this was quoted a while ago “If your cameras orientation is determined by pitch and yaw angles (quake-style) then you will be better off setting it up with 2 glRotate calls - one around the x axis, one around the up axis.” is there a different style of doing camera movement besides the 2 glRotate calls?

If so is it better?

# cameras

Do a search on “quaternions” and “Euler angles” on the web. There are many ways of applying a viewing transformation. It’s just that in a first person shooter you only need the five degrees of freedom until you fall over dead hence the recommendation you read.

The actual implementation of camera viewing transformation is completely arbitrary, for example, if I was sitting on a seat on a fairground ride where the seat was swinging and was attached to an arm a few meters out from a spinning hub, and the arm was pitching under hydraulic control then the most appropriate way to apply a viewing transformation would probably be to concatenate all the above transformations and apply the inverse of the resulting matrix to the modelview stack, that way the eye would follow the transformed seat when I drew it later.

There are a few well exercised ways of doing viewing transformations, just understand at the outset that these are arbitrary conventions which get used and pick the one most appropriate for your application. Don’t sweat it too much.

i was heading towards quake style… since that seems to be easy and at the moement is what i need… thx for the help