I'm sitting next to my hospital bed at 9am on a sunny Sunday 7 days after an 8 hour operation to remove tumours from my tongue and neck (and me a lifelong non-smoker too) plus facial reconstruction. When I go home I'll have many months with not much to do; however I'm not sure testing 68k cores is the way I'll want to spend it :o) Anyway I was thinking of going in a different direction. Obviously, assembler and the QL are like a horse and carriage so projects like these are definitely time well invested, but I think SMSQ/E could have a bright future not locked into one cpu or limited by emulation. Look how easily Linux gets ported to other CPUs. Taking the Raspberry Pi as an example, could we make, rather like me, a Frankensteinesque creature comprising Linux from boot to loading the kernel, redoing SMSQ/E as a replacement kernel, SBASIC as the shell, ditch X it's too slow anyway - go for direct screen handling? I could do some of the coding if in C (but not the integrating into Linux or the screen handling). This would not be an emulator but a completely new version "based on" the original, with necessary differences take them or leave them.
Ian. It's taken me half an hour to type this on their worn-out bouncy rubber keypad. ----- Daniele Terdina <[email protected]> wrote: > Marcel, do you know if any free VHDL simulation/analysis tools exist that may > be close to a code debugger (breakpoints etc., if that makes sense)? > > I wouldn't mind giving the occasional look and fix some instructions, but > only if there is an efficient way of doing it. Going through countless > iterations of experiments and/or needing the hardware would exceed my > available time. > > Thanks, > > Daniele > > > Date: Sat, 22 Mar 2014 01:35:26 +0100 > > From: [email protected] > > To: [email protected] > > Subject: Re: [Ql-Users] New QL > > > > [email protected] wrote: > > > In case you are thinking about hiring someone to design a better > > > CPU, I'm not sure if software developers are the ones to ask. This > > > is not software, it's hardware design. I'm sure folks like Daniele, > > > Richard and Marcel know a lot about 68K instructions, but it seems > > > unlikely one of them would also like to learn chip design. > > > > Actually I have dabbled with VHDL in University and would love to do > > more with FPGAs, but no, I have enough pet projects going as it is. > > Perhaps in another life ;-) > > > > Without even having seen the code I guess the remaining bugs are in > > the processor flag handling. If you don't get them exactly right > > everything kind of works, but then after a few million flawless > > instructions some tiny thing breaks and things go down from there. I > > remember tracing thousands of lines side by side with a real QL to > > catch the remaining differences. But it did pay off, in the 18 years > > since the first release only a handful bugs have been found in the > > emulation core, most by George Gwilt. You should send him a board ;-) > > > > Cheers, Marcel > > > > _______________________________________________ > > QL-Users Mailing List > > http://www.q-v-d.demon.co.uk/smsqe.htm > > _______________________________________________ > QL-Users Mailing List > http://www.q-v-d.demon.co.uk/smsqe.htm _______________________________________________ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
