X-Received: by 10.224.5.5 with SMTP id 5mr5090743qat.4.1380722308713; Wed, 02 Oct 2013 06:58:28 -0700 (PDT) Path: csiph.com!v102.xanadu-bbs.net!xanadu-bbs.net!news.glorb.com!a6no816408qak.0!news-out.google.com!9ni3300qaf.0!nntp.google.com!Xl.tags.giganews.com!border1.nntp.dca.giganews.com!nntp.giganews.com!local2.nntp.dca.giganews.com!nntp.supernews.com!news.supernews.com.POSTED!not-for-mail NNTP-Posting-Date: Wed, 02 Oct 2013 08:58:27 -0500 From: Randy Yates Newsgroups: comp.arch.embedded,comp.dsp Subject: Re: Small, fast, resource-rich processor Organization: Digital Signal Labs References: <7x7gedqun7.fsf@ruckus.brouhaha.com> <7xa9j82g4a.fsf@ruckus.brouhaha.com> <7xhad4vhoi.fsf@ruckus.brouhaha.com> <87k3hw39ec.fsf@digitalsignallabs.com> Date: Wed, 02 Oct 2013 09:58:26 -0400 Message-ID: <87ob77bwql.fsf@digitalsignallabs.com> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/24.2 (gnu/linux) Cancel-Lock: sha1:2q0gQATq4pii8rQcGrNSkvefKb8= MIME-Version: 1.0 Lines: 25 X-Trace: sv3-ZKl7t7Yg9Nm7u/nz5FZN2AT376RUjxTSGIjq2mvKjl1dWZHS0fh6oaRWOvSVJLJ0KMQQoZ1F50F0Yjx!WYosGsNwGT3Txi0UVByteXdt2sKxKztedGEpyDhrSGUu5qy4c6UnFLSIkZ2LH7TJLGFL/zDY X-Complaints-To: www.supernews.com/docs/abuse.html X-DMCA-Complaints-To: www.supernews.com/docs/dmca.html X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your complaint properly X-Postfilter: 1.3.40 X-Original-Bytes: 2667 Content-Type: text/plain Xref: csiph.com comp.arch.embedded:14088 comp.dsp:19449 Tim Wescott writes: > On Tue, 01 Oct 2013 18:37:31 -0400, Randy Yates wrote: > >> rickman writes: >>> [...] >>> I'm just going to make one more point and then I'm done with this >>> topic. Floating point is just fixed point with a range extending >>> multiplier added. Anything you can do with fixed point you can do with >>> floating point. The opposite is *not* true. >> >> Here is something you can do with fixed-point that you can't do with >> floating point: perform a multiply-accumulate with 64 bits of precision. > > You're presuming that floating point ends at 64-bit. There's higher > (even arbitrary) precision floating point standards out there. YOU'RE assuming integer ends at 64-bit - there are higher precision integer libraries (e.g., GMP) out there... (See my follow-up to rickman for the basis of comparison I intended) -- Randy Yates Digital Signal Labs http://www.digitalsignallabs.com