Jump to content

Audio Processing Software?


Recommended Posts

Good morning all! I saw a post yesterday where someone mentioned the possibility of linking the equinox to an audio analyzer to get a visual readout on target tones and I found the idea fascinating. I've never seen this done and was curious what the old hands here thought about the idea. I'm sure the engineers at minelab (or any other shop) configured the sound of their machines with the human brain in mind, so our wetware may already be the best analyzer there is. But I can't help but wondering if a visualization of the audio would provide interesting heretofore undetected differences in targets that appear to be mostly indistinguishable to a rookie ear, namely uniform can slaw and pulltabs. Anyone here have any idea? 

  • Like 7
Link to comment
Share on other sites


El Nino77 may be using online language translation software which can somewhat obscure the actual meaning of his post.

It is possible to use up to 4 separate WM08 wireless modules with one Equinox in order to “analyze” different tones with spectrum analyzers and oscilloscope type hardware and software. That is definitely interesting but not something I would do.

You might want to send El Nino77 a private message to clarify things.

As for distinguishing uniform can slaw, pull tabs and other low to mid range non-ferrous targets from desirable targets like US nickels, lead, brass and gold..........after thousands of hours using the Equinox it is sometimes very possible for me to hear subtle differences in those targets using the detect mode and pinpoint mode audio and number characteristics and to call them correctly before I dig them. I stress sometimes! Explaining those very subjective differences to another person is something else I would not do successfully.

  • Like 5
Link to comment
Share on other sites

Whether or not @EL NINO77 misspoke I still think it's an interesting concept, and hearing folks such as yourself explain that you can in fact sometimes hear a differentiating audio signal between similar looking targets leads me to believe that there are subtle differences that might be readily obvious with spectrum analysis, if not to an untrained ear. I think one immediate issue is that while the equinox can connect to headphones with Bluetooth, is not clear whether you could make a connection to say a computer or smartphone the same way. 

  • Like 1
Link to comment
Share on other sites

I still think that this idea of using Equinox's sound output for graphic analysis of sound is still current ...👍

I still think that this idea of using Equinox's sound output for graphic analysis of sound is still current ...

  Although I have no problem using classic really good analog detectors / Tesoro Mojave, Vista Gold Gain, and Golden Mask-Coin killer / .. I still saw ... and I see with digital detectors that there is a possibility to achieve some progress in additional information not only from from the point of view of the VDI number itself but also with the additional Graphic signaling signal ...
  I have several detectors that also have a graphically showing identification signal ... / WHITES Spectra V3, AKA Signum MFD, Rutus ALTER 71 v.2, Minelab Etrac .. for which this graph definitely helps in the detection of so-called good signals ..

Since Equinox does not have such a possibility of "graphic display of the signal ... one of the possibilities can be the use and display of audio in a 50-ton multitone to display this signal in a graphic sound analyzer.

...we can use the Whites Spectra V3 detector as a good example of the possibility of graphical signal display ...

288706835_november1iphone5S2019004.thumb.JPG.3272bcd14bf3c09869d76ae4da1d64d0.JPG

  • Like 4
Link to comment
Share on other sites

7 hours ago, EL NINO77 said:

we can use the Whites Spectra V3 detector as a good example of the possibility of graphical signal display

Is that a histogram of VDI's from the three frequencies?  I'd be thrilled with just a single histogram of the digital TID's in a single mode with the Equinox.  I try to do that in my head but we humans are so biased -- we see what we want to see....

Here is a mockup of what I'd like to see on the Eqx screen:

522142567_Screenshotat2021-03-17202530.png.091eb2f5c547bf4bda4505e4868cddbd.png

You wouldn't need multiple channel outputs or audio signal processing.  It could be done inside the Equinox now, possibly with a memory upgrade and of course some extra lines of code.  Do the FBS detectors already do this?

  • Like 3
Link to comment
Share on other sites

3 hours ago, GB_Amateur said:

Is that a histogram of VDI's from the three frequencies?  I'd be thrilled with just a single histogram of the digital TID's in a single mode with the Equinox.  I try to do that in my head but we humans are so biased -- we see what we want to see....

Here is a mockup of what I'd like to see on the Eqx screen:

522142567_Screenshotat2021-03-17202530.png.091eb2f5c547bf4bda4505e4868cddbd.png

You wouldn't need multiple channel outputs or audio signal processing.  It could be done inside the Equinox now, possibly with a memory upgrade and of course some extra lines of code.  Do the FBS detectors already do this?

I know the Whites let's you chose your own adventure by displaying a probability histogram of target ID but it is also not processing the target signal with Multi IQ.

Interesting, but honestly if my Equinox was showing me that on every target I'd be sending it in for repair.  The tone ID is either stable  (singular metal, symmetric shaped target) or unstable (multi targets - e.g., mixed coin spill or multi alloyed target - e.g. bottlecap, irregular shape - can slaw), and solid (button, ring, coin) or hollow/distorted (e.g. can slaw) audio that's all I need out of the audio - visual aids can't achieve those nuances or are merely redundant to the audio.  Plus it violates the KISS principle that is mostly embraced by the overall Equinox design (vs. the etrac and ctx).

But perhaps I am misunderstanding the concept.

  • Like 3
Link to comment
Share on other sites

Chase .... I think you wrote very well 3 good considerations, where the graphical display of the signal shows a possible movement of the display of VDI in the case of less clean signals, or mixed signals in the iron separation.
Using the IRON bias function with EQUINOX would of course allow you to change the characteristics of the graphic display of the signal.

Another thing is that the graphic representation of the Equinox sound signal would show a bit of the history of the signal calculation and so it would show -more-less it would be too Graphically Multiton ..

With Spectra, the graphical renderer helps a lot in such situations ..- when you get only partial color sounds of the signal, but the graphical display still shows sufficient representation of the signal in the non-ferrous zone ..
Then I decide to dig such signals ..

 

  • Like 1
Link to comment
Share on other sites

1 hour ago, EL NINO77 said:

Using the IRON bias function with EQUINOX would of course allow you to change the characteristics of the graphic display of the signal.

Good point, but again, the iron bias tool is crude enough such that fine tuning or tweaking it for every situation to make a dig decision using such a visual tool is akin to using sledgehammer to kill a mosquito.  I use coil control and Equinox's ability to lock-in on an iffy mixed signal solely via the audio.  I can integrate audio and my swing dynamics better than using a jumpy visual display.  It's the same reason I don't bother with the X-Y display on the Deus.  I can hear via the target audio when the distortions that show up on X-Y.  Willing to see how it might be implanted in practice but I would rather ML consider matrixed Fe-CO TID and zDisc like the FBS2 machines

 

1 hour ago, EL NINO77 said:

Another thing is that the graphic representation of the Equinox sound signal would show a bit of the history of the signal calculation and so it would show -more-less it would be too Graphically Multiton

Presumes Multi IQ has a time domain based history to display.  Seems to me that Multi IQ is processing so quickly that any graphical signal history info would either be really stale data or would be invalidated by changes in any number of real time variables including sweep angle across the target.

Anyway - at this point I think that as far as depth, target ID, and ability to handle hot ground, we may be at the limit or plateau of the merger of target signal processing tech and the induction balance target detection principle.  Such visual concepts above are like strapping a 4k monitor onto a crude 8-bit color video signal.  

One of a number if things needs to happen at this point.  

A new detection principle needs to be fielded that is as portable as induced magnetic field metal detectors but which renders the ground visually transparent.

Or

Artificial Intelligence and/or augmented reality technology needs to be better applied in the hobby to solve other problems such as providing a visual representation of your search pattern and specificall coil coverage ensuring you have completey covered every inch of ground at your hunt site, as an example.

 

  • Like 2
Link to comment
Share on other sites

10 hours ago, Chase Goldman said:

Interesting, but honestly if my Equinox was showing me that on every target I'd be sending it in for repair.  The tone ID is either stable  (singular metal, symmetric shaped target) or unstable (multi targets - e.g., mixed coin spill or multi alloyed target - e.g. bottlecap, irregular shape - can slaw), and solid (button, ring, coin) or hollow/distorted (e.g. can slaw) audio that's all I need out of the audio - visual aids can't achieve those nuances or are merely redundant to the audio.  Plus it violates the KISS principle that is mostly embraced by the overall Equinox design (vs. the etrac and ctx).

I don't see it as being on all the time.  (I agree, that would be confusing.)  If I got an iffy target and would like to investigate the dTID's then I'd turn it on and take some data.

As far as not needing it, well maybe your ear (brain) is better trained than mine.  (In fact, I'm sure it is.)  IMO the value of this isn't to determine a single target's properties (for lack of a better word) but rather determining if a masked target is worth digging.

I must live in the trashiest place in the country because all my 'worthy' sites (meaning potential for old coins -- my desired targets) are loaded with nails, wire, etc.  It's like the nail factory provided everyone with free topsoil after they had salted it with their rejects!  That's why for me it's all about unmasking, not figuring out the symmetry of a target.

  • Like 2
Link to comment
Share on other sites

1 hour ago, GB_Amateur said:

I must live in the trashiest place in the country because all my 'worthy' sites (meaning potential for old coins -- my desired targets) are loaded with nails, wire, etc.  It's like the nail factory provided everyone with free topsoil after they had salted it with their rejects!  That's why for me it's all about unmasking, not figuring out the symmetry of a target.

I honestly do not think technology can provide a magic bullet here, at least not as long as the detection principle being applied is induction balance. The reason being the detection principle is not that sophisticated and the variables in play are too numerous.

  • Like 2
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Similar Content

    • By water spider
      speculation please. what does the future hold? what advancements might happen? keith southern are you out there?
       
    • By mh9162013
      I got in a few hours of metal detecting yesteday with my AT Max, which included some time at a park and a few permissions (private homes). Nothing of note was found, although I continued to struggle with trying to find good targets in high-trash soil. Given how I'm using the Garrett AT Max, I know have two primary options for finding good targets (silver coins) in these types of conditions.
      First, get a smaller coil, like the 5x8.
      Second, start digging the trash targets to clear up the ground and reveal possible good targets that are being masked or otherwise "overshadowed" by all the bits of aluminum, nails and other garbage.
      The second approach is not a viable option for most places I hunt (parks and private permissions). Not only do I not have the time to implement that strategy, my body can't readily handle that much digging. Also, I'm pretty sure digging almost everything is bound to lead to the loss of any good graces I have with property owners and park maintenance crews.
      Ok, so that leaves the first option. But before I go that route, I have to concede the possibility of getting an Equinox. Based on my experience with my Vanquish, limited time on the Equinox 600 and experiences with my AT Max and Fisher F2, I'm confident that one of the advantages of getting an Equinox will be more stable VDIs and more accurate VDIs at depth. And right now, I think I can live with that.
      I understand that getting a solid signal (a good, repeatabe signal from both swings and in 2 directions) on a dime or quarter at 6+ inches in my mineralized soil isn't always realistic with the AT Max. But I know the AT Max is at least capable of getting a decent signal (a good, repeatable signal from at least 1 direction and in 1 swing).
      Put another way, I get how the AT Max may not get me the "dig me!" type of signal that an Equinox can, but I at least need it to get me the "take a closer look, please" signal.
      All of that to say that I'm thinking about how my AT Max's target separating ability and recovery speed limitations (using the stock coil) will compare to an Equinox 600 and a stock coil. I came to this realization when running the AT Max with only iron discrimination set at 35 resulted in information overload for me and notching out everything below 70 was likely leading me to completely miss "take a closer look, please" signals that might lead to silver coins, dimes or quarters.
      Therefore, I want to use Monte's Nail Board. I know it's not ideal, and I plan on using Steve's approach of using both the AT Max and Equinox 600 on real-world targets. But I think the Nail Board will offer quantitative data when comparing the AT Max and Equinox.I also plan on using it with my Fisher F2 and Vanquish 340 to help put things into perspective. So how do I go about doing this test? Here's my approach so far:
      Step 1: Create Monte's Nail Board and use it with a modern, clad dime and new nails.
      Step 2: For each of the 4 passes, I will give it a rating: Will Dig, Maybe Dig, Won't Dig.
      Step 3:  I will set the sensitivities at either 50% or the highest possible given EMI
      Step 4: I will run each machine with zero discrimination and with enough notching so that it's only going to sound on dimes and quarters (and maybe copper pennies).
      Step 5: For the AT Max, I will also test it with iron discrimination set to 35.
      Step 6 (maybe): Run the test with the AT Max using both its stock and 5x8 coils.
      So here's my first real question: what changes or additions would you all make to my current approach? 
      My second real question(s): what "base" setting should I use with the Equinox 600. I'm thinking Park 1 with recovery speed set at the highest setting (3?) and a small or moderate amount of iron bias. Should I also run some tests with the Equinox 600 in 4KHz mode?
      My third real question: would it be benefitical to modify Monte's Nail Board so that the nails are replaced by either clumps of aluminum  or maybe pulltabs? A lot of my hunting is in parks and yards that are often littered with more aluminum trash than iron trash.
      Any insight is appreciated. Thanks!
    • By Steve Herschbach
      The whole depth with single frequency VLF detectors thing in my opinion has been nothing but a red herring for decades. I have read a thousand posts from people wanting VLF detectors with "more depth". Yet VLF detectors maxed out for usable depth by at least 1990 if not before. I have not used any single frequency VLF metal detector since 1990 that got more depth on coins than my old Compass Gold Scanner Pro.

      The only real improvement we have seen and are still seeing is in the ability to find and correctly identify items that are masked by the ground itself or adjacent undesirable targets. There are an amazing number of targets in the ground at depths achievable by any decent detector made in the last 25 years, but that are being missed because they are improperly identified and ignored, or just completely masked and invisible. This is an area where the Minelab BBS and FBS detectors have excelled. They do not go deeper. They simply get more accurate discrimination at depths exceeding what most detectors achieve. Machines like the DEUS and a lot of other Euro machines are excelling not for the depth they get, but this ability to acquire and accurately identify targets at shallower depths that are missed by other detectors.

      If we had a detector that could simply see through everything and accurately identify coins to 10" the ground would light up with countless missed finds. I get a chuckle out of all the deep coins I see people talk about on the forums when the best detectors made can't accurately identify a dime past 5-6 inches in my soil. Anything deeper just gets called ferrous. There is huge room for improvement in metal detectors still not by getting more depth, but by simply finding shallower targets that have been missed by other detectors made up until now.
      How To Make Yourself Crazy!
      U.S. Versus Euro Style Detectors
    • By Tony
      From what I can gather, higher frequency VLF detectors are more suited for smaller gold but ground mineralisation may be something to factor in. Would there be a “better” frequency for nuggets 1 gram and above in heavy ground?
      I’m not too concerned if I miss sub gram nuggets if there is a better suited frequency.
      The old Garrett Groundhog circuitry was legendary in this country…..I think it was around the 15 kHz mark. Is this frequency range a good starting point or do I need to consider other things such as better ground balancing capabilities or Garrett’s extra coil voltage. 
      My Minelab PI units will be mainstay detectors but as mentioned in another post, I have ground littered in man made iron junk and the ground mineralisation is severe. There are plenty of nuggets in the 1 gram to 5 gram range (maybe bigger) but the iron signals are as dense as 5 per square metre 🤬
       Thanks for any ideas.
       
    • By water spider
      maybe we could have a multi frequency coil, that recieves a single frequency or selectable single frequency and effectively distorts and amplifies the single frequency resulting in frequency variants up and down, mimicking or creating smf
    • By Skullgolddiver
      After the good new I realized when tested a few days ago my machine after It drowned and I've succesfully reanimated It....
      Now the horrible gasket Is fighting to stay out of the housing against any kind of attempt😒.
      So I'm in the middle of a headache manutention session with scarce results.
      That's the Mood guys😑
       

×
×
  • Create New...