Windows 10 having the same system requirements - holding it back?

I feel like removing or raising target system specs is the leading reason we see shoddy unoptimized code.

Keeping on with the same system requirements probably forces Microsoft to do a better job and have cleaner code. Even if it means not rolling out certain features I feel its better.
 
Do you believe that Windows could be advanced further if Microsoft did not try to cater to a wider audience?

I can't help but think that the designers and developers of Windows are being held back by business suits who tell them to cater to a lower common denominator.

No, i do not think so.
 
Do you believe that Windows could be advanced further if Microsoft did not try to cater to a wider audience?

I can't help but think that the designers and developers of Windows are being held back by business suits who tell them to cater to a lower common denominator.
The only problem I have is that Microsoft is still offering a 32-bit version of Windows 10. The 64-bit versions of Windows Vista and later had backwards compatibility with 32-bit applications.

But, then again, you have to take a look at the use cases. In some cases, the businesses only care if the computer can run the email client, a web browser, Microsoft Office, and their internal applications. Watching YouTube videos and playing games isn't on that list. And, you don't need a powerful processor or lots o'RAM to do that.
 
The only problem I have is that Microsoft is still offering a 32-bit version of Windows 10. The 64-bit versions of Windows Vista and later had backwards compatibility with 32-bit applications.

But, then again, you have to take a look at the use cases. In some cases, the businesses only care if the computer can run the email client, a web browser, Microsoft Office, and their internal applications. Watching YouTube videos and playing games isn't on that list. And, you don't need a powerful processor or lots o'RAM to do that.
The number of businesses out there still relying on ancient 16-bit Windows (or DOS!) applications for their line of business would depress you. And it's not like saying "use DOSBox" or "use a VM" is a working solution since plenty of them are relying on ancient hardware drivers to connect with devices whose manufacturers are long gone.
 
Not to mention that despite the 32-bit backwards compatibility layer, some 32-bit programs have issues running in 64-bit Windows while running fine in 32-bit Windows.

So it could be said that it is legacy software is what is holding Windows back, not weak hardware.
 
The number of businesses out there still relying on ancient 16-bit Windows (or DOS!) applications for their line of business would depress you.
I work in tech support. I am painfully aware of some of the legacy applications that are still out these that some of our customers run. Fortunately, it is outside my area of support.

Of course, in about 24 hours, we will no longer support Windows XP following last April's lead by Microsoft. I can't wait for some of our major customers to complain despite numerous notifications.

Nie mój cyrk, nie moje małpy.
 
The number of businesses out there still relying on ancient 16-bit Windows (or DOS!) applications for their line of business would depress you. And it's not like saying "use DOSBox" or "use a VM" is a working solution since plenty of them are relying on ancient hardware drivers to connect with devices whose manufacturers are long gone.

That kind of busineses deserve to die away if they didn't upgrade their hardware in the past 10 years.
 
Legacy 3rd party code is holding Microsoft back. That being said there are good reasons for that but its still not a Microsoft issue per say...you will have that with any modern OS that isn't recently coded from scratch.
 
I don't think holding down the system requirements is a big issue, or that it has much of an effect on how it works on higher end systems. The performance obviously degrades on lowest end configurations in case someone wants to run it on such an old system. Other features work better on systems with more resources, such as pre-fetching and resources available to higher end video cards when gaming.

MS chasing the "Windows Tablet" market consumers don't seem to want at least has been good to help slow bloat. If not for that idea I shudder to think what kind of installation footprint Windows would have 9 years after Vista. 50GB? 75GB?
 
The reality is that computers have NOT advanced so why should system requirements advance? This isn't about businesses its about consumers. The average person has moved from a desktop to a laptop to ultrabook and many to tablets and these smaller devices are less powerful than desktops from 5 years ago. Less is more kids, less is more keep drinking the koolaid while you get nothing more for your money and these tech companies smash record profits quarter after quarter.
 
The number of businesses out there still relying on ancient 16-bit Windows (or DOS!) applications for their line of business would depress you. And it's not like saying "use DOSBox" or "use a VM" is a working solution since plenty of them are relying on ancient hardware drivers to connect with devices whose manufacturers are long gone.

That's wonderful, except why are these same companies even considering a new version of Windows? If those antiquated apps and hardware are doing the job, what's wrong with Windows NT 4.0 or whatever?
 
That's wonderful, except why are these same companies even considering a new version of Windows? If those antiquated apps and hardware are doing the job, what's wrong with Windows NT 4.0 or whatever?

The antiquated computers are failing, the software is what is irreplaceable. Many instrumentation devices are decades old in factories and they don't usually get replaced until they physically break. Since those devices are no longer supported (for years) they have ancient drivers and code. Forget plug and play and USB, RS-232 is where it's at!

The process plants should upgrade their hardware before it breaks in order to keep up with OS advances but that would cost a lot of money. Stockholders wouldn't like that, see?
 
Unfortunately, the adage in business is "if it ain't broke, don't fix it". If the equipment is paid for and it works, why get something that provides an incremental improvement? There is also the training cost of transitioning the person from old version to new version of software, and the loss of productivity as a result. Tell that to a $alesperson that they have to spend a day or two in training, and they will scream bloody murder because they are not making money.

Now, add to the mix that these old operating systems are attached to systems that cost maybe ten grand or hundred grand. One well-known example is the ATMs which are running Windows XP Embedded, and won't hit EOL until 2016. Now, how long does that same ATM remain in operation? Years, and it's hardened against physical attacks. There is also medical equipment that is FDA certified at a certain configuration, and even applying a security patch may require FDA recertification.

Now, once again, the mantra for the past few years has been "increase revenue, reduce costs", and purchasing new equipment counts as a cost.
 
I can understand why this sucks for businesses. My mother bought a very expensive sewing machine that can embroider. It would cost 1000 to buy the box to connect it to anything newer than winXP but offer no benefit. So she has an old laptop. For many businesses upgrading hardware would be at a significant cost with no tangible benefit.

What good would upping minimum reqs for an OS do? If windows required a minimum 16GB of memory and a dedicated GPU it would only decrease market share. I can't think of any new features it would allow them to offer.

I agree, doing away with 32bit version would push developers to transition to 64bit programming though I'm not sure it would have any benefit.
 
In other news, computers doing 2 billion operations a second are still fast enough for most things even though computers doing 15 billion operations a second exist. Is that a problem? No. Minesweeper doesn't want or need high resolution textures and volumetric fog. Microsoft Word doesn't want or need 8GB ram. An email client shouldn't want a high end processor.

George R.R. Martin writes his multi-million dollar "A Song Of Ice and Fire" series on a DOS machine.
 
In other news, computers doing 2 billion operations a second are still fast enough for most things even though computers doing 15 billion operations a second exist. Is that a problem? No. Minesweeper doesn't want or need high resolution textures and volumetric fog. Microsoft Word doesn't want or need 8GB ram. An email client shouldn't want a high end processor.

George R.R. Martin writes his multi-million dollar "A Song Of Ice and Fire" series on a DOS machine.

When working on Word files that get into the hundreds of pages with multiple sections and a large number of pictures, Word crawls with even 8GB of RAM. 16GB and 64-bit Word is a whole lot faster.
 
George R.R. Martin writes his multi-million dollar "A Song Of Ice and Fire" series on a DOS machine.

He does this not because his DOS workstation is "fast enough", he does it because he dislikes graphical word processors on newer computers. He specifically cites features like spellcheck and grammar check as being intrusive to him and his writing. Yes, those features can be turned off, but he apparently isn't satisfied with that. So, he sticks with what he's used since early in his writing career.

In a lot of ways, he's like the [H]ers who refuse to upgrade to Windows 8.

That kind of busineses deserve to die away if they didn't upgrade their hardware in the past 10 years.

This is pretty narrowminded. You'd singlehandedly shut down the internal medicine and radiology departments of most hospitals if you enforced this. It's not feasible to buy multi-million dollar control rooms every few years just for the sake of swapping out some ports on a few devices.
 
When working on Word files that get into the hundreds of pages with multiple sections and a large number of pictures, Word crawls with even 8GB of RAM. 16GB and 64-bit Word is a whole lot faster.

I've not worked with jumbo documents in Word for a while so I just converted this PDF, 841 pages, into a DOCX file : http://nca2014.globalchange.gov/sys...s_in_the_United States_HighRes.pdf?download=1

Did it on my Surface Pro 3 i7/8GB/512GB and while it takes about 25 seconds to load into Word 2013 once loaded the performance is fine, and this is with a lot of other programs running.
 
This is pretty narrowminded. You'd singlehandedly shut down the internal medicine and radiology departments of most hospitals if you enforced this. It's not feasible to buy multi-million dollar control rooms every few years just for the sake of swapping out some ports on a few devices.

Then at least the people responsible for procurement should be fired as they didn't do due diligence on product lifespan. If you make a multi million investment you should make sure you don't have technological dead ends at half the planned product life.

If the control room is based on DOS at 2004 you're just doing it wrong! :)
 
Then at least the people responsible for procurement should be fired as they didn't do due diligence on product lifespan.
It's often not an issue of product lifespan, but the refusal to replace software or hardware when it's working perfectly fine for its intended uses.

The last place I worked still had a couple of systems running NT 4 and NT 3.5. The problem with upgrading had less to do with getting the software and interface cards used for devices connected to it than the costs involved. No one made a mistake by buying that hardware or continuing to use the old OS and hardware with it.

Back to the point, if some companies have almost 10 year and older systems with Pentium 4/D and Athlon 64/X2 processors and want to run Windows 7/8/10 on it, that's fine too. I'd just hate to work there. :p A system doesn't have to be 1337 to be (nearly) usable.
 
Back to the point, if some companies have almost 10 year and older systems with Pentium 4/D and Athlon 64/X2 processors and want to run Windows 7/8/10 on it, that's fine too. I'd just hate to work there. :p A system doesn't have to be 1337 to be (nearly) usable.

Hardware that old can fail without warning at any moment. Of course even new devices fail but the older they get the more probable a sudden catastrophic failure comes. The risk is especially big on components such as hard drives that contain moving parts.
 
Then at least the people responsible for procurement should be fired as they didn't do due diligence on product lifespan. If you make a multi million investment you should make sure you don't have technological dead ends at half the planned product life.

If the control room is based on DOS at 2004 you're just doing it wrong! :)

Most of the time you don't have a choice in the matter. Enterprise and commercial is a very different environment from consumer. Planned obsolescence is in greater effect, they don't want to give you clear and cheap upgrade paths.
 
Then at least the people responsible for procurement should be fired as they didn't do due diligence on product lifespan. If you make a multi million investment you should make sure you don't have technological dead ends at half the planned product life.

If the control room is based on DOS at 2004 you're just doing it wrong! :)

This is pure ignorance on display. You're limited to two or three vendors, and you have to make sure the product you buy plays nicely with the other multi-million dollar pieces of equipment you already have. Oftentimes, this creates lock-in to an extent not seen in the consumer market—GE makes Microsoft's Windows monopoly look like a blissful fairy land—and you're stuck with whatever they have to sell at the time.
 
The stupidity of all of this is the argument that multimillion dollar machines have anything to do with this. They don't, period. They are a small niche segment of the market that will always be served by specialized companies. If some MRI running on windows XP has a problem then people come in with old hardware and software to fix it. If an old machine needs to be connected to the internet then a specialized company will offer a support package for windows XP. MS does not make their next OS for 20 year old multi million dollar equipment they make it for the mass market. Every OS since the explosion of smartphones has seen this trend toward not needing any higher specs.
 
The stupidity of all of this is the argument that multimillion dollar machines have anything to do with this. (text deleted) MS does not make their next OS for 20 year old multi million dollar equipment they make it for the mass market. Every OS since the explosion of smartphones has seen this trend toward not needing any higher specs.
So, which distro of Linux would you recommend then? If it was up to me, then yes, if it's specialized equipment with long shelf life, I would prefer a hardened Linux distro over Windows.

However, where is the after-market support then? And, if something goes seriously wrong, who do you sue? Remember, a lot of companies adopted OpenSSL, and last year, it was discovered that there was the HeartBleed vulnerability. That OpenSSL code was maintained by how many people?
 
This is pure ignorance on display. You're limited to two or three vendors, and you have to make sure the product you buy plays nicely with the other multi-million dollar pieces of equipment you already have. Oftentimes, this creates lock-in to an extent not seen in the consumer market—GE makes Microsoft's Windows monopoly look like a blissful fairy land—and you're stuck with whatever they have to sell at the time.

In my experience when multi-million deals are at hand the vendor will bend over backwards to spec the product according to your needs. Its buyer stupidity to blindly take whats offered. If your hardware clearly exceeds the lifespan of typical software/OS then a proper maintenance plan has to be made with the vendor before the purchase.

It would be pretty dumb to pay millions for a device that stops working 8-12 years before its intended lifespan simply because nobody prepared for software support in the future. There is even an argument if Windows should be used at all in devices which require stability and long term support. I see Windows use in medical devices borderline criminal with all the security risks at hand.
 
Last edited:
To a point, yes. Similar to how Windows 9x was holding Windows back until NT was brought to the consumer level. It still had a lot of legacy code, but it worked great and was very compatible with previous versions. Right now, Windows 10 can run apps created for Windows 2000, and some for Win95. There is a lot of bloat in Windows, but take a lot of it out, and you break thousands of old applications or newer ones that use old, crap coding techniques.

To build a whole new Windows from scratch to be optimized, fast and efficient would break a lot of compatibility and piss off the majority of their user base. In the end, you'd have something like Linux - barely any market share, yet a very great OS.
 
In my experience when multi-million deals are at hand the vendor will bend over backwards to spec the product according to your needs. Its buyer stupidity to blindly take whats offered. If your hardware clearly exceeds the lifespan of typical software/OS then a proper maintenance plan has to be made with the vendor before the purchase.

It would be pretty dumb to pay millions for a device that stops working 8-12 years before its intended lifespan simply because nobody prepared for software support in the future. There is even an argument if Windows should be used at all in devices which require stability and long term support. I see Windows use in medical devices borderline criminal with all the security risks at hand.

They will only do so to a limit. That limit depends on how much you pay them. So if you want that easy upgrade path, the vendor may charge you double. Would double the cost be worth the easy upgrade path? Depends. Not to mention you would have to pay additional when the upgrades do come.
 
In my experience when multi-million deals are at hand the vendor will bend over backwards to spec the product according to your needs. Its buyer stupidity to blindly take whats offered.

Looking down on things you aren't involved with and don't understand does not make you correct or endear you to being right.
 
Looking down on things you aren't involved with and don't understand does not make you correct or endear you to being right.

I handle procurements projects all the time and work as a subcontractor for multi-million SAP etc. procurements so the area of interest is not completely unknown to me.
 
I handle procurements projects all the time and work as a subcontractor for multi-million SAP etc. procurements so the area of interest is not completely unknown to me.

Great--too bad your expertise doesn't extend to the specific market being discussed.

Don't know what to tell you. Your attitude that options somehow exist or different decisions could have been made is incongruous with the reality of medical tech companies. There are at best 2-3 options for a given product, and interoperability with existing equipment that is not being replaced is not guaranteed, further limiting choice.

I cannot replace a biomed tech's 8 year old Windows XP laptop because its serial port is necessary to interface with a nuclear medicine controller. USB to serial adapters do not work. The controller in question costs 3 million dollars to replace. Laptops cost $400. Thus we're stuck with it.

Hurray legacy.
 
Great--too bad your expertise doesn't extend to the specific market being discussed.

Don't know what to tell you. Your attitude that options somehow exist or different decisions could have been made is incongruous with the reality of medical tech companies. There are at best 2-3 options for a given product, and interoperability with existing equipment that is not being replaced is not guaranteed, further limiting choice.

I cannot replace a biomed tech's 8 year old Windows XP laptop because its serial port is necessary to interface with a nuclear medicine controller. USB to serial adapters do not work. The controller in question costs 3 million dollars to replace. Laptops cost $400. Thus we're stuck with it.

Hurray legacy.

To me that tells that the manufacturers of those machines are not really competing with each others and keep releasing sub-par solutions to their customers. Serial port was already legacy technology 8 years ago and Windows XP is a consumer OS, why was it selected in the first place? :)

I'm fairly sure that you could get the interface easily working again with current hardware if someone had access to the source code. The person who did the procurement to this multi-million hardware was too short sighted to guarantee access to the source in the event of the manufacturer going bust or technology advancing.

It's only a matter of agreement.
 
To me that tells that the manufacturers of those machines are not really competing with each others and keep releasing sub-par solutions to their customers.

The machines work. In fact, they work too well, which is why they're still in use 14 years after initial acquisition.
 
To me that tells that the manufacturers of those machines are not really competing with each others and keep releasing sub-par solutions to their customers. Serial port was already legacy technology 8 years ago and Windows XP is a consumer OS, why was it selected in the first place? :)

I'm fairly sure that you could get the interface easily working again with current hardware if someone had access to the source code. The person who did the procurement to this multi-million hardware was too short sighted to guarantee access to the source in the event of the manufacturer going bust or technology advancing.

It's only a matter of agreement.

lol man you are dense I am sure these companies are going to give up their source code, and oh yeah I am really sure a hospital wants to get into programing for a highly specialized multi million dollar machine. But hey that's right up there with your zero value for peoples time mentality.
 
So, which distro of Linux would you recommend then? If it was up to me, then yes, if it's specialized equipment with long shelf life, I would prefer a hardened Linux distro over Windows.

However, where is the after-market support then? And, if something goes seriously wrong, who do you sue? Remember, a lot of companies adopted OpenSSL, and last year, it was discovered that there was the HeartBleed vulnerability. That OpenSSL code was maintained by how many people?

I don't recommend any distro of Linux, I am only pointing out that any discussion of highly specialized multimillion dollar machines is a thread derail. Even with multi million dollar machines the companies that make them don't really care if they last long. If you keep the machine too long they will start charging you and arm and a leg every time you need support. If you replace it, they hope to win the replacement bid.
 
Back
Top