HD or not really doesn't impact dev times/team size/budgets significantly. Gamer expectations, driven by the experience/art quality possible on a console with more resources, has a much larger impact.
If only there was a rating system, so that parents could easily gauge what was appropriate to their children. Of course, that would require stores only sell age appropriate titles to kids directly.
Damn that record indus.. oh, wait? We are talking about games? They already have those.
On a serious note, the military use of games people keep bringing up is about training people to work as a team, not overcoming hesitation to kill people. I would be more concerned with people training children to use guns (hunting, target practice, paintball) than games.
So, assuming a team of 30 and a dev cycle of 2 years, they pay 18,000,000 in salary? While average games have a total budget 10,000,000? Care to mention the company name?
Couldn't the customer alsoturn around and give it away for free if they choose? The developer doesn't have any way to prevent this, do they?
Doesn't this effectively prevent companies from developing and selling middleware libraries? The developer can't control who uses the library, so they can't make money by selling it (as anyone who buys it could just put it up on the net for public download).
If they do a good job, have a clean, simple API and no bugs, then no one needs to pay for support.
Just curious. I am still trying to understand the different buisness models that are reasonable with GPL'ed code. After reading the GPL license (family) several times, I am still trying to figure out how it all plays out.
Apologies - my post was meant to be sarcastic. It was also intended as a response to Minna Kirai post. I definitely blew it on all counts.
I work in the game industry as well; been here about 4 years. The idea that an MMP ought to be simple to optimize, and ought to be nice and modular is definitely a magic-magic land suggestion in my opinion. Games are written against hard deadlines, with requirements that change all the time.
I will leave the arguments about how well software engineering handles these situations to those with more free time than I.
Right! The linux code should be just as simple to follow as your average bash script.
After all, most code out there is properly engineered, with all of the planning done up front, with requirements defined at the beginning of the project and remaining the same through the end.
It isn't like games get tweaked or redesigned daily, as people on the team realize what is fun and what isn't.;)
You want to make a city? Make a game level in your editor of choice like UnrealEd or QRadiant
Want to make a model? Pop open milkshape or max for games.
The Sims, Quake, Half-Life, Unreal. All of these games have huge communities and toolsets for creating new worlds.
Kids still are building things -- the blocks and tools just became more advanced. Instead of physical objects, they get simulated ones. It also happens to be EXTREMELY cheap by comparision if you have a PC.
The creativity isn't gone, it just relocated a bit.
Abolutely have to agree. I have a few extras sitting around just in case I lose one and suddenly have a good idea. After using these, I can't imagine using a bic. I love never having to scribble to get the ink flow started.
Re:It's not surpsing Carmack wants to get it right
on
No Doom 3 This Year?
·
· Score: 2, Insightful
Its called a feature set.
Doom isn't trying to be Battle Field -- so why spend resources trying to add a feature (like vehicles) that your game doesn't call for?
If the concern is licensees, anyone with the money to license the engine surely can afford programmers who can add vehicles. Hell, Gunman Chronicles, which starts as a mod and never even had engine source access added a vehicle, even if it was a bit simple.
Doom isn't targeting massive outdoor areas, so why does it need to support 64 clients? Scalablity costs time and money. If they aren't going to use it, then there is no reason at all to work on it. If there was a huge demand for 64 player games, either ID would do it if it fit in to their feature set and time line, or a licensee could add it.
If you are a licensee and you want to add an occlusion system, well, go for it. Same with LOD (though LOD can frequently be a slowdown if the CPU is doing it dynamicly).
Carmack and Id know their shit. I am not a fan boy (hell, I am part of the competition:) ) but all of shortcomings you mentioned have specific reasons and could changed by any decent coder.
Pardon my ignorance here, but if dumping is definied as:
'Export price that is "unfairly low," defined as either below the home market price (normal value) (hence price discrimination) or below cost.'
Then how is freeware and linux not considered dumping as it is 'below home market price'? To argue that point, it seems like one would have to argue that Linux and OSS had no production costs.
How could, say, Microsoft legally compete price-wise, as they couldn't lower their prices below the cost threshhold?
A vector of bools isn't a vector... of bools. Its an specialized object which is actually a proxy wrapping objects which it accesses individual bits in.
Because of this, it is not an stl compliant container; it doesn't support references it what it stores because you can't have references to bit fields.
As it is non compliant, a better substitute would be a deqeue of bools or a bitset.
For a much better description, see Effective STL by Meyers.
1 - So what? Isn't that true of pretty much every product? You can't make a game that appeals to everyone; you pick a niche and try to own it.
2 - What does story line have to do with diablo 2/doom 3? While diablo2 was very enjoyable (and I am sure Doom3 will be as well) that doesn't suggest that mixing the gameplay would be a benefit.
As you mentioned in point 1, someone who likes click fests might not like shooters. If you put both in the same game, you are making something that doesn't appeal to either audience.
Note that I am not saying such a game couldn't be successful -- instead I am suggesting that by missing genres, you face dangers as well. Your product loses focus; look at C&C/Renegade.
3 - There are coop games (NOLF2 being the first one that comes to mind). Unfortunately, not many people experience the fun of taking on a shambler with their mates. If coop sold enough titles to be profitable as a primary feature, there would be more such titles.
4 - Those standards are called GL and DirectX.:) Unfortunately, there are driver bugs, minor differences, etc. Always will be.
Specialized areas? All the math he can learn. Other than that, it depends entirely on what area he is interested in programming. AI? Sound? Graphics? Networking? Physics? Those are all specialized jobs.
If he can find one that he likes to focus on, all the better. Being a specialist first, a generalist second -- its the only way to stand out from all the other generalists.:) Oh, and learning to communicate with humans is generally a necessity, so be sure to focus on it.
Income? Depends entirely on where he lives. Anything from 15k to 40k starting, up to 35k to 100k once he has several years of experience. In general, take whatever a buisness c++ programmer makes in the area and lower it a bit. Of course, this assumes that the current tend of outsourcing to India/other standard locations slows.
Developers buy existing engines either as a kit or as a foundation. Many developers use it as a foundation to allow content producers to start working day 1 instead of having to wait for the programmers to write the renderer, get everything stable, etc.
Typically, huge chunks of the licensed engine are ripped out and replaced with inhouse code specialized to the title. Anything in the engine is fair game for rewritting.
The game logic generally is always written per game, so he can count on that.
Main suggestion: 1) Specialize in something like physics or networking which isn't saturated (like 3d). 2) Start out early, getting involved with teams making games as quickly as possible. Any of the mod communities out there are wonderful for building these skills.
Its wonderful to see slashdot celebrating its heros like John Carmak, Richard Stalman and Linus Torvads. I mean, with names like these contributing, who needs editors?
Most of them are clean, but that doesn't mean that you automaticly can develope any sort of following through them. The readers there are looking for information on particular games, and generally aren't looking for new things.
In addition, when on these sights, your game is side by side with pro titles. It is difficult to stand out, as people there are going to give games a visual once-over at best.
Now, this isn't to say that it is impossible, just that there is a limit to what these sites can do. They can start buzz well, but you have to be buzz worthy in the first place relative to everything else on the page.
So.. everyone is saying that if we are paying for it, we should have access to it. What about those who aren't payingfor it? Should people who are taxpayers in other countries not get access to the work?
Should Russia or China have free access to programs written by NASA? If programs were open sourced just because they are funded by taxpayer money, well, wouldn't that help people we still view as competition in some ways?
See the boost (http://www.boost.org) sharedptr. It works very well for handling it, as the STL is copy-based. Autoptr is pretty much exactly what you don't want.... Or so I have experienced in my work with it.
Slashdot Mirror
Weird, my 360 works just fine on my non-HD TV.
HD or not really doesn't impact dev times/team size/budgets significantly. Gamer expectations, driven by the experience/art quality possible on a console with more resources, has a much larger impact.
If only there was a rating system, so that parents could easily gauge what was appropriate to their children. Of course, that would require stores only sell age appropriate titles to kids directly.
Damn that record indus.. oh, wait? We are talking about games? They already have those.
On a serious note, the military use of games people keep bringing up is about training people to work as a team, not overcoming hesitation to kill people. I would be more concerned with people training children to use guns (hunting, target practice, paintball) than games.
So, assuming a team of 30 and a dev cycle of 2 years, they pay 18,000,000 in salary? While average games have a total budget 10,000,000? Care to mention the company name?
See:
http://en.wikipedia.org/wiki/Wtf/
Quick, someone email Bill. Slashdot wants Java.NET. ;)
Couldn't the customer alsoturn around and give it away for free if they choose? The developer doesn't have any way to prevent this, do they?
Doesn't this effectively prevent companies from developing and selling middleware libraries? The developer can't control who uses the library, so they can't make money by selling it (as anyone who buys it could just put it up on the net for public download).
If they do a good job, have a clean, simple API and no bugs, then no one needs to pay for support.
Just curious. I am still trying to understand the different buisness models that are reasonable with GPL'ed code. After reading the GPL license (family) several times, I am still trying to figure out how it all plays out.
Apologies - my post was meant to be sarcastic. It was also intended as a response to Minna Kirai post. I definitely blew it on all counts.
I work in the game industry as well; been here about 4 years. The idea that an MMP ought to be simple to optimize, and ought to be nice and modular is definitely a magic-magic land suggestion in my opinion. Games are written against hard deadlines, with requirements that change all the time.
I will leave the arguments about how well software engineering handles these situations to those with more free time than I.
Right! The linux code should be just as simple to follow as your average bash script.
;)
After all, most code out there is properly engineered, with all of the planning done up front, with requirements defined at the beginning of the project and remaining the same through the end.
It isn't like games get tweaked or redesigned daily, as people on the team realize what is fun and what isn't.
You want to make a city? Make a game level in your editor of choice like UnrealEd or QRadiant
Want to make a model? Pop open milkshape or max for games.
The Sims, Quake, Half-Life, Unreal. All of these games have huge communities and toolsets for creating new worlds.
Kids still are building things -- the blocks and tools just became more advanced. Instead of physical objects, they get simulated ones. It also happens to be EXTREMELY cheap by comparision if you have a PC.
The creativity isn't gone, it just relocated a bit.
...fixing the robots?
Abolutely have to agree. I have a few extras sitting around just in case I lose one and suddenly have a good idea. After using these, I can't imagine using a bic. I love never having to scribble to get the ink flow started.
Its called a feature set.
:) ) but all of shortcomings you mentioned have specific reasons and could changed by any decent coder.
Doom isn't trying to be Battle Field -- so why spend resources trying to add a feature (like vehicles) that your game doesn't call for?
If the concern is licensees, anyone with the money to license the engine surely can afford programmers who can add vehicles. Hell, Gunman Chronicles, which starts as a mod and never even had engine source access added a vehicle, even if it was a bit simple.
Doom isn't targeting massive outdoor areas, so why does it need to support 64 clients? Scalablity costs time and money. If they aren't going to use it, then there is no reason at all to work on it. If there was a huge demand for 64 player games, either ID would do it if it fit in to their feature set and time line, or a licensee could add it.
If you are a licensee and you want to add an occlusion system, well, go for it. Same with LOD (though LOD can frequently be a slowdown if the CPU is doing it dynamicly).
Carmack and Id know their shit. I am not a fan boy (hell, I am part of the competition
Hey now, lets be nice. Some of us devs are on the board. ;)
Pardon my ignorance here, but if dumping is definied as :
'Export price that is "unfairly low," defined as either below the home market price (normal value) (hence price discrimination) or below cost.'
Then how is freeware and linux not considered dumping as it is 'below home market price'? To argue that point, it seems like one would have to argue that Linux and OSS had no production costs.
How could, say, Microsoft legally compete price-wise, as they couldn't lower their prices below the cost threshhold?
Its about the OEMs as much or more than the consumer market. They watch the benchmarks closely -- and make decisions based on results.
This is where the money really is, and what is worth fighting for.
Its all about the mags! All the programemrs here have mags, all the artists have ... trinitrons or viewsonics.
;)
As most engineers tend to be paid more, that must mean that Mags are better.. right..?
A vector of bools isn't a vector... of bools. Its an specialized object which is actually a proxy wrapping objects which it accesses individual bits in.
Because of this, it is not an stl compliant container; it doesn't support references it what it stores because you can't have references to bit fields.
As it is non compliant, a better substitute would be a deqeue of bools or a bitset.
For a much better description, see Effective STL by Meyers.
1 - So what? Isn't that true of pretty much every product? You can't make a game that appeals to everyone; you pick a niche and try to own it.
:) Unfortunately, there are driver bugs, minor differences, etc. Always will be.
2 - What does story line have to do with diablo 2/doom 3? While diablo2 was very enjoyable (and I am sure Doom3 will be as well) that doesn't suggest that mixing the gameplay would be a benefit.
As you mentioned in point 1, someone who likes click fests might not like shooters. If you put both in the same game, you are making something that doesn't appeal to either audience.
Note that I am not saying such a game couldn't be successful -- instead I am suggesting that by missing genres, you face dangers as well. Your product loses focus; look at C&C/Renegade.
3 - There are coop games (NOLF2 being the first one that comes to mind). Unfortunately, not many people experience the fun of taking on a shambler with their mates. If coop sold enough titles to be profitable as a primary feature, there would be more such titles.
4 - Those standards are called GL and DirectX.
Specialized areas? All the math he can learn. Other than that, it depends entirely on what area he is interested in programming. AI? Sound? Graphics? Networking? Physics? Those are all specialized jobs.
:) Oh, and learning to communicate with humans is generally a necessity, so be sure to focus on it.
If he can find one that he likes to focus on, all the better. Being a specialist first, a generalist second -- its the only way to stand out from all the other generalists.
Income? Depends entirely on where he lives. Anything from 15k to 40k starting, up to 35k to 100k once he has several years of experience. In general, take whatever a buisness c++ programmer makes in the area and lower it a bit. Of course, this assumes that the current tend of outsourcing to India/other standard locations slows.
Developers buy existing engines either as a kit or as a foundation. Many developers use it as a foundation to allow content producers to start working day 1 instead of having to wait for the programmers to write the renderer, get everything stable, etc.
Typically, huge chunks of the licensed engine are ripped out and replaced with inhouse code specialized to the title. Anything in the engine is fair game for rewritting.
The game logic generally is always written per game, so he can count on that.
Main suggestion:
1) Specialize in something like physics or networking which isn't saturated (like 3d).
2) Start out early, getting involved with teams making games as quickly as possible. Any of the mod communities out there are wonderful for building these skills.
Its wonderful to see slashdot celebrating its heros like John Carmak, Richard Stalman and Linus Torvads. I mean, with names like these contributing, who needs editors?
Most of them are clean, but that doesn't mean that you automaticly can develope any sort of following through them. The readers there are looking for information on particular games, and generally aren't looking for new things.
In addition, when on these sights, your game is side by side with pro titles. It is difficult to stand out, as people there are going to give games a visual once-over at best.
Now, this isn't to say that it is impossible, just that there is a limit to what these sites can do. They can start buzz well, but you have to be buzz worthy in the first place relative to everything else on the page.
And how are you expect us to find the right software systems randomly with GAs without infinitely powerful machines? ;)
Of course.
I was just curious how the arguement 'we paid for it, so we should get it' applied when it would also be accessible to people who didn't pay.
Considering how much classified information gets around, I know that we couldn't keep something really important out of the hands anyone.
So.. everyone is saying that if we are paying for it, we should have access to it. What about those who aren't payingfor it? Should people who are taxpayers in other countries not get access to the work?
Should Russia or China have free access to programs written by NASA? If programs were open sourced just because they are funded by taxpayer money, well, wouldn't that help people we still view as competition in some ways?
See the boost (http://www.boost.org) sharedptr. It works very well for handling it, as the STL is copy-based. Autoptr is pretty much exactly what you don't want. ... Or so I have experienced in my work with it.