I cannot comment on the GDI soot size issue; not researched it enough yet to understand the differences aside from port injection.
However, soot is generally WAY smaller than any normal filter is going to catch, at least when it starts out. So filtration is meaningless in that regard.
As the add-pack degrades over time, it becomes overwhelmed and soot starts to agglomerate (co-join) into ever-larger particles. This is not an event that happens like a light-switch; it's not "off" for XXXX miles and then suddenly comes "on" when the next mile passes. Rather, it's a slow, cumulative effect that probably has some form of parabolic response curve. Think of plaque in your arteries and you get the idea ...
This is one reason I constantly try to remind folks that while filtration is very important, it is NOT the controlling entity in wear control. Once a finite level of necessary filtration is established, "more" of it really does not have effect on the wear rate curve.
Additionally, the SAE study I often reference shows that as the OCI goes up, the wear rate goes down, despite the continued production of soot. This indicates that until the add-pack is usurped, the tribo-chemical barrier is probably the MAJOR player in wear reduction. You cannot conveniently avoid this very important portion of the overall wear topic.
Soot production in an engine is generally a constant based upon your specific engine and type of driving. You'll generate "X" grams of soot for every 1k miles you move. So just like Fe, soot tends to be cumulative with the OCI.
But this is not a reason for those who employ a short OCI to jump off the couch and scream for victory ... Not at all. Because of today's very clean running engines, and good quality oils, you can easily run LONG OCIs and still have good low wear. Those of us who actually practice what we preach, can show long OCIs with soot/insolubles still at acceptable numbers.
What folks need to understand is the wear comes from several sources, and each component has a job to do. Silica intrusion comes from the air ingestion, and so air filters are very important. Soot comes from incomplete combustion and starts out VERY small, so the dispersents and detergents are very important. Modern fuel systems help reduce soot generation. Metal wear particles that are large, as well as silicates that are large, are caught by the oil filter. The tribochemical barrier helps keep metal-on-metal contact to a minimum, and can almost make it non-existent, even at start-up. They all work together in concert to provide a low wear environment. .
Now, if GDI does indeed produce more soot, then it's likely to have effect on the OCI duration. But only testing will confirm the duration that would be safe; conjecture has no place for those who treasure factual basis for solid decisions.
