
8/25/86
Rob Horning

I looked at several of the cache array signals to get an idea of what it
would take to increase the clock speed.  It appears that we can not run
at 27.5 Mhz with 25 ns SRAM's.  We would need to pick up about 3 ns in the
cache access.  We can only pick up a little less that half of this.  I did
ask a couple of the vendors if we would be able to get screened RAM's that
have 23 ns access time.  They have not gotten back to me.

I got everything together to build 8 more math boards.  The boards are back but not turned on.  I am going to let Murphy turn them on.

I have spent some time on Ultra-fox.  I have have started to read about what
has been done so far.  I went to a meeting on multi-layer ceramic/thick film.
It does not seem likely that we would use it, but it is good to know about just
in case we do get a need.

I have been keeping in touch with the SRAM vendors.  Cypress is still going to
try to get me some 25 ns parts this month.  Hitachi and Toshiba are at least
a month out.

I am expecting to get 1 Mbit DRAM's from both Hitachi and Toshiba next week.

