Using VALUES + ALL instead of just ALL. Resolution included - Need to know WHY it works

[prill22ta]

Hi there, 

I recently ran into the following issue. My team has since resolved it (included) but we don't know 'why' it is the resolution.

The ask: Calculate a Distinct Count of Values where the Primary Key  IN  a list of PK's defined by a subquery. 
The strange thing is, that when we ran the subquery through DAX Studio by itself, it returned the expected results. 
Thank you for taking a look!

The DAX that WORKED: 

VAR _subq =

CALCULATETABLE (

    VALUES(  table[field] ),

    ALL( table ),

    table[field2] = "Yes"

)

VAR _count = CALCULATE (

    DISTINCTCOUNT( table[field] ),

 table[field] IN _subq)

RETURN

_count

 
The DAX that DID NOT WORK: 

VAR _subq =

CALCULATETABLE (

    ALL(  table[field] ),

    table[field2] = "Yes"

)

VAR _count = CALCULATE (

    DISTINCTCOUNT( table[field] ),

 table[field] IN _subq)

RETURN

_count

[v-jialluo-msft · Accepted solution]

^{Hi @prill22ta ,}

^{Using a single column argument, ALL returns all the unique values of the column.}

CALCULATETABLE (
    ALL(  table[field] ),
    table[field2] = "Yes"
)

^{So this statement returns the unique value of the table[filed] column, and the following table[filed2]="Yes" does not play a filtering role for the previous statement, because the ALL function ignores all filters}

ALL – DAX Guide

^{Best Regards,}

^{Gallen Luo}

^{If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.}

[prill22ta]

Both versions are a measure.

[FreemanZ]

For the first dax code, is it a measure? what filter context does it have?