Windows is the absolute best OS out there for gaming at this point
That's a true statement - but it has to be carefully qualified:
It's NOT the best because it has technical features that make it the best.
It IS the best because you can buy more games for it than for any other (non-console) OS.
That distinction is critical because:
...if it's the best (for the second reason), it follows that more gamers will use Windoze than Linux, BSD, MacOS or the others that are out there.
...if pretty much all gamers run Windoze, games manufacturers will target that OS - to the exclusion of all others.
Hence, 'Windoze is best for games' is simply a vicious circle. It could be broken quite easily if just a handful of major games companies would take the step of releasing Linux clients at the same time and price as the Windoze version.
At that point, Linux's natural TECHNICAL and PRICE superiority would kick in and the market could flip-flop to the point where no games would be released for Windoze and all gamers would run Linux.
If you buy into the "Schroedinger's Cat" thought experiment, then you pretty much have to accept that there are multiverses.
In Schroedinger's experiment, there is a cat in a box that is either killed or not killed depending on the result of some test of quantum superposition. The cat remains in this hybrid state of being dead or not dead until someone opens the box to 'observe' the event. I think most scientists believe this - and it's being demonstrated right now in things like Quantum Computers.
That's all well and good - but suppose it's a human in the box? They are simultaneously both alive and dead - for that person, there are two parallel universes - one in which they are alive and the other in which they are dead.
Now, you can extend the cat experiment outwards. Inside the room where the cat experiment is being conducted is the Mad Scientist and his assistant (who happens to be a Cat lover). If the cat in the box dies, the assistant kills the Mad Scientist in a fit of rage...if the cat is OK, nothing happens.
So, for someone standing outside the lab, we have a second Schroedinger experiment. Only when the lab door is opened will we know the state of the Mad Scientist - is he dead or alive? Since that state directly depends on the quantum superposition, he's in the same hybrid state as the cat was.
By extension, the entire planet earth is superposed between the/. headline "Scientist murdered following insane Cat experiment!" and another boring day of RIAA news.
There are potentially VAST numbers of these 'superposition' experiments going on throughout the earth - none of which are resolved until some alien race happens to 'observe' us. Then, the aliens in turn are in a super imposed state that is not resolved until something also observes their reaction to that resolution.
So, there must be an almost infinite number of superimposed universes all waiting for some ultimate resolution. But life goes on - and we don't *feel* any different for being in a superimposed state.
It follows then - directly from Schroedinger - that there are an infinite number of universes with all possible outcomes of all possible events happening in them.
Now here is the biggie: If you also believe in the 'strong anthropic principle' - then I exist in *this* universe because it just happens to be one of the universes where I didn't die in a car crash last week. If that's true then I must live forever. There will always be an infinite number of universes in which I miraculously avoid every possible accident - where immortality is somehow invented before I get too old.
By definition, that's always going to be the universe I'm living in right now because I have to be here in order to observe it.
Freedom of speech is important - but so is the freedom not to have to listen.
My time on this planet is finite - and is arguably the most valuable thing I have. Choosing not to waste that time by being forced to listen to people whom I don't want to hear is a freedom I cherish.
Spammers and Cold callers are entitled to speak - that is their freedom - but I should have the right to say that I don't want to listen to them anymore.
I think the TV broadcasters had a perfect right to say "I choose not to repeat what those people said at the Oscars". I would have preferred that they would have broadcasted it uncut - but that's not the same thing. (As it happens, that was the ONLY part of the Oscars I saw - because the act of cutting it was so newsworthy! So that actually worked out rather well.)
If someone else had come along and TOLD the broadcasters not to transmit that part of the Oscars - then that would be an entirely different matter. Forced censorship of the media is "A Very Bad Thing" - but if they chose not to do it because they felt it was inappropriate - then that's their call.
Just think about it for a moment. If broadcasters were required to broadcast all events in their entirety without editing out any of anyone's words, things would get out of hand rather quickly. Phone-in radio shows would have to broadcast crazy lunatics without editing - they could be stuck with broadcasting hours of ranting about nothing because they wouldn't have the right to edit it for relevent content.
What happened at the Oscars SHOULD have been broadcast - in context - un-cut - but I defend the TV stations right to cut it if they so desired.
I'm not *so* concerned about MY government watching me. What bothers me (and bothers me a LOT) is who else can get in and find out things about me that I'd rather they didn't know.
* Big business (I don't think Microsoft should be able to find out what software I run on my PC for example).
* Other people (Identity theft is a HUGE problem).
* Other governments (I don't think we'd want Iraqi government officials finding out too much about our citizens).
* Small businesses (I don't want to be Spammed, Cold-called or Junk Faxed anymore - and I CERTAINLY don't want those people to be able to find out a lot about me and thus target me more precisely).
Now, if the price of being private from all those people is to also be more private than is convenient for my government - then I'm sorry that has to be the case. Dunno about you but I'm much more worried (in a cold, hard statistical sense) about having my life wrecked by identity theft than by a terrorist.
There may be 'engine' technology produced along the way - and that may be re-used for the next game and the one after that. There is definitely a desire to push as much of the work out into OpenSource as is reasonable - but remember that this IS a commercial activity and money has to be made at some point. It's far to early in the dev cycle to know what spin-off technologies may make it out to OpenSource.
I was annoyed by someone at work today telling me that whilst the British are helping the US, the Americans have made the greater sacrifice.
I wanted to set the record straight.
100,000 US ground troops fighting - from a total of 250,000 people out there. 40,000 British ground troops fighting (I don't know the total number of Brit's out there).
So, when you look at the guys who *might* get killed - that's 0.04% of the US population and 0.06% of the British population. By that measure (and of course there are ways to look at this), the British commitment is 50% larger than the US.
Against a background where 80% of Brits are against the war and only 40% of Americans are - I sure hope the US appreciates that!
Adding doors and windows after the concrete has set isn't easy.
Our builder forgot to install the outlet for the cooker hood ventilator through the kitchen wall - he had to use water lubricated diamond cutters to make the hole for it.
So, it's possible - but far from easy.
Modularity isn't too bad. The Greenblock system is pretty simple, you cut the bricks to suit where you need holes. Their limitations are things like only having 90 and 45 degree angles. You can't build a hexagonal structure and things like bay windows have to be 45 degree angles. Other manufacturers have a wider range of brick types that makes them more flexible. We have have three octagonal rooms at three corners of a square building with the garage at the fourth corner. That complexity wasn't really a problem for the system.
My house (which I designed myself) was the first in the Southern USA to be built with a technology first brought out (I believe) by a company called 'GreenBlock'. (http://www.greenblock.com)
This stuff looks like giant pale blue foam polystyrene Lego bricks.
In fact, each basic 'brick' is a 2' long by 9" by 6" block made out of two foam plates (each about an inch thick) tied together with carbon fibre 'webbing'. There are studs on the top and receptacles that they plug into underneath - so the analogy with Lego is not entirely without merit!
You lay conventional foundations (ours is a 'waffle' slab design) and then build the outside walls - quite literally like building a giant Lego house.
As you lay the bricks, cutting holes for the windows and doors with a hand-saw. You thread steel reinforcing bars down inside them - and when you are done, you hire a concrete pump to dump very runny concrete down inside the bricks.
(It's a little more complex than that - the ReBar in the walls is tied into the slab - so they become an integral whole with the slab rather than just resting on it.)
In about two days, the outside walls are done (although we are told that it'll take years for the concrete to COMPLETELY harden). You end up with steel reinforced concrete walls with the original foam bricks forming an inch of foam insulation both outside and inside the walls. You can then sheetrock the inside of the house and either brick, stucco or conventionally clad the outside of the house. This is essentially only for decorative purposes...you really don't want a giant pale blue foam polystyrene house!
Although the house is immensely strong, the primary reason for doing this is energy efficiency. Hence the interior of the house is then built conventionally...although you could do it with the same approach I suppose.
Your walls come out about a foot thick and have an 'R' value of about 50. Even in the height of Texas summers, our electricity bill for a 2500 sq.ft house is between 80 and 100 dollars. Most people I know get $300 or more electricity bills for equivelent sized homes.
We believe that this house will still be standing in 100 years - it's claimed that it'll be tornado-proof - although clearly this doesn't stop the windows from blowing out and the roof from being ripped off in the event of a direct hit.
Since our builder got into building this way, he's subsequently built dozens and dozens of houses in our area in the exact same way - as far as we know, all the owners are happy with them.
However, the bigger issue is how you'd demolish such a house when it's not wanted anymore!
In the 1970's, I could write a great game by myself in a day. You can code "Hunt the Wumpus" or Pong that quickly.
Later, it took a couple of weeks to write a good game, asteroids, pac man, space invaders, Colossal Cave.
Still later, it could still be done by one determined person in a year. You could write Doom or perhaps Sim City that quickly.
But then you started to need a lot of game depth - and depth doesn't come cheap.
I could write the software for Mario'64, or other early 3D games in a year - but the artwork would need several full time artists. I wrote "Tux - A Quest for Herring" (an OpenSourced game at about the level of Mario'64) in about six months. With another six, I could have it do everything that Mario'64 does - but without level designers and artists, I could not do it.
But now we are seeing games with immense, detailed 3D worlds with hundreds of 3D characters, vehicles, etc. There is no way you can do that with less than 50 people over several years. I'm too discouraged to even try.
With every new game, the ante is raised - and raised so high that only someone with a very large bankroll can hope to write a 'modern' computer game. Once you start needing a million dollars to develop it, people tend to become very focussed on getting it right. You can't take as big a risk with a megabuck as you can with a rainy weekend when you can't think of anything better to do.
Hence, the number of games being written shrinks and with no tolerance for risk, there can be no room for innovation.
We see this in movies and TV too - every movie is either a sequel, a remake or something based on an already very successful TV show or book. What movies made money over the last few years? Star Trek, Star Wars, James Bond, Harry Potter, SpiderMan, LOTR - every one a sequel or a spinoff. Almost every TV show is a small variation on something that already exists - and if a new idea ever does show up, it's cloned mercilessly until the airwaves are swamped with them and consumers get very pissed at the lack of variety. (Think about the recent waves of Jerry Springer clones, then Court TV shows, then Game show clones and now reality TV.)
That's because the cost of making a movie is becoming HUGE - you can't afford to take any risks at all. It's not that the industry lacks ideas or motivation.
Games are going the exact same way for the exact same reason. Government funding won't change that...it's just swapping one risk-averse source of venture capital for another even more risk-averse source.
I think it takes a change of heart from the gamers. They'll have to get bored with games that need a 3D model of an entire city - or a game that has 30 uniquely modelled levels. They'll have to be so desperate for something new that when something like that does arrive, it can be something small but addictive - downloaded from the Internet probably.
Since all but the very largest games companies will have gone bust by that time, such a thing can only come from a small independent games house with people working for free in the hope of making money only when their game sells.
However, independent games writers are becoming frustrated and discouraged by their inability to come up with games as impressive as the big companies - and that's a shame. However, it may only take one or two off-the-wall cult successes to spur people into greater efforts.
Will we see a "Blair Witch Project" of the gaming business? Only time will tell.
In the middle ages, the world would have seemed to be utterly unchanged - for the previous few centuries at least. In that situation, why would you ever expect change? Predicting a very different future back then would have been just silly.
We have seen such spectacular growth in just about every part of life in perhaps two lifetimes - we now see life in terms of change. Shall I buy an ATI Radion 9700 graphics card - or should I wait a few months and get an nVidia GeForceFX? (Oh - wait...bad example!)
I expect change - I *rely* on change. Predicting the future is now a survival trait and humans are nothing if not adaptable when it comes to surviving.
We have codified change into things like Moores Law. We are suprised and perhaps even a little fearful when things don't change fast enough (see dozens of/. articles about the immenent failure of Moores Law for example).
Actually, I think what's most interesting about this exhibit is just how LITTLE change he predicted. Cars still have enormous chrome fins - people still dress exactly the same as they did in the 30's, 40's and 50's - everyone still commutes to work. For us, looking at these, we see a weird mix of antique design with machines and buildings that we still havn't managed to engineer.
Placing the camera, arranging the scene and moving stuff all boils down to sending a 4x4 matrix to the hardware. One API entry...perhaps with some fancy syntactic sugar letting you set up the matrix using euler angles or something. Maybe a stack-based push/pop mechanism.
Lighting is now subsumed into the vertex and pixel shaders.
Games don't use OpenGL lighting much anyway - they mostly use the 'lightmap' technology pioneered in the early versions of Quake.
In the last year, we have seen all the major 3D card vendors heading towards full programmability of their hardware. Also, the old mechanisms for feeding vertex data to the hardware has been changing to a scheme where instead of sending a 3D coordinate, a texture coordinate, a colour and a normal, you send a 3D coordinate and some arbitary data. Also, the mechanism for sending that data now boils down to "Please DMA this large data block via AGP".
Since the programming languages for the 'vertex' and 'pixel' pipelines is essentially identical between D3D and OpenGL - and programs can be compiled offline using tools like Cg (C-for-graphics) - these programs can work identically between D3D and OpenGL.
Hence, all the complicated API for setting up textureing, fogging, shading, lighting, etc that consumes most of the API spec for D3D and OpenGL has boiled down to "load these two programs"...and all the complicated glVertex/glColor/glNormal stuff has boiled down to "send this big block of data to the card".
Also, both API's have complex support for all sorts of pixel depths and packing mechanisms - when we all know that 8 bits R/G/B/A is everywhere in the future.
Now there is very little left of either API except for legacy stuff (that programmers can happily ignore) and initial setup.
The new ground for standards is the 'shader languages' we'll be using to write the vertex and pixel programs. There is a standardization battle going on between three or four of these languages (all of which look a lot like C and bear homage to the venerable RenderMan interface). Cg (nVidia), OpenGL 2.0 (ATI, 3Dlabs), are the main contenders. Cg works identically for OpenGL and D3D - but isn't well supported for non-nVidia hardware. OpenGL 2.0 hasn't been finalized yet and obviously isn't intended to work with D3D - however it may well be possible to use the OGL2 programming languages to compile programs you could load at the machine code level into D3D.
It irony here is that all programmers really NEED here is a much simpler API than either D3D or OpenGL. A way to clear the screen, a way to download texture maps, a bulk vertex data transfer mechanism and a way to download shader programs.
Writing thin wrappers for those basic functions would allow programs to run identically on D3D and OpenGL - and would make it quite possible to drop OpenGL support for everything except legacy applications in favor of a MUCH simpler API.
What needs to be standardized is the shader programming language - but that has no dependancies on the Operating System, graphics API or graphics hardware (presuming you can write back-ends for any of the mainstay 3D cards).
Since Microsoft never implemented a new version of OpenGL after their initial 1.1 implementation (the rest of the world is up to 1.4 and heading for the big 2.0 sometime towards the second half of '03) - they clearly weren't getting much out of their presence on the ARB.
Now they won't get to veto things and slow everyone else down either.
I wonder how long it'll be until they stop shipping OpenGL with Windoze? I guess so long as Quake/Doom need it they are somewhat bound to keep shipping it.
'man has just exactly what you need in exactly the right order.
First, the bare minimum - the name of the program or function an a one sentence description of what it does.
Secondly it's usage with a well thought out meta-language - that is generally enough to nudge your memory if you already know the command. For functions, it also tells you what odd-ball header files you might need.
Thirdly, a *slightly* more detailed description - and a concise list of the options/parameters - not spread out over many pages...right there.
Fourthly...more stuff...that you may or may not care about.
The information cleverly gets more and more detailed - so you generally get 99% of what you need right there in the first screenful.
If I want more info than a two screenful man page can delivers, I want it on the web in a browser in HTML. I don't want to have to learn another markup language - or another navigation scheme - and I want a choice of a dozen convenient browsers.
info does neither of these things - it sucks and needs to *DIE*.
As the boss of Silicon Graphics once famously said: "Linux is the Future of UNIX". UNIX isn't dead - it's just had a major rewrite/cleanup. That's hardly suprising for a 30 year old software package.
The code has changed completely - but the core ideas are exactly as they were back in 1976 when I used UNIX on a PDP-11.
There are more people using UNIX-like OS's now than there have ever been.
> Early science fiction movies had people being shot to the moon with guns,...and the idea of using a large, smooth-bored cannon to shoot satellites into orbit is still a pretty trendy idea. Remember Saddam's space cannon, captured during the last round of the gulf war?
Probably a few too many g's to send people to the moon though.
Which prediction?
on
Ask Larry Niven
·
· Score: 2, Interesting
Which of all the predictions of the future in your books do you most wish would come true?
Slashdot Mirror
That's a true statement - but it has to be carefully qualified:
That distinction is critical because:
Hence, 'Windoze is best for games' is simply a vicious circle. It could be broken quite easily if just a handful of major games companies would take the step of releasing Linux clients at the same time and price as the Windoze version.
At that point, Linux's natural TECHNICAL and PRICE superiority would kick in and the market could flip-flop to the point where no games would be released for Windoze and all gamers would run Linux.
No - it's not going to happen - but we can dream.
If you buy into the "Schroedinger's Cat" thought experiment, then you pretty much have to accept that there are multiverses.
/. headline "Scientist murdered following insane Cat experiment!" and another boring day of RIAA news.
In Schroedinger's experiment, there is a cat in a box that is either killed or not killed depending on the result of some test of quantum superposition. The cat remains in this hybrid state of being dead or not dead until someone opens the box to 'observe' the event. I think most scientists believe this - and it's being demonstrated right now in things like Quantum Computers.
That's all well and good - but suppose it's a human in the box? They are simultaneously both alive and dead - for that person, there are two parallel universes - one in which they are alive and the other in which they are dead.
Now, you can extend the cat experiment outwards. Inside the room where the cat experiment is being conducted is the Mad Scientist and his assistant (who happens to be a Cat lover). If the cat in the box dies, the assistant kills the Mad Scientist in a fit of rage...if the cat is OK, nothing happens.
So, for someone standing outside the lab, we have a second Schroedinger experiment. Only when the lab door is opened will we know the state of the Mad Scientist - is he dead or alive? Since that state directly depends on the quantum superposition, he's in the same hybrid state as the cat was.
By extension, the entire planet earth is superposed between the
There are potentially VAST numbers of these 'superposition' experiments going on throughout the earth - none of which are resolved until some alien race happens to 'observe' us. Then, the aliens in turn are in a super imposed state that is not resolved until something also observes their reaction to that resolution.
So, there must be an almost infinite number of superimposed universes all waiting for some ultimate resolution. But life goes on - and we don't *feel* any different for being in a superimposed state.
It follows then - directly from Schroedinger - that there are an infinite number of universes with all possible outcomes of all possible events happening in them.
Now here is the biggie: If you also believe in the 'strong anthropic principle' - then I exist in *this* universe because it just happens to be one of the universes where I didn't die in a car crash last week. If that's true then I must live forever. There will always be an infinite number of universes in which I miraculously avoid every possible accident - where immortality is somehow invented before I get too old.
By definition, that's always going to be the universe I'm living in right now because I have to be here in order to observe it.
OK - now I have a headache.
Freedom of speech is important - but so is the freedom not to have to listen.
My time on this planet is finite - and is arguably the most valuable thing I have. Choosing not to waste that time by being forced to listen to people whom I don't want to hear is a freedom I cherish.
Spammers and Cold callers are entitled to speak - that is their freedom - but I should have the right to say that I don't want to listen to them anymore.
I think the TV broadcasters had a perfect right to say "I choose not to repeat what those people said at the Oscars". I would have preferred that they would have broadcasted it uncut - but that's not the same thing. (As it happens, that was the ONLY part of the Oscars I saw - because the act of cutting it was so newsworthy! So that actually worked out rather well.)
If someone else had come along and TOLD the broadcasters not to transmit that part of the Oscars - then that would be an entirely different matter. Forced censorship of the media is "A Very Bad Thing" - but if they chose not to do it because they felt it was inappropriate - then that's their call.
Just think about it for a moment. If broadcasters were required to broadcast all events in their entirety without editing out any of anyone's words, things would get out of hand rather quickly. Phone-in radio shows would have to broadcast crazy lunatics without editing - they could be stuck with broadcasting hours of ranting about nothing because they wouldn't have the right to edit it for relevent content.
What happened at the Oscars SHOULD have been broadcast - in context - un-cut - but I defend the TV stations right to cut it if they so desired.
I'm not *so* concerned about MY government watching me. What bothers me (and bothers me a LOT) is who else can get in and find out things about me that I'd rather they didn't know.
* Big business (I don't think Microsoft should be able to find out what software I run on my PC for example).
* Other people (Identity theft is a HUGE problem).
* Other governments (I don't think we'd want Iraqi government officials finding out too much about our citizens).
* Small businesses (I don't want to be Spammed, Cold-called or Junk Faxed anymore - and I CERTAINLY don't want those people to be able to find out a lot about me and thus target me more precisely).
Now, if the price of being private from all those people is to also be more private than is convenient for my government - then I'm sorry that has to be the case. Dunno about you but I'm much more worried (in a cold, hard statistical sense) about having my life wrecked by identity theft than by a terrorist.
The end product is definitely a game.
There may be 'engine' technology produced along the way - and that may be re-used for the next game and the one after that. There is definitely a desire to push as much of the work out into OpenSource as is reasonable - but remember that this IS a commercial activity and money has to be made at some point. It's far to early in the dev cycle to know what spin-off technologies may make it out to OpenSource.
Blair is a member of the Labor party - that makes him a Democrat...and then maybe a bit further in that direction.
More secure than a safe - if you leave the door open.
I was annoyed by someone at work today telling me that whilst the British
are helping the US, the Americans have made the greater sacrifice.
I wanted to set the record straight.
100,000 US ground troops fighting - from a total of 250,000 people out there. 40,000 British ground troops fighting (I don't know the total number of Brit's out there).
So, when you look at the guys who *might* get killed - that's 0.04% of the US population and 0.06% of the British population. By that measure (and of course there are ways to look at this), the British commitment is 50% larger than the US.
Against a background where 80% of Brits are against the war and only 40% of Americans are - I sure hope the US appreciates that!
The coders have made this abundantly clear - this will get fixed.
The team were only informed that they'd been selected sometime late last night!
It's a bit early to expect anything other than "getting to know each other" chat via email.
Adding doors and windows after the concrete has set isn't easy.
Our builder forgot to install the outlet for the cooker hood
ventilator through the kitchen wall - he had to use water lubricated
diamond cutters to make the hole for it.
So, it's possible - but far from easy.
Modularity isn't too bad. The Greenblock system is pretty simple,
you cut the bricks to suit where you need holes. Their limitations
are things like only having 90 and 45 degree angles. You can't build
a hexagonal structure and things like bay windows have to be 45 degree
angles. Other manufacturers have a wider range of brick types that
makes them more flexible. We have have three octagonal rooms at
three corners of a square building with the garage at the fourth
corner. That complexity wasn't really a problem for the system.
My house (which I designed myself) was the first in the Southern USA to be built with a technology first brought out (I believe) by a company called 'GreenBlock'. (http://www.greenblock.com)
This stuff looks like giant pale blue foam polystyrene Lego bricks.
In fact, each basic 'brick' is a 2' long by 9" by 6" block made out of two foam plates (each about an inch thick) tied together with carbon fibre 'webbing'. There are studs on the top and receptacles that they plug into underneath - so the analogy with Lego is not entirely without merit!
You lay conventional foundations (ours is a 'waffle' slab design) and then build the outside walls - quite literally like building a giant Lego house.
As you lay the bricks, cutting holes for the windows and doors with a hand-saw. You thread steel reinforcing bars down inside them - and when you are done, you hire a concrete pump to dump very runny concrete down inside the bricks.
(It's a little more complex than that - the ReBar in the walls is tied into
the slab - so they become an integral whole with the slab rather than just resting on it.)
In about two days, the outside walls are done (although we are told that it'll take years for the concrete to COMPLETELY harden). You end up with steel reinforced concrete walls with the original foam bricks forming an inch of foam insulation both outside and inside the walls. You can then sheetrock the inside of the house and either brick, stucco or conventionally clad the outside of the house. This is essentially only for decorative purposes...you really don't want a giant pale blue foam polystyrene house!
Although the house is immensely strong, the primary reason for doing this is energy efficiency. Hence the interior of the house is then built conventionally...although you could do it with the same approach I suppose.
Your walls come out about a foot thick and have an 'R' value of about 50. Even in the height of Texas summers, our electricity bill for a 2500 sq.ft house is between 80 and 100 dollars. Most people I know get $300 or more electricity bills for equivelent sized homes.
We believe that this house will still be standing in 100 years - it's claimed that it'll be tornado-proof - although clearly this doesn't stop the windows from blowing out and the roof from being ripped off in the event of a direct hit.
Since our builder got into building this way, he's subsequently built dozens
and dozens of houses in our area in the exact same way - as far as we know, all the owners are happy with them.
However, the bigger issue is how you'd demolish such a house when it's not wanted anymore!
100mbps ... 100 millibits per second? Wow! That's 1/10th baud. You'd better
type R-e-a-l-l-y s-l-o-w-ly!
Maybe they should consider shooting for 100Mbps?
In the 1970's, I could write a great game by myself in a day. You can code "Hunt the Wumpus" or Pong that quickly.
Later, it took a couple of weeks to write a good game, asteroids, pac man, space invaders, Colossal Cave.
Still later, it could still be done by one determined person in a year. You could write Doom or perhaps Sim City that quickly.
But then you started to need a lot of game depth - and depth doesn't come cheap.
I could write the software for Mario'64, or other early 3D games in a year - but the artwork would need several full time artists. I wrote "Tux - A Quest for Herring" (an OpenSourced game at about the level of Mario'64) in about six months. With another six, I could have it do everything that Mario'64 does - but without level designers and artists, I could not do it.
But now we are seeing games with immense, detailed 3D worlds with hundreds of 3D characters, vehicles, etc. There is no way you can do that with less than 50 people over several years. I'm too discouraged to even try.
With every new game, the ante is raised - and raised so high that only someone with a very large bankroll can hope to write a 'modern' computer game. Once you start needing a million dollars to develop it, people tend to become very focussed on getting it right. You can't take as big a risk with a megabuck as you can with a rainy weekend when you can't think of anything better to do.
Hence, the number of games being written shrinks and with no tolerance for risk, there can be no room for innovation.
We see this in movies and TV too - every movie is either a sequel, a remake or something based on an already very successful TV show or book. What movies made money over the last few years? Star Trek, Star Wars, James Bond, Harry Potter, SpiderMan, LOTR - every one a sequel or a spinoff. Almost every TV show is a small variation on something that already exists - and if a new idea ever does show up, it's cloned mercilessly until the airwaves are swamped with them and consumers get very pissed at the lack of variety. (Think about the recent waves of Jerry Springer clones, then Court TV shows, then Game show clones and now reality TV.)
That's because the cost of making a movie is becoming HUGE - you can't afford to take any risks at all. It's not that the industry lacks ideas or motivation.
Games are going the exact same way for the exact same reason. Government funding won't change that...it's just swapping one risk-averse source of venture capital for another even more risk-averse source.
I think it takes a change of heart from the gamers. They'll have to get bored with games that need a 3D model of an entire city - or a game that has 30 uniquely modelled levels. They'll have to be so desperate for something new that when something like that does arrive, it can be something small but addictive - downloaded from the Internet probably.
Since all but the very largest games companies will have gone bust by that time, such a thing can only come from a small independent games house with people working for free in the hope of making money only when their game sells.
However, independent games writers are becoming frustrated and discouraged by their inability to come up with games as impressive as the big companies - and that's a shame. However, it may only take one or two off-the-wall cult successes to spur people into greater efforts.
Will we see a "Blair Witch Project" of the gaming business? Only time will tell.
> Butt its amazing how many people they're our who don't no how to use the write homophone. :)
:)"
Tsk, tsk. That should of course have been:
"Butt its' a mazeing how many people they're our who
don't no how too ewes thee write homophone.
English would be a better language if there was a homophone for homophone.
In the middle ages, the world would have seemed to be utterly unchanged - for the previous few centuries at least. In that situation, why would you ever expect change? Predicting a very different future back then would have been just silly.
/. articles about the immenent failure of Moores Law for example).
We have seen such spectacular growth in just about every part of life in perhaps two lifetimes - we now see life in terms of change. Shall I buy an ATI Radion 9700 graphics card - or should I wait a few months and get an nVidia GeForceFX? (Oh - wait...bad example!)
I expect change - I *rely* on change. Predicting the future is now a survival trait and humans are nothing if not adaptable when it comes to surviving.
We have codified change into things like Moores Law. We are suprised and perhaps even a little fearful when things don't change fast enough (see dozens of
Actually, I think what's most interesting about this exhibit is just how LITTLE change he predicted. Cars still have enormous chrome fins - people still dress exactly the same as they did in the 30's, 40's and 50's - everyone still commutes to work. For us, looking at these, we see a weird mix of antique design with machines and buildings that we still havn't managed to engineer.
The name change was necessary - UNIX was always a registered trademark of Bell Labs. Linux had to be called something else.
"Linux is UNIX spelled sideways"
"Plan 9 from Bell Labs" sounds suspiciously like the worst SciFi movie ever made - a curious choice for a name!
Placing the camera, arranging the scene and moving stuff all boils down to sending a 4x4 matrix to the hardware. One API entry...perhaps with some fancy syntactic sugar letting you set up the matrix using euler angles or something. Maybe a stack-based push/pop mechanism.
Lighting is now subsumed into the vertex and pixel shaders.
Games don't use OpenGL lighting much anyway - they mostly use the 'lightmap' technology pioneered in the early versions of Quake.
In the last year, we have seen all the major 3D card vendors heading towards full programmability of their hardware. Also, the old mechanisms for feeding vertex data to the hardware has been changing to a scheme where instead of sending a 3D coordinate, a texture coordinate, a colour and a normal, you send a 3D coordinate and some arbitary data. Also, the mechanism for sending that data now boils down to "Please DMA this large data block via AGP".
Since the programming languages for the 'vertex' and 'pixel' pipelines is essentially identical between D3D and OpenGL - and programs can be compiled offline using tools like Cg (C-for-graphics) - these programs can work identically between D3D and OpenGL.
Hence, all the complicated API for setting up textureing, fogging, shading, lighting, etc that consumes most of the API spec for D3D and OpenGL has boiled down to "load these two programs"...and all the complicated glVertex/glColor/glNormal stuff has boiled down to "send this big block of data to the card".
Also, both API's have complex support for all sorts of pixel depths and packing mechanisms - when we all know that 8 bits R/G/B/A is everywhere in the future.
Now there is very little left of either API except for legacy stuff (that programmers can happily ignore) and initial setup.
The new ground for standards is the 'shader languages' we'll be using to write the vertex and pixel programs. There is a standardization battle going on between three or four of these languages (all of which look a lot like C and bear homage to the venerable RenderMan interface). Cg (nVidia), OpenGL 2.0 (ATI, 3Dlabs), are the main contenders. Cg works identically for OpenGL and D3D - but isn't well supported for non-nVidia hardware. OpenGL 2.0 hasn't been finalized yet and obviously isn't intended to work with D3D - however it may
well be possible to use the OGL2 programming languages to compile programs you could load at the machine code level into D3D.
It irony here is that all programmers really NEED here is a much simpler API than either D3D or OpenGL. A way to clear the screen, a way to download texture maps, a bulk vertex data transfer mechanism and a way to download shader programs.
Writing thin wrappers for those basic functions would allow programs to run identically on D3D and OpenGL - and would make it quite possible to drop OpenGL support for everything except legacy applications in favor of a MUCH simpler API.
What needs to be standardized is the shader programming language - but that has no dependancies on the Operating System, graphics API or graphics hardware (presuming you can write back-ends for any of the mainstay 3D cards).
Since Microsoft never implemented a new version of OpenGL after
their initial 1.1 implementation (the rest of the world is up to
1.4 and heading for the big 2.0 sometime towards the second half
of '03) - they clearly weren't getting much out of their presence
on the ARB.
Now they won't get to veto things and slow everyone else down
either.
I wonder how long it'll be until they stop shipping OpenGL
with Windoze? I guess so long as Quake/Doom need it they
are somewhat bound to keep shipping it.
Oh - for a creative way to say "Me Too".
'man has just exactly what you need in exactly the right order.
First, the bare minimum - the name of the program or function an a one sentence
description of what it does.
Secondly it's usage with a well thought out meta-language - that is
generally enough to nudge your memory if you already know the command.
For functions, it also tells you what odd-ball header files you might need.
Thirdly, a *slightly* more detailed description - and a concise list of
the options/parameters - not spread out over many pages...right there.
Fourthly...more stuff...that you may or may not care about.
The information cleverly gets more and more detailed - so you generally
get 99% of what you need right there in the first screenful.
If I want more info than a two screenful man page can delivers, I want
it on the web in a browser in HTML. I don't want to have to learn another
markup language - or another navigation scheme - and I want a choice of a
dozen convenient browsers.
info does neither of these things - it sucks and needs to *DIE*.
As the boss of Silicon Graphics once famously said: "Linux is the Future of UNIX". UNIX isn't dead - it's just had a major rewrite/cleanup. That's hardly suprising for a 30 year old software package.
The code has changed completely - but the core ideas are exactly as they were back in 1976 when I used UNIX on a PDP-11.
There are more people using UNIX-like OS's now than there have ever been.
> Early science fiction movies had people being shot to the moon with guns, ...and the idea of using a large, smooth-bored cannon to shoot satellites
into orbit is still a pretty trendy idea. Remember Saddam's space cannon,
captured during the last round of the gulf war?
Probably a few too many g's to send people to the moon though.
Which of all the predictions of the future in your books do you most wish would come true?
So nobody thinks to rename the files when they offer them for piracy?