My roommates and I have been PC PVR'ing for about a year. It quite hard to get it setup even for a software guru, but once you do its pretty cool.
One tip I've learned is to get good software. ATI's in the box PVR software is in currently pretty bad shape. It crashs all the time, and generally just sucks.
The best PVR software I've found so far is iuVCR: It's nice because it lets you choose the codecs and it uses the win2k scheduler so it works. It also is very stable.
If you combine that with PICVideos MJPEG codec at setting Q19 then it looks pretty good at 2GB/hr.
And then if you want it small then use VirtualDub to post-process recompress it with DivXNetworks's Divx v5.02 at 600bps to get it down to 0.35GB/hr.
If you like to automate it you can use the scheduler to run a batch file VirtualDub and pass in cmd line args to make it automatically recompress all AVI files from one dir and output them to another. Just run vdub and setup the settings the way you want, and "Save Processing Settings" to a file called settings.vcf (say) then run this at night:
There have a lot recent games that support user created art & content for example: the Quake series, the Sims, Morrowind for the PC, Neverwinter Nights, Mind Rover, Rollercoaster Tycoon, and more.
And most of these games that allow user content creation are among the highest critical reviews: http://www.metacritic.com/games/pc/highscores.shtm l
However you must also realize that creating game art is a lot more complex today. It takes a team of artist years to create the average game. In 1987 games like "Adventure Construction Set" (http://www.the-underdogs.org/game.php?id=1923) were more accessible simply because only a few pixels need to be set per sprite for the art to be competitive with professional quality games.
If you had lots of money could by Visual Matrix's XBox for your MS Xbox. And then if couldn't afford the Xbox HDTV addon pack, you could convert your Xbox's RGB output to 1080p. What fun!
Of course, the UBox may not be so useful.
First off we need to establish how the mainstream PC game industry works. Let me state that the industry's goal is not only out to make really awesome games, but they are sadly also out to make money. Most of the companies featured at your local EB are publicly traded, so they have to post a profit every 3 months (quarterly) or else they will enter a downward spiral of falling stock price, employees leaving, and a lack of funds for new games. So money is very important to the management of these companies.
For example Ken Williams, retired founder of Sierra, in a recent talk said that he cared a lot about the money aspect of the business because if he did he could grow the company and make 500 games (which he did before retiring), but if he didn't then he could only make a few games. When asked what the biggest mistake was he made during his career he said that it was going public. Because he did this he was forced to ship early, and met the expectation of rising profits every quarter. On the plus side, it was able to start a company that grew to 1000+ employees and made "Sierra" well-known industry name
Now how does all of this relate to Linux game development?
To answer that, let's look at how you make money in the industry. Ken Williams in this talk explained how difficult it is to make money in the industry. He said a typically a game sells for ~$50. Of that money, Sierra saw around $25. Of that $25, it is divided between all aspects of running the company (marketing, R&D, etc). By the time its all said and done, you're leave with around $3-4 profit and you have typically around $5 of money to spend on R&D (paying the developers and artists). So if you sell 100k units, you only have $500k to pay the artists & developers. This will pay a team of 6 people over a development time of 2 years, but only if the game sells 100k units. According to PC Data, the top PC games in 1999 sold 80k units (top consoles games sell ~2mil units). The industry has grown a little since then in general is still the same but with larger teams and longer dev time. Now if you go over budget, you're eating your profits, and if you don't sell (>10k units), then you're operating at a loss.
Once we realize the situation the mainstream game companies are in, it becomes clear that Linux game development is very difficult. The sales of Linux games are really small. This is mainly due to a small install base (as compared with Windows or console platforms).
Aside from small sales numbers, you also have to consider the "easy of use" factor and how it affects sales. The console market (which it has fixed hardware and OS) making a game work right out the box is easy because there is only one test matrix. On the Windows platform, it gets exponentially harder, because now you have to deal with the various MS OSs as well as support for all the possible HW the user could have. While the DirectX API does make it much easier than the old DOS days, it doesn't solve all the problems because there are still a huge number of configurations possible. Many non-hardcore users will be running out-of-box drivers and many times these aren't the ones that work. So the problem gets increasing complex. On Linux where there's not only a huge number of hardware configurations, there's also a huge number of Linux OS versions out there so supporting them all of that is even harder than Windows (where you have 2 mainly two different OS configs -- Win9x & Win2k). On Windows and Linux making these combinations work, takes time (money which is cut out of profits). If you ship without making it work right you get a mass of complaints, as seen in nearly every 3D PC game shipped, and compatibility complaints make users less likely to buy your products in the future (creating a downward spiral).
There are of course small development houses (shareware, startups, etc) that are not publicly traded so they can afford the luxury of not worrying about money as much as the big guys do, but in order to make games that arrive at an EB near you, they eventually end up inside "the system".
Slashdot Mirror
My roommates and I have been PC PVR'ing for about a year. It quite hard to get it setup even for a software guru, but once you do its pretty cool.
/wait virutaldub /ssettings.vcf /b"d:\in\","d:\out\" /x /r
One tip I've learned is to get good software. ATI's in the box PVR software is in currently pretty bad shape. It crashs all the time, and generally just sucks.
The best PVR software I've found so far is iuVCR: It's nice because it lets you choose the codecs and it uses the win2k scheduler so it works. It also is very stable.
If you combine that with PICVideos MJPEG codec at setting Q19 then it looks pretty good at 2GB/hr.
And then if you want it small then use VirtualDub to post-process recompress it with DivXNetworks's Divx v5.02 at 600bps to get it down to 0.35GB/hr.
If you like to automate it you can use the scheduler to run a batch file VirtualDub and pass in cmd line args to make it automatically recompress all AVI files from one dir and output them to another. Just run vdub and setup the settings the way you want, and "Save Processing Settings" to a file called settings.vcf (say) then run this at night:
start
move d:\in\*.avi d:\in-done
Good luck
There have a lot recent games that support user created art & content for example: the Quake series, the Sims, Morrowind for the PC, Neverwinter Nights, Mind Rover, Rollercoaster Tycoon, and more.
m l
And most of these games that allow user content creation are among the highest critical reviews: http://www.metacritic.com/games/pc/highscores.sht
However you must also realize that creating game art is a lot more complex today. It takes a team of artist years to create the average game. In 1987 games like "Adventure Construction Set" (http://www.the-underdogs.org/game.php?id=1923) were more accessible simply because only a few pixels need to be set per sprite for the art to be competitive with professional quality games.
tpun
If you had lots of money could by Visual Matrix's XBox for your MS Xbox. And then if couldn't afford the Xbox HDTV addon pack, you could convert your Xbox's RGB output to 1080p. What fun! Of course, the UBox may not be so useful.
First off we need to establish how the mainstream PC game industry works. Let me state that the industry's goal is not only out to make really awesome games, but they are sadly also out to make money. Most of the companies featured at your local EB are publicly traded, so they have to post a profit every 3 months (quarterly) or else they will enter a downward spiral of falling stock price, employees leaving, and a lack of funds for new games. So money is very important to the management of these companies.
6 74 371,00.html
For example Ken Williams, retired founder of Sierra, in a recent talk said that he cared a lot about the money aspect of the business because if he did he could grow the company and make 500 games (which he did before retiring), but if he didn't then he could only make a few games. When asked what the biggest mistake was he made during his career he said that it was going public. Because he did this he was forced to ship early, and met the expectation of rising profits every quarter. On the plus side, it was able to start a company that grew to 1000+ employees and made "Sierra" well-known industry name
Now how does all of this relate to Linux game development?
To answer that, let's look at how you make money in the industry. Ken Williams in this talk explained how difficult it is to make money in the industry. He said a typically a game sells for ~$50. Of that money, Sierra saw around $25. Of that $25, it is divided between all aspects of running the company (marketing, R&D, etc). By the time its all said and done, you're leave with around $3-4 profit and you have typically around $5 of money to spend on R&D (paying the developers and artists). So if you sell 100k units, you only have $500k to pay the artists & developers. This will pay a team of 6 people over a development time of 2 years, but only if the game sells 100k units. According to PC Data, the top PC games in 1999 sold 80k units (top consoles games sell ~2mil units). The industry has grown a little since then in general is still the same but with larger teams and longer dev time. Now if you go over budget, you're eating your profits, and if you don't sell (>10k units), then you're operating at a loss.
Once we realize the situation the mainstream game companies are in, it becomes clear that Linux game development is very difficult. The sales of Linux games are really small. This is mainly due to a small install base (as compared with Windows or console platforms).
Aside from small sales numbers, you also have to consider the "easy of use" factor and how it affects sales. The console market (which it has fixed hardware and OS) making a game work right out the box is easy because there is only one test matrix. On the Windows platform, it gets exponentially harder, because now you have to deal with the various MS OSs as well as support for all the possible HW the user could have. While the DirectX API does make it much easier than the old DOS days, it doesn't solve all the problems because there are still a huge number of configurations possible. Many non-hardcore users will be running out-of-box drivers and many times these aren't the ones that work. So the problem gets increasing complex. On Linux where there's not only a huge number of hardware configurations, there's also a huge number of Linux OS versions out there so supporting them all of that is even harder than Windows (where you have 2 mainly two different OS configs -- Win9x & Win2k). On Windows and Linux making these combinations work, takes time (money which is cut out of profits). If you ship without making it work right you get a mass of complaints, as seen in nearly every 3D PC game shipped, and compatibility complaints make users less likely to buy your products in the future (creating a downward spiral).
There are of course small development houses (shareware, startups, etc) that are not publicly traded so they can afford the luxury of not worrying about money as much as the big guys do, but in order to make games that arrive at an EB near you, they eventually end up inside "the system".
Cheers,
Jason
References:
http://www.zdnet.com/zdnn/stories/news/0,4586,2
http://www.feedmag.com/vgs/data.html
http://www.pcdata.com/