Alternate Light Gem

[sparhawk]

What would this thing have to do with the limitations of computers? It would only significantly complicate the calculation of the visibility for every AI, and this is nothing unthinkable for today's computers. Also, it's not limitations of today's computers, but of computers of at least a year from now we're talking about, no?

 

I think the arguments were all given already. You said yourself you read it, so what would it change if we explained it yet another time?

[OrbWeaver]

It would only significantly complicate the calculation of the visibility for every AI, and this is nothing unthinkable for today's computers.

 

And you'd know this because...?

 

If you have some technical knowledge about how this could be implemented using the Doom 3 SDK, feel free to share it. Even better, put together a proof-of-concept demo that shows such a feature is (1) not a major performance issue and (2) improves the gameplay. Nobody is interested in listening to Joe Schmoe carping about what "should" be possible based purely on a gut feeling, with no actual knowledge of the limitations of the available technology -- particularly when there are far more pressing issues to deal with and a small number of people to deal with them.

[ZylonBane]

Anyway, 90% of the benefit of a silhouette system could be conferred by just giving the player a stealth bonus when standing against a wall.

 

As an example of how this might be implemented, TDM could continuously cast, say, six rays out from the player's position in a circular pattern for a distance of about a foot. If three or more adjacent rays strike terrain, the player would be considered to be "wall-hugging", and their visibility would be slightly decreased. Something like that.

[New Horizon]

It feels as though the Mod is just recycling the old stuff.

 

It's a lot of extra work, for little benefit. It also alters the gameplay drastically. It's going to be a hell of a lot more time consuming for us to tweak and balance a gameplay feature that has never been tried and tested in the original series, at least with the current features...we have something to base our work on and test against. Wherever and whenever possible, we are not only recreating the original features but tweaking and improving upon them as well....and our AI are certainly going to have some abilities above the originals.

 

Sorry to disappoint you, but we're making a classic Thief inspired toolset. It's supposed to play and feel similar to the originals...90% of what we're doing is going to be similar to the originals. We have systems that are also going to be different. We weigh the pros and cons. Then we make decisions based on where our time is best spent, and where system resources are best allocated. What's the point of implementing a system that will hinder performance? We saw enough of that shit with the renderer in T3. It's tough enough getting features we're familiar with ingame, let alone try implementing yet another feature to balance the gameplay against. It will be plenty challenging without silhouette detection.