Hey Everyone! When starting out with DSA, many of us get confused about why our solutions get Time Limit Exceeded. We often don't know how to estimate whether our solution will pass the time constraints or not. Let's break this down and understand it clearly!
The Golden Rule
Most online judges (including LeetCode) have a time limit of 1-2 seconds per test case. Modern computers can perform roughly 10^8 (100 million) simple operations per second.
This gives us a simple way to estimate if our algorithm will be fast enough:
If your algorithm performs more than ~10^8 operations for the given constraints, it will likely give TLE (Time Limit Exceeded).
How to Use This Rule
Step 1: Identify the Constraint
Look at the problem constraints. For example:
1 <= n <= 10001 <= nums.length <= 10^51 <= n <= 10^9
Step 2: Calculate Operations for Different Complexities
Here's a quick reference table:
| Time Complexity | Max Input Size (n) | Operations for Max n | Will it Pass? |
|---|---|---|---|
| O(1) | Any | 1 | ✅ Always |
| O(log n) | 10^18 | ~60 | ✅ Always |
| O(n) | 10^8 | 10^8 | ✅ Yes |
| O(n log n) | 10^6 | ~2×10^7 | ✅ Yes |
| O(n²) | 10^4 | 10^8 | ✅ Borderline |
| O(n²) | 10^5 | 10^10 | ❌ TLE |
| O(n³) | 500 | 1.25×10^8 | ✅ Borderline |
| O(n³) | 1000 | 10^9 | ❌ TLE |
| O(2^n) | 20 | ~10^6 | ✅ Yes |
| O(2^n) | 30 | ~10^9 | ❌ TLE |
| O(n!) | 11 | ~4×10^7 | ✅ Borderline |
| O(n!) | 12 | ~5×10^8 | ❌ TLE |
Practical Examples
Example 1: Two Sum
Constraint: 2 <= nums.length <= 10^4
Brute Force (O(n²)):
Operations = (10^4)² = 10^8✅ Will pass (but just barely)
Hash Map (O(n)):
Operations = 10^4✅ Will pass easily (and is optimal)
Example 2: Subarray Sum Equals K
Constraint: 1 <= nums.length <= 2×10^4
Three nested loops (O(n³)):
Operations = (2×10^4)³ = 8×10^12❌ TLE guaranteed
Two nested loops (O(n²)):
Operations = (2×10^4)² = 4×10^8⚠️ Risky - might TLE on some test cases
Prefix sum + hash map (O(n)):
Operations = 2×10^4✅ Will pass comfortably
Example 3: Permutations
Constraint: 1 <= nums.length <= 6
Backtracking (O(n! × n)):
Operations = 6! × 6 = 720 × 6 = 4,320✅ Will pass easily
But if the constraint was nums.length <= 10:
Operations = 10! × 10 = 3,628,800 × 10 ≈ 3.6×10^7✅ Still passes (factorial grows slower than exponential for small n)
Quick Decision Guide
When you see a constraint, think:
| Constraint | Think | Likely Approach |
|---|---|---|
| n ≤ 10 | Exponential/Factorial is fine | Backtracking, DFS with pruning |
| n ≤ 20 | 2^n is acceptable | Bitmask DP, subset enumeration |
| n ≤ 100 | O(n³) might work | Dynamic Programming |
| n ≤ 1,000 | O(n²) should work | Nested loops, DP |
| n ≤ 10,000 | O(n²) is risky | Optimized nested loops |
| n ≤ 100,000 | Need O(n log n) or better | Sorting, heap, binary search |
| n ≤ 1,000,000 | Need O(n) or O(n log n) | Hash map, two pointers, sliding window |
| n ≤ 10^9 | Need O(log n) or O(1) | Binary search, math formula |
Summary
Before coding, always:
- Check the constraint (the max value of n)
- Calculate how many operations your approach will make
- Compare to 10^8 - if you're over, optimize!
- Choose the right algorithm for the constraint
NOTE: The 10^8 Rule Is an Estimate, Different judges have different time limits and hardware. Some use 10^9 operations/second. When in doubt, aim for a safety margin.
This simple mental math can save you hours of debugging and rewriting code!
Happy coding! 🚀