DAX 'ALL' Function

[dicken]

Hi,
I have a question regarding the ALL function, in MS DOCS they have the construction,

Titel :=
SUMX ( ResellerSales_USD, ResellerSales_USD[SalesAmount_USD] )
    / SUMX ( ALL ( ResellerSales_USD ), ResellerSales_USD[SalesAmount_USD] )

But if you use ALL is returns a table of distinct values and removes filters so
eg SUMX( ALL( Table, Table Numbers),Table Numbers ) where the numbers are 1, 2,, 2, 3, 4 would see 1,2,3,4 so I don't see how this can
be correct? What am I missing?

Richard.

 

[GraH]

It's pretty logical isn't it? Why would you want/need duplicated values in a specific column returned for a filter context?

Still somewhat confused.
"ALL(Table) returns all of the values in the table". All of the values <> all of the distinct values.
I never paid much attention to it since I hardly need to use ALL at all. Seems it is time to practice some more.

 

[RoryA]

I think the point is that the situations where you would use ALL(Table) you would typically just want all the records in the table (and creating a unique set based on all columns could be very computationally expensive). When you use ALL(column) you would typically only need a unique list because you'd normally be using the result as a filter table.

 

[dicken]

It's pretty logical isn't it? Why would you want/need duplicated values in a specific column returned for a filter context?

Well yes, but if you just put ALL (Table) , you do
Below two accounts of item b unit 5 .

This really just started as wanting a measure SUM( units ) / SUMX(ALL( Table), Table units, and I put
Table [ Units ] into ALL and of course did not get the correct result.

Richard.

 

[RoryA]

My point was about the ALL(column) syntax. You would not typically want or need duplicated values back from that.

 

[dicken]

My point was about the ALL(column) syntax. You would not typically want or need duplicated values back from that.

True , but it just got me thinking.

Richard.

 

[RoryA]

Never do that with DAX; you'll go mad.