On Tue, Mar 15, 2011 at 1:26 PM, Dave Park <plasticu...@gmail.com> wrote: > On Tue, Mar 15, 2011 at 1:14 PM, Dilwyn Jones > <dil...@evans1511.fsnet.co.uk>wrote: > >> Geoff Wicks wrote: >> >>> From then on the project faded from view. >>> >>> Can it be revived? I know very little about hardware, but I sense there is >>> a new interest in native hardware and some projects that people thought >>> would never come now seem feasible. It is for you hardware specialists to >>> determine what is possible and what is not, >>> >> I'd go as far as to say that it is too late in the day for any conventional >> QL replacements now, with one exception. >> >> To stand any chance of success it will need: >> >> 1. Low component count >> 2. Commonly available components (e.g. memory, keyboards, case, power >> supply) >> 3. Sufficiently simple and low cost to allow small production runs and >> assembled by hand by one or two individuals. >> 4. Able to use either Minerva or an existing (or easily modifiable existing >> SMSQ/E version) >> 5. Cost no more than about £200 - £300 >> 6. Hires display with at least 16-bit colour. >> >> To clarify what I mean I'll call it "QL On A Chip or two". >> >> Q40, Q60, Aurora, Goldfire... great designs in their time, but now we need >> a radical leap forward to stand a chance of success. > > > Indeed an FPGA implementation of an m68k chip, or emulation of an n68k on > some other lightweight chip are the only two economically sensible > solutions. > > I preferred the ARM embedded option because in part it is smple and quick > and a software problem, and in part because it matches my skills and ability > to make it happen. > > However, Peter Graf showed that a 68000 in a FPGA is possible (not easy, but > possible) and therefore if we can find someone to take an existing vhdl > 68000 core and build a QL around it in a FPGA, we can build a new system > with a 2- or 3- chip board and VERY low cost. The biggest advantage here > being that the boards would be easily reconfigurable/updatable at a later > date. >
How important is the original keyboard/case to this product design? I'm contemplating gutting one of my old original QL's and refitting with uQLx on a Beagleboard or Pandaboard, and maybe an arduino for keyboard/microdrive interfacing. This solution would preserve the original look/feel, but isn't really reproducible on a large scale. I'm also researching keyboard enclosures that could house a SFF ARM board. Regardless of whether a solution like this went the FPGA or ARM w/emulator route, how do you house it? Or does it matter? One could produce a decent looking retro case for a BB or PB and just require a USB mouse/keyboard from the end-user. The uQLX route is interesting to me because I plan on finishing this experiment in the next couple of weeks and move on quickly to building driver software for the additional interfaces (MMC, USB, network ,etc). FPGA is more interesting to me personally due to the greater challenge, but it'd have to be a coordinated effort and will take longer. Mark _______________________________________________ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
