[email protected] (Paul Gilmartin) writes: > Isn't this telling me that relative to the "machine cycle" (perhaps > a moving target), the 168 outperforms the 3033, and the 165 is > better yet?
re: http://www.garlic.com/~lynn/2014d.html#17 Write Inhibit http://www.garlic.com/~lynn/2014d.html#19 Write Inhibit http://www.garlic.com/~lynn/2014d.html#20 Write Inhibit sorry finger slip avg 370 was 2.1cycles for 165, 1.6cycles for 168 and 1cycle for 3033 165 had same cycle time as 168 ... but went from 165 2mic access memory to approx. 400ns access memory for 168 (so cache miss latency was less). recent posting in thread about IBM ACS http://www.garlic.com/~lynn/2014c.html#104 IBM ACS 3033 started out Q&D effort to 168 logic mapped to 20% faster chips (from FS) http://www.garlic.com/~lynn/submain.html#futuresys i.e. the lack of 370 products during the FS period (which was completely different than 370 and was going to completely replace it ... internal politics had 370 efforts being suspended and/or killed off) is credited with giving the clone processors a market foothold. some logic optimization finally had 3033 approx. 50% faster than 168 (rather than just 20%) .... 3031 was 158 with new covers and the integrated channel microcode moved to a 2nd 158 engine; a 3032 was 168 with new covers and using channel directors (158 engine with just the integrated channel microcode) as external channels. the previous mentioned URL goes into more details about the FS failure and the Q&D efforts to turn out 3033 & then 3081. http://www.jfsowa.com/computer/memo125.htm With regard to ACS (IBM's advance computing effort) by Amdahl ... this has account of many of the features ... but was killed off by IBM management because they were worried it would advance computing state-of-the-art too fast and they would loose control of the market. http://people.cs.clemson.edu/~mark/acs_end.html Amdahl leaves sorthly after ACS was killed to start his own clone computer computing. The ACS "end" article lists several ACS features that don't show up in IBM processors until es/9000 over 20yrs later. -- virtualization experience starting Jan1968, online at home since Mar1970 ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
