K
Kani Math
Cosmic Abacus
Launch App →
One minute

Abacus reading speed test

How many soroban frames can you read in sixty seconds? Choose two, four or six rods, answer by tapping or by typing, and the next frame is already there.

⏱️ 60 seconds👆 Tap or type🎁 Free, no signup

Pick how wide the frames should be and how you want to answer, then read the number — a wrong answer costs you nothing but the second you spent on it, and the next frame comes straight away.

How many rods?
How do you want to answer?

What this measures

Reading an abacus is a glance, not a calculation. A fluent reader looks at a rod and knows the digit without counting the beads, the same way you read the word "seven" without spelling it. This tests how close you are to that, and nothing else — there is no arithmetic anywhere in the minute.

That is worth measuring on its own because every other abacus skill sits on top of it. If reading a frame still takes you a moment, then adding on one costs you that moment on every single step, and no amount of practice at addition will move it.

Why the frames never get easier

Every frame in the minute is the width you chose, and one frame in four has an empty rod. Neither changes with your speed, so a faster run is a faster run rather than an easier one — which is the only way two attempts can honestly be compared. Scores compare within a width: six rods is a different test, not a harder setting of the same one.

A wrong answer costs no extra time and no penalty. It also does not show you the answer: the misreads are listed at the end instead, where you can look at them without the clock running. The one thing to watch is your accuracy — with four options in front of you, a fast score with a low percentage is guessing rather than reading, and the keypad is there when you want to leave no doubt.

What counts as a good score

A learner who still counts beads one at a time usually lands in the low single figures, and that is the expected result rather than a bad one — it is the number that has the most room to move. Somebody comfortable at four rods tends to sit somewhere in the twenties. Past forty you are reading rather than working it out. Two rods will always score higher than six; compare like with like.

The number to beat is your own, a week from now. Comparing it to somebody else’s is not very useful: this is a test of one narrow skill, and a phone and a laptop do not answer at quite the same speed.

Common questions

What is a good abacus reading speed?
A learner who still counts beads one at a time usually reads a handful of frames in the minute, and that is the expected starting point. Somebody comfortable with three-rod frames tends to land in the twenties. Past forty, you are recognising the frame rather than working it out. The number worth chasing is your own from last week.
Does a wrong answer cost me time?
No. A wrong answer moves straight to the next frame, exactly like a right one, and there is no penalty and no pause. That is deliberate: any time cost on a miss would mix caution into the score at a weight nobody could justify. The misreads are listed at the end instead, where you can look at them without a clock running.
Do the frames get harder as I go?
No, and that is the point. Every frame is three rods wide and one in four has an empty rod, whether you answer five frames or fifty. If the frames widened as you sped up, a faster run would also be a harder run and two attempts could never be compared.
Is my score on a phone the same as on a laptop?
Close, but not identical. Four-option mode keeps the gap small on purpose, because on a phone a thumb is about as fast as a mouse; the keypad widens it again, since typing three digits on glass is slower than typing them on keys. Compare your own runs on the same device and in the same mode and the number means something; compare a phone run to somebody else's laptop run and it does not.
Why is my score low when I know how to read an abacus?
Check the accuracy line under it. With four options on screen, tapping fast without reading still scores about one in four, so a big number with a low percentage is a guess rate rather than a reading speed. The other common cause is the empty rod: 407 read as 47 is the single most frequent misread, and one frame in four is built to catch it.
See all questions →