4 minutes ago, wintertime said:

One unusual test is just a single measurement and does not imply that the whole of a complex system (DX in this case) is fast or slow.

You are completely misunderstanding.
Please read again the fact#1:  "1. As of now, just as of now, My PC, My test,at this moment, DX11 is faster than DX12 in a scenario where less(below 15) draw call with mass polygons."

This is a crystal fact.

7 minutes ago, wintertime said:

with ms you can better measure time single systems take,

single systems take. Of course I agree.   

wintertime, no offense I really want to know there is actual sample that runs DX11 faster in the scenario.
If you have 1 please let me know.
If you have not or do not know, also let me know.
My conclusion can't be finished :p

Thanks.

I was just pointing out you should try looking outside the box and not focus on a single flawed microbenchmark.

Newer API got the potential to be better, but it all depends on the application (how its using it).

I'm not interested in giving you another microbenchmark I don't believe in, even though you sadly still insist.

2 minutes ago, wintertime said:

I'm not interested in giving you another microbenchmark I don't believe in,

"I don't believe in" This tells me everything.

I had a little hope that you can answer, but it seems not.

As I expected. sadly. 

Thanks.

I meant you believe in them too much, but whatever, bye.

What do profiling tools tell you? (like PIX, GpuView, Visual Studio Gprahics Tools or your GPU IHVs tools.. )

Some guesses: PSO caching, DX11 driver implicit optimisations, wrong PSO settings....Or just simply the fact that with one draw call you are spending more time settings in a single thread mode things like root signature, descriptor heap, pso, etc, which costs are not amortised in the entire frame..