Is this a way to evaluate phase group delay ?

[Tepeix]

Just randomly, trying to compare some phase i get this.
is this a way to check if a system have linear group delay ?

It seams to works but i'm not 100% sure.. Also the graphic must be not in a log view to see a linear function, but it make more hard to clearly see witch frequency will be the more phase distorted..

[Tepeix]

Hum, some while after i discover that the module of the precedent post is a little stupid..
If i understand well, the bode plot i use was in log mode.
To see if the phase is linear we could switch to linear mode which help to see if there's a curvature;)


But now here's a virtual linear delay representation.
(i hope it works and present reality and not something else?)

I plan to use it to try to implement a 2th order allpass interpolation.
(One thing i learn about this is that the 2th order could correct the phase of some frequency, but at the price of making worse other's. (not 100% sure of this but almost) generally the 2th order will make the <1,x/2 nyquist better but >1,x/2 nyquist worse.. ...EDIT Thinking more about this, it's maybe wrong, the phase will be less linear at the end but each frequency will have less additional delay ?)

So providing a fractional delay it give the virtually perfect (i hope) phase response and make possible to compare with a true delay phase response.