Looks as if the characters are 3D models and everything is else 2D. This isn't all that difficult to achieve just need the assets and program it like a regular 2D game with a 3D model art pipeline for the characters and frame animation.

To me, it looks as most of the environment elements are in 2d raster images and the characters and some dynamics elements are polygonal.

Technically, is not that complicated to do this kind of rendering. You can just render the 2d element on a quads and the polygonal models as usual.

I haven't actually tried the game and am not familiar with it, but reading the responses I notice they appeared to have mixed a bunch of different functions from the graphics API. This reminds me of the experimental programming I used to do.

I would find some new functions, a new API or SDK or whatever, and start programming with no design. I would learn what the functions could do by playing with them and watching the output. Then I would mix them together into something that was "interesting" or "looked neat". While doing this, I would also be thinking about what I could theoretically do with these functions given a lot of time and/or effort....and would write that as comments usually, or even as output options in the functions I was using, like as text under objects or even variable names.....and start designing fun gameplay loops and other game parts that looked neat and were personally interesting and novel to me.....at this point I would usually write one or more longer design documents, sometimes in a text editor like word or notepad, but usually in a comment block ( /* and */) at the top...or elsewhere. Typically at this point i would hand it off to a builder caste human trained in programming......although I've been busy and there's a shortage of soldier caste humans, so all the builders and miners must design their own programs for now. The builders are a little bigger and fatter, while the miners actually get stronger as they shrink (or the other way....it's a survival while mining deep with cave-ins and smaller passege traversal, and also a smaller minimum diameter requirement. They're pretty good at predicting earthquakes and cave-ins, and so the others tend to avoid caves. builders are great at long repetitive tasks and shaping clay and clay-like materials, including metals that turn to clay with sufficient fire, assembled from mined fuels and metals, also some stone and 'fuze', the powder mined element that explodes when dried and lit, and is called 'flint' at a low density. Anyways the miners might be better at design, but both should probably be in on it, and both should be familiar with the functions and variables and scripting, while adding as much as they can, and the builders and diggers work well in groups together doing slightly different tasks, with relevant overlap, so it's best to be unlimiting and natural, and if a builder feels the urge to wander around deep in a mine, they should probably be facilitated as much as possible, and the mixing yields relevant results typically, with good value for the risk. The 'other' caste, the female caste, is also pretty awesome, and while I probably wouldn't have 50% females on the team, having 0% would also be a bad idea. Some are better than others, and some have problems that can render them more than worthless, so be careful who you choose, this applies to all others too. Desperate people are the ones who cause problems, and that can be caused by a lot of things.....I tend to try and give them charity to help them get going in a nice life, and they will try for that if they can find a way....while also recognizing that they can and will pursue violence when starving every time, and other factors that are not starvation can cause similar desperate results in more complex and subtle ways, as welll......

But I digress, finding new gameplay styles means remixing the old. I always like powerful functions for experimenting, so using something like OGRE 3d Engine SDK to invent a new design might work great, especially since you may simply be inspired by the open source code, and notice your "random" or 'haphazard' remixing of the functions while playing with them both in code and in run-mode, along with possibly remixing open source or other available code that uses familiar, connected, relevant, or random other function/variable sets for your preferred language, (note some languages have more code available, also some have more awesome SDKs and whatnot, also note the licensiscing limitations, although those typically only matter if there's upfront cost or if your design uses a 'heavy' feature exclusive to that SDK)

Anyways get the most awesome functions/variables you can for the graphics or whatever, remix them into something nifty, then either post it for free to build reputation, maybe try for a patent (although I personally object to these, they exist and are worth a ton of money....note people kill over that stuff, also lying is common, so be careful around it!), or build it as an SDK or game and try and sell directly to a demographic, such as programmers or gamers. It might even make a good resume, and you can offer it to a company in exchange for money and/or promises of employment....although be careful of plans that "cannot" change, and be ready for anything that is in the future to be unexpected. I should note you might remix powerful open source functions to make something fun, then find when you have something fun all the open source you have used could be re-implemented, either by you or with a more liberal lic. package that does something similar, so that may require swapping out.....also I was looking at the xbox live indy games market, looks like it's only $100 to submit 10 games for consideration, that could be a neat high-school accessable lottery there if you can make a game in XNA, which is in visual basic or C#, and for a small indy game it's probably best to take those xbox-friendly functions and remix them into a novel programming gameplay loop that's personally fun, and try and make a neat $1 or $5 diversion, with the option of posting the failures and crappy ones (probably the first bunch, it takes a while to get up to speed on anything new!, be sure to play the games you make a lot (or features), and be honest with yourself comparing them to other profitable and unpopular games on the market....also play and read about those!)....anyways the nice part about remixing functions in a fast loop is that is the 'core' of the gameplay, from which new styles may come about, and since graphics are a big part of the game, mixing those in innovative ways should be 'interesting' and 'inspiring' like fine art to the people who view them, giving them new ideas about how images and moving game pictures are and can be, and hopefully inspiring new gameplay loops in the future that become popular and well refined with time and use and a lot of human effort.

Proofreading edit: Note if microsoft rejects your first few efforts, maybe post them for free online and get free player feedback and testing, to make another attempt. Also note that all the code can be copied and pasted, a lot of people try and make their first project epic, and fail, when the first project is the novice project.....you gain skill by doing, like in a skill-based mmo!, so you need to increase your programming and/or game-making skills a lot to make profit on the open market, even with charity possibly, although making a halfway decent program might let you get some money out of friends or relatives at first, then maybe actual businesspeople or investors or donating customer/player/testers, since you might impress someone rich with your attempts luckily....although if you post some free software it's not hard to get feedback, and if you take it to heart, fix the problems quickly, and update, you might get enough skill to win good eventually. Thus be sure to write easy programs at first, and maybe just show your friends or even just be honest with yourself, and make a few short ones. The skills you learn, the functions, variables, tricks, debugging techniques, and how the code affects your perceptions at runtime....those tricks stay with you, and by building up a big programming trick book you will be able to write & design better programs faster and more consistently with more past information to draw upon while pondering your moves. Teamwork is also good, and underutilized here, so team up when possible while acknowledging how you need to learn to do it.....and try and make it so you both or all win as hard as possible, while also limiting how much you risk and put out at first and watch the others to make sure they are really into it and good teammates, it might take a few tries to find good people to work with, although I should note tutoring and apprenticeship, in that order probably, is good...a tutor is paid to teach someone, while an apprentice does work for someone who already is good at programming (or whatever) and is trained while working, usually with a reduced salary until they are skilled enough to qualify for a regular job....although a good mechanic I knew was always telling his employees things, tricks, tips, car stuff, and they would sometimes tell him things too, although they clearly respected him as the best mechanic in the shop with the most knowledge....and the new guy looked like he could barely change the windshield wipers, something _I_ can do... , while also noticing that having extra strong hands can be useful provided they have a minimum level of common physical coordination....

Anyone know the specifics of this game?

You appear to be correct. But I've only seen the game in videos.

It really doesn't matter. There is no 2D/3D distinction that needs to be made. You only have a screen full of pixels. The underlying logic is most likely just an evolution of the same tile based platforming logic they have used since the beginning, but now it allows for slopes, spinning surfaces, and anything else they have in there.

Using 3D models allows the animations to be shared across all models and characters, allowing them to get much more out of each of them. Why draw sprites anymore when you can have smoothly animated models in their place? Sprites are memory hogs, are only useful for one character, and the animation is always very clunky as the in-between frames are always missing. Keyframed skeletal animations are memory efficient and reusable!

[font="'Arial Narrow"][size="1"]Typically at this point i would hand it off to a builder caste human trained in programming......although I've been busy and there's a shortage of soldier caste humans, so all the builders and miners must design their own programs for now. The builders are a little bigger and fatter, while the miners actually get stronger as they shrink (or the other way....it's a survival while mining deep with cave-ins and smaller passege traversal, and also a smaller minimum diameter requirement. They're pretty good at predicting earthquakes and cave-ins, and so the others tend to avoid caves. builders are great at long repetitive tasks and shaping clay and clay-like materials, including metals that turn to clay with sufficient fire, assembled from mined fuels and metals, also some stone and 'fuze', the powder mined element that explodes when dried and lit, and is called 'flint' at a low density. Anyways the miners might be better at design, but both should probably be in on it, and both should be familiar with the functions and variables and scripting, while adding as much as they can, and the builders and diggers work well in groups together doing slightly different tasks, with relevant overlap, so it's best to be unlimiting and natural, and if a builder feels the urge to wander around deep in a mine, they should probably be facilitated as much as possible, and the mixing yields relevant results typically, with good value for the risk. The 'other' caste, the female caste, is also pretty awesome, and while I probably wouldn't have 50% females on the team, having 0% would also be a bad idea. Some are better than others, and some have problems that can render them more than worthless, so be careful who you choose, this applies to all others too. Desperate people are the ones who cause problems, and that can be caused by a lot of things.....I tend to try and give them charity to help them get going in a nice life, and they will try for that if they can find a way....while also recognizing that they can and will pursue violence when starving every time, and other factors that are not starvation can cause similar desperate results in more complex and subtle ways, as welll......[/font]

What the hell was that huge post about??? Did you really write all that stuff yourself, or was it done using one of those random text generators?

Nope, aersixb9 is just epic.