Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!decwrl!uunet!sjsca4!poffen From: poffen@sj.ate.slb.com (Russ Poffenberger) Newsgroups: comp.sys.ibm.pc.hardware Subject: Re: Memory speeds Message-ID: <1991Feb13.205141.18582@sj.ate.slb.com> Date: 13 Feb 91 20:51:41 GMT References: <59051@eerie.acsu.Buffalo.EDU> Reply-To: poffen@sj.ate.slb.com (Russ Poffenberger) Organization: Schlumberger Technologies, San Jose, CA. Lines: 23 In article vcl@mimas.UUCP (Victor C. Limary) writes: >v119l94u@ubvmsd.cc.buffalo.edu (Dennis G Crawford) writes: > > >Actually, I have a 386DX, not an SX. I got a 16MHz one because it was a >DX, and it was cheaper than the SX's that I could find at the time (about >2 years ago). I understand that I won't be getting any significant >increase in speed with anything better than 80ns SIMMS, though. I'll >probably end up getting 80ns ones. Thanks for your help. > Unless you can select how many wait states are used for memory access, putting faster chips in will NOT make your system go any faster. The speed of memory access is fixed by the number of wait states, and putting in memory chips faster than is required by the number of wait states is a waste of money. You can probably get away with 100ns chips in a 16 Mhz system. Check the manufacturers recommendation. Russ Poffenberger DOMAIN: poffen@sj.ate.slb.com Schlumberger Technologies UUCP: {uunet,decwrl,amdahl}!sjsca4!poffen 1601 Technology Drive CIS: 72401,276 San Jose, Ca. 95110 (408)437-5254