1) good work, and don't take my comments as criticism.
2) how the smeg did you work out how the beam stuff worked? I tried to work it out (to fix the Shivan beams) but I couldn't make head nor tail of it. And I hope the answer isn't 'wiki', because then I'll be embarrassed.
1. Thanks, and criticism is always good IMO.
2. I got loads of help, and of course the wiki was helpful. Actually, my beams are a lot simpler then the current MediaVP ones. In most cases my beams consist simply of one layer animation, I just made it look like it has many levels. I *could* make it use up to three layers with different $Zadd values, but that would basically triple the memory use to no palpable effect in most cases.
The biggest problem in my beams is in my opinion is that when you view the beam from a very shallow angle, you can sometimes see the flatness. Most of the time they work fine though.
And iirc the tile factor in mode 1... IIRC it is the lenght in meters... that is it is not relative to beam width.
Well, that's what I thought, but as I said, I got some anomalous behaviour when I tried to use same aspect ratio for all same-coloured beams. I first calculated the tiling factor based on width so that it was same on all shivan beams, but surprisingly, some beams had very different texture aspect ratio when I did this. So I finally settled on using the (about) same tile value on all same-coloured beams, and that made the beams of same colour more consistent.
To clarify the previous problem, here's a screenshot.
Could someone find out what the code actually does with the tiling factor value? I'd really like to know. Obviously, I could've made some calculation error too.
I would really like it if you could really specify the texture length in metres if it would work properly...
NB: The picture attached is somewhat older stuff. You can see that even though the tiling factor and beam width *should* have given same aspect ratio to the beams in this case, the smaller beam had the texture tiled in much denser than the wider beam. Currently, that's not a problem any more.
[attachment deleted by admin]