You might be surprised by the hardware some people in the business cling to. I have a friend who's been a pro photographer for over 30 years and in addition to the Intel Macs and PPC PowerMac G5s he has, he still has two Powermac 9600s running Mac OS 8.6 and Photoshop 4 IN PRODUCTION. What amazes me is that the old SCSI drives in those machines seem to be bulletproof. I can't see modern SATA drives ever lasting 13 years.
Great point. Just because something new comes out, doesn't mean you actually need it. My LC 575 with a 33 MHz 68040 processor (Mac OS 7.6.1) loads Word 5.1 faster than my iMac G5 2.1 GHz loads Word 2004. Both Macs can print to my HP LaserJet 6MP over an AppleTalk/Ethernet bridge. The LC 575 also runs Photoshop 2.5.1 quite well too.
The announcement of going to Intel was made on the same day Apple released the 1st intel (iMac iSight - 10 Jan 2006)
Apple announced and released the PPC iSight iMac only three months before it was discontinued and replaced by the Intel.
Wrong. Apple announced the move to Intel during WWDC 2005, 6 months before they released the first Intel Mac (and that Intel Mac was released 6 months ahead of the original plan). Apple even offered an Intel Development Mac (A Pentium 4 living in a Power Mac G5 tower) to software developers so they can start writing for Intel before the new Intel Macs even arrived.
This is called a RUMOR site. Their main goal is to post bullshit stories, along with a mix of some real ones. They enjoy seeing how gullible people are to believe this stuff, and then comment on it. If you are looking for journalistic integrity, this is not the place for you.
This is not your typical rumor site. There is integrity in the fact that this particular site does make a point of detailing the source of what it posts, leaving it up to its audience, i.e. us, to consider the validity of what is being presented.
I find that of all the sites covering Apple this one is at the top of the heap as far as "rumor" sites go. A lot of what is posted here proves to pan out and the stuff that is highly speculative is not misrepresented. You do Appleinsider a great disservice by taking a cheap shot by grouping it in with a lot of weaker sites working in a similar vein.
I predict, at the upgrade price and the rate of new Macs sold (which would be high despite the economy), by the end of 2010 the adoption of Snow Leopard will be probably "the best ever rate of adoption" or something like that.
There is a much greater probability of Snow Leopard succeeding than Windows 7 mainly because of what has gone on before. Considering how expensive Windows 7 will be to upgrade to and considering that its predecessor Vista has been an unqualified disaster, how long will the line-ups be to scoop up copies of Windows 7 when it comes to market? Certainly if you spent a couple of hundred to buy a cheap PC, you would be rather reluctant to spend nearly as much to update your OS.
It's an embarrassment for Microsoft that right now a lot of people are running XP because of how unstable, expensive and power-hungry Vista has proven to be. In comparison Leopard has been an unqualified success, likely running today on the vast majority of Macs.
Even if the adoption rate is slowed somewhat because a few PPC owners out there cannot migrate to SL, at $29, Snow Leopard will seem far more successful than WIndows 7 which, you can be sure, will cause a three-OS mess on the PC side. By that I mean soon enough we'll have XP, Vista and Windows 7 out in the wild at the same time whereas on the Mac side, I predict that by early next year there will be Snow Leopard running on the vast majority of Macs. Certainly the perception will be that the state of Mac's OS is that there is Snow Leopard and nothing else of consequence.
Kudos to Apple for getting it right and total puzzlement over how a company with the resources of a Microsoft could mess up so thoroughly. Microsoft is one of the worst monopolistic giants in business history. Thank God there is an Apple, otherwise the state of personal computing today would be outrageously bleak.
This is not your typical rumor site. There is integrity in the fact that this particular site does make a point of detailing the source of what it posts, leaving it up to its audience, i.e. us, to consider the validity of what is being presented.
I find that of all the sites covering Apple this one is at the top of the heap as far as "rumor" sites go. A lot of what is posted here proves to pan out and the stuff that is highly speculative is not misrepresented. You do Appleinsider a great disservice by taking a cheap shot by grouping it in with a lot of weaker sites working in a similar vein.
No they don't. Most of their stories are from "an anonymous source that claims to work for X company". Half the stuff they post is bullshit.
No they don't. Most of their stories are from "an anonymous source that claims to work for X company". Half the stuff they post is bullshit.
When you are told that the claims come from an anonymous source claiming to work for X company, if that doesn't translate to "take this report with a grain of salt" the fault is not with the site offering up the rumor but rather with the person at the receiving end. It is legitimate to present a rumor if you let it be known clearly that it is an unconfirmed rumor and the credentials of the source are not misrepresented.
This site is excellent and easily the best source of Apple info at our disposal. Everything they publish is not presented as confirmed fact and it's absurd to accuses them of some dubious motives or incompetence because not all of the rumors prove to be based in fact.
There are other sites of great value to the Apple faithful and they should be utilized and valued accordingly. But as far as providing a steady flow of Apple info, rumors and otherwise, this site is the very best the Internet has to offer. To claim otherwise is to do a disservice to the people running this site.
It's not just about cross compiling. You still have to test the new code on the old platform. The code still needs to be continually tested, and maintaining that testing on a platform that's over the hill and only represents a small portion of your user base doesn't make sense. The big under the hood changes such as OpenCL won't work on video cards sold with or available for the G5. Grand Central almost certainly requires a fair amount of optimization that I'm betting is architecture specific and really only benefits the G5 quad, and the reports are that the people that bring in a G5 quad for repairs get a Mac Pro in return.
There's no point in adding all those optimizations for a platform that's been discontinued three years ago. Without those two under the hood optimizations, you're missing out on the biggest benefits from Snow Leopard anyway. The hardware is perfectly capable of most of the other changes, but there are still expenses in development that make it a less viable target.
The PPC server base is likely much more than 15%. There is a lot of PPC iron still cranking as xserves.
The PPC server base is likely much more than 15%. There is a lot of PPC iron still cranking as xserves.
And...? This isn't the first time Apple as obsoleted hardware, we get the same complaints every time, I've seen it with Tiger and Leopard.
Last I read, Apple makes about 12,000 (twelve thousand) Xserves a year. So 36,000 PPC Xserves in use is probably an optimistic estimate. That's a far cry from the tens of millions of Macs in use, Xserves are negligible in comparison. And I doubt any of them are going to stop working this Friday just because they don't have the latest OS.
You're not going to be getting OpenCL on G5 Xserves.
Wrong. Apple announced the move to Intel during WWDC 2005, 6 months before they released the first Intel Mac (and that Intel Mac was released 6 months ahead of the original plan). Apple even offered an Intel Development Mac (A Pentium 4 living in a Power Mac G5 tower) to software developers so they can start writing for Intel before the new Intel Macs even arrived.
Problem is, there are plenty of xServes running like crazy in many places, as reliable as they were three years ago, many of those educational, who will now have to re-purchase a lot of very expensive server hardware to be up to date. As recently as this year's Ed update meetings, Apple reps claimed there was no decision yet to make Server an Intel-only app. So now a (larger than you might think) number of IT need to tell our bosses that we need to spend $2700 to get a $500 (retail) update.
Why can't they just keep running what they have? It's not like their servers will turn into pumpkins on Friday when SL ships...
They wait until they get new hardware, then they get SL or whatever is beyond SL at the time they buy. Big deal.
Apple had to make the transition to Intel and break from PPC someday. SL is as good a reason as any to put a stake in the ground and say "Here"!
Even doing so, they will still have to maintain two processor architectures for OSX - Intel and ARM. I imagine maintaining three (PPC too) was a bit much.
Those are very different things. Rosetta allow for legacy apps to be supported on the modern OS by using an app for emulation. MS is maintaining more native support for legacy apps with legacy OS underpinnings. In MS defense, legacy apps will run natively, but since the HW has increased so much since the PPC days and Rosetta is well done one is not likely to see much, if any slow down.
Actually, Rosetta is a transparent layer of the OS rather then a separate application.
Not all legacy applications will run natively in Windows 7. Windows 7 offers XP emulation by actually running a hidden XP virtual machine (think parallels). It's technology they acquired when they purchased a company called Kidaro and then baked Kidaro's seamless application publishing into Windows 7. If you have an Enterprise Agreement with Software Assurance and you subscribe to the Microsoft Desktop Optimization Feature Pack (MDOP), with the full version of Microsoft Enterprise Desktop Virtualization (MED-V) you can also expose the XP desktop (if you want) . MED-V is what they renamed Kidaro to (I really wish they would stop picking lame names for cool technology they acquire).
BTW - for whoever said they still needed IE6 for enterprise applications at work, you can get that through the XP emulation baked into some version of Windows 7. Well, as long as your CPU supports virtualization extensions \
So see, Apple isn't the only one that has OS/hardware restrictions
Yup, three versions of Windows, some emulation here and there, and 3 web browsers on top of it which vary far from ACID tests and so on. The craptacular IE6, the mediocre IE7 and the wannabe IE8.
Windows 7 is a definite improvement from Vista but yeah, it ain't all as smooth for Microsoft as it is for Snow Leopard.
Quote:
Originally Posted by DocNo42
Actually, Rosetta is a transparent layer of the OS rather then a separate application.
Not all legacy applications will run natively in Windows 7. Windows 7 offers XP emulation by actually running a hidden XP virtual machine (think parallels). It's technology they acquired when they purchased a company called Kidaro and then baked Kidaro's seamless application publishing into Windows 7. If you have an Enterprise Agreement with Software Assurance and you subscribe to the Microsoft Desktop Optimization Feature Pack (MDOP), with the full version of Microsoft Enterprise Desktop Virtualization (MED-V) you can also expose the XP desktop (if you want) . MED-V is what they renamed Kidaro to (I really wish they would stop picking lame names for cool technology they acquire).
BTW - for whoever said they still needed IE6 for enterprise applications at work, you can get that through the XP emulation baked into some version of Windows 7. Well, as long as your CPU supports virtualization extensions \
So see, Apple isn't the only one that has OS/hardware restrictions
Quote:
Originally Posted by Carmissimo
There is a much greater probability of Snow Leopard succeeding than Windows 7 mainly because of what has gone on before. Considering how expensive Windows 7 will be to upgrade to and considering that its predecessor Vista has been an unqualified disaster, how long will the line-ups be to scoop up copies of Windows 7 when it comes to market? Certainly if you spent a couple of hundred to buy a cheap PC, you would be rather reluctant to spend nearly as much to update your OS.
It's an embarrassment for Microsoft that right now a lot of people are running XP because of how unstable, expensive and power-hungry Vista has proven to be. In comparison Leopard has been an unqualified success, likely running today on the vast majority of Macs.
Even if the adoption rate is slowed somewhat because a few PPC owners out there cannot migrate to SL, at $29, Snow Leopard will seem far more successful than WIndows 7 which, you can be sure, will cause a three-OS mess on the PC side. By that I mean soon enough we'll have XP, Vista and Windows 7 out in the wild at the same time whereas on the Mac side, I predict that by early next year there will be Snow Leopard running on the vast majority of Macs. Certainly the perception will be that the state of Mac's OS is that there is Snow Leopard and nothing else of consequence.
Kudos to Apple for getting it right and total puzzlement over how a company with the resources of a Microsoft could mess up so thoroughly. Microsoft is one of the worst monopolistic giants in business history. Thank God there is an Apple, otherwise the state of personal computing today would be outrageously bleak.
Why can't they just keep running what they have? It's not like their servers will turn into pumpkins on Friday when SL ships...
That's a great statement for all the PPC owners right there. Guess what. Next week, your machines will all run as they have been running. There is an iBook G4 in my family, it will soldier on as a Leopard machine. It's almost into its 5th year so far.
I agree that one reason why I like OS X is that they aren?t supporting nearly as much legacy code as Windows does. However, most businesses need that legacy support and MS? Windows business oriented OS means that they they have to support legacy HW and code. Apple will never be able to get a substantial foothold in the business sector as long as they are quick to drop legacy support and don?t allow 3rd-party PC makers to license their OS. It?s obvious that we have our Mac preference and I wish that all the Carbon was removed from Snow Leopard (which it hasn?t) but we also need to realize that these different business models both suit their customer?s needs.
PS: Why did you reply to yourself?
Needs that legacy support? Who says. IFFs (like RTF, the old txt format) were old hat when I put on my Museum Computerization presentations on in 1991 and 1992 and they have only expanded since then. As for legacy HW what on earth are businesses using, ISA and EISA cards or something that goes through a ISA serial port?
Or is is more a case where businesses are trying to squeeze the last amount of value out of a version of software or hardware no one even supports anymore? There are exception but if you really look at is this backward comparability is largely just a smoke screen. for "we don't want to update this program because we are cheap,"
Back when I did my survey of museums I found two who were running their databases on 8088s--largely because they simply didn't have the money to upgrade the computer and even if they had the programs they used simply wouldn't run on newer OSes.
Comments
You might be surprised by the hardware some people in the business cling to. I have a friend who's been a pro photographer for over 30 years and in addition to the Intel Macs and PPC PowerMac G5s he has, he still has two Powermac 9600s running Mac OS 8.6 and Photoshop 4 IN PRODUCTION. What amazes me is that the old SCSI drives in those machines seem to be bulletproof. I can't see modern SATA drives ever lasting 13 years.
Great point. Just because something new comes out, doesn't mean you actually need it. My LC 575 with a 33 MHz 68040 processor (Mac OS 7.6.1) loads Word 5.1 faster than my iMac G5 2.1 GHz loads Word 2004. Both Macs can print to my HP LaserJet 6MP over an AppleTalk/Ethernet bridge. The LC 575 also runs Photoshop 2.5.1 quite well too.
The announcement of going to Intel was made on the same day Apple released the 1st intel (iMac iSight - 10 Jan 2006)
Apple announced and released the PPC iSight iMac only three months before it was discontinued and replaced by the Intel.
Wrong. Apple announced the move to Intel during WWDC 2005, 6 months before they released the first Intel Mac (and that Intel Mac was released 6 months ahead of the original plan). Apple even offered an Intel Development Mac (A Pentium 4 living in a Power Mac G5 tower) to software developers so they can start writing for Intel before the new Intel Macs even arrived.
Monday August 24, 6:46 AM
Yep, something big is coming....
If you have a PPC Mac then there's Leopard 10.5.
If you have an Intel Mac then there's a tuned version called Snow Leopard 10.6.
This is called a RUMOR site. Their main goal is to post bullshit stories, along with a mix of some real ones. They enjoy seeing how gullible people are to believe this stuff, and then comment on it. If you are looking for journalistic integrity, this is not the place for you.
This is not your typical rumor site. There is integrity in the fact that this particular site does make a point of detailing the source of what it posts, leaving it up to its audience, i.e. us, to consider the validity of what is being presented.
I find that of all the sites covering Apple this one is at the top of the heap as far as "rumor" sites go. A lot of what is posted here proves to pan out and the stuff that is highly speculative is not misrepresented. You do Appleinsider a great disservice by taking a cheap shot by grouping it in with a lot of weaker sites working in a similar vein.
I predict, at the upgrade price and the rate of new Macs sold (which would be high despite the economy), by the end of 2010 the adoption of Snow Leopard will be probably "the best ever rate of adoption" or something like that.
There is a much greater probability of Snow Leopard succeeding than Windows 7 mainly because of what has gone on before. Considering how expensive Windows 7 will be to upgrade to and considering that its predecessor Vista has been an unqualified disaster, how long will the line-ups be to scoop up copies of Windows 7 when it comes to market? Certainly if you spent a couple of hundred to buy a cheap PC, you would be rather reluctant to spend nearly as much to update your OS.
It's an embarrassment for Microsoft that right now a lot of people are running XP because of how unstable, expensive and power-hungry Vista has proven to be. In comparison Leopard has been an unqualified success, likely running today on the vast majority of Macs.
Even if the adoption rate is slowed somewhat because a few PPC owners out there cannot migrate to SL, at $29, Snow Leopard will seem far more successful than WIndows 7 which, you can be sure, will cause a three-OS mess on the PC side. By that I mean soon enough we'll have XP, Vista and Windows 7 out in the wild at the same time whereas on the Mac side, I predict that by early next year there will be Snow Leopard running on the vast majority of Macs. Certainly the perception will be that the state of Mac's OS is that there is Snow Leopard and nothing else of consequence.
Kudos to Apple for getting it right and total puzzlement over how a company with the resources of a Microsoft could mess up so thoroughly. Microsoft is one of the worst monopolistic giants in business history. Thank God there is an Apple, otherwise the state of personal computing today would be outrageously bleak.
This is not your typical rumor site. There is integrity in the fact that this particular site does make a point of detailing the source of what it posts, leaving it up to its audience, i.e. us, to consider the validity of what is being presented.
I find that of all the sites covering Apple this one is at the top of the heap as far as "rumor" sites go. A lot of what is posted here proves to pan out and the stuff that is highly speculative is not misrepresented. You do Appleinsider a great disservice by taking a cheap shot by grouping it in with a lot of weaker sites working in a similar vein.
No they don't. Most of their stories are from "an anonymous source that claims to work for X company". Half the stuff they post is bullshit.
No they don't. Most of their stories are from "an anonymous source that claims to work for X company". Half the stuff they post is bullshit.
When you are told that the claims come from an anonymous source claiming to work for X company, if that doesn't translate to "take this report with a grain of salt" the fault is not with the site offering up the rumor but rather with the person at the receiving end. It is legitimate to present a rumor if you let it be known clearly that it is an unconfirmed rumor and the credentials of the source are not misrepresented.
This site is excellent and easily the best source of Apple info at our disposal. Everything they publish is not presented as confirmed fact and it's absurd to accuses them of some dubious motives or incompetence because not all of the rumors prove to be based in fact.
There are other sites of great value to the Apple faithful and they should be utilized and valued accordingly. But as far as providing a steady flow of Apple info, rumors and otherwise, this site is the very best the Internet has to offer. To claim otherwise is to do a disservice to the people running this site.
It's not just about cross compiling. You still have to test the new code on the old platform. The code still needs to be continually tested, and maintaining that testing on a platform that's over the hill and only represents a small portion of your user base doesn't make sense. The big under the hood changes such as OpenCL won't work on video cards sold with or available for the G5. Grand Central almost certainly requires a fair amount of optimization that I'm betting is architecture specific and really only benefits the G5 quad, and the reports are that the people that bring in a G5 quad for repairs get a Mac Pro in return.
There's no point in adding all those optimizations for a platform that's been discontinued three years ago. Without those two under the hood optimizations, you're missing out on the biggest benefits from Snow Leopard anyway. The hardware is perfectly capable of most of the other changes, but there are still expenses in development that make it a less viable target.
The PPC server base is likely much more than 15%. There is a lot of PPC iron still cranking as xserves.
And OpenCL on a GMA? I can't wait to see that...
The PPC server base is likely much more than 15%. There is a lot of PPC iron still cranking as xserves.
And...? This isn't the first time Apple as obsoleted hardware, we get the same complaints every time, I've seen it with Tiger and Leopard.
Last I read, Apple makes about 12,000 (twelve thousand) Xserves a year. So 36,000 PPC Xserves in use is probably an optimistic estimate. That's a far cry from the tens of millions of Macs in use, Xserves are negligible in comparison. And I doubt any of them are going to stop working this Friday just because they don't have the latest OS.
You're not going to be getting OpenCL on G5 Xserves.
Wrong. Apple announced the move to Intel during WWDC 2005, 6 months before they released the first Intel Mac (and that Intel Mac was released 6 months ahead of the original plan). Apple even offered an Intel Development Mac (A Pentium 4 living in a Power Mac G5 tower) to software developers so they can start writing for Intel before the new Intel Macs even arrived.
Great... Just waking up, are we?
Follow the thread.
Problem is, there are plenty of xServes running like crazy in many places, as reliable as they were three years ago, many of those educational, who will now have to re-purchase a lot of very expensive server hardware to be up to date. As recently as this year's Ed update meetings, Apple reps claimed there was no decision yet to make Server an Intel-only app. So now a (larger than you might think) number of IT need to tell our bosses that we need to spend $2700 to get a $500 (retail) update.
Why can't they just keep running what they have? It's not like their servers will turn into pumpkins on Friday when SL ships...
They wait until they get new hardware, then they get SL or whatever is beyond SL at the time they buy. Big deal.
Apple had to make the transition to Intel and break from PPC someday. SL is as good a reason as any to put a stake in the ground and say "Here"!
Even doing so, they will still have to maintain two processor architectures for OSX - Intel and ARM. I imagine maintaining three (PPC too) was a bit much.
Those are very different things. Rosetta allow for legacy apps to be supported on the modern OS by using an app for emulation. MS is maintaining more native support for legacy apps with legacy OS underpinnings. In MS defense, legacy apps will run natively, but since the HW has increased so much since the PPC days and Rosetta is well done one is not likely to see much, if any slow down.
Actually, Rosetta is a transparent layer of the OS rather then a separate application.
Not all legacy applications will run natively in Windows 7. Windows 7 offers XP emulation by actually running a hidden XP virtual machine (think parallels). It's technology they acquired when they purchased a company called Kidaro and then baked Kidaro's seamless application publishing into Windows 7. If you have an Enterprise Agreement with Software Assurance and you subscribe to the Microsoft Desktop Optimization Feature Pack (MDOP), with the full version of Microsoft Enterprise Desktop Virtualization (MED-V) you can also expose the XP desktop (if you want) . MED-V is what they renamed Kidaro to (I really wish they would stop picking lame names for cool technology they acquire).
BTW - for whoever said they still needed IE6 for enterprise applications at work, you can get that through the XP emulation baked into some version of Windows 7. Well, as long as your CPU supports virtualization extensions
So see, Apple isn't the only one that has OS/hardware restrictions
Windows 7 is a definite improvement from Vista but yeah, it ain't all as smooth for Microsoft as it is for Snow Leopard.
Actually, Rosetta is a transparent layer of the OS rather then a separate application.
Not all legacy applications will run natively in Windows 7. Windows 7 offers XP emulation by actually running a hidden XP virtual machine (think parallels). It's technology they acquired when they purchased a company called Kidaro and then baked Kidaro's seamless application publishing into Windows 7. If you have an Enterprise Agreement with Software Assurance and you subscribe to the Microsoft Desktop Optimization Feature Pack (MDOP), with the full version of Microsoft Enterprise Desktop Virtualization (MED-V) you can also expose the XP desktop (if you want) . MED-V is what they renamed Kidaro to (I really wish they would stop picking lame names for cool technology they acquire).
BTW - for whoever said they still needed IE6 for enterprise applications at work, you can get that through the XP emulation baked into some version of Windows 7. Well, as long as your CPU supports virtualization extensions
So see, Apple isn't the only one that has OS/hardware restrictions
There is a much greater probability of Snow Leopard succeeding than Windows 7 mainly because of what has gone on before. Considering how expensive Windows 7 will be to upgrade to and considering that its predecessor Vista has been an unqualified disaster, how long will the line-ups be to scoop up copies of Windows 7 when it comes to market? Certainly if you spent a couple of hundred to buy a cheap PC, you would be rather reluctant to spend nearly as much to update your OS.
It's an embarrassment for Microsoft that right now a lot of people are running XP because of how unstable, expensive and power-hungry Vista has proven to be. In comparison Leopard has been an unqualified success, likely running today on the vast majority of Macs.
Even if the adoption rate is slowed somewhat because a few PPC owners out there cannot migrate to SL, at $29, Snow Leopard will seem far more successful than WIndows 7 which, you can be sure, will cause a three-OS mess on the PC side. By that I mean soon enough we'll have XP, Vista and Windows 7 out in the wild at the same time whereas on the Mac side, I predict that by early next year there will be Snow Leopard running on the vast majority of Macs. Certainly the perception will be that the state of Mac's OS is that there is Snow Leopard and nothing else of consequence.
Kudos to Apple for getting it right and total puzzlement over how a company with the resources of a Microsoft could mess up so thoroughly. Microsoft is one of the worst monopolistic giants in business history. Thank God there is an Apple, otherwise the state of personal computing today would be outrageously bleak.
Why can't they just keep running what they have? It's not like their servers will turn into pumpkins on Friday when SL ships...
That's a great statement for all the PPC owners right there. Guess what. Next week, your machines will all run as they have been running. There is an iBook G4 in my family, it will soldier on as a Leopard machine. It's almost into its 5th year so far.
I agree that one reason why I like OS X is that they aren?t supporting nearly as much legacy code as Windows does. However, most businesses need that legacy support and MS? Windows business oriented OS means that they they have to support legacy HW and code. Apple will never be able to get a substantial foothold in the business sector as long as they are quick to drop legacy support and don?t allow 3rd-party PC makers to license their OS. It?s obvious that we have our Mac preference and I wish that all the Carbon was removed from Snow Leopard (which it hasn?t) but we also need to realize that these different business models both suit their customer?s needs.
PS: Why did you reply to yourself?
Needs that legacy support? Who says. IFFs (like RTF, the old txt format) were old hat when I put on my Museum Computerization presentations on in 1991 and 1992 and they have only expanded since then. As for legacy HW what on earth are businesses using, ISA and EISA cards or something that goes through a ISA serial port?
Or is is more a case where businesses are trying to squeeze the last amount of value out of a version of software or hardware no one even supports anymore? There are exception but if you really look at is this backward comparability is largely just a smoke screen. for "we don't want to update this program because we are cheap,"
Back when I did my survey of museums I found two who were running their databases on 8088s--largely because they simply didn't have the money to upgrade the computer and even if they had the programs they used simply wouldn't run on newer OSes.
The big under the hood changes such as OpenCL won't work on video cards sold with or available for the G5.
Heck OpenGL won't run on some IntelMacs so why the groaning?