Pulse Induction & VLF In One Detector. Don't Piss On My Dream.

[Digalicious]

What kind of weight would be looking at with a PI / VLF combo?

How about a modular upgradable detector? Initial price for the bare bones detector would be very cheap. It wouldn't come with a coil or any technology in the pod. Basically, it would be a shaft of your choice along with the screen.

The manufacturer would make money by selling modular and upgradeable components such as PI coils, VLF coils, as well as switchable technology pods to convert the detector to a PI or VLF. 



 

[phrunt]

3 hours ago, Geotech said:

At White's I designed an MXT+SMPI circuit, got the TX, coil, and preamp working but then abandoned the project for the far more promising truncated half sine. With MXT+SMPI you get one or the other but not both at the same time. With half sine you get both at the same time, and the option to do multifrequency/multipulse. I got the half sine project running and that's when I left White's. No one ever continued the project.

So, you're saying Garrett has it sitting in their storage room in among the boxes marked "whites old stuff" and they could continue on with the project? It is something someone else could take over or is it something that would be near impossible without your involvement? 

Would the code you have written be able to run on today's hardware or would it need redone due to obsolete components? 

I don't understand why if this was in existence the project would not be resumed, as it does sound like a very beneficial product that would sell very well.  

A PI/VLF combo both are the same time would be an absolute weapon.  If Minelab for example was able to offer a GPX 6000 with a Manticore built in, I'd have my dream machine, although I'd really like a GPZ with Target ID.  It opens up a whole new world of use for a detector having discrimination, but not the simple discrimination of the early GPX series, real discrimination of a VLF.

The Algoforce target ID's are amazing to me, such a surprise it took so long for a PI to come to market with it.

So, what makes the half sine design so special over the current market offerings? What are the benefits of it over the detectors we currently have? I know very little to nothing about it, well before my time when this was taking place, a real shame it didn't hit the market though.

[EL NINO77]

I think that Garrett ATX is one of the first to use the VLF mod/even if it is very basic/ it depends on the discrimination..., at least I really believe that...
Definitely the next generation..it can go further...
As for the hybrid detector, the Aka Smart pulse detector has been on the market for a while, which is a hybrid PI-VLF detector...

but first of all, ask yourself a question.. what do you expect from such a hybrid PI-VLF detector?

[Steve Herschbach]

5 hours ago, Sourdough Scott said:

Since we are thinking outside the box.... Why couldn't a detector also incorporate some AI into it? Then based on what signals are processed it can determine whether you should dig the target or not. 

Most overrated idea ever. You can already use the AI between your ears to accept or reject items based on the very minimal information that an engineer like Carl really has to work with. AI does not change the fact that for every good target there are countless trash targets that look exactly the same to the metal detector. AI can’t change the underlying physics. You will get people who will start marketing “AI assisted” detectors, but it’s just not what people think it is. People should do like Carl and use oscilloscopes etc to see what the detector sees, and then this idea that AI is going to somehow “learn” from you as you tell it what it is finding will get put to rest. No, for every trash item you tell the detector about and it finally decides to reject, you will miss good items also. It won’t do any better than the best of us at that, and probably worse, as there will always be more trash than good finds. Want to get rid of them? Reject all targets. That’s where an AI bias against digging trash ends up. AI can only work with the information it has and at the electronic level there simply is not this massive amount of information people seem to think there is. Detectors are half blind people groping in the dark guessing at half felt items.

The problem with half sine is patents and no action from Garrett. Anyone can do a PI / VLF either or detector, and I’m sure many of us would be happy with a machine that can hunt in all metal PI mode, then switch to VLF disc mode once target found.

[phrunt]

11 minutes ago, EL NINO77 said:

I think that Garrett ATX is one of the first to use the VLF mod/even if it is very basic/ it depends on the discrimination..., at least I really believe that...
Definitely the next generation..it can go further...
As for the hybrid detector, the Aka Smart pulse detector has been on the market for a while, which is a hybrid PI-VLF detector...

but first of all, ask yourself a question.. what do you expect from such a hybrid PI-VLF detector?

Target ID, VLF type discrimination and more sensitivity to smaller targets than a PI offers, the 6000 has nothing on a VLF on small gold, it's more equal to a Gold Bug Pro than a higher frequency unit, but even then, a GBP can pick up a #9 lead pellet at an inch or so, the 6000 can't see one at all so perhaps its more like a 13kHz detector, or maybe lower.

I didn't know about the AKA, thanks for pointing it out I'll take a look.

[Digalicious]

19 minutes ago, Steve Herschbach said:

Most overrated idea ever. AI does not change the fact that for every good target there are countless trash targets that look exactly the same to the metal detector. AI can’t change the underlying physics.  

I gave up trying to explain that to a few who keep claiming that detectors need AI to determine trash from treasure. Some just have a fundamental misunderstanding of how AI works...to the point that they think AI has some sort of "magical" ability.

[Sourdough Scott]

Most overrated idea ever. You can already use the AI between your ears. Steve H.  🤣 

I have another brilliant idea 💡.  How about everyone has a dedicated digger person named Mongo that digs all targets for you. Then us egotistical detectorists can do what we do best. No AI, discrimination,  no 2 in one detectors needed. Just Scott & Mongo. 

[EL NINO77]

23 minutes ago, phrunt said:

Target ID, VLF type discrimination and more sensitivity to smaller targets than a PI offers, the 6000 has nothing on a VLF on small gold, it's more equal to a Gold Bug Pro than a higher frequency unit, but even then, a GBP can pick up a #9 lead pellet at an inch or so, the 6000 can't see one at all so perhaps its more like a 13kHz detector, or maybe lower.

I didn't know about the AKA, thanks for pointing it out I'll take a look.

Simon, I mean that the VLF discrimination mode ATX..was used more for the discrimination of iron than for the identification of a non-ferrous target..especially when we assume that we are working in a more difficult terrain..but it was still a good idea...
with the current development of the ability of VLF detectors to work in mineralization, this can translate into really improved possibilities of such discrimination.... I think there is a strong area of possibilities as to how it can be done...

[EL NINO77]

I definitely believe more in the natural intelligence of the designers, engineers and field testers of metal detectors... who model the detection properties of a given detector according to their ideas... so that the detector works as best as it should work...

artificial intelligence needs a sufficient amount of information for its reliable work... so that the result is really good,,, and such amount of information is often used in error detection...

among other things, you need a really good computing power... because at home I sometimes use a de noise program with AI when editing and processing photos... so I know very well the time it takes to process one photo even if the computer works with a powerful graphics card... which is very accelerates photo processing...

the processor itself needs 7-8 minutes to process 1 photo..!!!
But the computer processor in cooperation with the graphics card can do it in 1 minute....

. A designer...or a tester can work well even with one achieved information or result...and can really develop it sufficiently further...if this result turns out to be the right way to further improve the detection...or the analysis of the detected signal ...

Here I believe, for example, in better electronics of the detector, and the technical design of the detector, which can also provide the prerequisites for a better input signal from the target for the detector, which can process it better than it was before

[Steve Herschbach]

7 minutes ago, Sourdough Scott said:

Most overrated idea ever. You can already use the AI between your ears. Steve H.  🤣 

I have another brilliant idea 💡.  How about everyone has a dedicated digger person named Mongo that digs all targets for you. Then us egotistical detectorists can do what we do best. No AI, discrimination,  no 2 in one detectors needed. Just Scott & Mongo. 

That is the best idea ever. Digger kids? Detector caddy boy? I want one!