ANSI settings are different for that user than the others. Probably, but not definately, ANSI_NULLS.

That sure sounds like homework.

Generally, for homework type questions, most of us ask that you post what you've tried to do and where you're having problems. That way we're helping...

Not for OUTER JOINS though. With...

You know, either way works for me, just put...

Christian Buettner (7/11/2008)

---

Hi Grant,

what makes you sure this time that it does not make a difference?

Did you get this information from Microsoft ?

I am pretty sure that I had examples...

Well, eliminating all the excess JOINS does improve performance threefold in this example, so I think you can reasonably expect the something similar with larger data sets.

Because of the OR...

I've got nothing. The examples I thought I had, I can't find.

My heartfelt apologies for putting out bad information. For an INNER JOIN, it really makes no difference in terms...

I probably won't have the example, assuming I can find it, until tomorrow. I'm looking though.

I did find this in the BOL

This sequence can sometimes influence the results of the...

I'm going to start digging, but I've had instances where moving filtration from the WHERE clause to the ON clause does change the execution plan. I don't have the example...

There's an "Anti-Rbar Alliance"?

How do you join? How much are the dues? Does the initiation involve sheep or goats?

You got it essentially right. SQL Server 2000 support ended... in May I think. That's main stream support. You can still pay to get access to PSS. The same thing...

I sure don't mean to start a fight with my next statement....

Please don't do that. Learn to use the ANSI joins. There are a ton of good reasons. I'll give...

Hmmm... I just ran the test and received no difference between truncate and delete.

I think you're looking for something like this:

select c.date

from spdqty as c

inner join (select date, vialsOUT from spdqty) as a

on c.date = a.date

inner join (select date, vialsIN from spdqty)...

So, you've got two tables that have three different and distinct relationships between the tables? I'm pretty sure that violates one of the rules of normalization, but I'm not sure...