Matrix Settings and Values
Choose the matrix order and elimination controls. Then enter every matrix value.
Example Data Table
This sample shows how row elimination clears values below each pivot.
| Stage | Matrix form | Action |
|---|---|---|
| Start | [[2, 1, -1], [-3, -1, 2], [-2, 1, 2]] | Select 2 as the first pivot. |
| After column one | [[2, 1, -1], [0, 0.5, 0.5], [0, 2, 1]] | Clear values below the first pivot. |
| Final | [[2, 1, -1], [0, 2, 1], [0, 0, 0.25]] | Clear the final lower entry. |
Formula Used
Row replacement: Ri ← Ri − (aik / akk)Rk
Partial pivoting: choose p where |apk| is largest in the current column.
Determinant check: without pivot normalization, det(A) = (−1)s ∏uii.
The calculator repeats row replacement below each pivot. It swaps rows when a stronger pivot improves numerical stability.
How to Use This Calculator
- Select a square matrix order from 2 × 2 through 10 × 10.
- Choose partial pivoting for the most stable general calculation.
- Set display precision and the near-zero tolerance you need.
- Enter all values, including negative numbers and decimals.
- Select Convert to Upper Triangular Form to view the result above.
- Review the recorded row operations and download the finished matrix.
Understanding Upper Triangular Matrices
An upper triangular matrix has zeros below its main diagonal. The diagonal runs from the top left corner to the bottom right corner. Values above that line may be any number. Values below it must become zero.
Gaussian elimination creates this structure through row operations. The process starts at the first column. A suitable pivot is placed in the top available row. Every value below that pivot is removed. The method then moves one column right and one row down.
Why the Form Is Useful
Upper triangular form makes many matrix tasks simpler. A system of linear equations can be solved by back substitution. Start with the last row. Find the last unknown. Then move upward one row at a time. This is faster and clearer than solving every equation together.
The form also helps with determinants. For a triangular matrix, the determinant is the product of diagonal entries. A row swap changes the determinant sign. A row scaling changes the determinant size. This calculator reports the determinant from the original matrix, so pivot normalization does not confuse the answer.
Choosing a Pivot Method
Partial pivoting checks the available values in one column. It chooses the value with the greatest absolute size. This often reduces rounding problems. It is a strong default for decimal values, large values, and matrices with very small pivots.
First available pivoting uses the first nonzero value found below the current position. It can show a shorter path. It may be useful for classroom examples. However, it can be less stable when inputs contain decimals or values with very different sizes.
Precision and Tolerance
Real calculations can create tiny values, such as 0.00000000003. These usually come from floating-point rounding. The near-zero tolerance tells the calculator when to display such values as zero. A smaller tolerance keeps more detail. A larger tolerance gives a cleaner matrix.
Displayed decimal places only changes the shown result. It does not replace the internal calculation precision. Use four to six decimal places for everyday work. Use more places when checking sensitive engineering, science, or numerical analysis values.
Reading the Operation Log
Each listed operation explains one row change. A swap statement moves a better pivot into place. A subtraction statement removes a value beneath a pivot. Normalizing divides a pivot row so its leading value becomes one. Together, these steps create a traceable calculation.
Review the log when checking homework, preparing notes, or validating software output. It shows the exact elimination path. Different valid pivot choices can produce different-looking upper triangular matrices. They still represent equivalent row operations and can lead to the same system solution.
Exporting Results
Exported results support study and record keeping. The CSV file opens in spreadsheet tools for sorting, annotation, or comparison. The PDF file preserves a clean visual matrix for printing. Save both formats when a project needs an audit trail. Clear records make later calculations easier to review, repeat, and explain to others without hunting through handwritten notes or repeated captures.
Frequently Asked Questions
1. What is an upper triangular matrix?
It is a square matrix with zeros below the main diagonal. Entries on and above the diagonal can have any valid numeric value.
2. Can this calculator use decimal values?
Yes. Enter integers, decimals, negative values, or scientific notation supported by your browser. Choose a suitable tolerance when very small decimal results occur.
3. Why does the calculator swap rows?
A swap places a usable or stronger pivot in the active row. This avoids division by zero and can improve numerical stability.
4. What does partial pivoting do?
It selects the largest absolute available value in the current column. This choice often reduces rounding error during elimination.
5. Does normalization change the displayed matrix?
Yes. It divides each pivot row by its pivot. The result has leading ones where possible, while remaining an upper triangular form.
6. Why is the determinant zero?
A zero determinant means the matrix is singular. At least one pivot is missing, and the matrix does not have a full independent set of rows.
7. What does rank mean here?
Rank is the number of usable pivots found during elimination. It measures how many independent rows or columns the matrix contains.
8. Can I calculate a non-square matrix?
This tool accepts square matrices only. Square matrices support the determinant result and the requested upper triangular matrix layout.
9. What is near-zero tolerance?
It is the threshold used to treat tiny floating-point values as zero. It keeps harmless rounding residues from cluttering your result.
10. Are different triangular results possible?
Yes. Different valid pivot choices or pivot normalization can create different forms. They still come from equivalent row operations on the original matrix.
11. How can I verify my result?
Consistent practice builds confidence when handling unfamiliar matrix problems.