Datasheet legend
Ab/c: Fractions calculation
AC: Alternating current BaseN: Number base calculations Card: Magnetic card storage Cmem: Continuous memory Cond: Conditional execution Const: Scientific constants Cplx: Complex number arithmetic DC: Direct current Eqlib: Equation library Exp: Exponential/logarithmic functions Fin: Financial functions Grph: Graphing capability Hyp: Hyperbolic functions Ind: Indirect addressing Intg: Numerical integration Jump: Unconditional jump (GOTO) Lbl: Program labels LCD: Liquid Crystal Display LED: LightEmitting Diode Liion: Lithiumion rechargeable battery Lreg: Linear regression (2variable statistics) mA: Milliamperes of current Mtrx: Matrix support NiCd: NickelCadmium rechargeable battery NiMH: Nickelmetalhydrite rechargeable battery Prnt: Printer RTC: Realtime clock Sdev: Standard deviation (1variable statistics) Solv: Equation solver Subr: Subroutine call capability Symb: Symbolic computing Tape: Magnetic tape storage Trig: Trigonometric functions Units: Unit conversions VAC: Volts AC VDC: Volts DC 


Although not a programmable calculator, this SCM Marchant I deserves mention in my collection.
For one thing, it is one of the earliest portable calculators ever made. (Some sources describe it as the earliest, but that's probably incorrect.) For another, it's probably the only portable calculator ever made that used Nixie tubes for its display.
I found this machine recently in a thrift store. Almost missed it; because the sliding keyboard cover was in place, it did not look like a calculator at all. It would have been a real pity to miss it, too, because it is in surprisingly good condition, complete with the original wall adapter. Although the internal NiCd batteries were corroded, no corrosion appeared anywhere in the calculator itself; its insides are pristine, the outsides, nearly so, the SCM Marchant logo looks like it was affixed yesterday!
The calculator did, however, have a slight defect: the T key (somewhat essential, being the only means to display the total of any addition or subtraction) did not work. Disassembly, reseating some contacts, and reassembly was sufficient to fully revive the machine.
So what can I say about this early beast? Apart from the Nixie tubes in the display (beauties on their own right) it sports a number of unusual features. First, there's that analog meter next to the onoff switch, indicating the battery charge level. There's the onoff switch itself, which is mechanically connected to the display hood; you can turn on the calculator by lifting the hood, and turn it off by closing the hood again. Then there are the two indicator lights above the T and C keys;one is the negative result indicator, while the other blinks when the calculator encounters an error (overflow or division by zero.) These lights are actually tiny light bulbs! There are curious internal features as well; for instance, the calculator's two main chips are surface mounted on small ceramic circuit boards, which themselves are socketed edgewise in a fashion not altogether unlike the way used with PentiumII processors.
It is hard to imagine that 30 years ago, folks paid the price of a good used car for a fourfunction portable (but decidedly nonpocket) calculating machine that's almost as heavy as (but bulkier than) my PentiumMMX subnotebook computer, itself a couple of years old already. I wonder; will my subnotebook look just as ridiculous in another 30 years?
For a nonprogrammable calculator, I obviously cannot provide a programming example; what I can provide is an easy method to calculate square roots on the Marchant I. Easy inasmuch as it only requires you to enter the argument more than once, you do not need to note down and reenter intermediate results. For instance, to calculate the square root of ten, here's what you need to do:
1. Enter the argument: 10
2. Key in the sequence: + ÷ 10 = ÷ = = + T / 2 =
3. Repeat step 2 as many times as necessary, until the result remains unchanged except in the last digit.
Seven iterations are sufficient to generate an accurate result: 3.1622776. Since the algorithm is selfcorrecting, mistakes you make usually won't alter the final result, only increase the number of iterations required.