Couple of comments;
- AVL is proposing a concentrator to separate magnetite on site, with non-mag fractions containing (when fresh) sulphides including, viz. the base metal and also the hydrothermal auriferous sulphides
- The majority of the magnetite ore body is (at least partly) oxidised, meaning the non-mag tails would contain oxidised or partly oxidised sulphides that could be difficult to float
- Non-float portions of the tails could, when oxidised, contain free Au recoverable by a gravity circuit
So, this is an argument about fractions, or splitting a block of cheese into various bins based on overlapping Venn diagrams and Pie Charts.
At a 7.6:1 strip, this involves digging 276.7Mt of rock, of which 32.1Mt is massive magnetite.
There's not a lot of massive magnetite, by mass, in the total dig. eg, if you're talking about vanadium, the 3Mtpa of ore dug is 32.1Mt of the massive magnetite. This is what goes through the vanadium processing plant, and what ends up in the tails based on that throughput of 3.2Mt.
Mass pull through the mag separation process defines the tails that will go through to the supposed gravity circuit. AVL quotes mass recovery of between 52-62% (AVL 16th 03 2020). This means that the tails mass is ~1.2Mtpa which would go to the tails for Au recovery via gravity, and to the float plant for base metals. If this is oxidised, the base metal float is kaput. Oxide and transitional magnetite has lower recovery nett vanadium, and mass pull, so more mass needs to go through the plant to get the gold out the back end.
In terms of vanadium recoveries AVL expects, the pie chart looks like this;
This means that of the total dig across the 20 year mine plan, about 12Mt of tails from the magnetic separation gets produced.
if this has free Au, great. But what's the grade? Not 64g/t. And this assumes a uniform grade, which is clearly not the case from the re-assaying to date. So we have to consider, of that 12Mt of tails generated from the massive magnetite that's put through AVL's mill, what is the amount that is mineralised with Au? Considering the total number of pulps assayed is 464 out of >30,000m of drilling, so it's a very small percentage of the total population.
The logitudinal section (04/04/2020) shows fault damage zones where AVL has not assessed any of the mineralisation as resources. This would nominally show the areas of complex geology or dykes.
This is less than 5% of the 277Mt, and hence, ~5% of the 32Mt Reserves which will be thrown through the CMB plant. 5% of 12Mt is 0.6Mt. 
If you consider the grade at 64g/t, great, that's 1.2M ounces! Woohoo! At 1.5g/t, it's tiddles at 28K Oz, over 20 years or 1400Oz. At 0.5g/t it's 475 ounces per annum.
Now, is it worth it? What's a 1.2Mtpa gravity circuit going to set you back? If the total winnings is 28K ounces, or $40M, the pay back on this over 20 years would be iffy I'd think. If you get 0.5g/t, fauhgedaboudit.
Ipso facto, fuhgedaboudit
(20min delay)