Path: csiph.com!news.swapon.de!eternal-september.org!feeder.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail From: Paul Rubin Newsgroups: comp.lang.forth Subject: Re: The Wikipedia article on Forth Date: Tue, 28 Feb 2017 03:50:49 -0800 Organization: A noiseless patient Spider Lines: 19 Message-ID: <877f4aileu.fsf@nightsong.com> References: <680318eb-535b-4595-96c6-3bf7e7f2928c@googlegroups.com> <80bac9cc-dba0-430a-9a09-9d478e1e8f9f@googlegroups.com> <81c490f2-711c-4fe1-b055-a0c6b8d62bf1@googlegroups.com> <87r32kk3yp.fsf@nightsong.com> <90d87260-6074-41ee-89b2-4a61c8610a52@googlegroups.com> <87fuj0jz9d.fsf@nightsong.com> Mime-Version: 1.0 Content-Type: text/plain Injection-Info: mx02.eternal-september.org; posting-host="cd1e98fe3a45433ab5ee44680f27a824"; logging-data="6886"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX1/Xk9wY419R91zrDYnwXJxK" User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/24.3 (gnu/linux) Cancel-Lock: sha1:M6uQm1cHVrGQH2owQOBMpiv21po= sha1:wrAprVPl89j/U1SbVmI/VDDfsNM= Xref: csiph.com comp.lang.forth:55263 Ilya Tarasov writes: > Now I have a new product line requiring lowest cost chips, so my > latest CPU is multi-threaded and has small foorprint in FPGA. I'm still confused at what you gain by implementing your own cpu, instead of using a software Forth implementation on a hard silicon proessor. The hard cpu is almost surely cheaper/more efficient than a softcore, unless you're using a lot of block multipliers or some other special stuff. > What I want to do is replacing certain CPUs to a methodology which > allows medium-experienced programmer to write his own CPU. Maybe at > least to understand in details how it works. Understanding doesn't seem hard. I've looked at some simple Verilog processors and I think I understand what the code does. I've always heard that writing them is harder than it sounds, especially if you care about performance. So I don't claim to be able to do it. But I'd like to try someday.