Thank you all for the responses
The reason I am wanting to use the Retina is because of its apparent and reviewed engraving capabilities. The software side of things is of a value but the cost of constantly updating software can be prohibitive, and that takes a back seat. By the time software becomes an issue, there is a new piece of hardware.
The concerns with the Retina are:
1) As stated on their own website, no support for the DIY. It is understandable, considering the developers have "other" responsibilities, but that equates to making sure that all homework is done PRIOR to making the leap. Otherwise, you can end up with and expensive piece of useless hardware.
2) "1000 steps per inch". I like the R&P setup, especially on the X. On my router it saved me lots of $$$ over a quality screw. With a rack, though, I have not been able to figure out a combination to get an even 1000 s/i. I am assuming that it is because of the value of pi. Everything considered, will ".772 " make a real-world difference? Will the Retina go haywire?
I think I can compensate for the slight difference in part size. I am not sure that it will matter that much for what I do anyway - I deal with "nominal sizes".
3) I completely understand the resolution of the "artwork" on the Retina. Is the 1000 step requirement solely for the artwork or is it for the "operation" of the machine? I fully understand Mach and can deal with all of that.
With all of that said, I did locate belts that I feel can handle the distance of 8 feet. I have pitch choices of .200" (1/2" width), .375" (3/4" width) and .500" (widths 3/4" on up). The .500" calculates to a toothed pulley of 4 (I think), meaning that I think would be looking at a 16 tooth pulley with a reduction of 4:1. Building the reduction units are not a problem (my friend is one of the best fabbers I know) but going from a .250" shaft on a NEMA 23 to ultimately a .625" shaft for the final drive pulley just seems like a PITA, as requiring 3 (possibly 4) pulleys. Maybe just familiarity but the R&P just seems so much easier to implement on the larger machines (although belts seem like the smarter and easier choice for the smaller units).
Is there any way I can get some input from the makers of the Retina?