Thanks!

I needed to pad anything with less than 7 digits because the check function I had to use returned different results on padded versus unpadded.

Yes, the format for the...

Lutz,

The data was only in float because that is the format it is imported in.

Craig,

The ddl was transact SQL I think? It's sql server 2005.

Thanks again for all...

The funtion I needed was

UPDATE registrations

SET idnummod = stuff ('0000000', (8 - len(idnumber)), len(idnumber), idnumber)

FROM registrations

WHERE len(idnumber) < 7

After converting the float, I was able to get

UPDATE registrations

SET...

the data type is float. I am sure it's the actual line.

I ran a function on over 800,000 rows and only these two numbers has this weird aspect to them,...

Ok thanks. I guess I can run both on some test batches and check the times. You're probably right.

Either way, this is going to be running at like 4...

the runnable version without that talbe is below.

declare @TestData varchar(12);

set @TestData = '2545996';

SELECT

right((10 -((convert (int, (substring(@TestData, 1, 1))) +

convert (int, (substring ((convert(varchar(12), (2* convert (int, (substring(@TestData, 2, 1)))))), 1,...

Thank you for the response. I had not thought of just joining on no conditions. My worry is that, because I'm working with around 1 million rows of data, that...