Intel end-of-lifing BIOS and driver downloads for dusty hardware Intel is removing drivers and BIOS for its old desktop boards so anyone running an old Pentium-based PC has four days to get hold of anything they might need. The warning on Intel's download center page says: End Of Life - This download, BIOS Update [RL86510A.86A] P21, will no longer be available after November 22, 2019 and …
I'm sure we can all pitch in a few dollars for some disk drives so they can keep storing those old files.
I get that vendors don't want to keep actively supporting hardware that they might not even possess any more, but I cannot fathom why any manufacturer would actively remove download files from a backend system unless the sheer number of files involved manifestly impacted the performance of said backend.
It's such a shame that hardware that is still functional (old scanners, printers etc) that still have a useful life might have to be ditched for want of the proper drivers, technical documentation or user manuals. Let them shift the files to a secondary system that is slow but has a massive hard drive on it.
As a show of friendship, maybe the Raspberry Pi Foundation might like to help out poor, starving, Intel by posting them a Raspberry Pi with a suitably large SD card so that they can keep the ancient BIOS repository up and running?
I expect it would cope with the server load more than admirably…
And also that might make money
Geocities?
The website for epub specs erased when that crowd merged to the Web lot.
Some bosses just like clean desks and no dusty files in the cabinets.
The BBC used to destroy 16mm film and "save" £85 per episode recording over tapes when better practice and a fraction of production cost would be new tapes for each recording.
Probably a decision made by a manager or accountant that has never had to unexpectedly refurbish, restore, repair or commission old gear.
This will cost more in operator time than than leaving it alone and the risk that something important is removed.
Rules of IT:
Make backups
Make Archival copies that are not overwritten
Never delete archives.
Never EVER delete production files, ever, unless you have an entire system backed up, verified the backup restores on another box and someone is checking what you are doing.
rm is not your friend
rm -rf / is the devil's tool. Especially with --no-preserve-root
rm -rf / is the devil's tool. Especially with --no-preserve-root
It can, however, be quite fun when an utter cunstomer has absolutely assured you that they have backed up all their data and no longer need your service, and you are NOT to retain any backups you may have of their data.
Yeah, old kit having to be ditched because of no support.
Had to chuck out my old HP Touchpad tablet last year because even though it worked fine I'd tried to do a factory reset on it and it uses the wifi connection to phone home to authenticate against HP's servers - but HP had turned them off something like two or three years previously so it was essentially bricked. A shame because it was a nice piece of kit.
I agree wholeheartedly.
HP have to be one of the worst for this mercenary behaviour.
I have had to junk a perfect flatbed scanner and now I'm being told the latest Bios updates on my 250 G4 laptop require a new motherboard!
It appears to be sop by HP and it seems the only option is NEVER to buy their products.
An expensive lesson to learn but one I will adhere to.
HP have pissed on my chips for the last time, and I'd recommend to everyone to keep their chips away from them also!
I had a very nice HP gaming machine that randomly restarted all the time, twice whilst in the middle of my radio show and once in the middle of a DJ set it was about time for me to ask HPs help. To be honest their customer service was dire to say the least, after 4 months of going back and forth plus being unable to use my laptop I has paid a pretty penny for I gave up.
I had been a firm HP fanboy for 15 years, due to my bad experience they lost my loyalty to them. I now have an Acer which I have slightly upgraded but have not looked back since. I find Acer's customer service much more welcoming and helpful, though I have never used them for anything complex I must admit.
Not in my experience.
Even the last few years of IBM owned laptop production (when IBM owned the name but Lenovo actually made everything) produced many turkeys. I once had to refurb a few hundred Thinkpads from that era and they all had broken plastics around the keyboard and a nice big blotch in the middle of the screen where the nipple rubbed.
I'm sorry but anyone who still thinks Lenovo is anything special at all is living in denial.
Don't forget superfish either.
I worked for a very large company that decided to move to HP. The 1st print job on our new HP All-in-one was my resignation letter.
Never will I let that company darken my doorstep again.
Ironically, I think HP make good kit, their customer services and support on the other hand......
Oh I don't know - I've came across a couple even worse.
Though trying to find the drivers for an HP which happened to have been bought originally in Portugal, then outfitted with a UK keyboard, was an experience. According to HP's website that model didn't nor had it ever existed - until I looked at HP Brazil's website!!!!
For that reason I will not buy HP PCs - ever!
ASUS is another. As soon as they discontinue a product, i.e. stop selling it, it's like they try and erase it from history. I've had that with another laptop (not mine TF!), so that's another manufacturer who's PCs I won't go near with a galactic length barge pole.
Listen up manufacturers - this sort of attempt at rewriting history does NOT encourage us (well me) to buy your new kit. It just ensures I will NEVER buy your kit.
and the c***s still wouldn't release the bootloader, even though they scrapped the whole tablet line, and stopped all support.
A shame, cause I have two of those, sadly, foolishly, I let them be updated via wifi. Which automatically made them unable to be rooted and "alternative/newer" systems installed. Other than that, they work f... great :(((
I have a perfectly serviceable Canon printer, which only works with 32 bit drivers. I cannot use it on my modern computers, because they are all 64 bit, and no 64 bit drivers exist for the printer. Guess I'll just have to keep that old 32 bit laptop running connected to my LAN to act as a print server. :-(
I'm sure we can all pitch in a few dollars for some disk drives so they can keep storing those old files.
Indeed. If Intel don't want all their old IP they could just Open Source it all ...
Methinks the real reason may be to encourage people to ditch the old hardware and buy shiny new stuff.
I get that vendors don't want to keep actively supporting hardware that they might not even possess any more, but I cannot fathom why any manufacturer would actively remove download files from a backend system unless the sheer number of files involved manifestly impacted the performance of said backend.
It's like Crapcast/suXfinity deciding to *delete* their old forums when they (presumably) set up a new system, rather than simply archiving it to an older server. There was still plenty of relevant information on it. Perhaps because there were plenty of customer complaints, and they wanted to claim there weren't any complaints on the forums? That's the way LeapFrog dealt with customer complaints on the forums; they deleted the forums.
Just how much demand is there for 20 year old Intel motherboards?
I would have expected Intel motherboards to make up <1% of the market once OEM/third parties were taken into account and OEM/third parties to be more likely to still have stock if a unit failed.
Or does anyone else have more experience of Intel mother boards than I do? I don't believe I have ever seen one outside of a review.
Is it?
We lose significantly more important pieces of computing history than this with barely the blink of an eye.
My suspicion is that Intel P4 BIOS files are an insignificant piece of history and that numerous equivalent files are available from the likes of HP/Dell/Lenovo/ASUS etc that actually made up the majority of the market.
While the argument against holding a few MB of files for a few more years has some value, I'm not convinced it's a significant value. If it turns out there are more Intel motherboards out there than I thought (my optimistically high 1% of desktop sales over 3-4 years would amount to total sales of around 1 million units of which >95% will be in landfill) doesn't leave a lot of potentially affected users.
Losing P4 BIOSs before they are fully emulated in something like QEMU? Bad but as long as hardware exists, no worse than pre-flashable BIOS hardware that required other steps to get a BIOS image. If the hardwares gone but other P4 hardware from other manufacturers exists and has BIOSs available? Well there's no real change for the vast majority of users.
Just how much demand is there for 20 year old Intel motherboards?
I still have a few machines in service that are almost that old, but they're Gigabyte motherboards.
Some of them still have 10k Maxtor SCSI HDDs too, these are original fit so the same age. They're all running XP (no external access).
No, not really critical things, we can replace them fairly promptly in most cases, though there are one or two which have rather ancient hardware attached which we have not made work successfully on the machines we're currently building. There is a power saving argument for updating too. The original systems, fitted with 3GHz Pentiums, consume somewhere north of 120W even when idling. Our current fit takes less than a quarter of that, though it ramps a bit under load.
M.
"I still have a few machines in service that are almost that old, but they're Gigabyte motherboards."
This is effectively my point - I have HP/IBM machines of a similar era and a google search suggests that Intel motherboards around this time were reference samples for testing/base designs or used in appliances for some vendors. i.e. the likely number of affected users is likely in the tens or hundreds if I'm being optimistic.
I'm not saying ditch all support from all vendors at 20 years because the majority of the market is continuing to support older products (unless I'm just lucky with HP and Lenovo), I'm saying a very minor player in the motherboard market 20 years ago removing BIOS support isn't something worth worrying about....
Unless someone actually does use these Intel motherboards and I'm wrong, but I remain unconvinced that such unicorn poo exists...
That is clearly the case, but suggests people didn't read what I wrote - I asked how many people had P4-era Intel motherboards.
The majority appear to have read that as motherboards supporting Intel CPU's or something similar. The joys of asking technical questions on a general forum I guess.
And a few fake Internet points never hurt anyone although it does make me wonder if there is anyone out there besides a single Reddit user who was affected.
Rounding errors...quite common on Pentiums in those days... (the impact of FDIV for me was 1995, these motherboards appear to be from 2003-2004 so I'm out quite a bit...)
The archive is a little slow to download (70GB @ <<1MB/s for me) but the directory listing appears to include chipsets in the 845 to 915 era all of which were mainly P4s but also included support for Celeron and Pentium EE CPU's - I say appears to as some of this is documentation only and Intel have retained all the documentation from what I can tell:
https://ia800303.us.archive.org/33/items/2014.01.download.intel.com/2014.01.download.intel.com.tar.txt
This post has been deleted by its author
I assume the parent post was suggesting that I hadn't been around long enough to notice "old hardware" when the case is that I have been around so long that most of this stuff fits in the "recent history" category. I'm still younger than some though having only started on VAXen.
And still no one has owned up to having a P4-era Intel branded motherboard.
The "20 years ago" appears to have been 15-17 years ago, so it was the end of the Athlon XP years and the start of the Athlon 64 years.
Or more memorably - when Intel realise Rambus wasn't the way forward in the consumer market and released chipsets supporting DDR and then SDRAM.
If that is the BIOS for the 440BX motherboards, those things were awesome pieces of kit back in the late 90's and early 2000's. Rock solid, stable, reliable. Maybe not the fastest, most cutting-edge board available, but damn, they just kept going and going. I wouldn't doubt that thousands of them are still chugging along in service. I sold probably a hundred of them in various systems, and issues were scant.
Although, I guess the better question is why would folks still need a BIOS update this late in the game? I would think that the folks still using and maintaining systems using these aging boards would likely have already updated the BIOS, or would have several copies of Intel's update on thumbdrives and CDs scattered all over the shop.
Re: Just how much demand is there for 20 year old Intel motherboards?
Probably not much. Where I work, one of our suppliers pretty much exclusively uses Intel motherboards, and we probably do have PCs in storage somewhere with motherboards that are 20 years old. We are going through our storage recycling stuff we can, so they may well have gone.
We still have a PC with a 15 year old Motherboard in use though. It runs XP Pro, and has been removed from the network. We can't get rid of it though, as it's connected to a machine that requires custom software that has been specifically configured for that PC. The machine would cost around £35,000 to replace and apparently the cost to upgrade the PC is not far off that. Well, the cost of the PC itself is probably less than £1,000, what costs the money is getting the company who provided the machine (some sort of medical scanner) in to configure the new PC.
Or does anyone else have more experience of Intel mother boards than I do?
I built a few PCs around 20 years ago. At the time Intel motherboards were regarded as the "gold standard", but they came at a price. I always went for reputable third-party brands like Asus and Supermicro.
"I built a few PCs around 20 years ago. At the time Intel motherboards were regarded as the "gold standard", but they came at a price."
I accept there were reference boards and were expensive, but I'm not sure about a "gold standard" - while they we very reliable (I've seen a few in appliances like PIX firewalls with non-flashable BIOSes and suspect other appliance products at the time were similar as they fitted with Intel's bulk sales model), they lacked any features above the base platform (i.e. additional USB ports via bridges, additional IO slots, the majority of the tweaks for improving memory/CPU/IO performance) and by and large they weren't available via retail.
I don't think there's any doubt that the "cost" of hosting these old BIOSs is insignificant. The "cost" is probably more in the management of the website. If you're maintaining a site as large as Intel's, you want to keep it as trim possible so you stand a chance of it all working.
We're all geeks here so I'm sure we've heard about "code bloat". Well, the same thing happens with websites too.
I guess that would make sense if Intel didn't make motherboards that required BIOS packages any more such that an entire class of product was not longer relevant to the business unit.
However, that doesn't seem to me to be the case. If the problem is that the support pages' "motherboard model" drop-down is getting unusably long, then they probably have a website design problem anyway.
"The "cost" is probably more in the management of the website."
What Intel only needs to do is place all files on an FTP site, and keep a FILE_ID.DIZ (I'm nostalgic) with description for all downloads. Keeping all the BIOS and related files for all Intel motherboards (pre-2010) would take what, a couple gigabytes?
NB: my Intel Zappa (1994-1995?) was a great motherboard with a 3rd party BIOS. The stock BIOS was limited to 8GB hard drives, because Intel didn't bother to code LBA support in it. The MRBIOS update had that and many other features as well.
I came here to suggest stuffing them on an FTP site too ... Simply mark it as a never to be updated archive and UseAtOwnRisk, and allow other sites to mirror it, and be done with it. Or better, add to it over time as systems hit EOL. If more companies did this, us tinkerers and restoration nerds would be a lot happier!
For a good example, see ftp://tuhs.org/
I rather suspect the overhead for such an FTP site would be a LOT less expensive than the loss of goodwill they are going to receive for sending it all to the bitbucket in the sky at the other end of the big green earth ground.
This post has been deleted by its author
If you're maintaining a site as large as Intel's, you want to keep it as trim possible so you stand a chance of it all working.
I don't think that's accurate.
Many moons back I did my first SQL-based website for someone. And my eyes were very widely opened to the scalability of the backend. The same queries written for a few dozen items could easily manage hundreds of thousands of items. The database could actually contain small files (eg photos), and the front end was some simple PHP that populated the relevant portions of the web pages/forms as needed (so if for example you select "meat products" then the next page is populated with data related to meats).
There's not a lot of code really to let a user put in a search query (eg a model number) then pass that off to the backend and retrieve the data and display it. With decent use of site design and CSS you don't need to do any work for anything other than the primary page.
And look at other larger sites. How many "whitepages" or "yellow pages" have millions of entries? Does Intel really have that much data on its products?
Yes, there is a data cost - not only the size of the files but also backups. But as another poster said, it's going to cost more in the effort to purge the system of this data than it will be to keep it.
Unless their site design practices are worse than mine. In which case, I think I shall be making more use of AMD products in future (I do have one or two Intel CPUs still, not due for retirement for at least another 5 years)
"The database could actually contain small files (eg photos)"
Your contention that you'd have easily scaled it to 100s of 1000s of items falls apart right there. Sure, if no one was looking at the site. You'd quickly have found it falling apart with the traffic levels a site with 100s of 1000s of products gets.
"The database could actually contain small files (eg photos)"Your contention that you'd have easily scaled it to 100s of 1000s of items falls apart right there. Sure, if no one was looking at the site. You'd quickly have found it falling apart with the traffic levels a site with 100s of 1000s of products gets.
Hmm.. Better tell all the sites that use a similar setup and get large numbers of customers that their systems are falling apart even though none of the thousands of hourly customers have complained! (they probably use more than one server and some form of load balancing, but I'll bet each server has a full copy of the DB)
Oh, and considering we're talking updates for a very old MB, I doubt there's going to be that much traffic - but if you think so there's other posts with people suggesting donating a RPi to Intel so they can use that as their server for the old kit - those posts need your expertise to correct them as well.
Me... I have a couple of owncloud servers, don't get much traffic (just me and 3 or 4 family/friends). The files in them are around 500GB/each and around 200,000 items each. Admittedly few of these files are actually contained in the DB but it only takes the piddly little laptops (Tosh M200 running a 32bit Debian (due for retirement) and Dell D630 running a 64b Devuan, each IIRC with 2GB ram (though the Tosh may be 1 or 1.5gb). The Dell also runs OpenVPN for me when I'm on otherpeople's WiFI. If/when I get Fibre on, then I may see traffic levels increase but I doubt the machines will get anywhere near full CPU or RAM use even with the line maxed out.
(Of course, if you can actually back up your claims... (article from 2000, first paragraph is most relevant as we're talking static pages)
Well, fuck you Intel.
The Reddit post linked in the article contains downloads which combined weigh in at less than 2MB. Two fucking megabytes.
The last version of their new UHD drivers for Windows 10 64 alone weigh close to a fucking gigabyte:
https://downloadcenter.intel.com/download/29195/Intel-Graphics-Windows-10-DCH-Drivers?product=122139
Also, these older drivers require no maintenance, no updates, nothing. I really don't understand the need to remove them.
In a perfect world, maybe, just maybe, Intel has a count of downloads per file and date of latest download.
Just for a minute, let's imagine that the last download date of said BIOS files was, oh, November 12, 2003.
Maybe, just maybe, sixteen years wait time for no new activity is enough to consider that Pentium II BIOS downloads are not really something worth hanging on to ?
Of course, all of this depends on what the last download date actually was, and if Intel is actually tracking that.
That being said, there is one problem : Pentium II motherboards are still on sale at Ebay - if not elsewhere. I hope they have the latest BIOS version.
"Just for a minute, let's imagine that the last download date of said BIOS files was, oh, November 12, 2003."
I downloaded it about 3 months ago (replacing hardware on a Bridgeport mill). I seriously doubt I'm the only one. I'll bet there has been at least a download a week (average), if not several per day.
It's not like you to play the devil's advocate. Presumably the last people with any sense at Intel have retired or the few that are left aren't enough to outweigh the critical mass of stupidity and the lunatics have finally taken over the asylum.
On a similar note, go Ryzen 9!
The Reddit post linked in the article contains downloads which combined weigh in at less than 2MB. Two fucking megabytes.
The last version of their new UHD drivers for Windows 10 64 alone weigh close to a fucking gigabyte:
Interesting. A video card back then would have had about 2MB video RAM, compared to one today which would have something of the order of 4-8Gb.
Apples and oranges I know, but at a simple level would you not expect driver complexity to increase roughly proportionally with the complexity of the device.
Interesting. Yet another reason why we shouldn't use this technology. I am still running my steam engine from 150 years ago. But my computer from twenty years ago will no longer be maintainable by by grand kids.. Wish I would have known that when I bought it. Would have kept my abacus. I could still get support on that.
The entire download archive was mirrored by archive.org a few years ago, so you don't have to resort to downloading unsigned executables from totallylegitdrivers.ru for all your driver needs:
https://archive.org/details/2014.01.download.intel.com
Jason Scott, of the Internet Archive, is working to ensure that that mirror is current before Intel makes it go away forever:
https://twitter.com/textfiles/status/1196440682401058816
The Internet Archive is an invaluable site. The big problem with the Internet is just how easy it is to remove a site and have that information promptly lost forever.
If you think back 50, 100 or more years, much of our understanding of news, technology etc comes from old printed articles, documentation etc. that historians can research. Go forwards 50-100 years from now and just think about how much information that was only ever "posted online" will have been lost forever. Its actually quite frightening...
Most of the information posted online is either plain junk or various forms of reposts but that doesn't negate the fact that there have been countless invaluable websites whose information has been lost forever and there will be thousands more.
E.g. I like archiving applications and on my drive I already have at least three dozen applications which are nowhere to be found on the internet and I just don't know how to preserve this info for posterity. Maybe no one actually needs it - it'll all end anyways. We, planet Earth, the Sun, and ultimately this universe. Everything will be like we've never existed.
E.g. I like archiving applications and on my drive I already have at least three dozen applications which are nowhere to be found on the internet and I just don't know how to preserve this info for posterity.
Well... The first and biggest potential hurdle you'll face is licensing.
If that is not a problem, there's many free web hosts, or you can put stuff on the likes of google's cloud thingy (or dropbox or mega.nz or...) and posts the links somewhere (pastebin? Here? Your favourite farcebork page?)
You could go really weird and rare and set up your own server on a Raspberry Pi with a SD card of suitable size (imaged somewhere else for backup purposes as well).
It's pretty easy these days to get simple hosting done, especially if you're careful to retain your own copies in case (/when) the host disappears or stuffs you around (eg photobucket, google, MS, copy.com, flikr.......).
And there's also the likes of GOG (Good Old Games) - there may be an apps version of them out there?
The only thing that would slow me down (maybe) is the licensing issue. But I'd be hoping to find the original authors and get them on side anyway, though that is often very hard to do.
I used to run a BBS. I had several of the Fidonet 'file echos' feeding into my system. One day I realised I hadn't heard the modem answer a call in weeks so hit the power switch and plugged a phone in. A while later, with no calls coming in, I packed up the drives and got rid of the old case. The drives and SCSI card are sitting in a box in my office desk (yeah, I actually do have an 'office' I visit from time to time) just waiting for me to be board of El Reg/Games enough on a wet day to actually sit down, test and old mobo/psu I have (ISA SCSI card) to make sure it won't blow things up, then plug them all back in and recover the data. Even have an ISA network card, though I may have to hunt back through time to find a protocol I can use to upload the data. Plan is to image the drives and see if I can run stuff in a VM. I have a huge archive of Fido messages I've been meaning to ship off to someone as well, I may even have the biggest surviving collection - IF the drives still run.
So if you do find a place that will take old programs let me know, I may have some they're interested in :)
The first and biggest potential hurdle you'll face is licensing.
Even old Nintendo games are still on archive.org, which surprises me but maybe there's somewhere even they won't go because they must be aware that, as the Wii Shop is dead, the Wii U Shop can't be far behind, and the Switch Shop is pretty bare, it would be ridiculous to even suggest those as a legal method of preserving old games.
The Wii shop isn't dead - you can't purchase new games, but you are still able to download games you've already bought.
Whilst the Wii U's days are undoubtedly numbered, eShop titles are still selling tens of thousands, so I would have thought it's worth keeping. There's some lovely games for the system even if the platform wasn't a great success.
(Currently 200 hours into Breath of the Wild, Yoshi's Wooly World is stunning, and really need to try Wind Waker)
"If you think back 50, 100 or more years, much of our understanding of news, technology etc comes from old printed articles, documentation etc. that historians can research. Go forwards 50-100 years from now and just think about how much information that was only ever "posted online" will have been lost forever."
I mostly agree with you, but let's also consider the amount of paper documentation that was lost for one reason or another during that time. Because paper took up so much space, lots of it was recycled, discarded, or burned. And not all of it had an archival copy somewhere. Digital records at least make it easy to copy them, such that something like the Internet Archive can exist without requiring thousands of employees to copy and file stuff. Paper records are great when they need to survive something massive that creates a gap in custody between whoever has them now and future historians, unless that gap is created by fire. Digital records, however, make it possible for a small group of people to retain a massive set of data. Of course, they also make it possible to create even larger amounts of data which only gets backed up if someone thinks to do so.
the thing with paper records was that, even though it was so much more costly to produce and retain, once it was done, it was done in quite a few copies, and yes, those copies could have been, and have been and are being shredded and otherwise disposed of. But, this also involves cost. And, ironically, while electronic records are cheap and easy to create and maintain, they're equally cheap and easy to erase. And, with the avalanche of new data, previous records ARE being erased, never mind an accident or purposeful "eraser weapon" that goes bad.
you used to be able to trace back history by going through older and older maps. Not with mapping software, not any more. There's a gap, and it's growing. Likewise newspapers. Now any politicians will laugh in your face if you point out he said something stupid 5 years ago. And don't tell me that you will find in "on the internets". You won't.
I took it as a reference to a relevant situation where the need to support old hardware went beyond the normal life-cycle, not that El Reg was attempting to break this news.
I found it interesting anyway. Massively idiotic to tie the fortunes of a car, with expected lifespan measured in decades, to the operation of a single model of IT equipment that may have only been produced over a 6 to 12 month period before being replaced with a newer model. It's bad enough if you limit yourself to one OS, let alone a single bit of kit.
Doesn't affect the service life expectations. Actually, you'd expect cars not driven daily to last longer, and expensive ones to be serviced regularly. In this case though, it seems this was about dongle support, in the form of a proprietary Conditional Access card. The reporting about the CA card actually running the software or being an interface to the car appears to be inaccurate. Which is why they were able to later write software to replace it (if it had been a hardware interface, that wouldn't be true).
There's a hole in my software library from about this time, where I can still run and use pre-2000s software in DOSBox emulation. And I can run later software from the mid 2000s with standard OS support. The gap is from the era of proprietary driver support and weird DRM schemes like bad sectors. This history is getting lost, like 60's TV episodes. Warez crackers are actually doing a public service here in the long run.
With Apple you're lucky to get ten.
Redownload Archived macOS Installers to Address Expired Certificates
Not had to grab ancient intel files, but have had to download other MOBO manufacturer drivers that are extremely old when dealing with hardware issues.
Its always reassuring to get them from manufacturers site rather than some 3rd party site as (though still a risk) you would expect manufacturer site downloads to be legit whereas 3rd part sites are always a risk of getting a doctored file.
.. Yes I know some sites mirrors will be legit, run by honest people but some might not be.
My "newest" home computer is 12 years old - I try and run kit until it dies as don't need super power PCs (gaming done on non PC platforms) as nothing too resource hungry done on computer
I can't wait for some "totally reputable" third party sites popping up, claiming to host old Intel drivers, but actually delivering malware that will end up on machines which are probably thought to be secure because they're kept offline, running some old industrial process..
Similar thing happened on a Windows roll-out program for a local council a few years ago. Updating the systems would have bricked all the equipment used in the vehicle testing centre. Luckily, we caught that before they were rolled-out. As far as I know, they are still running their tests on XP with USB 'sneaker-net'.
> Mechanics wanting to service one of the hundred or so remaining McLaren F1 supercars face a similar problem. They were reliant on ancient Compaq LTE 5280 laptops in order to service the cars.
Well I find it funny how we see the laptop as !ancient" but the car as not.
If the cars are worth so much and they wont go for the build of a new car, perhaps they should put the money into making a special box that simulates the CA card thats specifically usable only in these laptops. The you just give the box USB, which can easily be converted to rs232 or parralel if needed by the hardware simulation and it will be usable with any laptop that can run this software in a VM.
It really cant be that hard, pay some developers to build a newer interface. Or build a new car.
Actually, another thought. Why not just build a new laptop to the same specs as the old one? Someone has the original schematics, you could even just build spare parts. For the kind of money these cars pull in, surely is not too expensive to commission some custom hardware build. There is nothing wrong with using the laptop or DOS as long as you can keep it running. It still does the same job, no matter how old the bits and bytes get.
As noted by the article, there are loads of industrial systems in service with 20 year old CPUs. Industrial flow computers use 486 and Pentium era hardware extensively, mostly on top of DOS (although I have even seen win98 in such use).
Closer to home, some of us that occasionally fancy breaking out win95/98 era titles that don't reliably work on anything else. There are few options to run say, Interstate 76 on anything but period hardware. True, not many titles are particularly deserving of preservation, but there's a decent chunk of computing history lives in the 3dfx/Win98 window. GOG has openly said that creating support for that period of time is probably it's biggest challenge.
I get Intel wants a shiny website that doesn't return "obsolete" search results (maybe don't try putting your 1999 BIOS update on a "modern" Pentium?) but they could simply archive off materials. Even better could they open source them? Help train future computer scientists on how things "were" and probably still are done?
How relevant is that? If those "industrial" systems have not been updated in the last 20 years, and they use Intel *desktop* boards, which were not made for industrial systems, and furthermore, they haven't a support contract, the disappearance of the website won't impact them: they do not give a fuck.
I'm sorry that your little entitled feelings are hurt by big bad meanie Intel *hug*
So when your perfectly serviceable, 20 year old Chromatograph (with embedded hardware dumping out info over RS232 or similar), that needs a replacement controller interface (desktop hardware); do you replace the whole shebang, or do you try and fix the old system first?
Bear in mind the Chromat is probably a £50,000+ machine, whereas a spare board from ebay is in the worst case a couple of hundred quid. Is it right to dump a ton of extra cost and time on replacing the WHOLE system onto people paying bills? Or would you rather the network utility spend just a few quid getting a spare board; flashing the BIOS and carrying on? You talk of support arrangements. Nobody offers warranties covering that kind of service life.
When your 1990's $15,000 CNC machine needs a new floppy drive, do you replace the CNC machine? Or do you try and fix the old system first? Perhaps with a Gotek USB floppy (like everyone and his dog has done).
Yes, I think we are perfectly entitled to be pissed off by this un-necessary move, that serves mostly only to tidy up search results on their web page.
I must have touched a nerve for someone to throw such angry commentary at losing a service that genuinely has saved everyone money.
Not just Human medical equipment ...
I keep my Vet's blood machine running. It's an original Idexx Vettest 8008, and is about 30 years old. It's up-to-date, though ... it runs a version of FreeDOS, because MS-DOS is no longer supported.
I know of about fourty of these things still in use here in the SF Bay Area, because I have worked on them. I suspect there are several hundred more. (I bought a bunch of used units 20 years ago from Vets with Y2K worries ... I was speculating on the need for spares in the future. It was a good guess on my part. I have since augmented my stock by purchasing them whenever I run across 'em.)
This post has been deleted by its author
Whilst everyone has been frothing about HP or how little storage is required and how Intel are asshats, I wonder exactly what they mean when they say
"Intel recommends that users of BIOS Update [RL86510A.86A] P21 uninstall and/or discontinue use as soon as possible."
Why would they recommend people uninstall an update just because it will no longer be available?
They mean uninstall or discontinue using the hardware itself because now it's not officially supported. And buy a nice new shiny Intel thing to replace it, please.
ghacks: Intel started to remove old downloads from its Download Center
Pretty shameless and/or desperate.
This article is about the BIOS software on a motherboard.
*
Surely most people want to preserve some functionality defined at THE APPLICATION LEVEL. If this is true, then old CP/M applications (say 1970s and early 1980s) can be run pretty fast on emulators on modern machines. DOS applications have similar emulators on modern machines. Even here at Fedora Mansions there are WIN31 applications running perfectly well using WINE on cheap modern laptops. (Note that this last example has 16-bit application software from 1992 running perfectly well on a newish x86-64 machine running the latest Linux kernel.)
*
So can someone explain why there needs to be any concern about obsolete BIOS software for obsolete hardware?
My feeling about this sort of thing extends far beyond computer software/hardware. It extends over the entire industrial/Intellectual property base.
As I see it. Patents/Copyright give the inventor/licence holder the opportunity to exploit the concept (Whatever it is) for as long as they wish to do so (Or for a limited period as in Patents).
However, should a license holder/patent owner chose to discontinue the product or use of concept, then the information, ALL the information pertianing to it, should be handed over to a government operated archive whereupon it all becomes open source for anybody to use freely.
Patent/Copyright holders should NOT be permitted to obsolete technologies simply by virtue of refusing to support them any more.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
