Clearing Up Confusion
I read the article Dave Bursky wrote and it's very interesting ["New Interfaces Squeeze Top Performance From SRAMs," May 27, p. 59]. Yet there's a slight error in the FCRAM description. The address bus is actually multiplexed, not nonmultiplexed as indicated in the article. Originally, the FCRAM design had a nonmultiplexed address bus (described at the VLSI conference in 1998). But actual devices adopted a multiplexed ad-dress bus based on customer feedback and performance/cost tradeoffs. The Infineon RLDRAM device uses a nonmultiplexed address interface. Perhaps the confusion arose from RLDRAM and old FCRAM material. Also, FCRAM is described as Fast Column instead of Fast Cycle.
Adrian Cosoroaba, Fujitsu
FCRAM Product Marketing Manager
History Of Resistor Values
Dear Mr. Pease: I have a historical question about resistor values. It's fairly common knowledge (at least among serious analog types) that all 96 resistors of a decade of the 1% tolerance resistors can be easily de-rived by rounding to two decimal places in the equation, 10(n/96), where n is an integer from 0 to 95. It also is true that most of the 24 resistors of a decade of the 5% series can be derived by rounding to one decimal place in the equation, 10(n/24). But the following values from the logical equation differ from the actual values we use: 2.6, 2.9, 3.2, 3.5, 3.8, 4.2, 4.6, and 8.3. Except for the 8.3 value, the other values are 0.1 below what's actually used. I've always wondered why.
There's bound to be an interesting story behind this. Did someone make a mistake way back when that was cast in concrete? Did someone have an overriding preference for certain values (i.e. an emotional management decision versus a logical engineering decision)? Did this result from some scheme to frustrate a competitor many years ago?
Kenneth A. Kuhn, Senior Engineer
OI Analytical / CMS Field Products
Pelham, AL
I've heard some of the stories of how these R values were chosen. Is 3.0, 3.3, 3.6, 3.9, etc., a set of "round numbers?" Are the colors easier to read than the theoretical ones you expected? I'll ask around.—RAP
Reprints
