I seem to be the only person on Slashdot who really liked Luigi's Mansion.
I loved Luigi's Mansion.
But it was way, way too short for a launch title that was supposed to be a placeholder for a Mario game. Keep in mind Mario, Zelda didn't come out until basically a year later. There wasn't a good platformer for the console until... well, until Sunshine, and by then, people were bitter.
A console launch needs to keep you from putting the console away and forgetting about it for months. The GameCube launch didn't do that. If you're like me, and didn't buy a Cube until almost a year later, hey, it was great, and I still play the GameCube more than my PS2. But that was me, and most people had already formed their opinion of the GameCube by then.
The Wii launch is primed to do that. They're going to be missing Mario again, most likely, but this time they substituted a much, much better filler than Luigi's Mansion: Zelda and Metroid. They can't let Mario slip past the end of the year, though. Wii, with Mario, Zelda, and Metroid at Christmas, is a killer launch. I think they can even survive without SSBB for a year, too.
The second party is YOU. What Gamecube games did you release?:)
Second party is a wholly-owned subsidiary of the first party, typically published by the first party. Rare was a second-party publisher for a while before Nintendo sold them off.
I thought Factor 5 was a second party as well (that's why I put Rogue Squadron in there) - turns out they're not, so that was a mistake.
anyone who thinks the GameCubes launch lineup was bad forgot that the N64 launched with only Maro64 and Pilotwings64 available.
Mario 64 was a better title than all of Nintendo's GameCube offerings combined. Nintendo's GameCube offerings were the grand total of Luigi's Mansion and Wave Race: Blue Storm.
The GameCube's launch was worse than the N64's, in my opinion, and that's hard to do when you're competing with only two games.
Seriously though...I've seen a few articles like this for the Wii, but I can almost remember them saying nearly the same things about how the learned from mistkes with the N64 and promised for 3rd party support for the 'Cube.
Note what he's saying, though: it's not just 3rd party support (honestly, I couldn't care less about where the game comes from) this time. Here he's saying "yeah, the launch titles weren't diverse enough - period."
That's a big deal. And actually, I personally think that the problem with the N64 was 3rd party support, and the problem with the GameCube was first-party support at launch. The Nintendo launch titles for the GameCube were.. well.. less than stellar. Note here that I'm going to glom first and second party together, as I'm never sure which ones are first or second.
Here's Nintendo's at launch titles for the GameCube:
Luigi's Mansion
Wave Race
Rogue Squadron
Even afterwards, it, uh, wasn't that much better. Soon-after-launch titles:
SSB:M
Pikmin
Uh. Yeah. That's... not that stellar. Not even that many, either. It's not that they were bad games. It's just that they were... well, few. And not Mario. Or Zelda. Instead, Wind Waker showed up a year later. Mario, nearly a year later as well. Both felt like too little, too late.
There's still plenty of time for Nintendo to screw up, mind you, but delaying Twilight Princess to the Wii launch (which virtually everyone thought that was what they were doing when they announced the delay) is actually quite smart. If they can get a good fraction of the titles at E3 out the door by launch, they're going to have a much, much more successful launch than the GameCube.
I don't doubt that it's wrong (like I said, I know from experience that it's of order 1 ms, not 100 ms) but the article is the one that's wrong, not the story summary.
(Seriously, haven't people heard cut-through switches which just look at the first part of the header and switch based on that... store-and-forward switches are so "1990s")
Even still - low 100 ms for store-and-forward ethernet switches? That seems really, really high. I would've said more like single milliseconds, which is still high, but it isn't 100 ms.
I know from experience that I've used store-and-forward ethernet switches with much, much better latency than 100 ms.
Maybe they should invest in basic research trying to solve the "energy input > output" for controlled fusion instead?
Huh? They have. From a basic research point of view, JT-60 passed breakeven in 1998. From an experimental point of view they didn't, but that's due to their inability to use D-T fuel (which ITER can handle). Note that their inability to handle it is a political/radioactivity safety issue, not an engineering issue.
ITER very likely might not be commercially viable (i.e. the costs will exceed the benefits) but the basic design is theoretically viable. It should generate more power than it takes in.
i dunno about the old xbox though, since that might just go away completely.
New, at least. Part of the problem here is that the Xbox is decent COTS equipment: it is just a PC, after all, and hacking it at this point is really, really trivial. So unfortunately, Xboxes still command a hefty resale point on eBay: you'll never get a working one for much under $100, for instance. Heck, a broken one sells for $50 on eBay.
It's fairly insane how well the Xbox is maintaining its value compared to a GameCube and a PS2.
Better than that - equations. My personal favorites are "heart square equals the sum of all mushrooms" and "fireflower is greater than or equal to square divided by warp pipe."
Let me rephrase that for you, "Driver support for a beta version of any OS may be lacking, so you might have trouble."
Really? There doesn't seem to be significant driver support issues for -pre versions of Linux.
And given that Vista is supposed to be able to use Windows XP drivers, there really shouldn't be an issue.
The only way this would indicate a "Major Problem" would be if we were at RC1 or so.
Well, that I'd agree with. Especially if the guy was just saying that the machine had to reboot after every driver, and it took 10 minutes to install each new driver. If that's all, it's not a big deal. If it's crashing every 10 minutes after the new drivers are installed, that's different. It's a bit ambiguous in the story.
Kind of hypocritical given Nintendo's track record of Mario platformers, racers, beat-em-ups and tennis games, along with the neverending Zelda series and the myriad Pokemon versions.
Not really. Most of Nintendo's sequels are much, much different than previous ones - they just keep some of the style and names the same. The biggest example I can think of is Majora's Mask for the N64. Sure, it's a Zelda game, but it feels wildly different than Ocarina of Time. Wind Waker felt different than both (though not in a good way).
This is as opposed to, say, Kingdom Hearts II, which... okay, I like, but it just feels so much like the original one for most of the game so far. Some of the combat's changed a bit, but that's kinda minor, as the way in which the combat's changed isn't that integral to the game.
Now, this isn't true for all of their games (Super Smash Bros. or Mario Kart, for example) but for Mario and Zelda it doesn't seem that similar.
Yeah, I'd really like to know which it was. If he means he needed to reboot after each installation, okay, that's a bit of a pain, but not that bad. If that means the machine's unstable, that's a much bigger concern.
Really? Then why is MS distributing this paper to XP driver writers about backward compatability issues? If you have evidence that MS has indicated that all XP drivers are supported then please enlighten me. Even if they have said so the cited paper (directed to driver developers) leaves some ambiguity.
Because nothing's perfect. There was a similar paper distributed when XP was released regarding driver compatibility then, too, and there are even Windows 2000 drivers that don't work in XP. But that's not the case for the majority of them.
Besides, if you read the paper, you realize that most of the problems listed there are install issues. These are much, much easier to fix than actual interface changes. The other changes are limitations rather than utter failures - for instance, WinXP drivers can't run the Aero interface, or deprecating interfaces that existed pre-XP (i.e. Vista doesn't support two-generation old drivers).
Most large provider's customers are HP, Dell, Gateway, etc., not you and I.
Who, unsurprisingly, will continue to release computers with old hardware with Vista on them. Thus demanding support for it.
If there's one person championing the openness of drivers it's Theo de Raadt, and he thinks otherwise.
Yeah, I disagreed with de Raadt when he said it, and I still do now. I buy a lot of small-vendor crap (like Ezonics stuff, for instance), and they simply don't care. I've emailed several companies about trying to get info to write drivers for it, and it's just been utter silence, or a stamped response "Thank you but we are not interested in Linux support at this time."
I've got a feeling de Raadt's opinion of 'small' is different than mine.
Could it not simply mean that it changes in size? I'd be surprised if it *didn't* change in size, based on all the variable energy in the solar system. The sun changes, the planets change place, etc.
That's actually very likely taken into account. When Voyager 1 found the heliopause, they were pretty sure that the termination shock was moving inward fairly rapidly due to it being past solar max, and so Voyager 2 would catch it pretty soon. This sounds like it happened quicker than their predictions expected. See here for more details. Note that they thought at the time that Voyager 2 might encounter the shock in 2005, but it might actually not catch it at all (as it might start moving outward again).
I saw the guy mentioned in the article (Ed Stone) give a talk on Voyager 1's results, so I'm pretty sure he'd know, especially since he, y'know, mentioned it.
I said that for an overwhelming majority of users, it doesn't matter.
No. You said for an overwhelming majority of users, it won't matter. In time. When Vista is released, for an overwhelming majority of users, it will matter.
Then Microsoft doesn't care about you.
When Vista ships, everyone will be me, unless people go out and buy new computers specifically for Vista.
If Microsoft really didn't care about backwards compatibility of hardware, they wouldn't've gone through such hell to support it in Windows XP.
Now, Microsoft doesn't give a crap about supporting hardware that only has drivers from Windows 98, and that I understand. But one system back? Yeah, I don't buy that. The majority of users don't buy all new hardware when a new OS comes out.
It means that they can't change the driver model while simultaneously stating that their new operating system will run on any existing system that has drivers for the old model.
But... they are doing that. Vista is supposed to run with XP (and 2000) drivers. This guy is trying to run it with XP drivers, and it fails.
It's the vendor's fault in both cases and the vendor doesn't care in either case.
It's not the vendor's fault in Microsoft's case. Microsoft has said that Vista will work with XP's drivers, so why should a vendor need to release new drivers in time for Vista to come out? Releasing them eventually would be nice, but it doesn't need to be done before Vista's released.
They don't give a rat's ass about supporting the crap you already bought on a new Windows platform (unless they're still selling the same crap).
You're a little... um... pessimistic about hardware providers. Most large providers do care about their customers and provide Linux drivers, future OS version drivers, etc. It's the small vendors who typically don't.
I don't agree that drivers necessarily have to contain trade secrets. That's my main issue. The hardware itself contains the trade secrets. Everything else should just be interface code.
I kindof see NVIDIA and ATi's point here, though - it's much like the old softmodems: it's cheaper to make dumb hardware (although in this case, it's not so dumb) and do a bunch more work in software than it is to make properly working hardware. What I wish is that NVIDIA and ATi would work with the Linux kernel to move more of that stuff into userland, and properly support the stuff that has no trade secret issues at all (i.e. 2D).
No one (or at least, few) would care if NVIDIA required a "NVIDIA Accelerator Program" running in userland to run games - especially if that NVIDIA Accelerator Program contained a shim against all non-protected code to allow recompiling for different system libraries.
they don't have to worry about rogue chip manufacturers selling knockoffs of their products
There is nowhere near enough info in a driver to allow someone to make a cheap knockoff of a graphics chip - or really, any chip. There are some trade secrets embedded in graphics drivers nowadays, but that's just because OpenGL/DirectX are too abstract to implement cheaply completely in hardware, and so you do some tricks. But if all you know is "here are the chip registers, here's what they do, here's the way the card's supposed to respond, here are the interrupts..." - that doesn't make it significantly easier to implement the hardware. After all, you still need to figure out how it does what it does.
I don't think its Microsoft's responsibility to keep the driver api identical for a new release of windows. Many of us complain about Microsoft's lack of security including drivers. It might have been more difficult for Sony DRM or many rootkits to work if the driver model were more secure and more drivers could run in userland.
There's a warning in Windows XP: "This driver has not been digitally signed"... blah blah blah if you insert an old or unsigned driver. That's all Microsoft has to do when they change the driver interface, because they don't release all of the interface with it. Therefore if something happens (like the Sony issue) Microsoft is perfectly fine in saying "this is a temporary issue, it will go away in the next iteration of Windows after Vista, when all drivers have been migrated over - oh, and talk to YOUR_HARDWARE_VENDOR for a new Vista driver, if you want it fixed sooner."
As an example: the SiS 530 is an AGP video card from the end of the Super Socket 7 Era. These machines were fairly old, but they were well within Windows XP's requirements: up to 550 MHz machines, 256 MB of RAM, etc. But SiS never redid the drivers for them for Windows XP, and has no intention to, and so using it in Windows XP is a *disaster* - they're ungodly slow, because you're just using the generic VGA driver. For some weird reason, WinXP can't use Win2K's driver (note that it *should* be able to), and so WinXP is basically off the table.
That's a very niche case, and so I don't hold it against Microsoft. There's probably something really weird going on there in the way that SiS implemented the driver - I've had plenty of luck using Win2K drivers in WinXP, so it's not a general problem. But if it were the common case, I'd be seriously pissed.
Note that I'm not saying they have to keep the driver interface (it shouldn't be called an API, should it? you're not writing an application) fixed. They just have to at least have nearly perfect compatibility with the previous version's drivers. Well, that, and not change the interface that fast.
Its ok for linux developers to drastically change a minor kernel release 2.6.x but not ok for microsoft to change on a major release?
Yes. It's because the Linux developers provide virtually all the drivers needed along with each change. Microsoft doesn't - it relies on the other vendors, and they have other work to do.
It's not a double standard. I'm just requiring "shipping an OS that works" - including drivers. In Microsoft's case, they do it by having compatibility with previous versions. In Linux's case, they do it by providing all the drivers themselves.
but by the time it ships, users won't have to look for drivers, Lenovo will have it preconfigured already.
Does Lenovo ship preconfigured versions of future operating system revisions when they come out for older hardware? I don't think they do...
Now, if you're saying "future hardware will work fine with Vista" that's fine, I agree, but I'm concerned with how my current hardware will work. I stopped chasing the bleeding edge years ago.
BTW, isn't the Slashdot mentality great? Poor driver support for Linux: "Broadcom/ATI/whoever Is The Devil." Poor driver support for Windows: "Vista Beta 2 has Major Problems." Go Figure.
It's a driver model thing. In Linux, the OS maintainers maintain the drivers, so they need info from the manufacturers. For a niche OS, this is probably intelligent - Broadcom isn't going to make drivers for Linux, just like they didn't make drivers for BeOS. It's better to ask them "hey, this niche OS isn't that niche anymore, it's pretty widely used. Give us the info, we'll maintain the driver for you" than to ask them to provide the driver, although for some reason they think it's easier to do the latter.
Add in the fact that Linux makes no claim of maintaining driver compatibility, and it's easy to see why "the manufacturer is the devil" there.
Microsoft, however, doesn't do that. They don't claim to even try to do that. They leave the driver programming to the vendors. Now, that means that Microsoft now has a responsibility to make sure that when they change the driver model, they either maintain driver compatibility, or make sure all the vendors have new drivers. The latter is basically impossible (vendors die, after all), so the former is absolutely critical.
Now, it's a little unfair to criticize vendors because they can't support Linux, because Linux's driver model is a little difficult for them to support the same way they do under Windows. So yeah, maybe we should actually be criticizing Linux as well for not having a stable driver interface. But if the information from the vendor was available, that wouldn't be important, so... it's a tough call.
I guess what I'm saying is, if both driver models (stable interface, vendors do all the work, and changing interface, OS maintainers do all the work) should be supported by vendors, then yes, the vendor is at fault in Linux's case, and Microsoft is at fault in Microsoft's case. Whether or not both driver models should be supported is another question - but it's not necessarily as hypocritical as you're making it out to be.
Yeah, I don't really see how that's any different from XP.
It's different from XP because once you get the drivers, XP works. Apparently according to the article, once you get the drivers, the machine still reboots every 10 minutes.
The article writes this off as "well, the drivers weren't Vista-compatible" but Vista-compatible drivers aren't all going to be available at launch - they better be able to use XP drivers. And it better not crash every 10 minutes when doing so.
Now, it's entirely possible that that compatibility system is the one thing they're still working on, but that's gotta work at launch. Otherwise most people won't touch Vista for a while - most likely, not till their next computer.
How dare they not support something that isn't widely avalible for the public and could change at the drop of a hate for any reason!
Drop of a hat, I'm guessing you mean.
However, you're spewing a good deal of vitriol at me for apparently criticizing Dell, when I'm not criticizing Dell. I'm criticizing Microsoft.
It depends exactly how far along Vista is supposed to be, and how compatible its drivers are supposed to be. Is it supposed to work with XP drivers? If so, there's no reason it shouldn't work. Also, is it a year away? two years away? six months away? Do the drivers exist and not work? If they're not compatible with previous drivers, they sure as hell better make sure that vendors like Dell, Lenovo, etc. have drivers available already. It'll take a while to work the bugs out of them.
Yes, in fact, I would expect an OS to support hardware that's been out, and working, fully qualified, on a previous revision of the OS - if by no other mechanism than providing a compatibility layer that accepts previous drivers (and yes, I have issues with the fact that Linux doesn't work that way).
Otherwise, if the drivers aren't ready now, how do you know they'll be ready - and working - at launch? Because if they aren't, then that is a problem at launch.
And what I'd like to know is what this meant:
Downloading the drivers from the Lenovo Web site took a long time, but after a while I had downloaded and installed the package. Unfortunately, not everything I downloaded is Vista-ready so rebooting had to occur after every 10 minutes of computer use.
Why did he have to reboot every 10 minutes? Was it crashing? If so, that is a big concern. Unless Microsoft makes a miraculously huge effort to make sure all of their driver providers have Vista-ready drivers at launch, they better have decent backwards-compatibility for drivers.
Laptops have so much custom hardware these days that it's a Bad Idea(TM) to attempt an OS installation from anything but restore CDs.
Er?
I've installed Windows XP on all of my laptops over the past few years, and everyone else in my office does the same thing, too. Laptops come with too much cruft installed by default, and in general, it's silly for us to pay to upgrade to XP Pro when there's a site license available for next to nothing here. So wipe the drive, in goes a new installation of XP Pro, alongside Linux, typically. I've never run into a problem.
Jumping to Dell's site for the laptop I'm on now, all of the drivers are right there ready.
Now, there aren't Vista drivers. But if what he's saying is "driver support for Vista may be lacking, so you might have trouble", I don't really see that as a problem. A lot of people only have laptops nowadays, so not being able to install Vista on a laptop easily means a lot of people aren't buying Vista.
DVD was an emerging standard with a huge and noticeable advantage over the popular storage medium of the time (VHS), while BR is not
And, of course, DVD had already proven that it has a market - that is, there were people willing to pay $250-300 for a DVD player.
Blu-ray drives aren't on the market yet. They might debut at $1000, but if no one buys them there, someone's gonna have to work real fast to drop the price down.
I sure as hell wouldn't buy a glorified DVD player for $1000.
Of course, I do wonder if the PS3 will suck as much as a Blu-ray player as the PS2 did as a DVD player.
But it wasn't a uge leap forward and we knew that the consoles coming out soon after were going to be amazing.
What consoles coming out after it? The PlayStation and the N64? The Saturn was just as powerful as both.
Both of them were basically in the same class as the Saturn - exactly like the PS3. This wasn't a case of the market waiting for the later consoles. It was a case of the market saying "are you crazy? that's too expensive for something that's not that much better!"
Similarly, the reason there are 171 miles of wires is that most are point-to-point single purpose. If you completely overhauled all of the networks on a jet to share a common multiplexed bus
Then you'd have a single point of failure. The advantage to point-to-point wiring is that a panel that comes loose and shreds a wire only takes out one system, not half a dozen.
The benefit of wireless in this case is that you lose the wiring, but keep the reliability. Now, granted, you've still got to have powered cables running to the things, so unless they've developed pulsed microwave power, I don't see this as a win at all.
Slashdot Mirror
I seem to be the only person on Slashdot who really liked Luigi's Mansion.
I loved Luigi's Mansion.
But it was way, way too short for a launch title that was supposed to be a placeholder for a Mario game. Keep in mind Mario, Zelda didn't come out until basically a year later. There wasn't a good platformer for the console until... well, until Sunshine, and by then, people were bitter.
A console launch needs to keep you from putting the console away and forgetting about it for months. The GameCube launch didn't do that. If you're like me, and didn't buy a Cube until almost a year later, hey, it was great, and I still play the GameCube more than my PS2. But that was me, and most people had already formed their opinion of the GameCube by then.
The Wii launch is primed to do that. They're going to be missing Mario again, most likely, but this time they substituted a much, much better filler than Luigi's Mansion: Zelda and Metroid. They can't let Mario slip past the end of the year, though. Wii, with Mario, Zelda, and Metroid at Christmas, is a killer launch. I think they can even survive without SSBB for a year, too.
The second party is YOU. What Gamecube games did you release? :)
Second party is a wholly-owned subsidiary of the first party, typically published by the first party. Rare was a second-party publisher for a while before Nintendo sold them off.
I thought Factor 5 was a second party as well (that's why I put Rogue Squadron in there) - turns out they're not, so that was a mistake.
anyone who thinks the GameCubes launch lineup was bad forgot that the N64 launched with only Maro64 and Pilotwings64 available.
Mario 64 was a better title than all of Nintendo's GameCube offerings combined. Nintendo's GameCube offerings were the grand total of Luigi's Mansion and Wave Race: Blue Storm.
The GameCube's launch was worse than the N64's, in my opinion, and that's hard to do when you're competing with only two games.
Note what he's saying, though: it's not just 3rd party support (honestly, I couldn't care less about where the game comes from) this time. Here he's saying "yeah, the launch titles weren't diverse enough - period."
That's a big deal. And actually, I personally think that the problem with the N64 was 3rd party support, and the problem with the GameCube was first-party support at launch. The Nintendo launch titles for the GameCube were.. well.. less than stellar. Note here that I'm going to glom first and second party together, as I'm never sure which ones are first or second.
Here's Nintendo's at launch titles for the GameCube:
Even afterwards, it, uh, wasn't that much better. Soon-after-launch titles:
Uh. Yeah. That's... not that stellar. Not even that many, either. It's not that they were bad games. It's just that they were... well, few. And not Mario. Or Zelda. Instead, Wind Waker showed up a year later. Mario, nearly a year later as well. Both felt like too little, too late.
There's still plenty of time for Nintendo to screw up, mind you, but delaying Twilight Princess to the Wii launch (which virtually everyone thought that was what they were doing when they announced the delay) is actually quite smart. If they can get a good fraction of the titles at E3 out the door by launch, they're going to have a much, much more successful launch than the GameCube.
It says 100 milliseconds in the article.
I don't doubt that it's wrong (like I said, I know from experience that it's of order 1 ms, not 100 ms) but the article is the one that's wrong, not the story summary.
(Seriously, haven't people heard cut-through switches which just look at the first part of the header and switch based on that... store-and-forward switches are so "1990s")
Even still - low 100 ms for store-and-forward ethernet switches? That seems really, really high. I would've said more like single milliseconds, which is still high, but it isn't 100 ms.
I know from experience that I've used store-and-forward ethernet switches with much, much better latency than 100 ms.
Maybe they should invest in basic research trying to solve the "energy input > output" for controlled fusion instead?
Huh? They have. From a basic research point of view, JT-60 passed breakeven in 1998. From an experimental point of view they didn't, but that's due to their inability to use D-T fuel (which ITER can handle). Note that their inability to handle it is a political/radioactivity safety issue, not an engineering issue.
ITER very likely might not be commercially viable (i.e. the costs will exceed the benefits) but the basic design is theoretically viable. It should generate more power than it takes in.
Just what "various factors" does it take into account, anyway?
Sales tax versus VAT.
i dunno about the old xbox though, since that might just go away completely.
New, at least. Part of the problem here is that the Xbox is decent COTS equipment: it is just a PC, after all, and hacking it at this point is really, really trivial. So unfortunately, Xboxes still command a hefty resale point on eBay: you'll never get a working one for much under $100, for instance. Heck, a broken one sells for $50 on eBay.
It's fairly insane how well the Xbox is maintaining its value compared to a GameCube and a PS2.
There are MATH SYMBOLS in the CLOUDS!!!
Better than that - equations. My personal favorites are "heart square equals the sum of all mushrooms" and "fireflower is greater than or equal to square divided by warp pipe."
There's some fundamental truth there, I tell you.
Let me rephrase that for you, "Driver support for a beta version of any OS may be lacking, so you might have trouble."
Really? There doesn't seem to be significant driver support issues for -pre versions of Linux.
And given that Vista is supposed to be able to use Windows XP drivers, there really shouldn't be an issue.
The only way this would indicate a "Major Problem" would be if we were at RC1 or so.
Well, that I'd agree with. Especially if the guy was just saying that the machine had to reboot after every driver, and it took 10 minutes to install each new driver. If that's all, it's not a big deal. If it's crashing every 10 minutes after the new drivers are installed, that's different. It's a bit ambiguous in the story.
Kind of hypocritical given Nintendo's track record of Mario platformers, racers, beat-em-ups and tennis games, along with the neverending Zelda series and the myriad Pokemon versions.
Not really. Most of Nintendo's sequels are much, much different than previous ones - they just keep some of the style and names the same. The biggest example I can think of is Majora's Mask for the N64. Sure, it's a Zelda game, but it feels wildly different than Ocarina of Time. Wind Waker felt different than both (though not in a good way).
This is as opposed to, say, Kingdom Hearts II, which... okay, I like, but it just feels so much like the original one for most of the game so far. Some of the combat's changed a bit, but that's kinda minor, as the way in which the combat's changed isn't that integral to the game.
Now, this isn't true for all of their games (Super Smash Bros. or Mario Kart, for example) but for Mario and Zelda it doesn't seem that similar.
Yeah, I'd really like to know which it was. If he means he needed to reboot after each installation, okay, that's a bit of a pain, but not that bad. If that means the machine's unstable, that's a much bigger concern.
Really? Then why is MS distributing this paper to XP driver writers about backward compatability issues? If you have evidence that MS has indicated that all XP drivers are supported then please enlighten me. Even if they have said so the cited paper (directed to driver developers) leaves some ambiguity.
Because nothing's perfect. There was a similar paper distributed when XP was released regarding driver compatibility then, too, and there are even Windows 2000 drivers that don't work in XP. But that's not the case for the majority of them.
Besides, if you read the paper, you realize that most of the problems listed there are install issues. These are much, much easier to fix than actual interface changes. The other changes are limitations rather than utter failures - for instance, WinXP drivers can't run the Aero interface, or deprecating interfaces that existed pre-XP (i.e. Vista doesn't support two-generation old drivers).
Most large provider's customers are HP, Dell, Gateway, etc., not you and I.
Who, unsurprisingly, will continue to release computers with old hardware with Vista on them. Thus demanding support for it.
If there's one person championing the openness of drivers it's Theo de Raadt, and he thinks otherwise.
Yeah, I disagreed with de Raadt when he said it, and I still do now. I buy a lot of small-vendor crap (like Ezonics stuff, for instance), and they simply don't care. I've emailed several companies about trying to get info to write drivers for it, and it's just been utter silence, or a stamped response "Thank you but we are not interested in Linux support at this time."
I've got a feeling de Raadt's opinion of 'small' is different than mine.
Could it not simply mean that it changes in size? I'd be surprised if it *didn't* change in size, based on all the variable energy in the solar system. The sun changes, the planets change place, etc.
That's actually very likely taken into account. When Voyager 1 found the heliopause, they were pretty sure that the termination shock was moving inward fairly rapidly due to it being past solar max, and so Voyager 2 would catch it pretty soon. This sounds like it happened quicker than their predictions expected. See here for more details. Note that they thought at the time that Voyager 2 might encounter the shock in 2005, but it might actually not catch it at all (as it might start moving outward again).
I saw the guy mentioned in the article (Ed Stone) give a talk on Voyager 1's results, so I'm pretty sure he'd know, especially since he, y'know, mentioned it.
I said that for an overwhelming majority of users, it doesn't matter.
No. You said for an overwhelming majority of users, it won't matter. In time. When Vista is released, for an overwhelming majority of users, it will matter.
Then Microsoft doesn't care about you.
When Vista ships, everyone will be me, unless people go out and buy new computers specifically for Vista.
If Microsoft really didn't care about backwards compatibility of hardware, they wouldn't've gone through such hell to support it in Windows XP.
Now, Microsoft doesn't give a crap about supporting hardware that only has drivers from Windows 98, and that I understand. But one system back? Yeah, I don't buy that. The majority of users don't buy all new hardware when a new OS comes out.
It means that they can't change the driver model while simultaneously stating that their new operating system will run on any existing system that has drivers for the old model.
But... they are doing that. Vista is supposed to run with XP (and 2000) drivers. This guy is trying to run it with XP drivers, and it fails.
It's the vendor's fault in both cases and the vendor doesn't care in either case.
It's not the vendor's fault in Microsoft's case. Microsoft has said that Vista will work with XP's drivers, so why should a vendor need to release new drivers in time for Vista to come out? Releasing them eventually would be nice, but it doesn't need to be done before Vista's released.
They don't give a rat's ass about supporting the crap you already bought on a new Windows platform (unless they're still selling the same crap).
You're a little... um... pessimistic about hardware providers. Most large providers do care about their customers and provide Linux drivers, future OS version drivers, etc. It's the small vendors who typically don't.
I don't agree that drivers necessarily have to contain trade secrets. That's my main issue. The hardware itself contains the trade secrets. Everything else should just be interface code.
I kindof see NVIDIA and ATi's point here, though - it's much like the old softmodems: it's cheaper to make dumb hardware (although in this case, it's not so dumb) and do a bunch more work in software than it is to make properly working hardware. What I wish is that NVIDIA and ATi would work with the Linux kernel to move more of that stuff into userland, and properly support the stuff that has no trade secret issues at all (i.e. 2D).
No one (or at least, few) would care if NVIDIA required a "NVIDIA Accelerator Program" running in userland to run games - especially if that NVIDIA Accelerator Program contained a shim against all non-protected code to allow recompiling for different system libraries.
they don't have to worry about rogue chip manufacturers selling knockoffs of their products
There is nowhere near enough info in a driver to allow someone to make a cheap knockoff of a graphics chip - or really, any chip. There are some trade secrets embedded in graphics drivers nowadays, but that's just because OpenGL/DirectX are too abstract to implement cheaply completely in hardware, and so you do some tricks. But if all you know is "here are the chip registers, here's what they do, here's the way the card's supposed to respond, here are the interrupts..." - that doesn't make it significantly easier to implement the hardware. After all, you still need to figure out how it does what it does.
I don't think its Microsoft's responsibility to keep the driver api identical for a new release of windows. Many of us complain about Microsoft's lack of security including drivers. It might have been more difficult for Sony DRM or many rootkits to work if the driver model were more secure and more drivers could run in userland.
There's a warning in Windows XP: "This driver has not been digitally signed"... blah blah blah if you insert an old or unsigned driver. That's all Microsoft has to do when they change the driver interface, because they don't release all of the interface with it. Therefore if something happens (like the Sony issue) Microsoft is perfectly fine in saying "this is a temporary issue, it will go away in the next iteration of Windows after Vista, when all drivers have been migrated over - oh, and talk to YOUR_HARDWARE_VENDOR for a new Vista driver, if you want it fixed sooner."
As an example: the SiS 530 is an AGP video card from the end of the Super Socket 7 Era. These machines were fairly old, but they were well within Windows XP's requirements: up to 550 MHz machines, 256 MB of RAM, etc. But SiS never redid the drivers for them for Windows XP, and has no intention to, and so using it in Windows XP is a *disaster* - they're ungodly slow, because you're just using the generic VGA driver. For some weird reason, WinXP can't use Win2K's driver (note that it *should* be able to), and so WinXP is basically off the table.
That's a very niche case, and so I don't hold it against Microsoft. There's probably something really weird going on there in the way that SiS implemented the driver - I've had plenty of luck using Win2K drivers in WinXP, so it's not a general problem. But if it were the common case, I'd be seriously pissed.
Note that I'm not saying they have to keep the driver interface (it shouldn't be called an API, should it? you're not writing an application) fixed. They just have to at least have nearly perfect compatibility with the previous version's drivers. Well, that, and not change the interface that fast.
Its ok for linux developers to drastically change a minor kernel release 2.6.x but not ok for microsoft to change on a major release?
Yes. It's because the Linux developers provide virtually all the drivers needed along with each change. Microsoft doesn't - it relies on the other vendors, and they have other work to do.
It's not a double standard. I'm just requiring "shipping an OS that works" - including drivers. In Microsoft's case, they do it by having compatibility with previous versions. In Linux's case, they do it by providing all the drivers themselves.
but by the time it ships, users won't have to look for drivers, Lenovo will have it preconfigured already.
Does Lenovo ship preconfigured versions of future operating system revisions when they come out for older hardware? I don't think they do...
Now, if you're saying "future hardware will work fine with Vista" that's fine, I agree, but I'm concerned with how my current hardware will work. I stopped chasing the bleeding edge years ago.
BTW, isn't the Slashdot mentality great? Poor driver support for Linux: "Broadcom/ATI/whoever Is The Devil." Poor driver support for Windows: "Vista Beta 2 has Major Problems." Go Figure.
It's a driver model thing. In Linux, the OS maintainers maintain the drivers, so they need info from the manufacturers. For a niche OS, this is probably intelligent - Broadcom isn't going to make drivers for Linux, just like they didn't make drivers for BeOS. It's better to ask them "hey, this niche OS isn't that niche anymore, it's pretty widely used. Give us the info, we'll maintain the driver for you" than to ask them to provide the driver, although for some reason they think it's easier to do the latter.
Add in the fact that Linux makes no claim of maintaining driver compatibility, and it's easy to see why "the manufacturer is the devil" there.
Microsoft, however, doesn't do that. They don't claim to even try to do that. They leave the driver programming to the vendors. Now, that means that Microsoft now has a responsibility to make sure that when they change the driver model, they either maintain driver compatibility, or make sure all the vendors have new drivers. The latter is basically impossible (vendors die, after all), so the former is absolutely critical.
Now, it's a little unfair to criticize vendors because they can't support Linux, because Linux's driver model is a little difficult for them to support the same way they do under Windows. So yeah, maybe we should actually be criticizing Linux as well for not having a stable driver interface. But if the information from the vendor was available, that wouldn't be important, so... it's a tough call.
I guess what I'm saying is, if both driver models (stable interface, vendors do all the work, and changing interface, OS maintainers do all the work) should be supported by vendors, then yes, the vendor is at fault in Linux's case, and Microsoft is at fault in Microsoft's case. Whether or not both driver models should be supported is another question - but it's not necessarily as hypocritical as you're making it out to be.
Yeah, I don't really see how that's any different from XP.
It's different from XP because once you get the drivers, XP works. Apparently according to the article, once you get the drivers, the machine still reboots every 10 minutes.
The article writes this off as "well, the drivers weren't Vista-compatible" but Vista-compatible drivers aren't all going to be available at launch - they better be able to use XP drivers. And it better not crash every 10 minutes when doing so.
Now, it's entirely possible that that compatibility system is the one thing they're still working on, but that's gotta work at launch. Otherwise most people won't touch Vista for a while - most likely, not till their next computer.
Drop of a hat, I'm guessing you mean.
However, you're spewing a good deal of vitriol at me for apparently criticizing Dell, when I'm not criticizing Dell. I'm criticizing Microsoft.
It depends exactly how far along Vista is supposed to be, and how compatible its drivers are supposed to be. Is it supposed to work with XP drivers? If so, there's no reason it shouldn't work. Also, is it a year away? two years away? six months away? Do the drivers exist and not work? If they're not compatible with previous drivers, they sure as hell better make sure that vendors like Dell, Lenovo, etc. have drivers available already. It'll take a while to work the bugs out of them.
Yes, in fact, I would expect an OS to support hardware that's been out, and working, fully qualified, on a previous revision of the OS - if by no other mechanism than providing a compatibility layer that accepts previous drivers (and yes, I have issues with the fact that Linux doesn't work that way).
Otherwise, if the drivers aren't ready now, how do you know they'll be ready - and working - at launch? Because if they aren't, then that is a problem at launch.
And what I'd like to know is what this meant:
Why did he have to reboot every 10 minutes? Was it crashing? If so, that is a big concern. Unless Microsoft makes a miraculously huge effort to make sure all of their driver providers have Vista-ready drivers at launch, they better have decent backwards-compatibility for drivers.
Laptops have so much custom hardware these days that it's a Bad Idea(TM) to attempt an OS installation from anything but restore CDs.
Er?
I've installed Windows XP on all of my laptops over the past few years, and everyone else in my office does the same thing, too. Laptops come with too much cruft installed by default, and in general, it's silly for us to pay to upgrade to XP Pro when there's a site license available for next to nothing here. So wipe the drive, in goes a new installation of XP Pro, alongside Linux, typically. I've never run into a problem.
Jumping to Dell's site for the laptop I'm on now, all of the drivers are right there ready.
Now, there aren't Vista drivers. But if what he's saying is "driver support for Vista may be lacking, so you might have trouble", I don't really see that as a problem. A lot of people only have laptops nowadays, so not being able to install Vista on a laptop easily means a lot of people aren't buying Vista.
DVD was an emerging standard with a huge and noticeable advantage over the popular storage medium of the time (VHS), while BR is not
And, of course, DVD had already proven that it has a market - that is, there were people willing to pay $250-300 for a DVD player.
Blu-ray drives aren't on the market yet. They might debut at $1000, but if no one buys them there, someone's gonna have to work real fast to drop the price down.
I sure as hell wouldn't buy a glorified DVD player for $1000.
Of course, I do wonder if the PS3 will suck as much as a Blu-ray player as the PS2 did as a DVD player.
But it wasn't a uge leap forward and we knew that the consoles coming out soon after were going to be amazing.
What consoles coming out after it? The PlayStation and the N64? The Saturn was just as powerful as both.
Both of them were basically in the same class as the Saturn - exactly like the PS3. This wasn't a case of the market waiting for the later consoles. It was a case of the market saying "are you crazy? that's too expensive for something that's not that much better!"
Similarly, the reason there are 171 miles of wires is that most are point-to-point single purpose. If you completely overhauled all of the networks on a jet to share a common multiplexed bus
Then you'd have a single point of failure. The advantage to point-to-point wiring is that a panel that comes loose and shreds a wire only takes out one system, not half a dozen.
The benefit of wireless in this case is that you lose the wiring, but keep the reliability. Now, granted, you've still got to have powered cables running to the things, so unless they've developed pulsed microwave power, I don't see this as a win at all.