
You're far from alone in that, my friend. A bunch of people I know got the same kind of treatment. I stayed with Apple and Microsoft for business reasons. I might finally start looking at Linux again soon. It won't be because I think the fanbois' manners have improved, but because I really don't like the "you don't own it, you only rent it" approach to software the big boys are taking.
I've calculated my velocity with such exquisite precision that I have no idea where I am.
I really don't take an issue with this to be honest. They're supporting an old OS for a heck of a while, that hardware is incredibly incredibly old.
You can continue to use Windows 7, you just need slightly newer hardware. It's not that unreasonable.
I'm a pretty archaic nerd but hanging on to ridiculous old hardware has held us back. Look at 64bit support due to crappy Atom chips and netbooks. Windows Vista and 8, should never have had 64bit. But Windows 10, STILL shipping in 32bit? Cmon enough is enough.
So if they support Windows 7 until 2020 only on one machine in Redmond that is in the basement behind the doors with sign "beware of leopards" then it means they kept their promise "to support Windows 7 until 2020"?
XP Embedded Standard 2009. Btw, you do realize this only reinforced the point, right? People buy Microsoft because of the long support periods. If Microsoft wants to not support products for such long periods or on such old hardware, they had the choice to not offer such. Now that they've sold the product as such, they should be required to follow through.
"Of course, the real lesson here is to cut Windows products out of your environment as much as possible."
What should they be replaced with that will be supported for 9 years?
This kind of thing always leaves me angry. A PC uses a lot of physical resources (metals, water, energy) to make, so planned obsolescence, via 'no patches' or 'version upgrade bloat' (a game that IBM used to play in the 70s and 80s with mainframe software to sell or rent more memory) is ecological vandalism.
Forget that little Billy is 'curing' malaria and thus helping his share portfolio etc., Microsoft is doing a lot of avoidable damage with this, basically to improve revenue and 'shareholder value'.
Happily, my computers are about ten years old and don't use Windows. Yes, I'm a Linux person, not particularly rabid, but also a green person, so I hold the same amount of hatred for Apple's repairable-with-extreme-difficulty hipster toys as well. Computers should last for a generation.
On y va, qui mal y pense!
I expect it to last the other 2 years that Microsoft claimed it would be; that's the early part. 2018 is before 2020, that is "early" by any accounting.
Brick the Windows 10 machines too and be done with it.
Now if only they'd cut off Windows 10 support for newer computers...
with the way some of these OS patches have been going, I would like to just subscribe to network and driver stack patches somehow.
win 10 patches.. are they even patches? who the fuck knows. it seems they just do them to promote their apps and to reset settings from people so they could be tricked to using them.
as far as the support for win7, well, dunno. ask your money back I suppose, they're obviously lying about their pr in regards to it.
also win7, wtf are the patches for again? network stack? if you got local binary running you might just as well have admin rights anyways.
world was created 5 seconds before this post as it is.
Do a Google search for: Windows 7 EOL and you will get the following:
Microsoft ended mainstream support for Windows 7 on January 13, 2015, but extended support won't end until January 14, 2020. This applies as long as you have Service Pack 1 installed.
So, Windows 7 is already well beyond the end of the normal support cycle for consumers, and the extended support is going away in another 1.5 years. If people insist on holding on to what is soon to become a dead end, then you run into the problems you see with Windows XP, where getting it to run on a new computer is problematic and requires a virtual machine, because there are no drivers for the newer components. Want to put antivirus or other programs on there, nope, they won't run on anything older than Windows 7 currently.
The longer you hold on to an OLD product, the more difficult it will be to migrate your programs/data, and at some point, you just won't be able to get your old programs running on newer computers. Then, you end up needing to really hunt for parts to fix your old computer. If you are on a laptop, it becomes even more difficult to deal with a hardware failure due to a lack of standardization in the components in a laptop.
I understand that many people don't like some of the things in Windows 10, but the bulk of those things can be removed or turned off, and it is worth the effort to modernize NOW, before you end up stranded and without a way to move to a new computer when your old machine needs to be upgraded.
Also most that don't fit your criteria? Running Win 7 Ultimate, AKA "Windows Pirate Edition". Look on your local CL sometime and see how many $50 POS boxes are running a $300 OS, although now its going to Athlon 64s and Pentium Duals running Win 10 since there are ways to run it without a key.
You'd be amazed how many think adding a new OS to a POS suddenly makes it worth money, they have been doing it since the days of XP when I'd see Win 95 era hardware struggling to run XP Pro, aka XP Pirate Edition, way back in 02.
ACs don't waste your time replying, your posts are never seen by me.
According to some Steam Hardware Survey stats, ~0.44% of Intel CPUs are under 1.4GHz of which I'm not sure if that's a lot of Pentium 3s, misconfigured CPUs/incorrect answers, or what.
The clock rate is likely to be accurate as its likely coming from the various CPUID strings that come from the CPU itself.
Clock is a poor proxy for chip generation. I have an Intel Core based Celeron running at 1.2 GHz and its not a mobile device. When you consider mobile devices you get even more "under 1.4 GHz". Intel still sells CPUs "under 1.4GHz" today, for example a Celeron N4000 at 1.1 Ghz base, a Pentium Silver N5000 at 1.1 Ghz base. If Steam happens to report a base clock rate rather than a burst clock rate you may be looking at such current dual and quad core CPUs.
Point is, it's not very hard to imagine a bunch of old PCs doing some PoS or other job for which they get monthly or quarterly patch updates and for which there are millions that exist.
"old" in a general sense yes, "old" as in Pentium III, no, that is too old. Pentium III's are going to be an aberration far far below 0.5%. Pentium III's do not constitute 1 in 200 computers.
As a company, you don't pull out of supporting something you promised to support without a good reason. Even if there is only one user on a Pentium 3s (which is highly unlikely), they're still obligated to continue support. Why is that so hard to understand?
That's a topic change. The criticism was about removing windows from your environment. The truth is every supported OS drops support for CPUs at some point and businesses have to be ready to adapt whether they are windows, os/2 or linux.
Why is it unreasonable to expect an OS to be supported until the date published by its developer a very long time ago?
Maybe those old systems are redundant or obsolete. Maybe they're not. I don't know, and neither do you, and neither does Microsoft.
If you disagree, post your argument. (-1, Overrated) isn't your personal censorship tool for views you don't like.
They're going to support it on a broad range of reasonable CPU's. Yes, that means some will no longer be supported, because the user base just isn't there or frankly needs to move on to something modern. The majority of people installing Windows 7 now are corporations who are slow to update to 10 and want to put 7 on new hardware until the last possible moment. NEW hardware. These are the people paying for support that is then shared with everyone else. That means bugfixes will eventually make running on older hardware impossible, and nobody's paying to support those ancient ass Pentiums.
Actually the indicator in the Steam hardware survey of a PIII would be lacking SSE2, currently Steam is showing 100% of computers surveyed as having both SSE2 and SSE3. Even then the vast majority are post Core2 even with support for SSE4.1 and SSE4.2. I honestly suspect this is a non-issue that affects primarily industrial applications, and honestly most of them are probably running XP without a care in the world. I was actually quite surprised by this story in the first place because I had been under the impression windows has required SSE2 since Vista.
Any and all content posted above may be ignored, considered irrelevant, or otherwise dismissed.
Probably less people running Win7 on a PIII than the number of people complaining about it on /.
Do you know anybody using WIn7 on a PIII, or are you just complaining because it is Microsoft?
Don't ask me, I most definitely would not do that. But these devices are out there and in use.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Not all of them, no. Development cycles can easily force use of a CPU already a few years old.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
You responded far more politely than I would have to a person is a living, breathing example of the kind of fuckwit who for years helped to keep Linux from being widely accepted.
I have to assume you mean linux on the desktop because if you are talking anything besides desktop PCs it's hard to imagine how it could be more accepted since it provides the core of Android and quite a lot of other mobile, server, and IoT devices. And the reasons linux fails on the desktop are mostly due to network effects of installed bases than anything else. People generally need/want to use what the people they interact with need/use and on the desktop that became mostly Windows long before linux was a thing. I'm sure you know the history as well as I do.
The fact that there were a bunch of asshats who would should "RTFM" at newbies asking legitimate newbie questions was a real problem but pretty far down the list of reasons linux never conquered the desktop.
Why is it unreasonable to expect an OS to be supported until the date published by its developer a very long time ago?
It's not unreasonable but it might be ill advised.
If it is mission critical and that old, then you've got problems anyway, but the usual solution is to take it off the internet and segregate it. All that medical stuff needs to be validated, and ensuring it can continue to be patched is part of maintenance. The system manager is probably thanking the heavens they have a good excuse to replace the system.
You are forgetting about embedded systems. An MRI-machine, for example, has a lifetime of > 20 years.
The companies that make and operate such devices are more than capable of paying Microsoft for support so that isn't really the problem. The real hassle is that the FDA certification process makes supporting such devices problematic but that isn't a problem caused by Microsoft nor one that Microsoft can solve.
And probably haven't received patches in quite a long time anyway. A 20-year old PIII-powered system isn't running Windows 7, it's walking it.
You haven't been following the official Microsoft propaganda. Windows 7 was advertised as being faster than XP, which was faster than ME, which was faster than Windows 98, all the way back to Windows 95 being faster than Windows 3.11.
Just like Windows 10 is faster than Windows 7. So fast that it should be able to start up in less than a second on 386-sx16.
If so, the development cycle would also force the use of an old version of the OS, in this case XP.
RHEL 5
SLES 11
Solaris 10
The shoocker for me was that I can't put any IBM OSes on the list (AIX, z/OS).
XML is like violence. If it doesn't solve the problem, use more.
SUSE Enterprise Linux or RedHat would be supported.
XML is like violence. If it doesn't solve the problem, use more.
One, a lot can change in 20 years in a community.
Two, Windows support communities can similarly be terrible.
The interesting difference is that for Microsoft, you can't (legally) be using it without some recourse for professional support. For Linux, you can either pay and have support or use it and have to resolve in the community.
Of course, based on what I've seen in MS community boards and formal responses, if you hit something 'weird', the community in MS has a better shot at figuring out than the officials from Microsoft...
XML is like violence. If it doesn't solve the problem, use more.
I will say that searching a forum/google can be rather spotty in terms of results. Often you can't find the prior thread unless you would use search terms that show you have the answer.
Also, speaking from experience, getting the same question over and over is a good indication for developers to know what they need to address.
XML is like violence. If it doesn't solve the problem, use more.
You can't upgrade the os on those systems anyway. The software combination isn't certified and the device manufacturer won't support it.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
Also with Windows 10, Windows 10 is 'eternal', but your device running Windows 10 can at any point have it's support yanked, but "windows 10" is still supported, it's just that your device can no longer get updates....
XML is like violence. If it doesn't solve the problem, use more.
Agreed. Who the fuck is still using a P3? They were released in 1999, we're almost at the two decade mark already. Throw that piece of shit away and drive around on trash night for something better.
Only the State obtains its revenue by coercion. - Murray Rothbard
I remember those days, using Linux with a hardware modem was awful. I had a 56k modem and could get 5kb/s in Windows but Linux PPP was super slow and big downloads would just stop. Never did figure out why it was shit.
Only the State obtains its revenue by coercion. - Murray Rothbard
The (non-mobile) P3 hadn't been updated for seven years when Windows 7 came into being. And it's doubtful any devices were built in 2009 that ran Windows 7, which at the time was a new, untested, operating system.
I would assume you wouldn't see an MRI or other safety critical device running Windows that run Windows 7 for several years after 2009, probably 2012 or later. And it's hard for me to imagine that anyone would manufacture a device after 2012 that used a chip that most probably wasn't even in production at that point - using older hardware is one thing, but using discontinued hardware with no support from the manufacturer is quite another. Where are you going to get the chips? eBay?
You are not alone. This is not normal. None of this is normal.
I'm predicting that support will be extended to at least five years past the deadline. There are two many big businesses with fleets of desktops that only upgraded to 7 after 8 was released.
Only the State obtains its revenue by coercion. - Murray Rothbard
1) Win 10 is not a real OS, it goes way beyond the mandate of an OS. 2) There's no reason why Win 7 could not get driver updates and patches for newer types of devices and live forever other than MSFT wanting to make more $$$ from Win 10
Embedded systems like MRI-machines that still run such old software should either not be networked (and thus not be updated so they're not affected by this move) or they should have been designed in a way that would easily allow upgrading any Pentium 3 to something not entirely ancient.
It's not a design problem. The problem is that the FDA only certifies the device for a very specific set of hardware and software so in most cases you really cannot change anything and still use the device. It's not a technical problem in most cases, it's a legal one.
And you really HAVE to network the device to do a lot of useful things unless you plan to do some sneakernet which is idiotic and wasteful. There are ways to do it which involve segmenting off the device to its own little private network but all of them are a pain in the ass.
From 2009?
They're acting like a bunch of whiny cheap-asses who would be willing to spend FAR more money fighting fine-print support bullshit in court than they would simply upgrading their hardware.
Some companies run multi-million dollar hardware/software to control their factory machines. An upgrade will have a multi-million price tag.
And what OS are they going to update too? Windows 10??
Operator to foreman: "The reason we're all sitting around on our thumbs, sir, is the Windows 10 controller PC is doing updates. Might be another hour or so."
This would also affect any Athlon XP systems as well, since they don't support SSE2.
Not that they shouldn't be replaced by now, but there would be a lot more of those systems running Windows 7 then Pentium III's simply because AMD sold them throughout the Pentium 4 Era and they were fast enough to run windows 7..
In Soviet Russia, Trojan exploits YOU!
Any Linux distros supported for free for almost 9 years? Any of you dorks planning to answer the question or is it tantrum time again?
Many, many people agree with what you said. Microsoft is shockingly self-destructive. A few of the many, many negative articles:
Windows 10 is possibly the worst spyware ever made. "Buried in the service agreement is permission to poke through everything on your PC."
7 ways Windows 10 pushes ads at you.
Microsoft again forced upgrades on Win10 machines specifically set to block updates (March 12, 2018)
Yes:
https://en.wikipedia.org/wiki/...
RHEL5 released in 2007 and is in extended support until 2020.
https://en.wikipedia.org/wiki/...
SLES11 came out in 2009, LTSS ends in 2022.
For fun, Solaris from 2005 is supported until 2021:
https://en.wikipedia.org/wiki/...
So while OSX and community distributions and Ubuntu LTS don't go that far, there are OSes that do. This is one reason why people snicker at Ubuntu's proclamation of 'Long Term Support', when the competitors have so much longer support cycles.
Of course, I wouldn't wish a RHEL5/SLES11 desktop on anyone, they are missing so many features in the current distros. Then again, XP also was pretty pathetic desktop experience wise when it was still popular relative to contemporary Microsoft desktops.
XML is like violence. If it doesn't solve the problem, use more.
Many ATMs run Windows 7.
Linux came out between versions 3.0 and 3.1 of Windows.
Linux was nothing more than a hobby project for several years though a promising one. I was in college when it came out and for desktop use it wasn't even remotely competitive until after Windows 95 dropped. The only people to touch it were the most serious of unix geeks who loved it. (myself among them) The earliest distros were useful but weren't even close to ready for use by the general public.
Lots of PCs were coming with Windows at that point, but most software was still DOS (since Windows was an app rather than an OS) and very few file formats required non-DOS software.
Windows after version 3.0 was not an application. To call it one really misrepresents how it worked. It was really an OS layer that ran on top of DOS. Calling Windows 3.1 an application layer is as incorrect as calling the World Wide Web the Internet. And if you think there wasn't a lot of Windows specific software in 1991 (when linux 0.1 dropped) then you weren't there.
I understand your point, but you should make it without revisionism.
Not a problem since there isn't any. I used some of the earliest versions of linux that came out when I was an undergrad. Linux was in no way, shape, or form ready to supplant Windows on the desktop at that time. By the time it was ready, Windows 95 had already dropped and the game was effectively over for control of the PC desktop from then on.
Large corporations that are still demanding support for 20-year old systems are not relying on a reasonable support cycle. They're acting like a bunch of whiny cheap-asses who would be willing to spend FAR more money fighting fine-print support bullshit in court than they would simply upgrading their hardware.
I can't disagree, but you are missing the forest for the trees.
Whether you think they are acting like crybabies, or someone else is pissed that support is being dropped a couple years early, the act is what it is.
It simply shows that putting trust in Microsoft is one of the acts of stupidity you can add your your rant:
Capital equipment that can be controlled by computer often lasts for decades.
The computers controlling it can also run for decades
The operating system for those computers won't be supported for the length of time the manufacturer of the OS said it would.
This might have an impact on the decision for the next controllers purchased.
If you have a bit of machinery that is expected to last 50 years, you might look at a PC running Windows 10 as a controller to be a bad decision, and build a custom controller that will be expected to last as long as the machinery.
The shepherds did so well protecting the flock that the sheep no longer believed that wolves existed.
> Some companies run multi-million dollar hardware/software to control their factory machines. An upgrade will have a multi-million price tag.
Sounds like something that should be set in stone and not subject to the whims of CONSUMER ideas on how systems should be updated.
This is an installation that shouldn't be set up to be vulnerable to the usual sorts of nonsense that plague Windows machines. Doesn't matter how new of a version of Windows they are running.
A Pirate and a Puritan look the same on a balance sheet.
It's amazing to me how often Windows is 10-20 years behind Linux in such basic features. Here's a screenshot from Gnome 1.0 in 1999. The 2x2 panel at the bottom left is for switching between virtual desktops (workspaces). It was included in 1.0 because it was considered a basic feature in Linux by 1999.
https://en.m.wikipedia.org/wik...
And it makes for a real interesting conversation for what you will use for your next generation machinery controllers.
Yeah I'm kind of amazed how little consideration purchasers of this expensive equipment typically give to these sorts of important questions. I run a manufacturing company and I'm SUPER careful about investing in software or hardware that I think even has a chance of not being supported in future years. If something runs Windows that's potentially fine if I'm only expecting the equipment to have a service life of 3-5 years. Anything longer than that and there had better be a very specific service contract involved or it needs to run a system that doesn't depend on a third party for support. The presses and other heavy equipment we use in our company have software written by and for the company that sells the press and they can support it 100% for the expected life of the device. No third parties are involved and that's to our benefit.
Microsoft has decided they don't have to offer support for as long as they say they do.
Which is a risk you take whenever you depend on a third party who is not a signatory to the equipment purchase. You're basically making an assumption unless someone who actually works for Microsoft is in the room and signs a commitment obligating Microsoft for support.
No. It could, but these are not tied together.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
VMs on endpoints were not as viable a solution when XP EOL'd as they are now so while I don't recommend this path,
Windows 7 runs fine in vmware under Linux, or under windows for that matter. Even 3d graphics work reasonably well these days. My PC has more than enough oomph to do the job.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
Inventory. Some companies stock spare parts for decades.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Bingo.
BTW, was this change in support reported by Microsoft, or just inferred from secret documents a third-party claims to have seen?
The patches offered for Windows 7 today are very infrequent, almost non-existant, and to be honest, unless your PIII computer is supporting a browser and is connected to the internet, Windows updates are likely irrelevant at this point - the vast majority of OS bugs were resolved long before this support rumour started.
Ken
What patch is likely needed after 10 years of updates to Win 7?
I'll play along, I've got a 20 year old medical infusing device running on PIII hardware, and the manufacturer updated the os ten years ago to Win 7, and the machine has run find for nearly 20 years, first running Win XP, then ten years later to Win 7. Unless my medical infusing device (which largely predates the public internet) now relies on internet access to function, what is the isdue?
Ken
Back in the earlier days of Linux, the problem often tended to be that all the "RTFM" and "do a search before posting" responses would overwhelm the one forum thread that everyone was telling you to search for to find out how to actually fix the problem.
The truth is every supported OS drops support for CPUs at some point and businesses have to be ready to adapt whether they are windows, os/2 or linux.
While true, it raises the question of why. In OS/2's case for example, the minimum currently is i686 (Pentium Pro) due to so much current software needing the atomic instructions that weren't available on older CPU's and in this day of multiple cores, SMP support is important.
Generally with SSE or other simd instructions, there are slower code paths that can be easily taken that don't depend on simd instructions.
https://en.wikipedia.org/wiki/Inverted_totalitarianism
Somewhere Mozilla has a lot of info on which CPU's Firefox users are using. I watched development for quite a while and every time they considered having SSE2 as a minimum requirement, they were surprised at the number of machines still not supporting SSE2, mostly Athlon's I believe. The numbers finally got low enough that after 52ESR (soon), the minimum is now a Pentium M. It's also the end of XP support.
I have a T42 here which I still use now and again, nice piece of hardware for its time. It's a Pentium M at 1.4 GHz and runs stuff satisfactorily. Things like Firefox take a long time to start but run fine once started.
https://en.wikipedia.org/wiki/Inverted_totalitarianism
"Linux Set To Shed Nearly 500k Lines Of Code By Dropping Old CPUs"
This is not comparable.
Linux dropped support for very obscure CPU architectures. Consumers are unlikely to possess these architectures at all. Corporate customers can still get the code from old Linux versions which are still available online under the GPL.
Microsoft is dropping support for extremely mainstream architectures which still have millions of users. And there's no alternative to download the code and recompile it in the future.
They're acting like a bunch of whiny cheap-asses who would be willing to spend FAR more money fighting fine-print support bullshit in court than they would simply upgrading their hardware.
Some companies run multi-million dollar hardware/software to control their factory machines. An upgrade will have a multi-million price tag.
The newest Pentium III chips are almost twenty years old now.
You're either smart enough to bake planned obsolescence into your business decisions, or you're stupid and ignorant enough to believe you don't need to.
And what OS are they going to update too? Windows 10??
Windows 7 to Windows 10 isn't exactly an unproven or impossible upgrade path, but ultimately this would depend on the manufacturer. Those who are manufacturing multi-million dollar systems running Windows should also know a thing or two about planned obsolescence and the OS vendor they're reliant on.
Who the fuck runs Windows 7 on a goddamned P3?
Still waiting on Serviscope_minor to wake up to fucking reality and realize that Jessica Price isn't going to fuck him.
Doesn't change the fact that most malware is actually installed by its users, voluntarily.
Wow, I remember the 1990s. Good times.
Meanwhile, in this decade, malware is almost entirely browser-focused, and installs via malicious ads. I guess you could claim that users "voluntarily" had javascript enabled, if you were a silly person.
Socialism: a lie told by totalitarians and believed by fools.
Perhaps Athlon's that never supported SSE2 and was manufactured to 2005, so perhaps still being sold new in 2006.
https://en.wikipedia.org/wiki/Inverted_totalitarianism
When you buy a machine, it comes fully functional, and as long as the 100 million dollar device doesn't get hacked into the internet, and is instead run in a closed ecosystem, it can run forever.
Expecting a machine that runs a specific version of Windows to still be running 20 years later is extremely foolish. If the company that sold you the machine doesn't have access to the source code for everything on that machine then you are playing with fire. Maybe you'll get lucky but only a fool depends on luck with big capital purchases.
Where are you going to get the chips? eBay?
Craigslist.
I tend to rant.
Window snapping, 'expose clone', virtual desktops, the nice window switching enabled by compositing (mostly eye candy, but there is some utility).
Non-functional wise, I hated the plastic design language, but at least you could go to Win2k style if you wanted. Shame about windows 10 not offering that or otherwise having a decent look to it.
XML is like violence. If it doesn't solve the problem, use more.
Thankfully some other distros have ignored that and the dreaded systemd disease.
Artix
Your Linux, your init.
For comparison purposes, consider that a Pentium III, at its peak, went to 1.4 gHz (wiki), and were no longer made after 2003.
I have a couple of dozen computers, with the oldest one being a Q6600 -- came out in 2006. I readily admit I have pushed it several years beyond its useful life. I think I do it more as a challenge and curiosity than anything else. Needless to say I don't update it -- those broke some 8 or 10 years ago.
I think I am comfortable with letting go of PIII's.
I come here for the love
The truth is every supported OS drops support for CPUs at some point and businesses have to be ready to adapt whether they are windows, os/2 or linux.
While true, it raises the question of why. In OS/2's case for example, the minimum currently is i686 (Pentium Pro) due to so much current software needing the atomic instructions that weren't available on older CPU's and in this day of multiple cores, SMP support is important. Generally with SSE or other simd instructions, there are slower code paths that can be easily taken that don't depend on simd instructions.
OS/2 may be used in more "embedded" applications than Win7. PIII min would possibly be a larger issue in the OS/2 community than the Win7 community.
"Linux Set To Shed Nearly 500k Lines Of Code By Dropping Old CPUs"
This is not comparable.
Linux dropped support for very obscure CPU architectures. Consumers are unlikely to possess these architectures at all.
When it was announced various users did complain. "Obscure" is not constant across platforms. With Linux being a re-use path for old PC hardware dropping a PIII would be a larger issue in Linux than in Windows. Windows boxes actively used still running on a PIII based system may be as rare as the platforms dropped by Linux. Things are a bit more comparable than you suggest.
"very obscure CPU architectures for which no maintainers would step up, and nobody had promised to maintain in exchange for money for a fixed amount of time.
Nobody else could, even if they would want to, step up and keep Microsoft's code up to snuff, nor was any such help solicited. And why would anyone? Microsoft took the money and promised support until 2020.
Some problems are in drivers and a driver for an obsolete unsupported version of Windows is quite doable.
Don't ask me, I most definitely would not do that. But these devices are out there and in use.
I've seen references to X-Ray and MRI user consoles in other responses, I don't think of them as "embedded", rather as "industrial control". Perhaps its personal bias, I once worked in telecommunications and the big expensive boxes we sold had plenty of embedded software (my part) and an industrial grade PC running the manager's interface software that generates reports and lets them configure the embedded software. The manager's software was desktop OS. I expect something similar (embedded and PC-based user software) in the X-Ray, MRI, etc. I consider the distinction because the PC based user/manager software is far more replaceable than the embedded software it communicates with. The embedded could care less if the managers' box is WinXP, Win7, Win10, OS/2, Linux or Mac. The manager's box was bundled with the system, sometimes a standalone PC and sometimes a PC installed in our equipment's rack, yet it remained a replaceable PC at its heart.
Many ATMs run Windows 7.
I thought that was OS/2, the world finally moved on? In any case that's more industrial control than embedded, and the PC component of the system may be upgradable/replaceable compared to the reset of the system. In telecommunications software I once worked on the big racksized box with all sorts of processors and embedded software inside also had an off-the-shelf industrial grade PC running a desktop OS with the user's interface software (reports, configuration). This PC board and its hard drive could be replaced and the embedded software could care less what OS it was running. I'd be surprised if an ATM's "PC" hardware is not similarly replaceable and the OS it runs similarly irrelevant to the actual embedded software. Much like with the ATM's transition from OS/2 to Win7.
Generally the need to keep these around comes from the combination of a lack of budget to get an alternative running, and the lack of knowledge of how to get an alternative. Ie, a lot of those old must-have DOS programs can be run in an emulator. And a lot of those single-source third party production codes should be thoroughly documented when first installed, backups made, etc.
One problem is that something that seems like just a temporary solution very often becomes mission critical over time.
What it really means, is that Microsoft doesn't really support its operating systems by any reasonable definition of "support".
My experience back when I was beginner - and even these days if I run into something I can't figure out - is to list all the references I looked into before posting.
So not
"I have a problem X, please help"
but instead
"I have a problem X. I have looked into this document (link) and this FAQ (link) + found some bug reports (link, link) that appear to have relation, but cannot figure it out. Could you please help".
Seems to go a long way to get rid of the "RTFM" stuff. Of course, asshats remain, but if you show that you've tried to solve the issue yourself the RTFM responses seem to go down quite a bit.
Whoever runs Win7 on a Pentium III or below.
Just kidding, I pity whoever runs any version of Windows on any hardware.
It is pitch black. You are likely to be eaten by a grue.
Thanks for sharing your experience! ReactOS seems very much alive (e.g., they have a GSoC intern adding BTRFS support). If you wish to hasten its production-ready status, why not join the project? ;)
'He who has to break a thing to find out what it is, has left the path of wisdom.' -- Gandalf to Saruman
from the post:
Someone has managed to install Windows 7 onto a 266 MHz Pentium II processor, 96 MB of SDRAM memory, and a 4 MB video card. But even a Pentium III system took 17 hours to install Windows 7, and it takes 17 minutes to boot the machine. Someone else claims to have... "installed Win 7 RC on a Pentium III 850 MHz notebook with 512 MB RAM and 100 MHz FSB in slightly less than 1 hour and it works exceptionally well." Monday, July 06, 2009 7:49 AM
https://social.technet.microso...
I see in the near future... 1 - Group of hackers list the names of libraries / executables with SIMD extension 2 instructions 2 - Another group develop patches for disk or a daemon for memory patches 3 - Windows become supported in old hardware :-)
Well, as long as a PIII still runs Win98, my retro gaming rig is intact. I can't say I even THOUGHT of running Win7 on a PIII. I think it has an XP partition, but that's it.
The day we post that Windows 10 won't get updates on certain devices because support is suddenly yanked, conflicted minds all over Slashdot will be blown.
This already happened.
Intel Clover Trail based platforms got end-of-supported in 2017.
Such devices started life as Windows 8/8.1 devices. When the users were upgraded to Windows 10, they actually got their support *shortened* in the process (Windows 10 moved to 'lifetime of the device' rather than a period of time and MS decided those devices were 'dead', because Intel decided it wasn't worth supporting MS in doing so).
XML is like violence. If it doesn't solve the problem, use more.
Heck, I'm thinking about restoring a TI-99/4A for my mother to use, albeit with a Raspberry Pi instead of a P-Box. A 30 year old machine that hasn't had an official manufacturer OS update since 1983.
But then again, it is limited to telnet.....
SJW: a person who perceives an injustice, and while correcting it, commits a greater injustice.
Well, let us know when Microsoft rolls out an automatic MSDOS update to 5.0 that cuts support to Pentiums without MMX support.
"Let's gamble by making our device dependent on a part that might or might not be being stockpiled by a third party for a decade" said nobody responsible ever! (Mind you, if they were responsible, would they make a consumer desktop operating system the UI/controller?)
You are not alone. This is not normal. None of this is normal.
It seems reasonable to me that if they promise to support some software until a certain date and give me a list of hardware that the software is supported on, that the support date applies to any and all hardware on that list. It's not like I managed to get Windows 7 running on some old CPU that was never supported, and Microsoft broke that (in which case... tough). This is hardware that Microsoft has officially stated Windows 7 supports, and they have also stated that Windows 7 is supported to 2020. Apparently which is no longer true.
well they will push windows 10 to any computer running windows 7 to 2023. ..otherwise, they don't care.
world was created 5 seconds before this post as it is.