As you can see if you have waded thru all 6 pages of this now 4-year-old thread, I've been thru a lot of changes with this 9-string guitar design.
I have had the new build for a year and a half.
As reported, I had serious challenges - surprise - merely with the physical strength required to play the thing.
This gives the lie to my old idea of "effortless playing" - ha! I have now spent the last year bulking up, doing calisthenics,changed my diet, gained a few kilos and strengthened arms, shoulders and psoas, & am back to a "normal" practice schedule and making progress. This whole process surprised me quite a bit. I had no idea. In order to take the strain off of my left wrist and bar muscle, which I injured repeatedly last year, I have to hold the guitar firmly with my right arm, which is something I intentionally trained out of my practice when I used to play six string. (Picking up a six-string now is like picking up a ukulele used to be.)
Having finally decided I was done with setting the action and the string diameters, I put my mind toward adjusting the intonation. Using a Seiko SAT800 tuner which has a cents readout, I made multiple fingerboard maps recording the variance in cents for every string on every fret, and averaged the results (there was a 2 to 4 cent variance in repeated measurements). At the twelfth fret, the G3 was the worst at about 12 cents sharp. (The A4 and F#1 were close enough to perfect, so there was no error in the basic saddle setback.) Some rough calculations indicated that the G3 would require about 2 mm more setback to correct. The existing saddle, however, was less than 2 mm wide. So I took a razor blade and chisel and widened the saddle slot to 4 mm and made a new saddle from scratch on the table saw from a cow femur (about 15 c by 4 mm), with individual (variable) setback notches for 7 of the strings. This took all day. The intonation at the 12th fret is now within 3 cents or so for all strings; only the A4 and F#1 are left uncompensated.
Another surprise: with the wider saddle the bass response was immediately enhanced. I can't quantify it, but it was very noticeable. Builders of ERGs might take note that a wider-than-usual saddle will permit necessary intonation adjustments and also improve the bass.
The area of the fingerboard above the twelfth fret still all runs a little sharp, probably no worse than any other guitar, but I am scratching my head about how to possibly quantify and correct this. I think that it is due to the extra stretch when pressing down the strings. If I can quantify it with confidence, I might possibly consider ripping the high frets out, filling and replacing with adjusted locations. This is a rather daunting prospect! so I will think about it for a _long_ time before I get into it.
There is also the usual problem with tempering the various octaves E3-E4, G3-G4, etc. between open and stopped strings in first position, but this is not too bad, and I have done no compensation at the nut. I will consider this carefully for a possible future adjustment.
I continue to be very happy with the design. The right angle fret at #3 seems optimal for first position work. As the treble strings are all less than 650 mm in length, it is very easy to "play requinto" in the upper registers, once having learned the necessary chord forms for the lute-tuning. The regular tuning of the bass strings in fourths makes it possible to "play bass" in a very idiomatic way, given that the four lowest strings duplicate the usual double bass tuning although a whole step higher. I don't think this would be possible with the "alto-guitar" design with diatonic basses. Although there was this question hanging out about whether the low F#1 might not have been better-sounding at 75 c, two factors weigh in: (1) I had enough trouble with my left hand and arm already; (2) the wider saddle made the F# string improve noticeably.
One further detail: when I was experimenting with string gauges, at one point I loaded up the treble strings with fat diameter strings, adding a few kilos (can't quantify) of tension. At this point the point of the angled bridge, where all of this tension was concentrated, began to dig into the top and make a serious and scary dimple there (in contrast to the belly-and-bulge pattern of an overloaded right-angle bridge). I took the heavier strings off PDQ. I don't know that this can be considered a weakness in the design, just another small surprise. Considering my left hand muscle difficulties, it is better I remain with the lightest gauge strings I can work with. Currently I have a rect. nylon .022" for the A4, and the rest about proportional, at about 6 kilos tension for each string.
I have read thru the Bach 2-part inventions many times, and even the 3-part ones. I will probably never play the 2-part ones at any acceptable performance level - this is a job for some young buck in his 30s who might take up this design, the new Elliot Fisk type, not me. But I have confirmed the possibility to my own satisfaction. I have a bunch of original pieces I am trying to record, poco a poco, and continue to extend my regular bread-and-butter duet repertory to incorporate the extended bass lines and more use of the A4 string. All of that legacy repertory I simply transferred intact from the 6 (and 7) string, without changing any fingerings and just ignoring the extra strings, to save the huge amount of work, but many of the sonorities are improved by gradually re-fingering onto other string sets. I haven't played any other instrument for more than about two minutes since I got this new build in April 2016, a year and a half ago.