Erik, you say: "The literature speaks at how great it is for a single coin and tiny targets". Yes, I'm sure the literature says that. I'm sure it will pick up coin-sized targets, just as any metal detector can

That's not the issue. The problem is that ...... yes it can pick up coins perhaps, but no, you will not be able to discipher anything in small enough pixels to be of any use to you. Ie.: you can't tell a square tab shape, from a round-coin shape, from a ring shape, from a bracelet shape, etc... All such smaller things will be one-pixel across.
My explorer says it's "great for single coins and tiny targets" too, but doesn't show shapes either. I think the advertisers took a little liberty to allow readers to be duped, by that statement, into thinking that if it works "great" for coins and tiny targets, and "if it shows shapes", then THEREFORE, "it must show shapes
down to tiny targets". I do not think that's the case. Yes you will hear tiny targets (just like any coin/relic machine), but no, the shape ID feature will only be of use for larger targets.
But in any case, I'm not so sure that machines that could show shapes down to tabs verse rings, verses foil globs, etc... will be of much good for m'ding. It would only be of use if targets were lying flat. If they even had the SLIGHTEST tilt, you could kiss meaningful information down the drain.
I knew a guy, back in the 1980s, who milked lots of silver coins out of a certain park. Naturally, since it was the "silver rush" days, he had cranked the disc, and passed all the tabs, foil, etc... One day, he got a hunch that there certainly MUST be lots of gold rings, good nickels, etc... that he'd been passing. So he made it his mission, to strip-mine every single signal out of a certain area, that had given him the most old coins. He lowered his disc. control to barely knocking out nails, and painstakingly dug every single beep, no matter how small. After dozens and dozens of hunts, over many many months, and hundreds and hundreds of holes, he did indeed find several gold rings, a few pieces of gold jewelry, and a few worthless orange buffalos and V's. He concluded that it just wasn't worth the time in the turf, and that his time would've been better spent, if jewelry were his goal, just trolling swimming beaches. But he made an odd discovery while he was on this experiment: since he was going slow, and making painstaking written records of all TIDs, depths, angles, etc... (thinking he was going to break the "ring code"), he noticed that whenever he got a gold ring with any sort of a crown, that they always tilted slightly towards the heavier crown end. In fact, some with a crown were vertical on edge. Only bands laid fairly flat. All others tilted.
So you can see that if a lot of gold rings are going to lay tilted in the ground, then shape-showing is going to be a hassle. And as for bracelets, how would the tangled shape of that, be any different than strips of can-slaw or whatever? Not meaning to be a kill-joy, but just bringing this up for conversation. What we REALLY need, is
NOT a machine that shows shape or conductivity, but rather: A machine that shows actual COMPOSITION. Ie.:
gold verses aluminum 