New Minelab Manticore

[Steve Herschbach]

32 minutes ago, Chase Goldman said:

Cal thanks for asking this because I knew it was going to come up eventually and here we are 4+ years later trying to put to bed a huge misconception that Minelab initiated with their original pseudo tech Equinox Multi-IQ marketing.  The 5 (now 6 on Equinox with the addition of 4 kHz)) single frequency set points have no relationship to the Multi-IQ SMF frequency spectrum for the various modes either in terms of frequency range or in the number of simultaneously transmitted frequencies.  Minelab continues to remain mum about which specific frequencies comprise the constituents of the SMF spectra and the absolute range covered.

As noted by Pimento previously, similar to Equinox and XP’s D2, it is likely that only two or three waveforms of different frequencies are being combined during transmission (certainly not 5, as many mistakenly believe).  But until we can hook a spectrum analyzer up to Molybdicore, we are not going to know and ML is probably not going to say.  

I will say this.  Just because FullMontycore is utilizing the standard 5 “Nox” single frequencies, doesn’t contradict the ML claim of an expanded SMF range or higher upper end frequency simply because the 5 SF settings have nothing to do with the SMF range, so Goldfield in Multi-IQ could still conceivably have a top end of 80 kHz and be heavily weighted to favor the target response from the 80kHz waveform.

 

It's all about reception and processing, not transmission per se. You have to transmit to set the buffet, but when you actually end up eating, you can pick and choose what to eat, and how much. It's easy to see what a detector transmits. It's very hard to see what is going on with the processing. I would posit that the processing algorithms have been "expanded" or improved.

[CPT_GhostLight]

I'm going to throw my own speculating into the mix and guess that the Manticore is positioned to replace the aged-out E-Trac and that there will also soon be another higher end machine to replace the soon to be aged-out CTX.

[Steve Herschbach]

18 minutes ago, CPT_GhostLight said:

I'm going to throw my own speculating into the mix and guess that the Manticore is positioned to replace the aged-out E-Trac and that there will also soon be another higher end machine to replace the soon to be aged-out CTX.

I don't see how that is speculation. E-Trac was discontinued. Manticore at same price replaces it. CTX continues to be flagship machine.... for now. Don't place many bets on it being any time soon however. Minelab flip flops between prospecting, and the rest. So what's up next? A Multi prospecting detector to replace the Gold Monster? Or the fabled GPZ 8000? More likely that instead of CTX replacement, but there is nothing that says they can't do the CTX thing next. It honestly just boils down to looking at internal sales figures, and replacing the weakest performer. They sell them until sales run out of steam, then replace with something else. As long as sales remain strong, it is foolish to replace a good seller, when something else is not selling as well. It's all about making money folks. Maybe Manticore finally undermines CTX sales enough that it ends up next on the chopping block.

[Daniel Tn]

Competition drives new models from manufacturers for sure.  You can almost guarantee one of the first new Minelabs will be on its way to Turkey to be taken apart to be reverse engineered. It didn't take them too long to make a competitor model to the Nox, and might not take long to do a competitor model to the Beast.  You can guarantee Minelab knows this and probably isn't letting the R&D and Engineer dept take too long of a breather after releasing the Beast.  

[Chase Goldman]

48 minutes ago, Daniel Tn said:

It didn't take them too long

Eh, 4+ years is kind of a long time…

But they did deliver, finally.

[Cal_Cobra]

1 hour ago, Geotech said:

Generally whatever you can do in time-domain system you can do in a frequency-domain system, just differently. You could determine target ferrousity in MIQ by looking at phase linearity through the response so I wouldn't be surprised to see that the Manticore does FE/CO with much the same waveform as MIQ. Also, it's difficult to mix simultaneous MF with sequential MF due to signal discontinuities that will screw up the channelization filters in the simultaneous side.

 

A breath of fresh air, thank you Carl 👍

Any speculation on this statement from Mark Lawrie when he said "the MultiEQ+ frequencies were expanded" ?

 

[Geotech]

19 minutes ago, Cal_Cobra said:

Any speculation on this statement from Mark Lawrie when he said "the MultiEQ+ frequencies were expanded" ?

If I read it right, someone posted that the single F modes were the same as the Eq. But the Eq MF modes use different frequencies than the SF modes (2.6, 7.8, 39) and I suspect that 2.6 is for ground analysis. If I were to guess then I would say it's expanded on the low end to favor deep silver. Maybe there's a new boosted 1.56kHz component. That would also explain the incompatible coil. But then, I would also expect to see a lower SF mode, maybe 2kHz.

[steveg]

17 hours ago, GB_Amateur said:

Hate to change the subject (although my smoke alarm did just go off -- better check my coils 😏).

Since no one here (who has gone public, anyway) has swung the Megawatt-Core, all we have besides mumbo-jumbo speak are specs (the ones ML has shown so far on their webpage) and a couple videos.  (BTW has anything else been uploaded since the Dutch one on Monday?)  What I'm wondering about is the graphics.  I feel a bit like I did when the Equinox (and Deus 2) came out, where people familiar with predecessors (FBS family and Deus 1 respectively) recognized things that us detector ignoramuses didn't.  On the Dutch video they seemed to indicate they were getting target shape information.  How accurate is that?  For example, does a coin on edge look oblong, round, or ??  Does an aluminum nail look different than an oblong pendant (both horizontally oriented)?

Also, is there ferrous-nonferrous info in that plot, and if so what does it say?  I didn't see any numbers so it's not quite what some of the FBS/FBS2 detectors reported, unless there is a configuration that hasn't been shown.

GB_Amateur,

I can take a speculative shot at this, having been a pretty long-time FBS user.  

I am going to guess that this X,Y screen is FE in the vertical (y-axis) and CO in the horizontal (x-axis).  The horizontal line across the screen is probably similar to the "12-line" on an E-Trac or CTX. 

What the 12-line means is that when moving from the Explorer series (where FE numbers were different for each different target), they were "normalized" for "good" targets to a "12" FE number with the E-Trac, and again on the CTX 3030.  For instance, on an Explorer, a nickel was like 11 (FE) - 06 (CO), a copper penny was 04-28, a dime 03-29, a quarter 01-29, etc.  Meanwhile, a nail might ID in at 25-29.  So, while the "CO" number of a nail may be the same as a dime or quarter, the FE number was the "giveaway."  It was MUCH higher than a coin's FE number.  THEREFORE, many folks would set up the machine so as to cue their tones off of the CO number (you could choose to cue your tones off of either -- the CO number or the FE number), and then set their "iron bias" at somewhere around 20 FE.  What that then means, is you are discriminating based on the FERROUS number, so any target with a FE ID of 20 or greater would be disc'd out.  The reason it was set around 20, is that no good targets air test with a FE number higher than 11 or 12 (nickel), BUT -- the interesting thing with Explorers is when bad/irony dirt would start to "screw with" the ID numbers, THE VAST MAJORITY of that effect was limited (through FBS wizardry) to the FERROUS side; the CO number stayed pretty steady, but the FE number would up-average on a deeper coin, in bad/irony dirt.  SO, that 03-29 air-test silver dime, down at say 8" deep, might read, on successive sweeps, 15-29, 13-28, 09-30, 16-28, 06-29, 05-28, 11-30.  The CO number would only vary by a digit or so either way, but the FE number might range all the way up into the teens.  THAT is why you had to set your "iron bias" (FE discrimination) up to at least high teens or around 20.  Make sense?  Meanwhile, on the 2D screen, where the FE-CO numbers represent an x,y coordinate, those "bouncing" FE numbers as I illustrated above, would create a similar "bouncing pattern" on the screen, if you watched the movement of the cursor.  Number guys (like myself) would watch how the NUMBERS bounce, but folks who ran the 2D smart screen, instead of the numbers screen, would watch how the cursor would "bounce," and they learned over time that certain locations of the cursor, and how the cursor would "bounce," on successive sweeps of the target, would indicate things about the type of target.

But then, along came the E-Trac, and Minelab felt it would be "easier" for users, instead of having to memorize TWO numbers for each target, and ALSO learn how the FE number would change, on deep coins, to instead just memorize CO numbers.  To allow this, they decided to "normalize" the FE number of any good target; they chose "12" as the number to "normalize" to.  So, even a deep coin was supposed to maintain a FE number very close to "12".  This was the case with most "good" targets (silver dollars, and half dollars, to some degree, would give lower FE numbers).  So, a copper penny that would read about 12-43 or so, a dime 12-44 to 12-45, silver dime maybe up to 12-46, and a quarter 12-46 to 12-47, would maintain that "12" FE number, even at depth -- maybe dropping to 11 or increasing to 13 at depth.   Meanwhile, moving over to the smart screen, the obvious change with the E-Trac, compared to the Explorers was that now, a good target should not display the "cursor bounce" like you'd see on an Explorer.  Instead, you would look for targets where the cursor stayed roughly "fixed" along the "12-line" (i.e. a horizontal line drawn through "12" on the y-axis, which is the FE axis).  KEEP IN MIND, though, that "bad" targets would have FE numbers NOT normalized to 12, so the FE number of a "bad" target will usually be higher -- i.e. low on the screen.  So, focusing on targets that fall along the 12-line, was a visual aid that helps imply what are "good" targets, and others not falling on the 12-line being generally junk.

NOW -- applying this, speculatively, to the Manticore...  If you look, you'll see a horizontal line through the screen, about 2/3 of the way up toward the top of the 2D coordinate system.  Let's call that the "12-line."  That's essentially what it is, as the videos I have seen paint a good target around that horizontal line.  So, it's essentially, very likely, a "12-line."  And, all that target trace is, instead of your cursor readout being "instantaneous only" -- i.e. where the cursor location at any moment, corresponds to the target ID AT THAT VERY MOMENT (as it was on the E-Trac, and the CTX also when not in "target trace mode") -- the target trace is simply a "memory" of cursor positions over the last roughly 5 to 10 seconds.  It plots the cursor as a pixel on the screen, and so if you do the "Minelab wiggle" over a target, you are getting an increasingly dense plot of multiple, successive cursor positions (ID readouts of the target).  And yes, those cursor positions will form a "shape."  But, you can't think of the shape as something like a ground-penetrating radar, where it is showing you the "outline" of the target.  Let me explain further...

Again, recall that what those cursor positions are, are plots of TARGET ID.  So, you can imagine that a shallow, easy-to-ID coin is going to read VERY CONSISTENT, in terms of ID, almost like an air test.  SO -- imagine a shallow penny, on a CTX, that is IDing at 12-43, or VERY close to that.  This means that a "plot" of every cursor position is going to be very small, and nearly "round."  The "round" is representing the very slight "x-axis" variation in the target (CO) and the very slight y-axis variation (FE).  Thus, you get a small, round plot of cursor dots in target trace (and, centered along the "12-line").  

BUT, imagine a 2" long, 1/2" diameter brass cartridge casing.  Sweeping across the "short" axis, will give a tight ID readout, similar to a coin.  But, turning 90 degrees, and sweeping the "long" axis, will result in a DIFFERENT ID, and probably less "tight" (more variation in the CO number).  So, you can imagine that a plot of all those IDs, as you rotate the target, would "elongate" the plot...narrow when sweeping in one direction, but "longer" -- i.e. a more variable range of IDs, and focused in a different place along the "12-line" when sweeping across the target in the other direction.  SO -- you get a "shape" that is more "oblong," as opposed to tight and round.  MEANWHILE, a nail that is "falsing" with a high tone at times, is going to give you an elongated trace vertically, as it GENERALLY gives a correct FE ID down well below the "12-line," but other times, when it's "false high-toning" and masquerading on that sweep as a "good" target, you'll get some cursor plots near the "12-line."  Again, that would result in an elongated target trace -- suggestive of a "non-coin" target, some plots near the 12-line, but most being farther down the screen, and thus "elongated.")  

SO -- while a coin should look "tight" and "round" (if no other object is nearby that the machine is detecting simultaneous to the coin), the "round" is not DIRECTLY proportional to the fact that the coin is round, but the INDIRECT relationship is there, because of the fact that a round non-ferrous object generally IDs consistently/accurately on a high-quality ID machine.  And thus, with little variation in any direction of the target ID (FE, OR CO), the plots of the ID will all be concentrated in a small, generally circular shape.  Meanwhile, an elongated target that IDs differently when swept in one direction, versus another, will give an elongated trace...make sense?

Sorry for being so long-winded, but that's it, in a nutshell, what I feel is likely going to be the case with Manticore.

Steve

[GB_Amateur]

45 minutes ago, steveg said:

I can take a speculative shot at this, having been a pretty long-time FBS user.

Thanks, steveg!  I understood a lot on the first reading and I'll reread after checking out more screen photos of the Mambocorp.  The ML Explorer (digital) readout seems more intuitive but I guess in practice the later normalized methods were more informative?  (Or is it the typical "dumbing down" for the least common denominator type of detectorist?)

Good stuff.

[canslawhero]

6 hours ago, Steve Herschbach said:

Equinox gave us ice cream, and now we are getting different flavors of ice cream. Manticore is a very nice detector, for sure, but as a pretty good operator I can honestly say I can use a Equinox, Legend, and Deus II, and get similar results, and Manticore is not going to change that. You want to make more and better finds, then put yourself on better locations, and put in more hours of detecting. Those two things will do far more to improve anyone's results, than the latest new whiz bang detector.

but I simply can't help myself being inexorably sucked in to the new 'Manhood' marketing hype....I WILL BUY ONE