"Rosenzweig, Joel B" wrote:
> scanners aren't that accurate, yes we could break them up into
> Regions, but to
> get where they intersect to sync consistantly
> scanners are analog devices... take a pencil, draw a zig zag line
> down a
> page... then draw a second one right next to it and try to get all
> the points at
> the ends of the lines to meet... if there is a game that is R G &
> Blue only,
> we could seperate that way as long as any object isn't multi-colored
>
> If the system was that inaccurate, you'd have terrible video quality. They
> are apparently accurate enough that you can draw multiple overlapping video
> frames without making your line widths inappropriate due to movement of the
> lines. No doubt you'd need to calibrate your scanning/projection system
> just like any other device of this nature, but given the inherent qualities
> of the display device you are already using, I think you'd be ok.
High Speed scanners have good repeatability. I can do a stable image no
problem. Vector Games are not Raster. every image isn't static to the
Scanner.... they are variable.. so yes I could take a static single frame and
get it to align, no problem... scanners are analog, not digital, they are not
stepper motors. the DAC's are 16-bit and to align basically to a single point
level. we are running at like a 10 degree scan angle... per point to align
up... I'd have to be accurate & Precise within .01 degrees
next thing.. the beams are not coming from the same place.. so they are going
to be coming from different angles and hitting a flat surface (skew issues)
at this point it would be easier to rasterize the whole thing...
>
>
> the MAME driver is a interperter, you don't have code for what is
> what... that's
> all handled in the ROM code it self....
>
> which is being emulated and running on your host machine. All the data is
> there.
>
> and yes in MAME there is one catch all vector to raster file that
> you can pull
> the vectors and pallet with ease.. that is exactly what we did with
> LaserMAME,
> we had to massage the data a bit to reduce some vectors and blanked
> lines
>
> The reason you can pull out the vectors for your own use is because the
> original machine generates and uses a display list for all the vectors. So,
> any machine that is emulating the vector game will also be emulating this
> display list in some form. The whole point is that you do have access to
> the display list, so you can perform whatever post processing steps you like
> to this data before sending it to your scanning devices. In my scenario,
> you would do the processing to split the display list and you'd skip the
> rasterization.
yes, if it was possible to get multiple projectors to align, this is where it
would be done.
>
>
> now for some games like Battle Zone, there are defiently segmented
> regions...
> like the top scoring area... if that was broken off to a different
> region/scanner I could probably eliminate flicker completely with
> laser.. but
> would have to add another $8k US of equipment to display it.
>
> No one said that high performance was cheap! It's not unreasonable to me to
> believe that if you can spend $8K on a prototype and you're looking to
> improve your performance now, and just can't wait for the scanning system
> that does 200K points per second, that buying more hardware that *is*
> currently available would go a long way. It all depends on your goal.
we are making a board right now that will work with MAME and output as
vector... I think partswise it's going to be under $200 on a ISA board... it's
a frame dumping board, designed to work with MAME and output on Laser or Vector
Monitor and not have to rely on the PC to generate re-tracing... it's the same
exact type of signal... It would be easy enough to test your theory on even a
Vector Monitor.. by changing the vector Order to work in Sections... and with
even using a single Gun Vector Monitor... this will either prove me right or
wrong.. but I bet there will be seperation of corners of Rocks, ships etc...
even on a Vector Monitor as it crosses the borders.
Later,
Rob
---------------------------------------------------------------------------
** To UNSUBSCRIBE from vectorlist, send a message with "UNSUBSCRIBE" in the
** message body to vectorlist-request@synthcom.com. Please direct other
** questions, comments, or problems to neil@synthcom.com.
Received on Tue Nov 21 15:09:40 2000
This archive was generated by hypermail 2.1.8 : Fri Aug 01 2003 - 00:32:34 EDT