Dan,

I can't remember if you have discussed your current employment etc. on our
board but I for one would like to know what you have been up to of late.
It's interesting to me to know what each of the creators of AR are doing
these days. Kindly disregard this message if you would prefer to remain
private ;)

That info about the mip-mapping is quite interesting. I like the idea the
16 bit crew had about decimating rows and columns to save memory (1 texture
instead of many).

Cheers,

Darryl Giors

Lance, of the City of Xebec's Demise

Tyr, wielder of the Eater of Evil, wrought by the Dwarven Smithy of the
Dungeon.

At 02:21 PM 9/21/03 -0700, you wrote:
>Uhhh, I forget. Ken would remember, he wrote that part. I think it uses a
>simple Z test but it may have been optimized to use a simple ray casting
>type tests at certain intervals. This is how the Dungeon was able to
>support arches , windowed doors and windows. We had windows in wall at one
>point which were supposed to give you a teaser view in a hard to get area,
>but we didn't like the look. Without bumpmapping or anything else they
>looked rather odd.
>
>I say simple because it had to be, getting the priority and order of what
>the viewer sees from what he is able to see is very CPU intensive or memory
>intensive for pre-generated tables of what the user can see. The Dungeon
>doesn't use the vertical wall mirroring trick the City was able to, arches
>would look funny flipped upside down.
>
>I'm not sure how easy it would be to make the 8 bit Dungeon view pan 360
>degrees, but I did see it on 16 bit Amiga and Atari ST versions. One
>thing Ken did which I still think is very cool is coming up with the idea
>of mip-maps. The 16 bit guys went a different route. They opted for
>single "texture" that had a table for which columns and rows would be
>decimated as it was scaled. To my knowledge the Dungeon is certainly the
>first game that ever used mip-maps, we didn't even have a name for
>it. Perhaps the first use of it ever.
>
>I'll go kick Ken's chair on Monday and see if he can shed some light on the
>raycasting issue.
>
>Dan
>