So... Next gen?

My journey begins... and it really couldn't start any more exciting for me as having the possibility of looking at next gen processes. I Don't know what I might get out of this (given the limited access that I have to Next Gen engines and tools), however my focus will lie upon emulating what I think will be good next gen workflows and techniques for artists using the engines available to us today.

My first meeting with Josh had me shooting for what I thought what would be possible to do in the near future (with the dawning of the release of the next gen engines just around the corner), and where I thought the next gen might be going. Together we started brainstorming and with a little research I had done prior to the meeting, we decided to look at tessellation techniques and set-ups that could help us prepare us to generate "Future Proof assets". 

What do we mean by that?

With "Future Proof Assets",  we aim to develop assets that could be easily transferred to a next gen engine and have it work with all the elements of said engine, without extending production time for a studio , and thus, without generating more cost for that studio. to achieve this we must create this asset on an engine that is available to us now. A good example of why this would be very useful would be any game that would release close to whenever the next gen arrives. To maximize profit and efficiency, naturally, a studio would want to release his game on all possible hardware. However, the production cost and time implications of re-generating all game assets so that they work on the new engines and justify the game being on a next gen engine would typically be very large.

We aim to cut the cost and time of production, using assets that we can create on a "present engine" (such as UDK). These assets must work on both present, and next gen engines. However the next gen engine version must take advantage of the technological advancements of  the new engine.

  

 (Excuse the crudity of my drawings! :s)

To do this I will be looking at dynamic tessellation on in-game objects. I believe this is what the industry is moving towards. Understanding how to set this up and how to keep it cheap on the performance side can be hard... but I guess it wouldn't be worth doing if it wasn't a challenge!

 

So stick around, hopefully I will have some sort of approach to this sometime soon, and hopefully I will receive feedback from anyone who might read this.