.

HP codename, series | Unknown |

Type, Precision, Input mode | Scientific, 12 BCD digits, exponent ±499, Algebraic or Reverse Polish Notation |

Programmable | Yes. Keystrokes, labels A-Z. |

Performance Index | 44 |

Memory | 31277 bytes available by default. According to the manual the memory is shared by programs, formulas and variables. 26 number storage registers (A-Z), index register (i). Interestingly, the free memory indication does not change when numbers are stored. So it seems the available bytes are available to programs and equations alone. Assuming 32 kB of RAM 1491 bytes are used for number storage and internal data. There are 26 direct number registers (A-Z), the index register, LastX, 4 stack registers and 6 statistics registers, in total 38 registers. Assuming 8 bytes per number this adds up to 304 bytes. So approximately 1 kB is used by the calculator internally. |

Display | LCD with 2 rows and 14 digits each, 5x7 pixel per digit, annunciators |

Special features | RPN
and algebraic entry mode, equations, hex, binary and octal mode,
equation solver, integration, statistics, complex functions (but not a
complex stack), fractions, unit conversions, many physical constants,
program and formula checksums (used to verify transcriptions from
printed listings). |

Original Pricing, Production | 61.95 Euro in 2007 |

Batteries | 2x large button sized cells |

Dimensions | Length 15.8 cm, Width 8.3 cm, Height 1.6 cm |

Links | Manual: HP-33S Scientific Calculator User's Guide (PDF, English, 387 pages, 3rd edition, Nov 2004) Manual: HP-33S Wissenschaftlicher Taschenrechner Benutzeranleitung (PDF, German, 408 pages, 2nd edition, Nov 2004) HP-33s info on Finseth.com. The Cosine bug. |

Comments | In the equation (EQN) and memory (MEM) menu it is possible to display the length of a program or an equation. The "V" shaped keyboard together with the tilted labels and a wealth of commands make it very hard to find the right keys. In general, the design of this calculator doesn't look very appealing to me. |

For one thing the identity sin(x) = cos(90 - x) is observed.

But for "somewhat small" arguments the iteration seems to be aborted too early. The critical point is between 5.7E-8 (correct) and 5.8E-8 (incorrect).

In Radians this corresponds to around 1E-9.

x |
sin(x) |
cos(90 - x) |

1E-1 | 1.74532836590E-3 |
1.74532836590E-3 |

correct | 1.74532836589 83E-3 | |

1E-2 | 1.74532924306E-4 | 1.74532924306E-4 |

correct | 1.74532924313 34E-4 | |

1E-3 | 1.74532925091E-5 | 1.74532925091E-5 |

correct | 1.74532925190 57E-5 | |

1E-4 | 1.74532925000E-6 | 1.74532925000E-6 |

correct | 1.74532925199 34E-6 | |

1E-5 | 1.74532920000E-7 | 1.74532920000E-7 |

correct | 1.74532925199 43E-7 | |

1E-6 | 1.74532900000E-8 | 1.74532900000E-8 |

correct | 1.74532925199 43E-8 | |

1E-7 | 1.74532000000E-9 | 1.74532000000E-9 |

correct | 1.74532925199 43E-9 | |

1E-8 | 1.74532925199E-10 | 1.74532925199E-10 |

correct | 1.74532925199 43E-10 | |

1E-9 | 1.74532925199E-11 | 1.74532925199E-11 |

correct | 1.74532925199 43E-11 |

Back To Contents |