Honestly what would be really awesome if some Japanese documentary maker made a deep dive documentary series about MSX, with interviews with all involved parties (not just Mr. Nishi). Engineers and managers from ASCII, Yamaha, Panasonic, Konami, etc.
That would be phenomenal.
For sure opinions from different parties would give us a clear overview. Anyway, you could ask Nishi yourselves in Barcelona meeting, now that attendance is confirmed 
:
https://twitter.com/nishikazuhiko/status/1605759351267815424...
Now i ask you: what were the alternatives made in japan in terms of VDP? None. They choosed what was available.
This is precisely one of arguments I said: Since Yamaha's version was not ready, there was no alternative made in japan. 
the evidence is that the v9978 specs are exactly the same of v9990 specs. There is no single row that mention some kind of compatibility with v9958
There was no documentation of V9978. Only a vague little description. It's normal (and common) that it only describes what's new.
Now i ask you: what were the alternatives made in japan in terms of VDP? None. They choosed what was available.
This is precisely one of arguments I said: Since Yamaha's version was not ready, there was no alternative made in japan.
I agree that there were no alternatives, but i think for msx1 nishi not even tryed to seek for a new VDP staying to TMS.
Strangely, the MSX parternship with yamaha started with MSX2 and the need of a more performant VDP.
This is proved by the fact that MSX1 is made with all standard, available on the market, components.
Z80, TMS VDP, General Instruments, PPI.... There is nothing custom. Some MSX do not have even a MSX-Engine....
It is clear what the original philosophy was and also how this changed on MSX2
the evidence is that the v9978 specs are exactly the same of v9990 specs. There is no single row that mention some kind of compatibility with v9958
There was no documentation of V9978. Only a vague little description. It's normal (and common) that it only describes what's new.
No, if you read the documentation you see that IS EXACTLY the same we read on V9990, same words, same points. Very weird, too much weird. And if this chip was intended as an upgrade of existing vdps with compatibility it's at least inusual that it nowhere say "compatible with v9958'.
Instead, there is no single evidence that proves that this chip existed, even as a prototype (as a V9990 with V9958 compatible core). This is only an urban legend caused by the fact that yamaha changed the model number from v9978 to v9990. It is common when a project was under development, maybe the V9978 signature was an internal or non definitive model number.
The V9958 docs clearly say that is almost fully compatible with v9938, highlighing only what is changed.
On the V9990 there was no need to say that is compatible or not, because it was clear simply by reading tech specs.
The V9990 IS the V9978 "renumbered".
To me it does sound like nishi had required the v9958 compatibility on the same chip, but yamaha refused to do it because of tech reasons, maybe proposing a new MSX-ENGINE with v9958 inside. After things broke, the V9978 was renamed to V9990...
Maybe also that they changed the model number to highlight the fact that this is a VDP that has nothing to do with predecessors. And even from a tech point of view it would be extremely complex to achieve compatibility. they never tryed to achieve this. it is more likely that a MSX3 would have got v9958 integrated into a new MSX-ENGINE instead.
I think Nishi already told that:
- MSX3 would be R800 + 9978 and also some more features (I think R800 DMA would be used too)
- TURBO-R was just a way to try to not lose all the money invested, waiting any longer for 9978 would cause R800 to no longer be that hot and it was what they could do to try to keep MSX afloat, Nishi didn't like calling it msx3 because that wasn't what they thought for it
-Yamaha asked ASCII permission to use parts of THE cancelled 9978 design in a new chip so they could at least try to recover some of their own investment
So, it is fair to assume 9978 would have 9958 features + 9990, MSX was always backwards compatible and 9990 alone doesn't have text modes and emulating 9958 is something unthinkable for that era. 9990 doesn't have any 9958 legacy because it didn't make sense to pay royalties for that legacy that is based on TI property for a design no longer being target for an MSX (they sold to whoever was interested, seems it was part of some PC video cards and of some Pachinko or arcade machines, not sure). Also, on a MSX3 it doesn't make much sense to have a 9958 + 9990 (separate chips) as this would cost quite a bit more, you would need vram for both, you would need some kind of video switching or to compose images in the same output, a really expensive proposition.
I think this is the first time that I have ever heard someone thinking that 9978 would be only 9990 / new features and have no legacy.
Note: I fully agree that 9958 + 9990 on a single chip is a challenge and most likely the reason 9978 never got out, probably ASCII had a contract not much favorable for them in terms of delivery date and delays so it would be cheaper, and Yamaha took advantage of that to have resources allocated to more lucrative business (I.e.: sega), this is surely speculation but it is the only way to understand those delays, ASCII not pocketing money from Yamaha because of it and Turbo-R being born with 9958 