I'm sorry, but I just couldn't help it. I kind of want to see somebody pull that stunt on a Linux forum now!

Bah! Missed that part. I fail. :p

Still, switch to XNA. :P You get Windows, 360 and WP compatible code with one go. :p

Ooh, sounds fun! Kind of a Civilization-style setup, where an entire team takes their turn at once, and turns don't shift until all said actions have been set? That could really REALLY be neato mosquito.

I know that the concepts would have to be laid down in a lot more detail, but in the long run, if there ain't a team then there ain't no point, is there? :D

As for the pricing? About half the games I've got in my head I'd imagine as being released free, as a doorway to distribution of the paid games.

I see some big heads with good ideas here and I think reality has to come into play here.

People mentioning XNA: it's a good idea but it's terribly implemented and has no cross platform compatibility with -- for example -- Mac, Linux, PS3, Wii, iPhone, Droid. That drops out huge segments of the market, and you're not going to make up for those easily. For example, for any given game released on Mac, Linux and Windows, Mac and Linux together will make up around 35% of sales despite only making up something like 6 percent of market share. If you're planning on a freeware game or don't care about market size, go for it. Otherwise, PLEASE look at something else.

To those suggesting the Quake 3 engine -- I love that engine, but it's hell for newbies due to some iffy netcode and age. I suggest the UDK for any sort of 3D engine. It's very flexible, I'm finding. Also, it supports PhysX, has enormous cross-platform compatibility if you're willing to fudge a bit -- Android is the only thing I haven't heard of it supporting, and I'd be surprised if that were actually the case -- and the sheer quality of games that can be near-painlessly made for it is astounding. I have made a dodecahedron into a work of beauty in the test room for my UDK game, and I don't even have models or textures in yet, just materials, lighting and post-process.

HPL1 is a great engine -in theory- but nigh impossible to force to do what you want. The open-sourcing of it gives me high hopes for its future, but it's a horribly implemented wreck. That might not be fair to Frictional, because it was their first game, but I think open-sourcing that trainwreck might have been the only way it'd be usable without coaxing the code to do exactly what you want over many, many frustrating hours. Also, yeah, the non-support of Intel cards. (Which isn't actually the dev's fault, but due to the crazy hacks normal engines that support Intel use to get around that, people are sort of spoiled.)

Python: it is hell. Others have had far different experiences, but to this day I have not been able to figure it out. Give it a shot, but don't put too much time into it if you don't get it right off the bat.

I don't know much about 2D game engines. I know that there is some seriously amazing stuff coming from the guys at the forums over at TIGSource.com, some of whose work you may even know already -- Aquaria, for example. They're the go-to guys for questions on 2D games, and I highly recommend heading over that way if that's your aim.

Not sure how much got I'd be at that, but I'd try. I'm not much for writing, I'm decent at it, but what I meant was that I spent almost all of my childhood trying to learn things from books to the point that I had very few friends. I'd love to help if I could in some way, but honestly, I have no talents. I can't think of a single thing I can do that anybody else couldn't do better.

I've dabbled with map editors for first-person shooters, as well as a few incomplete projects for RPG Maker 2, and small games for ZZT and Megazeux, Klik N' Play, and writing games and programs for TI-83+, C64, Arduino, all of it ultimately amounting to very little, but some experience in designing games.

You forgot the 360 and WP market. ;)

While I'm not in favour of XNA myself and agree that it isn't the language of choice if you want to reach decent performance, I wouldn't say that it really harms sales that much. Look at Magicka and it's success, even despite the very, very harsh requirements concerning graphic cards.

But I think it's very very safe to say that if it could be ported to Mac and Linux, it'd be an extreme sales boost.

You don't see the lack of Mac and Linux support hitting Magicka hard precisely BECAUSE it's doing well.
No, actually I didn't. The first list was of platforms XNA explicitly does not support. The second list was of three platforms which are the only places used by a hypothetical game for release. (These are also the most likely three for an indie developer, IMO.) I don't think that the 360 market would be a huge enough gain to obviate the Mac and Linux markets, anyway.

I can only see an owner of both a reasonably powerful computer and a 360 getting one copy of the game, certainly not one per system. An extra one on Windows Phone isn't beyond the realm of possibility though, and that might be enough of a gain to push the Mac and Linux sales back down to around 10% of all sales -- which is still a pretty big percentage.

Bringing up the issue with XNA is a bigger deal mostly due to where we're at -- if the games' target audience is mostly this site's members and those with similar tastes, then compatibility with Linux would be fairly important I'd think.

Whatever turns you on, guys and girls. :D

XNA is very easy to develop with, especially if the guy coding has no or just a small previous experience. Plus, the syntax is beautiful, and lots of code samples and examples are available straight from MSDN. Also, there is this: http://www.dreambuildplay.com/Main/Default.aspx

But if you want to use something else then its okay. :p

I'm not about to deny the ease of development, even if you're coding-challenged. UDK is similar, however, and MUCH MUCH better to use for reasons of speed, ease of use (many things can be done almost as easy as Game Maker), quality and portability.

XNA is locked to three platforms, is inefficient, REQUIRES coding to a certain degree (much in the UDK can be altered by simply replacing assets and changing config files), has no built-in mod capability (which is an innate part of the UDK) and requires ridiculous effort to reach the same result as a UDK game.

The UDK can be used on that sort of hardware as long as you don't mind developing without seeing the final product at its best. Turn off post-process, don't render shaders you can afford not to, disable real-time shadows, disable ambient occlusion. I've got a comparable setup (3GB RAM, ATI Radeon Mobility HD 4250) and don't even need to do that though. I'd say check it out. (Just don't enable anything with the word "volumetric", your fps will nose-dive even if otherwise you'd run well. For the record, Lightmass is volumetric lighting.)

Thanks! I'll give it a try too, then. After all, if it can be used to build a fully-realized Sonic The Hedgehog game from scratch, not to mention Borderlands, it should be good enough for anything I wanna do.

They seem to teach Python at MIT, at least at an introductory level: http://www.youtube.com/watch?v=k6U-i4gXkLM
Good lecture series if ur interesting in learning some CS in ur spare time :)