But Nobody's Going to SEE the underside of the table...

The ever-awsome Will Wheaton posted this image through Twitter and then his blog a couple of days ago (http://wilwheaton.typepad.com/wwdnbackup/2010/02/the-cooperhofstadter-coffee-table-proof.html).   What you’re seeing here is a background element.  A table that can be found on the set of “Big Bang Theory”.  Its not hugely important to the story, it’s not something the camera is going to linger on.  As Mr. Wheaton himself said, it could have been filled with any handful of magazines, they didn’t have to go so far as to make sure everything there fit the story and the characters.

The Cooper-Hoffstadter Coffee Table Proof

This is, interestingly enough, one of the things that separates a great game (or TV show) from an AWESOME one.  The level of detail in the world.

People operate under the misconception that if you can’t get a clear look at it, it doesn’t matter.  That taking a line of text and reducing the pixel count until it is unintelligible for a mipmap is just a pain in the a** and not the best way to handle the process, you can just hit it with a few clicks of the mouse in Photoshop and get some wordy looking blobs in place.

But the human eye is a wicked, amazing instrument.  Man-made optics cannot hold a candle to what the human eye can actually pick up on.  The trick is all in the perception, what the mind does with the information once it runs up the optic nerve.  So even though you may have taken a pretty 1024x1024 texture and reduced it down to 32x32, if you’ve stair-stepped it down, those little pixels, their placement in relation to one another, the colors chosen, are going to give the eye a ghost of an image, a hint, a kiss of what was there before.  You’re giving the mind the opportunity to make a connection to the larger, more complex texture, and it will make that connection.  The mind, given just enough information to do so, is looking for a way to fill in the blanks and add it’s own level of resolution to the images coming in through the hardware of the eye itself.  If you take your 1024x1024 texture and say…  well, I dunno, we could get away with just a flat grey panel here, and a black line there, you’re taking away those little details that allow the mind to jump the gap.  It *will* save on memory, don’t get me wrong, taking out those details can take a texture image down to the kb range of filesize, but it’s going to cost in terms of user experience and perception.

Players are the kinds of people (if they're a *fan* of the game and not just playing it to beat it and move on to the next) who are going to dig into corners.  Who are going to somehow, anyhow manage to shoehorn their player character into position to see whether or not you've gone to the trouble of texturing the back of the soda-machine, or whether the magazines on the table have nudy-pictures on the cover.  Details like this often get glossed over out of necessity, because the game ran out of time, because the dev-cycle was just too tight, but often as not they also get left out simply because no-one thought about it.