Club100 owns the bulk of the option rom rights used in REX. I would suggest that anyone who DIYs a REX should consider making a contribution to Club100 to help support the site. That is one of the primary reasons why I was selling through Club100.
I think it would make sense to have some people in the community who can own various aspects of design- 1) PCB layouts and designs 2) VHDL code 3) REX Manager software (big fixes and new features) Keep in mind that there are NEC and Tandy variants as well. Steve On Tue, Feb 21, 2017 at 10:05 AM, Brian White <[email protected]> wrote: > I say it's totally doable. I can write up my experience into a more > monkey-see-monkey-do recipe that should give the next guy a lot more > confidence to give it a go. > > The components were something like $60 or so for 3 boards. > > The only tools I used were a plain solder pen, fine pitch silver solder, > flux, wick, reading glasses, a xilinx usb platform programmer, a test clip, > a wire, a cut off usb cable and usb charger for 5vdc, and a 10 year old > laptop running linux. > > Oh and one helpful part was an empty molex socket to stick the board in to > for programming just to supply the 5vdc. You could manage that a number of > ways. Maybe the next easiest is just stick it into a model 100. But as it > hapoens, I have about 135 sockets, and I'll give at least one to anyone who > wants, free. > > The software was all free, from the OS to the xilinx tools to the tpdd > server to the rex files. > > It's not all presented in a complete recipe or procedure yet, but everything > is there. The files Stephen posted DO contain everything needed to produce a > working REX. I didn't get any special software by private email or anything. > I only used what was publicly available. > > > The next thing I would ideally like to do is copy the files to github so > that I and others can start hacking on them a little. > > A few basically trivial changes would help a lot. > > * Provide some better way to pull PORT_EN high. If there were room this > would be a jumper that is left open except for programming. But it could > also be as simple as just a pad that's easier to touch with a probe, and > easier to get it right without shorting your 5v supply or the on board 3.3v > regulator, than using the side of R3, or worse the correct leg right on the > cpld. > > * Router path that cuts the chip to the right size, so you don't have to > sand or grind the board down to make the half-holes. > > * Maybe a couple other clip-able test points to use as an alternate way to > provide 5v for programming. So you don't need a special molex socket. > > * Silk screen that indicates the proper direction of the module better. > Maybe even adjust the router path to provide a key tab the sticks out of the > center on one end to fit in one of the slots in the socket. I have an old > commercial rom that was built exactly that way. > > -- > bkw > > On Feb 20, 2017 12:12 PM, "Josh Malone" <[email protected]> wrote: >> >> This is awesome news! Perhaps there's hope for REX leaving the >> "unobtanium" status? >> >> On Mon, Feb 20, 2017 at 2:16 AM, Brian White <[email protected]> wrote: >>> >>> I followed the 4.9 install procedure and it seems to have worked! >>> >>> I used linux dlplus and the teeny loader that comes with it. >>> >>> Now I'm going over the docs to *use* it. I don't even know how to load an >>> option rom yet. I don't know how to excercise it either, like a memory test >>> to test all of the flash etc. So far all I've done is install and run the >>> initial backup. >>> >>> But so far so good. It went through every step in the 4.9 install doc >>> with no unexpetcted events. >>> >>> Twice actually, both boards. >>> >>> Thank you very much for creating this insane thing! I am amazed by what >>> all goes into it. >>> >>> -- >>> bkw >>> >>> >>> On Sun, Feb 19, 2017 at 7:05 PM, Brian White <[email protected]> wrote: >>>> >>>> I just noticed I was doing 2 things completely bone-headed before. I was >>>> pulling PORT_EN high from the 5v rail on the breadboard instead of the >>>> on-board 3.3v. It worked both times, because luckily the data sheet says >>>> it's 5v tolerant. >>>> >>>> I also had the damned board plugged in the socket backwards, so I was >>>> feeding the incoming 5v to pin 15 instead of 1 !!! ahhhh. >>>> >>>> So I re-did everything with the stuff connected the right way and >>>> replaced some of the pics in the photo album. So one of those links in the >>>> previous post is dead now. >>>> >>>> Both modules erased and programmed successfully the 2nd time around, so >>>> maybe I didn't kill anything. >>>> >>>> https://goo.gl/photos/7wRjuTUDv6cojpjU9 >>>> >>>> -- >>>> bkw >>>> >>>> On Sun, Feb 19, 2017 at 6:26 PM, Brian White <[email protected]> wrote: >>>>> >>>>> >>>>> That worked! >>>>> >>>>> I don't know what to do next, but it says programming suceeded. >>>>> >>>>> call 63012 just hung, so I guess an image of some sort still needs to >>>>> be loaded. >>>>> >>>>> Should I now follow the 4.9 update instructions, which would also serve >>>>> as a first time clean install? >>>>> >>>>> >>>>> >>>>> https://photos.google.com/share/AF1QipPJNUW-liQ5gNreJ7yIVSpcRFLI1Rc5J2JilvVlWL7_6VLMefMwbKHelHpMLQsOzQ/photo/AF1QipOVigLQapQwIbTO7hMGsqlK3Wt9xPsXDzXh9-o-?key=WmdhblljUlA2UngydnV3WTVOS0NJTEVxQjEwMERR >>>>> >>>>> >>>>> https://photos.google.com/share/AF1QipPJNUW-liQ5gNreJ7yIVSpcRFLI1Rc5J2JilvVlWL7_6VLMefMwbKHelHpMLQsOzQ/photo/AF1QipOmROW_Xr6_R-0I1MOZwII3_M9Lc71z3DeT1Taq?key=WmdhblljUlA2UngydnV3WTVOS0NJTEVxQjEwMERR >>>>> >>>>> >>>>> -- >>>>> bkw >>>>> >>>>> >>>>> >>>>> On Feb 19, 2017 8:16 AM, "Stephen Adolph" <[email protected]> wrote: >>>>>> >>>>>> you probably have to pull port_en high to program it. >>>>>> >>>>>> >>>>>> On Sun, Feb 19, 2017 at 1:09 AM, Brian White <[email protected]> >>>>>> wrote: >>>>>> > Hey Stephen, am I missing anything here? >>>>> >>>>> ... >>>>>> >>>>>> > https://goo.gl/photos/7wRjuTUDv6cojpjU9 >>>>>> > >>>>>> >>>> >>> >> >
