IT's an idiotic article, for one thing Ferrari ran this design last year with the partial split with the mgu-h in between the compressor and turbine.
https://www.youtube.com/watch?v=jgKqE7mVikM
SO honda didn't take a different approach with that alone, also they managed to miss the biggest advantage of the split turbo it enabled Mercedes to offset most of the problems with having a bigger compressor and turbine, their compressor is massive compared to the other two engines, it means the turbo did more, the compressor was more effective, the engine was more efficient, it harvested more power and it used that power more effectively.
I haven't looked too deeply in to it but Ferrari had the worst engine and Honda have most closely replicated their partial split turbo design. I've heard, though have no idea if true, that Ferrari have moved to something closer to Renaults design with the mgu-h sitting in the V of the engine also like Merc for more efficient mgu-h power. Ferrari and Renault likely want a full split turbo but can't necessarily change enough of the engine in with around 30 tokens to get there in one step. It's quite possible, if not probable that Renault/Ferrari have made changes with moving to a full split turbo next year in mind.
EDIT:- a scarbs analysis thing says the Ferrari has reverted to the compressor/turbine together with mgu-h as in the picture, however that certainly isn't what was being used last year as he specifically stated they'd changed.
A major issue on the Ferrari was the turbine was too small, considering it's the furthest back piece if that was one of the reasons the turbine was smaller. Being a bit further forward puts more a bigger area to make it bigger. Also saying they are getting it much hotter, again last year they had a cooler running engine but that was largely because the turbine was smaller, generating less power. More heat, more power, more mgu-h harvesting. The small split turbo but at the back of the car may well turn out to be the single worst design. Ferrari clearly had the worst design last year.
I wonder if in both cases it was a failed Merc design turned into a salvaged design. Ferrari specifically stated they tried the full split turbo like Merc but couldn't solve the vibration/engineering issue of the extra long shaft between turbine/compressor. They described what they did as a small split turbo. I wonder if they built the entire engine assuming they could sort it, then found when they couldn't because everything else was done and due to lack of time they effectively had to bolt it all on the back because they had no other choice.
Now, did Honda run into the same mistake. We'd been led to believe all year they changed to make a split turbo with everyone believing it to be the best.. with the advantage of being able to start off with that design while Ferrari/Renault might not have the tokens to switch to it... being touted for months). Then like Ferrari simply fail to make that work and at some point running into also having to make the same compromise Ferrari did. I hope they intended to do it and whatever Ferrari did wrong to make that design not work doesn't kill the Honda as well. But Ferrari have gone away from that and improved drastically... not sure that bodes well if that is really the way Honda have gone. Even then, Ferrari's intercooler was much lower, the higher radiators on the Merc seems like an odd choice.
It's been stated all over the place that the long shaft connecting the split turbo on the Merc took something like a full year to crack, with a decent amount of cash spent on that one thing, it has obvious benefits and two teams have ended up with something that feels like a failed attempt at it. At no stage through the past 12 months has it sounded like Honda were planning to do a half split turbo in the Ferrari style, not even a hint of it.
