Talgo technology has a very low-slung body and is unlikely to be workable for a classic compatible because the UK loading gauge is very narrow below the standard UK platform height of 914mm above top of rail. This also means a classic compatible will have a very large platform gap on UIC-gauge track if the entrances are below about 1m, and a much smaller gap if they are above 1m.
Talgo has a floor height of about 760mm arl, so if we rebuilt it for something on order of 915mm arl, we would gain 155mm of extra above rail height to work with, which will probably lift us above the very low slung parts of the lower structure gauge. Assuming am reading it correctly from an RSSB document.
I think that gets us far enough off the rail that we only take a small chunk out of the bottom corner of the passenger cabin, which depending on your interior layout might or might not be workable.
To be honest I think the only way to be sure would be a mockup.
The shorter articulation also potentially gains us a bit of width from reduced overthrows.
But I don't really have the tools to analyse that properly unfortunately.
It's not just step free access, it's a significant saving in dwell time at high speed platforms. Dwell time at classic platforms is less of an issue because the number of trains per platform per hour will be much less.
There are only two stations where dwell time has a significant impact.
Birmingham International and Old Oak Common.
The cost of an additional four platforms, giving you one in each direction at each of those stations, is a negligible fraction of the cost of the scheme. It would also have drastically improved the reliability of the system - which given that a collapse in the HS2 service would collapse the core of the entire rail network seems like a very good idea.
This constraint appears to have been artificially imposed to provide a justification for the stock choice.
In return for this saving in platform costs, a ~30% reduction in line capacity has been imposed, if figures for the Avelia Horizon are anything to go by.
(It is a clean sheet redesign of the TGV Duplex with shorter power cars and more, shorter vehicles to make use of the extra length, with an interior redesign to fix the interior volume problems caused by it inherently being a 90s design)
If you absolutely have to have a double deck train at an 1100mm-ish platform then I'm pretty sure it could be done by starting with the TGV design and raising the end vestibule floor with more steps down to the lower deck and fewer steps up to the upper deck. Some of the equipment in the space above the vestibule could be shifted to the space below it, and the vestibule in some cars could be lengthened to provide wheelchair spaces. This would still be TSI-compliant, as the entrance would be at the same height as compliant single-deck high speed trains. Or at least a lot less non-compliant than making a train 600mm higher!
TSI-compliance is less important for rolling stock than it is for infrastructure.
Someone who buys a noncompliant train knows they have one, but the point of a TSI compliant train is that it can go anywhere!
So providing extra head room on the line does not make the line noncompliant with TSIs, at least until you reach the maximum extension of the standard height train's pantograph!
If you raise the vestibule height you lose the flat through deck, which removes one of the big advantages of the TGV system.
The combination of GC gauge and the higher platforms is what causes the problem, and the platforms are the only one we can really change at this stage.
I would prefer 760mm, but 915mm would be a decent compromise, because with the extra height of GC over GB+ we can probably manage a TGV duplex with near flat boarding.
This only becomes an issue because of the combination, as I said.