No, no, reading these data in list directed input as you proposed is the perfect thing to do. In fact, explicit input formatting does not make much sense to me - this is a thing of the past - except maybe in some cases with character strings.
I would be very thankful, if someone could explain how the above numbers should be formatted in a text file, so we would be able to read them with the above formatted read statement.
But if you want to use explicit input formats then
(1) make sure you specify the field width with a bigger number than actual digits are present in the inputfile. Then Fortran adapts the field width to the numbers in the file.
(2) Set the number of fractional digits to 0. Fortran adapts the number of decimal digits to the actual position oof the dot. But if there is no dot present in the input the results can be confusing. Then Fortran sets the number of decimal digits to the number given in the format.
12345 in a format f10.3 would read as 12.345
12345. in a format f10.3 would read as 12345.0
1.2345 in the same format would read 1.2345
(3) You should place the format in inner brackets so that fortran repeats this format as often as items are in the inputlist.
So the format would be
1004 format ((f20.0))
or including it in the read statement
read (5, '((f20.0))') (AC(I),I=1,NUCAR)
But having this done and having spent some brain you achieved what you get more easily avoiding a lot of possible pitfalls by coding
read (5, *) (AC(I),I=1,NUCAR)
Norbert
The optimist believes we live in the best of all possible worlds - the pessimist fears this might be true.
