DTP: "Handle Duplicated Record Key" vs "New Records Only"

[Former Member]

Ygdrassil uber alles, folks.

Using some field tests I tried to understand the difference between these checkmarks:

"Handle Duplicated Record Key" and "New Records Only".

Lets say we have characteristic with 3 initial values:

Experiment 01. Lets load it from PSA with such values:

Note: we have the same keys (Batch) but slightly different values of attribute Movement-Type.

For this experiment we use DTP with such settings:

And the Result-01 is:

Here we can see, that there was only replacement of records, having attribute difference.

Experiment 02.

Using this initial characteristic set (yellow values edited manually for creating difference):

and the same PSA request as in previous example:

we now will use DTP with the chackmark "New Records Only"

and the Result-02 is:

and here what we have in master data table:

Note, here we dont have D and I versions of records. Which means that we dont need Master Data Activation.

---

As you can see, we get only two differences:

1. Different logs of DTP loads, yellow messages for "New Records Only"

2. We dont need to activate master data for "New Records Only".

Which sometimes can fasten things... Process Chain for example.

I have read few other articales at this site, and information there wasnt confirmed!

For example I couldnt get the result when "New Records Only" gives red log errors for duplicates.

---

And here comes the Question. Is this the only practial difference between these options?

If so, this is somewhat redundant, dont you think?

We can always use first option, and forget the other?

Or, maybe, there are some more complicated examples, where New Records Only - is the only option?