Heh, right. And how many gamers are going to look at a surface in Rage and go "woah, streaming megatextures!"?
Of course the scenario makes sense to me and my cynical nature if I swap "people" with "investors". You can give Impressive Technical Numbers for megatextures, but a square highlight is just a square highlight. What's likely to impress the money guys more?
Well, plus the fact that square highlights won't give you an entire landscape's worth of unique textures... it's a completely different issue, not really worth comparing like you did here.
Sure MoP. I didn't mean to compare megatextures being a equivilant development to square specular. And I'm certainly not trying to suggest Carmack doesn't know what he's talking about. I'm not smart, or arrogant, enough attempt that.
If he had said something like "we had to choose between allocating resources on developing megatextures or realistic spec..." then I wouldn't have even blinked. The choice in that context is obvious, especially if you are developing an engine for outdoor environments. And yes the initial post was kind of flippant. At this point in real time 3D technology the textures are going to be much more noticeable to the gamer than the quality of reflection.
I just thought it was funny that the one of the features of megatextures is to not see something - tiling textures. Yet that's a goal being talked about for reflecting actual light sources (or soft shadows, or whatever), so you don't see dots or other hacks everywhere.
The point I had in my head is if they both work then the end experience should be that the user doesn't notice that they are working, because they are fixing something that stood out because it was breaking the illusion.
Yet it sounds like he was saying he didn't pursue a bunch of techniques because they wouldn't stand out. Which is funny! Ha ha? Okay, funny to just me then. And absolutely not a point I made clear at all.
Sorry to bring this thread back, but there was a post in a thread on the 3dcoat forums that I thought was very thoughtful, and would certainly refresh our views on specular:
Replies
I think they made the right call.
Sure. No doubt. But that's assuming there was a dilemma point where they had to decide between them because only one could ever be done.
Well, plus the fact that square highlights won't give you an entire landscape's worth of unique textures... it's a completely different issue, not really worth comparing like you did here.
If he had said something like "we had to choose between allocating resources on developing megatextures or realistic spec..." then I wouldn't have even blinked. The choice in that context is obvious, especially if you are developing an engine for outdoor environments. And yes the initial post was kind of flippant. At this point in real time 3D technology the textures are going to be much more noticeable to the gamer than the quality of reflection.
I just thought it was funny that the one of the features of megatextures is to not see something - tiling textures. Yet that's a goal being talked about for reflecting actual light sources (or soft shadows, or whatever), so you don't see dots or other hacks everywhere.
The point I had in my head is if they both work then the end experience should be that the user doesn't notice that they are working, because they are fixing something that stood out because it was breaking the illusion.
Yet it sounds like he was saying he didn't pursue a bunch of techniques because they wouldn't stand out. Which is funny! Ha ha? Okay, funny to just me then. And absolutely not a point I made clear at all.
http://www.3d-coat.com/forum/index.php?showtopic=3561&view=findpost&p=30378
The bold bits are quoted from someone who was much less knowledgeable on the topic of Specular, and as you can clearly tell, he was set right.