Fighting with chat gpt.

I've asked you five times to define the sample size for pass fail testing of mirror bonds with 90 percent confidence and 98 percent reliability with no failures, but been given a wildly different answer each time. It should be 114. Try again and show your work.

=========
Chat gpt responded:

To determine the sample size for pass-fail testing of mirror bonds with a 90% confidence level and 98% reliability, we can use the formula:

n = ln(1 - confidence level) / ln(reliability)

First, let's calculate the reliability:
Reliability = 98% = 0.98

Next, we calculate the confidence level:
Confidence Level = 90% = 0.90

Now, substitute these values into the formula:

n = ln(1 - 0.90) / ln(0.98)
n = ln(0.10) / ln(0.98)
n ≈ -2.3026 / -0.0202
n ≈ 114 (rounded to the nearest whole number)

Hence, the sample size for pass-fail testing of mirror bonds with 90% confidence and 98% reliability with no failures is 114.