home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!ferkel.ucsb.edu!taco!gatech!darwin.sura.net!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!eff!iWarp.intel.com|ichips!hfglobe!chnews!chnews!doconnor
- From: doconnor@sedona.intel.com (Dennis O'Connor)
- Newsgroups: comp.arch
- Subject: Re: RTX and SC32
- Date: 9 Nov 92 09:10:08
- Organization: Intel i960(tm) Architecture
- Lines: 36
- Message-ID: <DOCONNOR.92Nov9091008@potato.sedona.intel.com>
- References: <17102@mindlink.bc.ca> <1992Nov3.144748.21826@sobeco.com>
- <ADAMS.92Nov5001909@PDV2.pdv3.fmr.maschinenbau.th-darmstadt.de>
- <DOCONNOR.92Nov5085408@potato.sedona.intel.com>
- <ADAMS.92Nov7002912@PDV2.pdv3.fmr.maschinenbau.th-darmstadt.de>
- NNTP-Posting-Host: potato.intel.com
- In-reply-to: adams@pdv3.fmr.maschinenbau.th-darmstadt.de's message of Sat, 7 Nov 1992 00:29:12 GMT
-
-
- adams@pdv3.fmr.maschinenbau.th-darmstadt.de (Adams) writes:
- ] One should remember, i960 was out of discussion because of price, at
- ] least here in Germany at evaluation time (about 1990). 800 US $ /chip
- ] were too much, cost for development systems not included.....
-
- Intel i960(R) microprocessors run the price gamut, from under US$20
- for low-end commercial chips to a lot more for full-blown mil-spec
- chips. The CA I mentioned I think sells for well under $100.
-
- ] > And remember, it takes 4 16x16 multiplies and 3 32-bit adds
- ] > ( minimum ) to do a 32x32 bit multiply. How long does that
- ] > take on the FRP1600 ?
- ]
- ] They had well known requirements, multiplying 12-14 bit operands with
- ] 12-14 bit coeffizients. 32 bit multiplications were not necessary.
- ] Summing up 3 products, 16 bit were sufficient.
-
- Sounds like a DSP application. At best it sounds like a 16-bit app,
- not a 32-bit one. There aren't many if any 16-bit RISC processors,
- and comparing architectures with different word sizes is fraught
- with dangers. After all, narrower structures are inherently faster,
- tho not by as much as they used to be back in the 1-level-metal days.
-
- ] > Some older RISC architectures didn't implement fast multiply
- ] > at first but added it in later products. There's no causal
- ] > relationship between architectural class ( RISC, CISC, Stack, VLIW )
- ] > and multiply speed.
- ]
- ] No, but silicon already gone for cache is not available for
- ] multipliers ;-<.
-
- Every year, as yeild figures rise and device geometries shrink,
- there's more silicon available for a given price point.
- --
- Dennis O'Connor doconnor@sedona.intel.com
-