No, open source doesn't mean you can't make money selling it. But what it does mean is that you can't charge a software tax on sales of software made by third parties. Sony makes money on every single PlayStation game sold, by every single publisher. That model won't work here. So, question is, how do they make money?
How are you planning to break even? The traditional console business model is to sell the console itself at a loss, and make money in the form of a 'tax' on the software. With a truly open system, this won't be possible. Unless I've missed something, that leaves you with a profit margin on the console itself as your source of revenue - but to turn a profit on the console would mean charging far more for it than your competitors will charge for their machines.
There's one extra difference in the way Direct3D and OpenGL expose features. Although with both APIs you have to check for the presense of the feature (cap query for D3D, extension query for GL), D3D then has a standard way to drive that feature, whereas with GL you might have to check seperately for each of several essentially similar but irritatingly different vendor extensions.
Expect this problem to get worse rather than better as more esoteric features come on stream that stray farther and farther from the GL 'core'. Sure, eventually the ARB sweeps up and standardises everything, but that can take a pretty long time.
Example: NV_register_combiners. An insanely complicated extension for NVidia hardware that pretty much does the job that D3D pixel shaders are supposed to do. But the NV extension is built in such away that it's almost never going to be implementable by other vendors (since it's basically just a register dump rather than sensible abstraction). So with GL you'll have to contend with that, along with whatever schemes get cooked up by ATI, 3dfx, and the rest.
Essentially, GL is now an open API in name only. The hardware vendors are basically just building their own private little proprietary APIs out of extensions and evangelising them like crazy, all the while claiming to be supporting open standards.
Microsoft is full of people who worked on Windows. I'd be pretty damn surprised if none of them suggested that Bungie might like to try running their servers on Windows.
Just suppose that RedHat had acquired a game company that ran all their servers on NT. Then suppose a few RedHat employees gently suggested that they might like to run on Linux instead. Would we be in the least surprised? Would we think this was in some way evil? Of course not.
Why this whole story is in any way surprising to anyone is quite beyond me.
It's not cheating as in hacking the game, or typing in cheat codes. In fact, cheating is probably the wrong word. It's just the use of some very nifty and sometimes highly inventive techniques.
For example, a "rocket jump" involves firing a rocket into the ground and jumping simultaneously to give yourself an extra high jump as the blast sends you higher. You sustain damage in the process, but it enables you to reach otherwise inaccessible spots. It's not considered cheating, indeed, it's a skill that any seasoned Quake player will have.
These guys have taken it even further, and regularly pull off grenade jumps (same principle as rocket jumps but your timing needs to be better), and even the fiendishly tricky grenade and rocket jump combo.
To put it another way, have you earned your right to vote? How bout your right to a trial? What about making a living wage?
No, but I did have to earn my right to, for example, drive a car. I had to take a test to prove that I knew how to operate one safely, and was aware of the basic rules and regulations of the jurisdiction in which I mostly drive.
I'm not saying that we should necessarily require a test before people are allowed broadband access, I'm just pointing out that it's not intrinsically as ridiculous as you are implying. The idea that for some human activities we choose to restrict those permitted to engage in them seems very reasonable. Even voting (your example) is restricted by age, because society considers that people need to be a certain age before they are able to make a reasonable and informed decision about who to vote for.
Myself, I always thought of Diablo as being a blend between Rouge/NetHack and Gauntlet.
Now, I realise you were being flippant, but it is fairly obviously the case that there was a pretty large chunk of innovation that went into making Diablo different from its influences. It's certainly worlds away from copying Pac-Man verbatim and merely changing the name of the protagonist.
How exactly are computer games any different from other IP? No-one ever seems to think that it's unfair for song-writers to get royalities when other bands do cover versions of their work. It's an analogous situation here.
"I speak a little Japanese from what I took in college, and there are some unique features of the language that might effect syntactical structure.
"For instance, the acting verb in a sentence almost always occurs at the end of a sentence."
Sorry, I can't resist being nitpicky, but that's not unique to Japanese: Latin typically puts the verb at the end of the sentence too.
Re:Guinness didn't want to be remembered for Obi W
on
Sir Alec Guinness Dies
·
· Score: 1
I don't think it was so much the film itself that he objected to, so much as the obessive nature of the fans and the fact that it was all anyone ever wanted to talk to him about thereafter. His opinion that he did better work in other movies is perfectly valid, and certainly one that many (including me) would agree with.
As far as your claim that he had a "massive ego" goes, nothing could be further from the truth. Guinness had a reputation for being extremely modest and shy of publicity.
nVidia was one of the worst offendors in telling review sites to downplay 16-bit color benchmarks on games... even on ones where the textures were in 16-bit color (so no benifit to 32 bit!) for the sole benifit of making 3dfx's faster card look worse. (ATI was particuarly bad, as well).
Actually, even when you have 16-bit source textures, the end result can look a lot better when rendered to a 32-bit framebuffer. The reason is that you don't just do a plain copy of the texture, but you typically modulate it with an interpolated lighting value and possibly apply multipass effects as well. With a 16-bit framebuffer its quite easy to start to see banding and/or dithering artefacts that wouldn't be there with a 32-bit framebuffer, regardless of the bitdepth of the source artwork.
What makes this so interesting? It's basically just a big powerful server. I don't really see why this is a superior solution for media serving than some well tested heavy-iron running a well known operating system.
If you were an IT manager, making a media server purchasing decision, would you rather go with something that has a decent reliability record and tools support, or an untried offering from Sony which will have limited initial support?
Furthermore, unless I missed it, there was no mention of what operating system it's going to run. The only vaguely plausible option would seem to be Linux, but you do have to ask yourself whether you wouldn't want the choice to run Solaris or *BSD instead for critical heavy-duty applications like media serving. And in the event that they turn out to be crazy enough to try shipping their own homespun OS, then any IT manager worth their salt should run a mile (heck, I'd rather use NT/Win2K than any sort of Sony-built freakshow).
Problem with that is that it's difficult to see how the economic model would work. Game consoles are sold at a loss, with profits being made on licensed peripherals and titles.
With a spec that anyone could manufacture, how would manufacturers make money? They either have to sell the consoles at a profit (which is going to be hard to do, given that they'd have to be priced about twice the cost of competing 'closed' consoles), or there needs to be some scheme for them to get a kickback on title and peripheral sales.
"-- The average sound card in a computer is crap. Complete and utter garbage when it comes to the fidelity of the audio outputs and the quality of the DACs."
Crap it may be, but then MP3 is crap in the first place (at least at any reasonable bitrate). It turns the top end into nasty, tinny slush, and bass becomes incredibly flabby. Anyone with an ear for fidelity wouldn't be using MP3 to start with.
Of course, it turns out that lots of people don't have an ear for fidelity (perhaps, ironically, as a result of going to too many rock concerts?), hence all the oft repeated claims that "it's as good as CD", and "I can't tell the difference". For the lucky people who can't tell the difference, just run a cable from your PC and have done with it. The rest of us will stick with CDs.
They test on a dual processor system, but then test versus Windows 95 which, as we all know, doesn't support multiple processors. A fairer test would be to either test on a single CPU machine, or versus NT/Windows 2000.
Although Microsoft now owns the whole of Bungie, as part of the deal, Take 2 Interactive (who used to own 19.9% of Bungie) have acquired all the rights to Oni and Myth, as well as the rights to build two titles based on the Halo engine.
Bungie have also been quoted as saying that they will remain autonomous within MS, and may continue to develop titles for non-MS platform (e.g. Mac), although it remains to see how long that lasts. I suspect that Mac titles may be allowed to continue for a little while, but PlayStation 2 titles will be knocked right on the head in favour of X-Box.
Regardless of whether or not MS is a monopoly in the PC OS business, when it comes to the console business they most certainly are not. IANAL, but I thought that you had to consider the relevant market when applying anti-trust considerations. In this case, MS would be doing nothing any different to Sony, Sega, and Nintendo if they controlled access to the X-Box. Indeed, the whole console business model rests upon this fact - if you couldn't control content you'd have to sell the consoles at a profit, which would suddenly mean $900 PS2s.
Of course, when Sony brought out the Playstation they had no experience of the console market either. And everyone said at the time how they would be crushed by Sega and Nintendo.
1: Design a physics engine that really only works properly for rigid 1 meter cubes.
2: Populate the game with (non-cuboid) 10 ton lizards.
3: Substitute player's arm with a 10 foot rubber probiscus with a crazy control mechanism which the player is expected to use to manipulate all objects in the game, including guns which need to be fired accurately.
4: Add an out-of-date graphics engine and turn the whole lot loose on some designers who can't come up with a puzzle more sophisiticated than "stack box A on top of box B".
Et voila! Worst game ever made! What's that you say? Someone's thought of it already?
1: For performance/smoothness reasons you may well want to leave some decisions in the hands of the client and/or supply more information to the client that is strictly needed to be conveyed to the player at that instant. While decision making can ultimately be validated by the server, once you've sent information out to the client you can't call it back.
2: It's easy to construct examples where you have given the client the minimum amount of information to function correctly, yet cheating is still possible. For example, suppose you have a cloaking device in the game which renders the player almost completely transparent. The client needs to know where that player is in order to draw it. However, a hacked client could draw that player in bright, easy-to-see colors, rather than the almost-invisible transparency that was the intent. Similar sorts of cheating are evident in Quake, where cheaters replace player models with large, brightly lit boxes that show up in dark corners and poke through obstructions.
Now I guess you could argue that you just should modify your game design to avoid these kinds of situation, but (a) you'll always miss something, and (b) why the hell should I mess with my game design just to serve a source licencing ideology?
I assert that the best solution to these problems is currently security through obscurity. It's not a great solution, but it's the best one I've heard. If you can prove me wrong, I'd love for you to do it because I'm working on a multiplayer game and this is giving me a real headache.
Good question. Be interesting to hear RMS's answer. My view is that games are different, and here's why:
Part of many games is deliberately obscuring information from the player that is actually 'known' by the software. For example, the software knows the positions of enemies and states of game objects, that the player might not be supposed to be allowed to see at that point in the game. If the game were open source it would be a simple matter to modify the game to report that information - ruining the game. For a single-player game this cheating isn't really a problem - if the player has fun doing it, then who is to say it's a bad thing? For a multi-player game though, cheating can be fatal.
There doesn't seem to be any way of preventing this cheating, other than to make it really hard for a cheater to modify their game and/or data in an undetectable way, and one of the most effective ways to make it hard is not to give the source out. Of course, this doesn't prevent cheating in principle, but in practice it hopefully slows down people enough that the game will be stale before anyone does serious damage.
The key point is that games are different to other apps. In any other app if you have data then it's reasonable to expect the user to be allowed to view it in any way they wish, and if the app is modified to allow that data to be presented 'better' then this is a good thing. With games this is simply not true.
Slashdot Mirror
No, open source doesn't mean you can't make money selling it. But what it does mean is that you can't charge a software tax on sales of software made by third parties. Sony makes money on every single PlayStation game sold, by every single publisher. That model won't work here. So, question is, how do they make money?
How are you planning to break even? The traditional console business model is to sell the console itself at a loss, and make money in the form of a 'tax' on the software. With a truly open system, this won't be possible. Unless I've missed something, that leaves you with a profit margin on the console itself as your source of revenue - but to turn a profit on the console would mean charging far more for it than your competitors will charge for their machines.
Expect this problem to get worse rather than better as more esoteric features come on stream that stray farther and farther from the GL 'core'. Sure, eventually the ARB sweeps up and standardises everything, but that can take a pretty long time.
Example: NV_register_combiners. An insanely complicated extension for NVidia hardware that pretty much does the job that D3D pixel shaders are supposed to do. But the NV extension is built in such away that it's almost never going to be implementable by other vendors (since it's basically just a register dump rather than sensible abstraction). So with GL you'll have to contend with that, along with whatever schemes get cooked up by ATI, 3dfx, and the rest.
Essentially, GL is now an open API in name only. The hardware vendors are basically just building their own private little proprietary APIs out of extensions and evangelising them like crazy, all the while claiming to be supporting open standards.
Just suppose that RedHat had acquired a game company that ran all their servers on NT. Then suppose a few RedHat employees gently suggested that they might like to run on Linux instead. Would we be in the least surprised? Would we think this was in some way evil? Of course not.
Why this whole story is in any way surprising to anyone is quite beyond me.
For example, a "rocket jump" involves firing a rocket into the ground and jumping simultaneously to give yourself an extra high jump as the blast sends you higher. You sustain damage in the process, but it enables you to reach otherwise inaccessible spots. It's not considered cheating, indeed, it's a skill that any seasoned Quake player will have.
These guys have taken it even further, and regularly pull off grenade jumps (same principle as rocket jumps but your timing needs to be better), and even the fiendishly tricky grenade and rocket jump combo.
No, but I did have to earn my right to, for example, drive a car. I had to take a test to prove that I knew how to operate one safely, and was aware of the basic rules and regulations of the jurisdiction in which I mostly drive.
I'm not saying that we should necessarily require a test before people are allowed broadband access, I'm just pointing out that it's not intrinsically as ridiculous as you are implying. The idea that for some human activities we choose to restrict those permitted to engage in them seems very reasonable. Even voting (your example) is restricted by age, because society considers that people need to be a certain age before they are able to make a reasonable and informed decision about who to vote for.
Now, I realise you were being flippant, but it is fairly obviously the case that there was a pretty large chunk of innovation that went into making Diablo different from its influences. It's certainly worlds away from copying Pac-Man verbatim and merely changing the name of the protagonist.
How exactly are computer games any different from other IP? No-one ever seems to think that it's unfair for song-writers to get royalities when other bands do cover versions of their work. It's an analogous situation here.
"For instance, the acting verb in a sentence almost always occurs at the end of a sentence."
Sorry, I can't resist being nitpicky, but that's not unique to Japanese: Latin typically puts the verb at the end of the sentence too.
As far as your claim that he had a "massive ego" goes, nothing could be further from the truth. Guinness had a reputation for being extremely modest and shy of publicity.
We can only dream...
Actually, even when you have 16-bit source textures, the end result can look a lot better when rendered to a 32-bit framebuffer. The reason is that you don't just do a plain copy of the texture, but you typically modulate it with an interpolated lighting value and possibly apply multipass effects as well. With a 16-bit framebuffer its quite easy to start to see banding and/or dithering artefacts that wouldn't be there with a 32-bit framebuffer, regardless of the bitdepth of the source artwork.
If you were an IT manager, making a media server purchasing decision, would you rather go with something that has a decent reliability record and tools support, or an untried offering from Sony which will have limited initial support?
Furthermore, unless I missed it, there was no mention of what operating system it's going to run. The only vaguely plausible option would seem to be Linux, but you do have to ask yourself whether you wouldn't want the choice to run Solaris or *BSD instead for critical heavy-duty applications like media serving. And in the event that they turn out to be crazy enough to try shipping their own homespun OS, then any IT manager worth their salt should run a mile (heck, I'd rather use NT/Win2K than any sort of Sony-built freakshow).
I have just one word to say: "Quake"
With a spec that anyone could manufacture, how would manufacturers make money? They either have to sell the consoles at a profit (which is going to be hard to do, given that they'd have to be priced about twice the cost of competing 'closed' consoles), or there needs to be some scheme for them to get a kickback on title and peripheral sales.
Crap it may be, but then MP3 is crap in the first place (at least at any reasonable bitrate). It turns the top end into nasty, tinny slush, and bass becomes incredibly flabby. Anyone with an ear for fidelity wouldn't be using MP3 to start with.
Of course, it turns out that lots of people don't have an ear for fidelity (perhaps, ironically, as a result of going to too many rock concerts?), hence all the oft repeated claims that "it's as good as CD", and "I can't tell the difference". For the lucky people who can't tell the difference, just run a cable from your PC and have done with it. The rest of us will stick with CDs.
They test on a dual processor system, but then test versus Windows 95 which, as we all know, doesn't support multiple processors. A fairer test would be to either test on a single CPU machine, or versus NT/Windows 2000.
Er, only sort of. The standard library has a string type, but that's not entirely the same thing.
Bungie have also been quoted as saying that they will remain autonomous within MS, and may continue to develop titles for non-MS platform (e.g. Mac), although it remains to see how long that lasts. I suspect that Mac titles may be allowed to continue for a little while, but PlayStation 2 titles will be knocked right on the head in favour of X-Box.
If they are non-commercial, why are they in the .com TLD and not .org?
Regardless of whether or not MS is a monopoly in the PC OS business, when it comes to the console business they most certainly are not. IANAL, but I thought that you had to consider the relevant market when applying anti-trust considerations. In this case, MS would be doing nothing any different to Sony, Sega, and Nintendo if they controlled access to the X-Box. Indeed, the whole console business model rests upon this fact - if you couldn't control content you'd have to sell the consoles at a profit, which would suddenly mean $900 PS2s.
They're going to recoup the cost by charging a licence fee on games and peripherals. Exactly the same way that Sony, Sega, and Nintendo do.
Of course, when Sony brought out the Playstation they had no experience of the console market either. And everyone said at the time how they would be crushed by Sega and Nintendo.
1: Design a physics engine that really only works properly for rigid 1 meter cubes.
2: Populate the game with (non-cuboid) 10 ton lizards.
3: Substitute player's arm with a 10 foot rubber probiscus with a crazy control mechanism which the player is expected to use to manipulate all objects in the game, including guns which need to be fired accurately.
4: Add an out-of-date graphics engine and turn the whole lot loose on some designers who can't come up with a puzzle more sophisiticated than "stack box A on top of box B".
Et voila! Worst game ever made! What's that you say? Someone's thought of it already?
1: For performance/smoothness reasons you may well want to leave some decisions in the hands of the client and/or supply more information to the client that is strictly needed to be conveyed to the player at that instant. While decision making can ultimately be validated by the server, once you've sent information out to the client you can't call it back.
2: It's easy to construct examples where you have given the client the minimum amount of information to function correctly, yet cheating is still possible. For example, suppose you have a cloaking device in the game which renders the player almost completely transparent. The client needs to know where that player is in order to draw it. However, a hacked client could draw that player in bright, easy-to-see colors, rather than the almost-invisible transparency that was the intent. Similar sorts of cheating are evident in Quake, where cheaters replace player models with large, brightly lit boxes that show up in dark corners and poke through obstructions.
Now I guess you could argue that you just should modify your game design to avoid these kinds of situation, but (a) you'll always miss something, and (b) why the hell should I mess with my game design just to serve a source licencing ideology?
I assert that the best solution to these problems is currently security through obscurity. It's not a great solution, but it's the best one I've heard. If you can prove me wrong, I'd love for you to do it because I'm working on a multiplayer game and this is giving me a real headache.
Part of many games is deliberately obscuring information from the player that is actually 'known' by the software. For example, the software knows the positions of enemies and states of game objects, that the player might not be supposed to be allowed to see at that point in the game. If the game were open source it would be a simple matter to modify the game to report that information - ruining the game. For a single-player game this cheating isn't really a problem - if the player has fun doing it, then who is to say it's a bad thing? For a multi-player game though, cheating can be fatal.
There doesn't seem to be any way of preventing this cheating, other than to make it really hard for a cheater to modify their game and/or data in an undetectable way, and one of the most effective ways to make it hard is not to give the source out. Of course, this doesn't prevent cheating in principle, but in practice it hopefully slows down people enough that the game will be stale before anyone does serious damage.
The key point is that games are different to other apps. In any other app if you have data then it's reasonable to expect the user to be allowed to view it in any way they wish, and if the app is modified to allow that data to be presented 'better' then this is a good thing. With games this is simply not true.