Usualy when I make my uv templates and I scale an object down, I keep every dimension proportional. However, today I was making one item skinier so that I could fit things better into my template. But then I would have extra space in the height (I hate seeing unused space in a uv map). Then it occured to me, that perhaps if I only scale in the width, but leave the height as full, im maximizing my pixels and my models will have a tad better quality at least in the height on the mapping?
Or am I adding more noise by not scalling proportionaly?
As for the second qustion, i've been thinking on this one for a while. I used to rotate my stock photo to get it aligned right, then use the perspective tool to fix perspective distortions. However, this involves two processes, and I know even rotating 1 degree adds a bit of noise to an image. (Cant shift 1 degree in a pixel very well as regards to 90). It did occure to me that I could not bother rotating an image at all first, and just use the perspective tool right away using a 1 step process, but somehow it seems it doesnt do as well a job.
Thanks for any input.