Normally I'd agree with you, but that's because in many cases, vendors seem to tune their drivers for the benchmarks at the expense of everything else - Q3A scores go up, but real-world performance suffers. If they've figured out some way to boost Q3A performance without having some performance trade-off somewhere (and they aren't spending so much time on Quake tweaking that their drivers lag in other areas), then I say it's fair game...
I'd like to point out that if something as simple as changing "Quake" to "Quack" reduces performance by 15%, chances are that other Q3A-based games that don't contain the word "Quake" are probably not seeing the performance increase that Q3A is, even though the majority of the engine code may be the same. Therefore, if ATI is trying to convince customers that games based on the Q3A engine will run faster on their cards, this kind of optimization may be outright fraud.
It would be different if ATI created a driver that analyzed the manner in which the card was being used to trigger a Q3A-oriented optimization, but as I understand it, that's not what's going on here. ATI has improved performance for a single executable as opposed to a class of games for the the sole purpose of improving the performance of a game that is commonly used as a benchmark, and they did so without informing the public. I don't see how there can be any question that what ATI did here was unethical.
While it is true that the price of Windows hasn't dropped since 3.1, it has only become more expensive with the introduction of XP ($99 upgrade for XP Home, compared to $89 for win95 through winme). One rationalization for that could be that since XP is really based on NT, it's actually cheaper (standard NT upgrades run around $200). Also, that's ignoring the $49 upgrade specifically from Win98SE to WinME. On top of that, comparing the price of Windows XP Pro to Win9x is a red herring -- a better comparison would be Win2K Pro to WinXP Pro -- or roughly $200 upgrade vs. roughly $200 upgrade. And to top it all of, if you factor in inflation the prices have actually dropped.
First of all, WinME really didn't offer much over Win98SE, so the lower upgrade price was necessary to attrack customers. The new features just weren't attractive enough to attrack customers at a price of $89. (This is especially true considering WinME's reputation for instability.)
It has also been expressed that XP should be compared to NT in pricing. This is misleading. XP Home does not support the same security and networking capabilities, even though it is the same codebase, so it would be more accurate to compare XP Pro with Win9x as equivalent to an upgrade to NT.
Meanwhile, Microsoft is saving money by not supporting multiple code bases, DOS and Netscape plug-ins. They've managed to fit everything on a single CD-ROM, which farther increases their savings. They're also saving money by virtue of the fact that the kernel for XP is pretty much Win2k. So let's not pretend they're R&D has gone through the roof. Let's face facts: $99 for an upgrade to XP Home, a security-crippled OS that requires to call up Microsoft if you install for than three pieces of hardware in your computer, is hardly a bargin. Twice that much for XP Pro is pretty much the price of upgrading to Win2K Pro, which under the hood is pretty much the same thing without a boatload of bloatware that you have no way of uninstalling.
So, when you really look at what you get, and what it costs MS to make it, where are the savings?
Slashdot Mirror
I'd like to point out that if something as simple as changing "Quake" to "Quack" reduces performance by 15%, chances are that other Q3A-based games that don't contain the word "Quake" are probably not seeing the performance increase that Q3A is, even though the majority of the engine code may be the same. Therefore, if ATI is trying to convince customers that games based on the Q3A engine will run faster on their cards, this kind of optimization may be outright fraud.
It would be different if ATI created a driver that analyzed the manner in which the card was being used to trigger a Q3A-oriented optimization, but as I understand it, that's not what's going on here. ATI has improved performance for a single executable as opposed to a class of games for the the sole purpose of improving the performance of a game that is commonly used as a benchmark, and they did so without informing the public. I don't see how there can be any question that what ATI did here was unethical.
First of all, WinME really didn't offer much over Win98SE, so the lower upgrade price was necessary to attrack customers. The new features just weren't attractive enough to attrack customers at a price of $89. (This is especially true considering WinME's reputation for instability.)
It has also been expressed that XP should be compared to NT in pricing. This is misleading. XP Home does not support the same security and networking capabilities, even though it is the same codebase, so it would be more accurate to compare XP Pro with Win9x as equivalent to an upgrade to NT.
Meanwhile, Microsoft is saving money by not supporting multiple code bases, DOS and Netscape plug-ins. They've managed to fit everything on a single CD-ROM, which farther increases their savings. They're also saving money by virtue of the fact that the kernel for XP is pretty much Win2k. So let's not pretend they're R&D has gone through the roof. Let's face facts: $99 for an upgrade to XP Home, a security-crippled OS that requires to call up Microsoft if you install for than three pieces of hardware in your computer, is hardly a bargin. Twice that much for XP Pro is pretty much the price of upgrading to Win2K Pro, which under the hood is pretty much the same thing without a boatload of bloatware that you have no way of uninstalling.
So, when you really look at what you get, and what it costs MS to make it, where are the savings?