慎用Number.toFixed() about blog

[gdutwyg]

你好，这个怎么解释呢？如果n一样，取较大的那个，但是实际是取较小的那个

[saradean90]

Thanks for the information.. https://www.mygeorgiasouthern.org/

[xiaokunxu]

 您好，您的来信已收到，我会尽快阅读且回复，谢谢！ 

[Kerry12342]

When I tried to use IEEE 754 MIPS single precision representation (1-bit sign, 8-bit exponent, 23-bit fraction) to store the floating point decimal into binary and then convert back to decimal, if the last 3 bits of the fraction field are recognized as guard, round, and sticky, the calculated result is 1.334, which would round down to 1.33 like it showed up in your example.

I haven't test it so this is just a thought but would creating an algorithm with threshold for deciding which precision representation to use (use a larger precision when needed) to store the decimal be another way to solve this problem?