RE: R: R: Vector games on Laser show hardware

> Maybe galvos have gotten a lot faster in the last few years
> but I suggest you do the math on the speed of the galvos and
> the speed of a typical line drawn on the CRT before you spend
> any real money on equipment.

Yeah, the galvos are still really slow compared to the slew rate of the
original monitors. A vector generator type of signal will steer a galvo
just fine, but there's some black magic involved with getting a good image
from them. Normally the galvos work with endpoints-- just imagine the start
and end of a vector-- and as the mirror physically moves it draws the line
that connects them. The tricky stuff is that for the PC based 'laser show'
type software they take into consideration the mass and momentum of the
mirror so they transform the individual endpoints into multiple points that
both overshoot the original target (to pull the beam closer to the endpoint
before moving to the next one) as well as accelerate and decellerate towards
the target.

(Helps to think about the galvo as being like holding a book at arms length
and swinging it around. It takes a short time to get up to full speed, then
when you try to stop there's momentum and you can't stop on a time, but if
you slow down before the stop you can get a target with miminum overshoot.)

So two 'points' in an image can get processed by one of the PC laser-show
type systems to be:

Orignal:
     1 2

Processed:
     1. . . . . 2 .

With the '.'s being new points added to the original two to get a good laser
image. The how and when of adding those seems to be treated as 'secret
sauce' (ie, trade secret) by the guys that do the various PC apps.
(Although I suspect some time with a scope and it'd be easy enough to figure
out.)

All that lends itself much better towards emulation (with the LaserMAME-like
post processing of the vectorlist), especially when you consider that the
drawspeeds are WAY slower than a CRT, so having a really efficient 'path'
through the display list is important. Another thing you can do is
essentially remove the undesireable traits from the original hardware (like
for example, the Atari AVG returning to the center of the screen frequently
to reset the integrator caps) that would just waste laser time with no
benefit.

Real 'speed' of galvos seems super hard to quantify in any useful metric.
(I tried to pull this from the laser forum guys a couple years back. ;-)
They talk in 'points per second', which supposedly are endpoints, but that's
at a particular 'tuned' slew rate/update frequency on a standard test
pattern. So while faster is better, there's really no good way that I found
to equate a "50K PPS" galvo to anything usable in vector game terms. (And
of course the chinese have cloned some US galvo designs and claim probably
double the performance than they'll actually do. ;-)

-Clay

---------------------------------------------------------------------------
** Unsubscribe, subscribe, or view the archives at http://www.vectorlist.org
** Please direct other questions, comments, or problems to chris@westnet.com
Received on Thu Dec 22 13:27:58 2011