count ID trought time to display distinct ID on the lapse of time

[fran_perea]

Hi

I´m trying to biuld and aggregated app which ave space aggregating data by month. 

The important fields i load are:

-Sales, amount and date

The script is something like that:

OPs_tmp:

LOAD

    Date(MonthStart(Fecha),'YYYY MMM') as Month,

    Count("HASH") as Count_Hash,

    Sum(Amount) as Amount,

    Count(sales) as Ops  

FROM [PATH.qvd]

(qvd)

Group by 

    Month

OPERACIONES:

LOAD *, 

    Resident OPs_tmp;

Drop Table OPs_tmp;

But when I try to display by set analysis I don´t get the right amount of different ID I should.

I I do count(distinct Count_Hash) I get 585, if a do sum(Count_Hash)=9052363

The real value is 3393807

PS: I tried to do an AsOfTable following the HIC post, this will be usefull to do the accumulation of the sales. But no to do the count of the different ID through time

[Cascader]

 in OPs_tmp:

you wrote count("HASH") 

is HASH  a column? then not need for double qoute

[fran_perea]

Yes, I know. Thanks.

But what I´m trying to do is to get the number of different hashes/ID for each month, without having duplicates in different months. Is there any function or anything to achieve this on the load script in qlik?