RE:RE:RE:RE:Jordaan on KEReport today, outstanding interview... I'm not an expert on ore sorters, either.
FWIW, my sense is ore sorters work better, to the degree that the gold consists of relatively-large nuggets, and less well when the gold is in tiny particles uniformly spread throughout the ore.
What follows is my naive understanding of the state of sorters. This is not investment advice; the reader should do their own due diligence.
The sorters detect gold by the metal's ability to affect a magnetic field, and by radiography, and perhaps by other technology (gravity?). Each of those works better on larger particles of gold.
What might not be as obvious is that throughput is markedly different, depending on the particle size. If the particles are tiny, you have to run the conveyor belt much more slowly, and have the detector heads closer to the belt. When the particles are very small, this can result in processing just a few tons per hour, on a given sorter.
Lastly, the loss ratio goes up as the particles get smaller, and in fact there's an interplay between throughput tons/hour and percent of gold that gets send to the reject pile. If you want very high recovery, you have to run the sorter more slowly.
To the best of my understanding (which ain't much) those are the reasons sorting isn't working out for Novo. I do believe there are other mining applications where sorting does work.