Using the same QP, P-frame seems better than I-frame?

[Ahlucard]

I found that in the same QP, P-frame looks more clear than I-frame .
I really wanna know how it happening.

[MatLz]

As I am a newbie and as english is not my native language, maybe I miss something....but is not simply because it's not the same frame?

[Ahlucard]

English is not my native language too, maybe i didn't express my point clearly.
And i have do a expriment that i code the same sequence:
one sequence using QP = 35, and code as I I I I I I I I I I....
another sequence using QP = 35, code as I P P P P P P P....
Then i found that the second sequence seems batter than the first one.
And i wanna know why?

[G_M_C]

English is not my native language too, maybe i didn't express my point clearly.
And i have do a expriment that i code the same sequence:
one sequence using QP = 35, and code as I I I I I I I I I I....
another sequence using QP = 35, code as I P P P P P P P....
Then i found that the second sequence seems batter than the first one.
And i wanna know why?

I think you have to read up on how I-P-B frames reference each other, and how that referencing comes to a lower average QP.

(I-P-P-P-P @ qp 16-40-40-40-40 is AVERAGE 35 too)

[Ahlucard]

(I-P-P-P-P @ qp 16-40-40-40-40 is AVERAGE 35 too)

I mean I-P-P-P-P @ qp 35-35-35-35-35
batter than I-I-I-I-I @ qp 35-35-35-35-35

Regardless the bitsrates of the sequence
and i say QP = 35 is not meaning their average QP, i just force every MB in every frame using QP = 35

[Raptus]

This is expected.
P-frames are more efficient because they are reference based and minimize the residual (coding error).