A scientific calculator is really several small tools bundled into one keypad: trigonometry, logarithms, exponents and roots, factorials and counting functions, and usually some way to handle very large or very small numbers. Most of the confusion people run into isn't the math itself — it's forgetting to set the right mode before typing an expression. Below are the four areas that generate the most "why is my answer wrong" moments, plus what a well-built online tool does under the hood to avoid them.
Trig and the degree/radian/grad trap
sin(30) = 0.5. In radian mode, that same key expects radians, so sin(30) gives a completely different (and usually meaningless-looking) number. To get sin(30°) = 0.5 in radian mode you'd need sin(π/6) instead.This is the single most common source of "the calculator is broken" complaints, and it's almost never the calculator. A full circle is 360 degrees, 2π radians (≈6.2832), or 400 grad — three different rulers for measuring the same angle. A good online scientific calculator puts DEGREE / RADIAN / GRAD as visible tabs above the keypad rather than burying the setting in a menu, specifically so you can double-check it in the two seconds before you hit equals.
Inverse trig works the same way, just backwards: asin, acos, and atan take a ratio (a number between −1 and 1 for asin/acos) and return an angle in whatever mode is currently selected. So asin(0.5) returns 30 in degree mode but 0.5236 in radian mode — same underlying angle, different units. Hyperbolic functions (sinh, cosh, tanh and their inverses) are unrelated to circles and angle mode entirely; they're defined in terms of e, so they behave identically no matter which angle tab is active.
| Angle | sin | cos | tan |
|---|---|---|---|
| 0° | 0 | 1 | 0 |
| 30° | 0.5 | 0.8660 | 0.5774 |
| 45° | 0.7071 | 0.7071 | 1 |
| 90° | 1 | 0 | undefined |
tan is undefined at 90° and 270° because cosine is 0 there (division by zero); a calculator will typically show this as a very large number due to floating-point rounding rather than a true error.
log vs ln vs log2 — which one do you actually want?
log(1000) = 3 because 10³ = 1000). ln is the natural logarithm, base e ≈ 2.71828 (ln(e) = 1). log2 is base 2 (log2(8) = 3 because 2³ = 8). For any other base, use a base-conversion function like logn(base, number), e.g. logn(3, 81) = 4.A worked example: suppose you're estimating how many times you'd need to double a value to go from 1 to 1,000,000 (a hypothetical scenario, not a real financial or scientific claim) — that's log2(1,000,000) ≈ 19.93, so about 20 doublings. Swap to log(1,000,000) = 6 and you get the base-10 answer instead: six powers of ten. Same number, two very different-looking results, because the base changed.
Factorial, permutations, and combinations
5! = 5×4×3×2×1 = 120, and by definition 0! = 1. Permutation (order matters): nPr(5,2) = 20. Combination (order doesn't matter): nCr(5,2) = 10. Both nPr and nCr are built on factorial under the hood.Factorial only makes sense for 0 and positive whole numbers — there's no "half a factorial" on a standard scientific calculator. It grows extremely fast: 10! = 3,628,800 already, and 20! is over 2 quintillion, which is why factorial buttons are usually paired with scientific-notation display for larger inputs.
| n | n! |
|---|---|
| 1 | 1 |
| 5 | 120 |
| 7 | 5,040 |
| 10 | 3,628,800 |
A classic example that trips people up: how many ways can you draw 6 numbers out of 49 in a lottery, where the order they're drawn in doesn't matter? That's a combination, nCr(49,6) = 13,983,816. If order did matter — say, ranking the top 6 out of 49 candidates in exact 1st-through-6th order — you'd use the permutation instead, nPr(49,6) = 10,068,347,520, a much larger number because every different ordering now counts separately.
Scientific notation and base-N conversion
6.022×10²³ directly (type 6.022, press EXP, then 23). Switching to a BASE-N mode lets you type a number in decimal, binary, octal, or hex and see it converted into all four bases at once, plus run bitwise operations like AND, OR, XOR, and bit-shifts.Scientific notation exists because very large or very small numbers are unreadable in plain decimal — nobody wants to count zeros in 602,200,000,000,000,000,000,000. Typing the exponent directly (rather than manually multiplying) also avoids rounding mistakes on the way in. On the output side, a calculator will typically switch to scientific notation automatically once a result gets large or small enough, e.g. displaying something like 1.23×10^18 instead of an eighteen-digit string.
Base-N mode is a separate, more specialized case aimed at anyone working with binary, hex color codes, or low-level bit operations: 1010 (binary) AND 0110 gives 0010, for instance. It's a different keypad and display layout from the trig/log keys, which is why it usually lives behind its own mode toggle rather than being mixed into the main scientific keypad.