This was prompted by a question on the electronics site (which was asking why only an open collector variant of a particularly old part existed and how it could be part of a data bus).

This was a pretty important development as it revolutionised bus interfacing (such as memory systems) among other things.

I know that 3 state (tri-state in National Semiconductor parlance which was [is?] trademarked) came out sometime in the (early) 1970s but I am trying to pin it down to at least a year range.

I have tried searching for the date of the trademark without success so I am asking this here in the hope people here might remember their first parts with this feature.

I have found a citation at ACM from 1978 that references this, but I am not sure if actual parts were in the wild prior to this. My first parts using this were in the early 80s (so I know it existed by then).

Note that I am looking for the first commercial implementations.

I would appreciate any insight.

3 Answers 3


I don't know how to prove a first commercial implementation, but the Texas Instruments System 74 Designer's Manual (copyright 1973) includes the 74125 and 74126 quad 3-state buffers.

A Motorola TTL databook from 1971 doesn't include these circuits. Neither does a Signetics databook from 1972.

(I found all these references linked from this project5474.org page.)

  • Interesting; I used to have the dead tree editions of those at one time (lost many a move ago, though). Dec 12, 2019 at 16:28
  • I'd hoped I might have something contemporaneous on the dustiest shelf at home, but the oldest thing I've got is a 1976 TTL databook. In the late 70s and early 80s when I was a happy digital hobbyist, I wanted the newest books I could scavenge, not the oldest. Sigh...
    – jeffB
    Dec 13, 2019 at 3:29
  • This pins the dates down to a reasonably close estimate (around 1972 give or take a year) and is really very useful when educating others (as many of us try to do) as to just why open collector bus implementations existed (and indeed still exist today). Dec 14, 2019 at 14:38

I developed all the early Tri-State devices at National. Dale Mrazek in applications postulated the the need for something to replace the open collector, resistor pullup, which was the standard for buses at that time. I figured out the implementation and wrote at least one (can't remember) application note. Dale wrote some stuff too. There was no formal introduction the way you'd think of a new family. It was just an extension of the DM55/75 family at National. The first part was the DM7551 (4 bit tri-state latch) and I think it was introduced in 1969/70. At the time, National used to give seminars on the application of new parts and I remember making presentations on it in Japan, England, and France.

  • 1
    Thank you, Jeff, for the first-hand account. I'm always worried that some of those "oral" histories will be lost. Mar 11, 2021 at 18:19
  • Did any TTL chips use three-state logic internally for their internal structures, or was everything up to that point implemented with cascaded NAND and NOR gates?
    – supercat
    Nov 23, 2021 at 17:08

National Semiconductor had the name Tri-State and introduced the first commercial devices. Texas Instruments and others adopted three-state logic into their product lines afterwards. There will be multiple search hits on Dale Mrazek who invented the logic family and National Semiconductor. NatSemi was always recognised due to the hold on Tri-State name.


  • 2
    Thanks for taking the time to answer, but this doesn't actually say when three-state was commercially introduced, which is the question here.
    – cjs
    Dec 13, 2019 at 0:09
  • This is very useful background information and it helps me to dig a bit deeper as well. Thanks! Dec 13, 2019 at 9:39

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .