Comparative discrimination tests in multi-target situations...any interest?

[pyrogort]

Most detector-vs-detector performance comparisons I read about on this and similar sites are about depth of
detection. I would never refute that depth IS a very important factor to consider when comparing the various
machines but feel that discrimination characteristics are not given enough weight. If you are strictly an all-metal
hunter (relics, prospecting, etc.) the subject is moot and you can skip this post. Most of us here, however, I'll
wager are coin/jewelry hunters and the subject of discrimination differentiation is extremely important. Do you
really know what your unit is telling (or NOT telling) you when you turn it on, balance it, set your sensitivity and
discrimination then begin to hunt? Most likely all of us have buried coin targets in "test gardens" at various depths to see "how deep it will go", but have you ever tested your machine(s) discrimination personality using
a coin target IN CLOSE PROXIMITY TO a trash target? Years ago I came across a little booklet called "Metal
Detecting Tips & Techniques" by Larry Woolis. In it he suggested such a close proximity trash/treasure test that
was a real eye-opener for me. All detectors are NOT alike (or even close) in this critically important performance
aspect.
Consider the following scenario: Detector Jack & Detector Jill decide to both hunt a nice 8 ft. wide 60 ft. long
parkway. Both have equal skill/experience levels but different brand/models of machines in same price range.
They set their units up with roughly equal sensitivity and agree to set discrimination at just the point where
foil drops out. Jack begins first while Jill has a sandwich & makes a phone call. 20 minutes later Jill begins
her search at the same spot where Jack began. She doesn't expect to find much since Jack has just "cleaned
out" the grass strip in front of her with a machine of equal power, depth, and coil size as hers and has several nice coins to prove it. Halfway down the strip, Jill gets a nice "wonk-wonk" solid signal that repeats in all
directions and never crackles, clips, or breaks-up. "Huh..." she thinks, "Jack must have missed going over this
spot". She calls Jack over to check the spot with his machine. They take off their headphones so they can hear each other's machine's response. "Check this out" says Jill as she passes her coil over the spot. "Wonk-wonk-wonk". "Wow" says Jack..."I'm sure I didn't miss any ground coverage. Let me try it". He passes his coil over
the spot and........silence. They are baffled but of course turn to the final judge of any target, the digging tool.
Jill cuts a half circle plug and folds back the turf. She checks the plug with her pinpointer, gets a signal, scrapes
dirt from the plug bottom and uncovers a nice wad of foil. Then she runs the pinpointer over the plug again.....
nothing. She then sticks the pinpointer into the hole, gets another signal, takes one good scoop of dirt out and
there it is.....a dime. A silver dime. A 1916-D Mercury dime. Depth? About 4-1/2 inches. Jill is elated...Jack is
sick to his stomach. What happened? Why didn't Jack find it? Was it a depth advantage of Jill's machine? Not
at a 4-1/2 inch target depth. Sensitivity? Same as Jill's. Too much discrimination? Again, same as Jill's....besides,
no coin machine has a discriminator that can be turned high enough to reject silver dimes (what good would it
be?). No, the deciding factor here was DISCRIMINATION DIFFERENTIATION. The rejected foil target was too
close to the dime for Jack's machine to differentiate, so he got no signal. Jill's machine was much more effective
at "seeing" the dime close to the rejected foil and so produced a good signal. She even passed her coil back over
the wad of foil AFTER removing the dime and got no signal.....complete rejection.
Wouldn't you like to know if YOUR machine is like Jack or Jill's? Wouldn't you like to know how ALL of the
currently marketed coin/jewelry detectors would perform given the above scenario? There is a pretty easy way
for all of us on this list to find out. You'll NEVER see this kind of comparison in any of the treasure magazines
we like to read.....we must do it ourselves. Although none of us as individuals can own ALL of the various brands/models that are out there, COLLECTIVELY we do. If each person will test his/her machine(s) and report
the results on this forum, we can ALL have a terrific database that yields the discrimination differentiation
characteristics of pretty much every machine. The test is easy and requires no digging. If you're a coin/jewelry
hunter and use at least SOME discrimination this information is VITAL. Don't just ASSUME that because you're
only using minimal nail/foil discrimination that your machine will "sing-out" and tell you to dig that 1918 8over7
buffalo nickel at 4 inches. It may not (as Jack found out the hard way). A standardized database WILL tell you which machines will do it every time and which won't.
I know some readers of this post will pooh-pooh the idea with "what-if" subjects like coil size, VDI readings,
variable ground mineralizations at different parts of the country (world), operator experience, etc...etc...
But I'm convinced the minimal effort required to do it will yield a wealth of eye-opening information for all of
us serious coin hunters. I'll be glad to spell out the simple parameters of the test which only requires 1 coin
and 2 sticks of chewing gum......plus a little of your time.
So the original question remains........any interest?

 

Amazon Forum Fav 👍

A History of Mining on the Kenai Peninsula, Alaska - Grab it through Amazon!

[vanner1]

For sure I for one would be interested. It would be interesting to see to what extent Im missing good targets, and also how the detector is distorting the response to them

 

[U.K. Brian]

This has already been well covered and there's lots of material on YouTube but the terms to look for is recovery speed ie the time a detector takes to recovery from passing over a bad target before being able to register a nearby good one and target masking where even a tiny bit of metallic rubbish can mask a good find or at least skew the discrimination reading, tone or meter so that you don't dig.

There's also a number of detectors with good iron "see through" which will still pick up good targets even if they are lying under ferrous.

 

[Silver Searcher]

This has already been well covered and there's lots of material on YouTube but the terms to look for is recovery speed ie the time a detector takes to recovery from passing over a bad target before being able to register a nearby good one and target masking where even a tiny bit of metallic rubbish can mask a good find or at least skew the discrimination reading, tone or meter so that you don't dig.

There's also a number of detectors with good iron "see through" which will still pick up good targets even if they are lying under ferrous.

Hence the term dig everthing, especially on good sites.

SS

 

[The-Bone]

i'm in! i would love to see the results for various machines. i now own 4 units and bet they each give different results

 

[liftloop]

dude,

well written,and sounds good,all give info on all my detectors,So your saying the dime was on edge and not flat so jacks detectors ran it off as a pull tab and didn't beep to warn him of his treasure that laid bellow.

But if jack had a v.d.I.,and a threshold,he would of heard the break in his threshold,and looked at his v.d.i number and real lies there no pull tab has that number and he should of checked....I would of used an Indian head penny that's closer to a pull tab for an example.

especially when a targets hallow plays in to effect.I know that targets that you want to cancel out,you will hit on until you bring the target to the surface then your detector will reject it,all because of the good hallow that diapered .I can't see one detector keep discriminating every out.If you stopped at pull tab.

but that's me.

 

[pyrogort]

Alright......so there IS some interest in trying to build a database of the various detectors' discrimination
characteristics when faced with multiple trash/treasure targets. Great! Thanks Vanner 1, The-Bone, &
Lift Loop. Let's get it started.
First let me say in response to UK Brian that if I am unaware of lots of data covering this subject (disc.
recovery time, etc..) I apologize. I've just never seen it in print or on the net. I have watched a couple
of you tube videos where someone tests their unit over a coin with a trash target lying over it. Useful
& good to watch but we could glean so much more data if we agree on a standard test technique that
will be used by everyone and reported back here.
As for Silver Searcher's advice to "just dig everything" the question arises as to WHY someone who is
going to do so would buy a metal detector with discrimination in the first place.
My reason for starting this topic is to gain useful information for people who DO use discrimination when
hunting and do NOT intend to dig everything. They are coin/jewelry hunters who WILL be using at least
some minimal discrimination. They are the "target" audience (pun intended!).
There is no need to get complicated with this simple test and the resulting basic data. We will discuss
audio responses to the test "garden" only.....no VDI, no meter data, no fancy computer-controlled disc.
recovery time adjustments on-the-fly, no matching up of digital numbers to various targets, etc.......
JUST SOUND. Did the unit "beep" and tell you to dig it or not? Or was it unsure? For now we will confine
our trash level to FOIL only to keep the amount of work & data manageable. In the future, should we wish,
we can expand the trash data to pull-tabs, etc....but for now the foil data alone will raise enough eyebrows.

We will each make a data sheet on which to record our results. Across the top of the sheet from left to right
we will list 5 headings: "6 inch trash/target separation", "3 inch trash/target separation", "1 inch trash/target
separation", "0 inch trash/target separation", and finally "Negative trash/target separation".
Extending down the left side of the sheet we will list the Brand and Model of each unit we are testing along
with the coil size. We will assume all units are functioning properly, ground-balanced, and have the discriminator set to reject a FOIL-WRAPPED stick of chewing gum. Each unit MUST be adjusted individually to "zero-out"
the gum stick. Next we will find a section of ground with no metal targets or clear one out, if needed.
We will then place a DIME coin (silver or clad...makes no difference here) on this section of ground and place
a stick of foil-wrapped gum on both sides of the dime leaving 6 inches of open ground between the dime and
the gum. Just to be clear, this means you will have (from left to right) a stick of gum.....6 inches of open
ground......the dime......6 inches of open ground.....a stick of gum. You will now swing the machine being
tested in normal fashion back & forth, left & right (parallel) over this test bed and then rotate 90 degrees and
do the same sweeps perpendicular to the test bed. For recording the results, we will use 3 parameters only.
1) SRS. This means Solid Repeatable Signal. A good, solid hit that repeats in all directions with no breaking up.
(You would dig without question)
2) QS. This means Questionable Signal. A confusing signal that clips, breaks up, partially nulls, or doesn't
ring in from all directions. (You might dig it or you might not)
3) NS. This means No Signal. (You'd never know it was there)
You will then record ONE of these 3 data responses under the "6 inches" column for that particular detector
and coil. Do this for each machine you have available for testing. Next, repeat the test after moving the 2 gum
sticks so there is only 3 inches of open ground between them and the dime. Record the data under the "3 inches" column. Next, repeat the test after moving the 2 gum sticks so there is only 1 inch of open ground between them and the dime. Record the results under the "1 inch" column. Next, repeat the test after moving
the 2 gum sticks so that there is NO open ground space between them and the dime.....here the gum sticks
are actually touching both sides of the dime. Record the results under the "0 inches" column. Finally, repeat
the test after moving the 2 gum sticks so that they COMPLETELY COVER-UP the dime....meaning you can't
visually see the dime at all. Record the results under the "Negative inches" column.
When sweeping the coil perpendicular to the test bed, remember to pass directly over the dime as this is the
target we want to try to detect.
So now everyone GO FOR IT and report your results back here. I will have the data from 3 machines to report
on soon.

 

[vanner1]

Be an interesting test. Now, does anyone know a brand of chewing gum that each stick is wrapped in foil and not paper

 

[U.K. Brian]

One warning about You Tube. Lots is fixed. We had much moaning and wailing last year because one well known British detectorist died. Not from me. He specialised in showing various detectors by laying a length of wood on the ground and moving one of our hammered silver coins towards and away from a iron nail so you could judge how well or badly that detector would do when a good target was near a bad one.

Not a bad idea as the fact the plank he was sliding the items along was on the ground so there would be some ground effect and you could watch it all happening in front of your eyes so it must be true.
If fact it was all a con. Either the sweep speed or the discrimination level level used would be wrong or the recovery speed adjustment provided by the manufacturer would not be used. People were misled.

Any and everything can be faked and often is. Another example is iron on above or below a target where one detector apparently picks up the good and ignores the bad. They may place a small silver coin in the jaws of a pair of pliers and wave across the coil and the detector sounds off. They remove the coin and sweep the pliers across the coil and nothing. Great, its a design that can reject a good bulk of iron or steel yet still signal on a fraction of silver. Got to buy one of those ! Another fake unfortunately. What they do off screen is fractionally adjust the discrimination to the point where the pliers are just, just rejected. Then adding the extra bulk of even a small copper, silver or gold coin is sufficient to move the audio from silent reject to sounding off.

Silver Searcher is correct that really no discrimination should be used which doesn't mean everything is dug as the audio in all metal or minimum discrimination means much rubbish can be avoided. To cover more ground in a shorter period of time people do use discrimination but it promises something it still can't deliver so the less the better. Even if your on a beach and don't want to dig any rubbish you don't have to discriminate out everything just set so the rubbish items give a sound you can recognise as bad. (Then Silver Searcher will come along and pick up what you have still missed but that's life).

Other complications are detectors that "boost" their performance as discrimination is increased (which balances back any depth loss from using a higher level of discrimination) and those designs that are insensitive to silver foil. Though everyone seems to think that pulse machines pick up everything the principle they work on means insensitivity to silver foil and some don't pick it up at all and this is on machines with no discrimination adjustment.

I'm not saying that these tests are going to be a waste of time but the results are not going to be clear cut in any way.

 

[[email protected]]

No interest.

You see, I capitalize on following behind the herds of "top of the line" VDI toting, high discriminatory, cherry pickin' elitist's and would like to keep that my little secret.

Without those people (which are the majority) there would be very little left for me to find.

Man, the things I've seen in the ground would blow your mind! Millions of variables!

I wish you well on your test. I can't participate because it would give my competition an edge.

You're on the right track to success!

 

[pyrogort]

You are a shrewd fellow indeed Muddyhandz!
For sure having factual test data on the discrimination characteristics of the various brands of detectors
will strengthen your competition. I'm bettin' the majority of casual hunters will never see the data we're
attempting to compile here though, so not to worry.

Results for 3 machines tested today are as follows:

Fisher 1280-x Aquanaut: 6 inch spacing-SRS
10 inch coil 3 inch spacing-SRS
(factory original) 1 inch spacing-SRS
0 inch spacing-SRS
Negative spacing-SRS

Garrett GTI 2000: 6 inch spacing-SRS
9.5 inch coil 3 inch spacing-QS
(factory original) 1 inch spacing-QS
0 inch spacing-NS
Negative spacing-NS

Minelab Sovereign: 6 inch spacing-SRS
original model, circa 3 inch spacing-SRS
1992/3, 8 inch coil 1 inch spacing-SRS
(factory original) 0 inch spacing-SRS
Negative spacing-SRS

Remember, SRS=Solid Repeatable Signal, QS=Questionable Signal, NS=No signal. Also remember that
these test results apply to FOIL level discrimination only. Higher levels of disc. (ie. pulltab) will no doubt
yield different results & would require separate tests.
As you can see, the Fisher & Minelab units turned-in excellent results with solid "dig this now" signals in
all cases. Not so with the mighty Garrett GTI unit (maybe Detector Jack had one of these!). Perhaps the
more recent model GTI's (the 1500 & 2500) would perform better......hopefully one of readers will test
one and report back. Indeed, I'm hoping lots of folks will test their units and report the results back here.
Time will tell!

 

[gleaner1]

Comparative discrimination tests in multi-target situations are a mystery, wrapped in a riddle, stuffed deeeeeply inside an enigma. Very interesting stuff.

 

[U.K. Brian]

The test has already started to fail. The ability of a detector to scan deeper has been ignored (deeper in most cases also meaning that the coil is detecting wider and so is more sensitive to surrounding targets). This can be got round by an increase of the recovery speed of the detector but this is not an option with most U.S. detectors except a few Whites.

The Garrett has come out rather poorly against the other two machines but that could be because it hasn't a very strong bias against iron and other rubbish. The problem with detectors that do have a strong bias against unwanted metals is that they can reject them completely which sounds good but comes at the price of good targets being ignored along with the bad in the real world.
Clean rejection is not as good as you would think. The noiser detectors that may click or crackle over the rejected target still have audio that makes it plain if something is good or bad but have the advantage of recovering far faster from the bad thus being able to make good finds alongside bad.

 

[[email protected]]

The test has already started to fail. The ability of a detector to scan deeper has been ignored (deeper in most cases also meaning that the coil is detecting wider and so is more sensitive to surrounding targets). This can be got round by an increase of the recovery speed of the detector but this is not an option with most U.S. detectors except a few Whites.

The Garrett has come out rather poorly against the other two machines but that could be because it hasn't a very strong bias against iron and other rubbish. The problem with detectors that do have a strong bias against unwanted metals is that they can reject them completely which sounds good but comes at the price of good targets being ignored along with the bad in the real world.
Clean rejection is not as good as you would think. The noiser detectors that may click or crackle over the rejected target still have audio that makes it plain if something is good or bad but have the advantage of recovering far faster from the bad thus being able to make good finds alongside bad.

This is a good post Brian.

With the type of hunting that I do, the worst machine in my books is one that has a bias to certain "junk" metals. With all the variables in the ground, nulling any metal out is bad news!
This is why I still use a 1266X 20 years later. It is one of the nosiest machines ever made! I want to hear iron being rejected. That allows me to catch better targets hidden in a bed of rust. The recovery speed is lightning fast.

One of the things that I find utterly ridiculous is the VDI machines that lump big iron in the silver dollar range. I used to follow behind an XLT user (no offense Brian, it's in your list of machines) at construction sites and this guy always left his holes open. These sites are only a 100+ years old (city streets) and iron would not be anything desirable to dig. Half of this guys holes (deep craters) still had huge iron targets down deep and he would give up on them. One pass with my machine and I think to myself "why the hell would this guy dig this?"
The answer........it gave a reading as a silver dollar.

That's a technological improvement?

My friend bought one of those bounty hunter fisher's (I.D. Excel) and besides the hilarious turkey gobble signal for iron overload, he would often call me over with a high tone (high VDI number too) and say "here's a Queen Vicky silver quarter."
That always translated to a piece of tin in my head. I would give it a pass and say "sounds like dog sh*t!"
Sure enough most of the time it was tin. Actually I think he found more silver coins in the mid to low range than in the "silver range."

I noticed that several Tesoro's have a liking to iron and it can be difficult to tell the difference between ferrous and non-ferrous. This can be good for relic hunting but is sometimes annoying at other sites. They do have a great response to non-ferrous in the lower range and are sensitive.

Let's just say that the tin foil range is where I capitalize the most. Some detectors have a bias in that range. That's a problem.

A friend of mine who owns 3 different tesoro's, an ace 250, a ML sovereign, and a 1266x, did a series of backyard tests with a silver half dime and a large iron spike covering it.
The '66 did the best and believe it or not.... The Minelab was the worst! The cheapest tesoro (silver u-max) did better than the higher priced tesoro's. Heck even the Ace did better than the ML.
He spent a lot of time on these tests, so it wasn't a one time thing.

One alarm that goes off in my head is top of the line detectors that claim great depth and come with a 10 1/2 inch coil.
If it's such a deep seeking machine, then why doesn't it come with the standard 8 inch stock coil?

I say this because I've gone head to head against some of the big ticket machines that have larger coils and my machine with the 8 inch coil went deeper every time.

Hey people, use whatever works the best for you! I have no problem with that.

I do have a problem with those who think they know everything that is going on underneath their coil and claim they don't miss anything with their miracle machine.

Sorry, you're in fantasy land! I have seen areas so choked with metal that NO MACHINE could get every signal. That is why archies dig an entire square (small area) and pull out hundreds to thousands of metal artifacts.

Why do you think there are civil war pit diggers out there? Because they know the limits of metal detectors.

There's more than meets the eye in the ground! I have seen it for myself.

H.H.

 

[U.K. Brian]

I've got far more detectors than listed but not all get used except for specific tasks. Some I keep so that people can compare what they are told is the ultimate to highlight the faults or drawbacks.

I still like the XLT, not quite as good in performance as the older Spectrum and partly due to the lack of DD coils masks good finds with bad to the N'th degree but many don't know what they are missing and don't care. I also run it totally as not suggested in the instructions or in any of the "how to" books.

The problem is people accept it when they are told new, or more expensive, is always better and don't look into things closely enough thinking a magazine review has told them all they need to know. An example of this is two similar "beep and dig" machines at roughtly the same price. You might have reviews providing basic air or in-soil test so brand X is deeper then brand Y on a quarter but no attention is paid to the performance over the whole range of conductors. The less deep detector is often the one that works better over a whole range of targets rather than just one.

 

[vanner1]

OK ..here are my results for the AT PRO with the 11 inch DD standard coil .

6 INCH SPACING ..SRS
3INCH SPACING ..QS
1 INCH SPACING ..QS
0 SPACING ..NS
NEGATIVE SPACING ..NS

This is a surprise to me as I thought the AT would do better. When I get a chance will test the 8x5 coil. Im doing a pretty trashy site at the moment and what I heard outside with this test sounds like what i am hearing in the field,,,, how many good targets I am missing worries me, especially if other detectors are able to unmask so easily. We will see what the small coil brings .So come on guys ,its a simple test..

 

[vanner1]

Well I redid the test because something bugged me after posting yesterday, and it is this. I was so intent on making the foil disappear that I went past the AT Pro s best unmasking setting, which is 31/32 high resolution iron discrimination, as discribed by Thomas Dankowski in one of articles. Going to much past this setting can make the target masked again. So here are the results using 32 high resolution iron discrimination

6inch spacing ..SRS
3 inch spacing ..SRS
1 inch spacing ..SRS
0 spacing ..SRS
negative spacing ..SRS

 

[Silver Searcher]

I dig foil all the time, and coke.. I know it's one or the outher, but I like to make sure. But I don't miss anything good either

SS

 

[pyrogort]

Thanks Vanner-1.......now we can add the Garrett AT Pro to our list of machines & we'll all know just
a little bit more!
Keep 'em coming folks!

 

[deepskyal]

Being overly analytical of my machine kind of takes the fun out of the hobby for me. Time, practice and patience will give you all the information about your machine you'll ever need.
People are going to buy a machine based on what they find and what people recommend.

My last few detectors have so many variables and adjustments to make, one single test of discrimination is pretty moot. I'd have to adjust and re-adjust, gain, sensitivity, sweep speed, etc. to come to an average conclusion and I'm just not willing enough to spend that much time in my back yard when I can be out hunting.

With an older machine I was demonstrating to a few guys one time, I got the typical trash garbling response on my machine and told the guys with certainty it was a pull tab. I dug it just to prove it. Turned out to be 2 quarters laying side by side, one a 1964, the other a 1965.
Normally I would NOT have dug it.

Conversly, I've taken a brand new machine to a park and found 3 silver dimes in the first hour, and it was a tiny, "worked over" park that several guys hit before me.

You will never duplicate your home test responses to those in the field. Different soil, different metals other than foil and iron, etc.
Try a rusty, old bottle cap from a beer bottle and let me know how it masks. I dig wayyyy too many of em. Good, solid, repeatable....hate them suckers. Some of those old urban parks are thick with them. But a good sign that most detectorists have avoided the trasy areas.

I may pass over a few dimes but I've also found dimes in those "worked out" areas.

Truth is, it's all hit or miss. You'll just never know with 100% certainty what is causing your detector to go off unless you dig it.

Al