I’m new to Quick Sight and I’m trying to understand the difference between the “Float” and “Fixed” data types, especially in relation to SPICE datasets.
According to these articles:
It appears that the “Fixed” data type uses a decimal format with up to 4 decimal places (Decimal(18,4)). This is suitable for exact mathematical operations but can lead to rounding during data ingestion and when performing calculations. On the other hand, the “Float” data type offers about 16 significant digits of precision, supporting both large and small numbers with up to 15 digits after the decimal point.
However, when I create a calculated column using the “Float” data type, it seems to round values to only 4 decimal places, similar to the “Fixed” data type. So far, I haven’t noticed a significant difference between them, either in terms of visualization or storage.
Am I missing something? Is there a way to adjust the behavior of these data types, like setting the maximum number of decimals for “Fixed” columns?
Also, is there any advantage to using “Fixed” over “Float” in scenarios where I don’t need exact precision?
Hi @Pradip and welcome to the Quick Sight community!
From my experience, while the dataset preview on a ‘float’ data type may only show up to 4 decimals, when utilizing in your analysis, you have the ability to customize the decimal places to your desired amount and the numbers will populate accurately (more than 4).
Are you noticing that your field marked as ‘float’ is showing the 4 decimal places in the dataset view or when you utilize in an analysis?
Hi @Brett, thanks for the reply!
I have noticed in the dataset view. Will try that in the analysis and get back to you with my new findings, thanks again!