There is no step in Optimizer that will convert a subset into a fully embedded font as far as I am aware.
What you could do if you're working on the machine that has all fonts installed is unembed all fonts through Optimizer. Then fully embed the fonts.
Are you saying that it should be the case that is there are multiple subsets embedd in a PDF, that leaving just one subset embedded, Optimizer should then fully embed this font automatically? This is not the case. Would this be achievable - what would happen if I particular font could only be embedded as a subset due to licencing restriction...
Leaving only one subset embedded would, as you correclty point out, not necessarily fix the problem.
As it happens I am currently experimenting with one such PDF. I will let you know how I get on with it.
Since I am only working on someone else's advice I am just a theorist here, but no I was not suggesting that Optimizer can convert a subset into a full set automatically. I was just puzzled by your remark "remove the offending subset" since I could not see how I could identify the difference between an offending subset and a non offending subset that could be left.
So as I now understand it, I need to use optimizer to remove all subsets and then get optimizer to fully embed all fonts.
Interesting that a person I am supporting at the moment asked me this very question an hour ago. I'll try to get hold of the PDFs he's stuck with and give it a go myself on my new Pro 6.01 installation.
I'll look out for your post in this thread and lets hope this is all useful to the other participants in this thread.
Cheers
Ian