Since the contribution means in welldefined sense a breakthrough in the understanding of TGD counterparts of scattering amplitudes, it is useful to summarize the basic results deduced above as a polished answer to a Facebook question.
There are two diagrammatics: Feynman diagrammatics and twistor diagrammatics.
 Virtual state is an auxiliary mathematical notion related to Feynman diagrammatics coding for the perturbation theory. Virtual particles in Feynman diagrammatics are offmassshell.
 In standard twistor diagrammatics one obtains counterparts of loop diagrams. Loops are replaced with diagrams in which particles in general have complex fourmomenta, which however lightlike: onmassshell in this sense. BCFW recursion formula provides a powerful tool to calculate the loop corrections recursively.
 Grassmannian approach in which Grassmannians Gr(k,n) consisting of kplanes in nD space are in a central role, gives additional insights to the calculation and hints about the possible interpretation.
 There are two problems. The twistor counterparts of nonplanar diagrams are not yet understood and physical particles are not massless in 4D sense.
In TGD framework twistor approach generalizes.
 Massless particles in 8D sense can be massive in 4D sense so that one can describe also massive particles. If loop diagrams are not present, also the problems produced by nonplanarity disappear.
 There are no loop diagrams radiative corrections vanish. ZEO does not allow to define them and they would spoil the number theoretical vision, which allows only scattering amplitudes, which are rational functions of data about external particles. Coupling constant evolution  something very real  is now discrete and dictated to a high degree by number theoretical constraints.
 This is nice but in conflict with unitarity if momenta are 4D. But momenta are 8D in M^{8} picture (and satisfy quaternionicity as an additional constraint) and the problem disappears! There is single pole at zero mass but in 8D sense and also manyparticle states have vanishing mass in 8D sense: this gives all the cuts in 4D mass squared for all manyparticle state. For manyparticle states not satisfying this condition scattering rates vanish: these states do not exist in any operational sense! This is certainly the most significant new discovery in the recent contribution.
BCFW recursion formula for the calculation of amplitudes trivializes and one obtains only tree diagrams. No recursion is needed. A finite number of steps are needed for the calculation and these steps are wellunderstood at least in 4D case  even I might be able to calculate them in Grassmannian approach!
 To calculate the amplitudes one must be able to explicitly formulate the twistorialization in 8D case for amplitudes. I have made explicit proposals but have no clear understanding yet. In fact, BCFW makes sense also in higher dimensions unlike Grassmannian approach and it might be that the one can calculate the tree diagrams in TGD framework using 8D BCFW at M^{8} level and then transform the results to M^{4}× CP_{2}.
What I said above does yet contain anything about Grassmannians.
 The mysterious Grassmannians Gr(k,n) might have a beautiful interpretation in TGD: they could correspond at M^{8} level to reduced WCWs which is a highly natural notion at M^{4}× CP_{2} level obtained by fixing the numbers of external particles in diagrams and performing number theoretical discretization for the spacetime surface in terms of cognitive representation consisting of a finite number of spacetime points.
Besides Grassmannians also other flag manifolds  having Kähler structure and maximal symmetries and thus having structure of homogenous space G/H  can be considered and might be associated with the dynamical symmetries as remnants of supersymplectic isometries of WCW.
 Grassmannian residue integration is somewhat frustrating procedure: it gives the amplitude as a sum of contributions from a finite number of residues. Why this work when outcome is given by something at finite number of points of Grassmannian?!
In M^{8} picture in TGD cognitive representations at spacetime level as finite sets of points of spacetime determining it completely as zero locus of real or imaginary part of octonionic polynomial would actually give WCW coordinates of the spacetime surface in finite resolution.
The residue integrals in twistor diagrams would be the manner to realize quantum classical correspondence by associating a spacetime surface to a given scattering amplitude by fixing the cognitive representation determining it. This would also give the scattering amplitude.
Cognitive representation would be highly unique: perhaps modulo the action of Galois group of extension of rationals. Symmetry breaking for Galois representation would give rise to supersymmetry breaking. The interpretation of supersymmetry would be however different: manyfermion states created by fermionic oscillator operators at partonic 2surface give rise to a representation of supersymmetry in TGD sense.
For background see the chapter The Recent View about Twistorialization in TGD Framework or the article with the same title.
