Path: csiph.com!news.mixmin.net!aioe.org!.POSTED!not-for-mail From: "Rod Pemberton" Newsgroups: comp.lang.forth Subject: Re: Minimal wordset (was: New to the group) Date: Fri, 09 Oct 2015 23:11:42 -0400 Organization: Aioe.org NNTP Server Lines: 70 Message-ID: References: <4f5b572c-04ae-433e-bed0-9d7b754bc27b@googlegroups.com> <2015Oct8.083842@mips.complang.tuwien.ac.at> <7I-dnQl_lpiNhYvLnZ2dnUU7-RmdnZ2d@supernews.com> <2015Oct8.153955@mips.complang.tuwien.ac.at> NNTP-Posting-Host: n4wpt9zq8xR26Ttf9mo2BA.user.speranza.aioe.org Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit X-Complaints-To: abuse@aioe.org User-Agent: Opera Mail/12.16 (Linux) X-Notice: Filtered by postfilter v. 0.8.2 Xref: csiph.com comp.lang.forth:42536 On Thu, 08 Oct 2015 11:13:12 -0400, JUERGEN wrote: > If you factor out the existing word set to the max > - the result should be a minimum list by definition. Forths that build up from "primitives" usually need at least 35 for a Forth which isn't abhorrently slow. Some have gone as low as 9, but this wasn't ANS or a complete Forth wordset. These are usually ITC Forths created after C.H. Ting and Bill E. Meunch popularized this build-up idea with eForth. > When I discussed this minimum wordset with others it was to find > out about the possibilities of different levels for hardware conversion: You can make an AND using NANDs, but if the hardware has direct support for AND, then you'll AND for AND. It's the the lowest, supported functionality, i.e., the "primitive". You can still reduce the wordset buy implementing NAND and eliminating AND, but then you're doing extra work to get AND from NAND, i.e., slower. > Core set in assembler bow - and conversion to VHDL > [...] > I am still not sure why nobody in the Forth community has > tackled this. > [...] > Forth could compile to VHDL ( or Verilog ) / Gates directly. > [...] > This was my interest in the minimum wordset. > Nothing to do with Forth as such. Hasn't a basic minimum wordset which would work with VHDL already been found by Charles Moore for his various microprocessors? He has implemented Forth directly as the instruction set for various microprocessors. I would think that you'd start by looking at his designs. Whatever is present for most of them is likely to be what you need. You're probably going to find some basic microprocessor functionality, some basic Forth words, and maybe a few other things. Forth processor instruction sets: http://www.ultratechnology.com/mup21.html http://www.ultratechnology.com/f21cpu.html#cpu http://www.colorforth.com/inst.htm http://users.ece.cmu.edu/~koopman/stack_computers/sec4_3.html http://users.ece.cmu.edu/~koopman/stack_computers/sec4_4.html http://users.ece.cmu.edu/~koopman/stack_computers/sec4_5.html http://users.ece.cmu.edu/~koopman/stack_computers/sec5_3.html These are Forth VHDL related: http://www.ultratechnology.com/p16vhdl.htm http://www.ultratechnology.com/4thvhdl.htm http://web.archive.org/web/20041012145634/http://www.cse.cuhk.edu.hk/~phwl/msl16/msl16.html (last is one link) You can find more links here: http://www.ultratechnology.com/chips.htm If the link is bad, try Wayback archive: https://archive.org/web/ Rod Pemberton -- Just how many texting and calendar apps does humanity need? Just how many food articles from neurotic millenials do we need?