Slashdot Mirror
Act Like A Real Star Trek Captain: Talk
Snaller writes: "Partnering up with Game Commander developers Mindmaker, Interplay is promising the world that when their space sim Star Trek: Klingon Academy is released, Trekker happy players will be able to order the crews around simply by speaking to the game - the press release does not specify if it has to be in Klingonese." And if Loki ports it, perhaps they will also come up with a robust, generalized voice recognition system for Linux! :)
No!
I said WARP SPEED!
Not GIVE WORF SPEED!
Someone get that crazy Klingon under controll!
It seems to me the Loki does straightfoward ports. It appears that they have developed a pretty robust toolset and they are able to port Windows software without too much difficulty. While each game no doubt requires adding to the tool-kit, voice recognition seems a bit off.
I'll be the first to admit that I haven't messed with sound much on Linux. My old 1.x kernel based Linux install choked on my PAS 16, and currently my Linux box is without sound... it all runs through the X Server on my NT machine.
However, from what I have seen, sound still doesn't have the super flexible API that MS produced. Quality or not, MS finally has a system for accessing sound cards without too much trouble. I have no doubt that Linux is pretty far along by now, but I doubt that sound integration is THAT pleasant. IIRC, there are two sound driver models... which is an improvement, but I don't see anything approaching DirectSound for at LEAST 6 months to a year.
Loki would no doubt reinvent the wheel. If the system that they bought/licensed abstracts enough, maybe they can plug in Linux sound support. But if the Linux sound capabilities isn't as flexible as the DirectSound, I wouldn't expect it.
Also, while porting old games is possible, most Linux users/gamers maintain a Win9x dual boot for games, releasing a Linux version a year down the road sounds like a bad business. Even if initial sales are solid, the later games will fail. While the novelty of a Linux port is solid in the beginning, in the end, how many people want to buy old games.
To get Linux ports, the userbase must demonstrate that there is a market (I believe there is, while Linux marketshare > %Windows users that are games), and must demonstrate that it is important to satisfy this market.
The trick is, companies must believe that a Linux port 6-12 months down the road isn't good enough. That requires companies that simultaneously release their versions see a large portion of Linux sales. If that happens, than maybe companies start working with Loki to port their game once the engine is largely complete. I mean, if properly abstracted, the game can write code that Loki can rewrite with their tools, and then the game is a straight recompile.
In a group project I was on, one person basically spend a day abstracting 1-2 line commands into a simple API. If we had to switch systems, he could rewrite his API in a day on the new system, and the higher level code would work without changes.
Portable code isn't difficult, you just abstract out all the hardware dependent features, and then have someone reimplement them. This may result in more work in the initial development, but it makes porting easier. Much like abstracting assembly code carefully, DirectX and other OS specific calls need to be abstracted away, so that they can reimplement on another system without trouble.
Alex
Just in case anyone else was wondering, it can be downloaded here.
And here is a good "How to get started" -document.
When you smile, the world laughs at you.
Actually, there already is one. IBM's Via Voice should do everything you need. Freely available, IIRC, but not open source.
"The invisible and the non-existent look very much alike." -- Delos B. McKown
Also, there is an official Klingon/English, English/Klingon dictionary available, appropriately titled "The Klingon/English Dictionary" =)
.- CitizenC (User Info)