Machines known for this problem:
Okuma LB3000 EX II-MY C950
Okuma LB3000 EX II M C1000 with SMW SLUX-2 snug on steady rest
Okuma Genos M560R-V with Tsudakoma RNA 200R (A-axis rotary table)
Okuma Genos M560-V-E with Tsudakoma RNA 200R (A-axis rotary table)

As the manual describes: When interpolating one or more linear axis simultaneously with a rotary axis be it C or A for lathe or machining center , the feedrate must be input in the unit of degrees/minute. As it turns out, in some radical situations when the angular movement is very small like 0,01 degrees and the linear movement is fairly large like 2 mm (and thus thus the angular feedrate has to be pretty low) the control differ from the above described behaviour , instead it interprets and carries out the input feedrate in a mm/minute fashion so the driving factor to the feedrate is the linear axis velocity and the rotation feedrate follows that motion. In these cases the actual tool feedrate is
magnitude lower than the desired one.

Now I'm pretty sure that this phenomena originates from a mathematical/physical limitation of the machines axis control system/motor encoder division numbers.
And here we are closing on my main problem that may fancy the minds of some of us who like to learn each day.
In an ideal world the proper solution for the problem is Inverse Time Feedrate command.
As it looks like today, there is no chance of getting that function in a short manner of time and i still have some parts that needs to be machined with the above mentioned machines.

After I developed a macro in Edgecam to calculate the proper feedrate for the moves when I use rotary axis it turns out I get beautiful feedrate output, tested it in some ridiculously radical situations when really low angle and large axial movement crated an ultra small feedrate and the output is correct. I also make use of G119 on the lathes so i can input mm/minute feedrate for the circular interpolations and the motion is perfect in circles. but I don't have the G175 command on the machining centers and that makes the output always so grainy
Anyway all thing works fine but I would really need to know what is exactly the threshold of the applicability of the angular feedrate in terms of relation to the simultaneously interpolated linear axes travel distance (wonder why I didn't us this sentence as the title ) so that I could test it in the post processing macro and output mm/min feedrate when needed.

I really Hope that someone have some good Idea on this