I have noticed one interest thing from all the three methods ( My old method, new method , Chris's method ), i got the same output.

ID ...

Thanks Chris !

I am not offending you, i am asking this question politely.

Jeff's Sample code:

--===== Right way to do a "Joined" update

UPDATE TableA

SET SomeTableAColumn =...

I think i have followed the new join method.

Can you highlight where i did mistake ?

As you suggested, i have tested both the scripts with SET STATISTICAL TIME ON.

update #t1

set #t1.Name = T.Name

from T inner join #t1

on #t1.ID =T.ID

Execution Time 0.

SQL Server cpu time:...

After modifying Name column as Name = convert(varchar(50),'') , i have executed my modified code.

select distinct ID, Name =Convert(varchar(50),'')

into #t1

from T

update #t1

set #t1.Name = T.Name

from T inner join #t1

on #t1.ID...

Jeff,

Thanks for highlighting this area.

I have executed my previous code.

select distinct ID, Name =''

into #t1

from T

update #t1

set #t1.Name =...

In order for it to be "legal", TableA must also be in the FROM clause or the huge performance problem could take place. General rule to follow on joined...

Jeff,

Sorry for the delay !

create table T

(

ID int,

Name varchar(50)

)

insert into T

select 1,'John'

union all

select 2,'Mary'

union all

select 1,'Jerry'

union all

select 2,'Karthik'

union all

select 3,'Kumar'

union all

select 3,'Raghu'

select distinct ID, Name = CONVERT(varchar(50),'')

into #t1

from T

Here, i...

Because i saw more than 4 answer for this post so far.

which one is good when we think performance ?

Jeff,

how about my code ?

Wildcat,

Thanks.

Select distinct id, Name = ''

into #t1

from Data

update #t1

set Name = Name

from Data

where #t1.id = Data.id

Post your error message.

use distinct keyword