F-22 was the first airplane designed to exploit the lessons of ACEVAL/AIMVAL. See and shoot the enemy before the enemy sees and shoots you.

Erm. That has been the adage of combat pilots since oh…. 1914.

The high-level thinking behind the F-22 is not revolutionary, no matter how much LM might want you to believe otherwise.

The problem is the pesky bugg@rs on the other side of a major war seem to have a habit of finding ways to counter all the pre-war thinking and doctrine [which was based on the last major war].

Did you read Viannet Riller test of Rafale on Defesanet?

Nope…

Should I?

BTW, I don’t think comparing aircraft turn rates at airshows is a valid method for comparing sustained turn rates.

+10!

If you want to genuinely compare aircraft that are closely matched, you need access to the flight test data.

A few numbers can be indicative, but are unlikely to tell the whole story.

For instance, I round on the F-35 for its abysmal headline sustained turn rate – which raises the potential for disaster if air combat does not evolve as envisaged by the USAF – but I do recognise that it can get close to that with 4 AAMs and its instantaneous turn rate probably isn’t too shabby at all.

[Albeit others like the Su-30MKI, Eurofighter and to a lesser extent the Rafale won’t be drastically affected by carrying 4 AAMs due to carriage placement… and obviously the PAK-FA and J-20 are like the F-22 in that they don’t really notice it.]

What you say is correct- if the engine-duct was a straight, uniform rectangle like the Su-27’s.

However, irrespective of flow separation resulting from oblique shocks, slightly viscous incompressible*fluid flow (i.e. air @ subsonic speeds) undergoes 3D rotational slip separation without the high pressure of potential flow. In the T-50’s engine-duct I have identified at least two low pressure gradient regions namely the convergent-divergent intake and the bulge where the MLG wheel stows for which the graphic depicting 3D turbulent flow is accurate and relevant.

The flow is highly unlikely to ‘reattach’ due to the curve and short length of the duct. I admit that there appear to be several ‘passive’ elements to minimise vortices, wake and boundary layer ingestion by the inlet- especially emanating from external transonic flow of the aerodynamic structure. These are the trough in the intake mouth, the bump/brake on the inner intake side wall, a ridged contour on both sides of the intake side walls, BL diverter cavity & spill gauze, but we are primarily concerned with flow that is downstream and has already been rendered subsonic.

Its adverse pressure gradient that causes separation and it doesn’t matter that the duct isn’t straight and uniformly rectangular.

Every civilian turbofan nacelle inlet has a (non-choked) throat behind which the area expands. They are designed to avoid significant flow separation.

With regards the graphic being relevant… it is a relatively sharp intrusion into the channel, assumes a very small radius of curvature with respect to the intrusion height and the intrusion occupies a significant percentage of the maximum cross-sectional area of the channel. That is not comparable to the geometry of the T-50 inlet.
With regards the graphic being accurate. You have no idea what they used as a mesh, as boundary conditions or for modelling turbulence, neither do I.

[assuming sensible engine design]

-Can do the Kessel run in less than 12 parsecs…

The aim is not to choke the intake so you won’t have shock induced BL separation on the diffuser.

If the aircraft is subsonic, then the engine capacity/requirement will determine the massflow into the inlet, avoiding choking.

If the aircraft is supersonic, then the intake ramps will determine the massflow into the inlet, avoiding choking.