I've pretty well settled on double stud walls (12 " outside of sheating to outside of drywall), filled with dense pack cellulose, for the main level as the most economical way to get r40+ walls.  I've explored many ways of doing the unheated basement. Easi-Wall is apealing but far from cheap and has limited contractor support.  The foam shotcrete (scip?) approach is even more pricey and also has even more limited contractor choices & I worry about the strength of the relatively thin concrete.   ICF puts the insulation under my main level's walls & floor and therefore poses a significant thermal bridging issue.  What I'm now thinking is 8" poured concrete with the outer surface in line with the outer surface of the sheeting of the outer stud wall that bears the roof & ceiling loads,  4" of xPS on the inside of the concrete walls from top of footings to 4" above the concrete wall's top, with 4" Xps between slab & footings.  On top of the slab an inner 2x4 stud wall (filled with un-faced cellulose bats)  with the top of it's top plate aligned with the top of the foam will support the floor system witch will cantilever over the foam with the outer surface of the band joist essentially above the outer surface of the foam.  The inside of the band joist to foam will be spray foamed primarily for air sealing.  The subfloor will stop at the band joist except in doorways.  This will put 8" of cellulose between the band joist and sheating except where the studs are where it'll be 4.5" so the band joist will be about r30, the main floor walls r40 and the basement walls about r 30 and the slab r20.    The only thermal bridging will be at windows, doors, where the roof trusses cross the inner wall and any pipes through the walls or ceiling.  Extending the foam on the inside of the basement above the concrete and making the inner basement wall taller than the outer eliminates a significant thermal bridge.  Has anyone seen or done this?  What do you think of it?

I need a picture to follow the detailing of top of the interior foam & studwall to the band joist.

While I'm a big fan of cellulose in most applications, very leery of using it in any below grade app where there's even a remote chance of flooding or groundwater penetation, due to the extreme wicking capacity of cellulose. Rock wool is a safer choice here.

XPS at anything more than about 2" usually has a higher lifecycle greenhouse gas potential than the source-emissions of additional energy savings. Most of that is due to the very high greenhouse potential of the HFC245fa blowing agent. At equivalent-R EPS has more than an order of magnitude less greenhouse gas footprint, since much more benign pentane is used for the blowing agent with EPS (iso too). Where you can, it's generally greener to go with EPS than XPS, and it's not a subtle difference. But even there less is more, and unless you need the air-impermeable aspects to avoid condensation at the foam/fiber interface in winter at your foam/fiber R-ratio, going with 2" and a 2x6 or staggered stud studwall to get the same whole-wall R value.

How much/what type of foam did you have going under the slab? (Typically with R40-ish whole-wall R above grade it calls for R20-ish foundation walls, and R10 under slab to balance out the heat losses.

Some high-R contractors in cold weather states pour the top of the foundation wall with a step in it to provide a means of thermally breaking the foundation top. See figure 37 p. 41 (p.45, in .pdf pagination), and figure 50 on p.56 (p.60 in .pdf):

http://www.buildingscience.com/documents/reports/rr-0903-building-america-special-research-project-high-r-walls

As drawn both of those show the interior studs as the structural studwall, but it could be either. When using cellulose in the double-studwall in a stepped thermally broken foundation it's important that there's a good capillary break (EPDM roofing or something) between the concrete and cellulose fill.

If you did something similar, and made the exterior studwall 2x6" with 2" cantilevered off the foundation you could use a minimal ICF or 2" of exterior EPS/XPS to thermally break the exterior at least down to the frost line.

Dana1,
Thank you!!
Definitely will incorporate a strip of EDPM membrane roofing over the concrete, up the 4" step to the top of foam over the foam and the inner basement stud wall. Floor trusses rest on the inner basement stud wall but extend beyond it so that the outer surface of the rim joist is directly above the outer surface of the foam. From what you've said I should use 2" ISO (foil side out ?) on the insides of the concrete walls stick with 2x4 stud wall & fill it's stud cavities with un- faced rock wool. This will, no doubt save some money & still exceed r20. The under slab foam I was thinking of is 2 layers of 2" 100 PSI rated XPS. This is to minimize deformation. The under slab foam would extend over the footings to the concrete wall and the ISO would start up the concrete wall atop the under-slab foam and serve as an expansion joint. From your statement that only r10 is needed under the slab I should drop to 1 2" layer, again saving some money. I looked at the site you referenced and they put the rim joist over the outside of the foundation with sub floor going all the way to the outside. I have moved the rim joist so that it's bottom outer corner is 4" above the inner surface of the foundation (basement concrete) to remove the thermal bridge of the sub floor and joist. The complication is having to build a taller By 4' + joist height + sub floor thickness- sill plate) exterior stud wall & erect it over the edge of the sub floor. BTW the 4" of basement wall between the top of the concrete and bottom of the floor above is over r 50 but that's the price of eliminating some thermal bridging.

Dana1,
Actually figure 50 is close to what I'm planning except it shows larsen truss walls supported by the inner wall which has serious thermal bridging to the concrete. What I'm proposing is like that above the basement but with the concrete under the outer wall and the inner wall & floor system supported by an added interior basement stud wall with foam between the concrete and stud basement walls.

Why not run the EDPM over the concrete, up the outer surface of the foam continuing up the outside of the band joist and over the subfloor, under the inside wall's bottom plate for an air seal, possibly eliminating the spray foam inside the band joist?