Separation, Recovery Speed, & Discrimination Level. A Somewhat "Different" Way To Look At How They All Factor Into Unmasking Ability, & How People Often Confuse The Terminology Of What Each Does.

Some recent events of in field comparison has made me re-think some beliefs I've always held, namely in just what factors I always that played the most important or perhaps only role in unmasking ability. It forced me to sit down and re-think the aspects involved in unmasking, and also helped me to clarify in my head just what the definition of each (recovery speed, separation, and discrimination level) is. For instance, some people confused recovery speed and separation as being the same thing. They are NOT if you really nail down and define them. These topics and further things are discussed in this little article I wrote below...

People sometimes get confused about recovery speed, separation, and discrimination level, in terms of what they each do to all add up to unmasking ability. Really unmasking ability is a combination of all three of these factors, and so they all play an important part in it but for very different reasons. And, if one is lacking it can be compensated for by the others to still add up to good unmasking ability.

Separation, plainly put, all falls down to how sharp the detection field of the coil is to see the good and not the bad beside it. Separation boiled down to it's purest meaning is the ability of the coil to see the good and not the bad right next to it. At first this may sound wrong to people but if you think about it this is the most exacting description of separation ability. It's all about the coil and how sharp it is.

Recovery speed is an issue if you like to sweep faster, so make sure then that you are using a fast recovery machine in that case. With a slow recovery speed you need to compensate for it by slowing down your sweep speed.

Some other pros/cons in how these all relate to each other in terms of the sum of unmasking ability are for instance....Recovery speed does not automatically mean better separation. A fast machine can be using a rather dull bulky coil detection field and thus have no real better ability to see between targets. The key to separation, again in it's purest form or description, is the sharpness of the coil's detection field. Recovery speed can of course alter the ability of the machine to see between targets, but this is all dependent and addressable by your sweep speed. Use a slow sweep on a slow machine and it's largely a non-issue.

I think of discrimination level in other terms than probably most people. Many think say for instance that low or no iron rejection gives the machine more ability to "see" a coin mixed with iron at the same time. I think this is wrong myself. I personally don't believe the machine has the ability to see both at the same time, unless they are laying at the same depth and right up against or on each other.

Rather, with lower iron rejection there is less lag time between the two targets...As the machine doesn't have to go from "rejection mode" of the discriminated out iron to "accepting mode" of the accepted coin. There is a certain lagish processing that takes place in a machine from going from one to the other, so by accepting iron it's staying in "accept mode" and thus ready to sound off more readily when it sees the coin next.

However, lacking the ability to lower iron rejection on some machines, you are then forced to find other ways to increase unmasking ability. Mainly, to me that would be using a "sharper stick", or in other words a coil such as the 12x10 that has a very sharp DD detection line. That or a small coil of course. It gives the machine more time to distinctly go from "seeing" to "not seeing" the iron, and so it has more time to reset from rejection mode and be ready to accept the coin.

In a sense it's in effect speeding up recovery speed, but not really. It is, though, giving the machine more time to reset from seeing the iron and so be ready to see the coin. Kind'a like cleanly turning the page in a book, so that it's ready to read or see the coin on the next page.

Over several months of hunting we've extensively compared the Etrac using low iron rejection (the popular Andy Savage open screen pattern) and a Sovereign GT, which of course has high built in iron rejection, but is using the sharp detection field of the 12x10, and surprisingly (never would have believed it myself) the outcome of both on the same undug targets has been exactly the same in how those coins were seen. Both machines were EXACTLY seeing these coins as good (or as bad) as each other from various angles without a dime’s worth of difference between them. I’m talking coins so badly masked that some were a complete null all the way around them but in one tiny spot from a tight angle, or even a few coins that were on edge, deep, and right up against iron. We compare notes on these targets extensively, asking (and sometimes showing) each other how they are acting from all angles, and these targets are acting EXACTINGLY the same from all directions to both of us.

A bit off topic, but in one instance I was running the GT at full blast sensitivity, which the 12x10 often allows, and I hit upon a super deep whisper of a coin signal that sounded well into the foot or deeper range. My friend was running Auto +3 and his sensitivity was very high but he could not hear the target. He then put it into manual and raised it the last few clicks to full blast and then could hear the target. That tells me both machines were pretty evenly matched depth wise at this particular site.

Back to the topic of this thread…

Until I saw this with my own eyes and ears I would have thought the only true way to unmask better was by lower iron rejection, but now I see there are two ways to skin that cat. Not saying one way is better or has more merit than the other, just saying that at least what I've seen so far has been a total wash between the two.

In summary, recovery speed can be address by proper coil use and speed. Discrimination level can also be addressed by proper settings. Separation can not be changed by how the user sets up or uses the machine, but rather by coil choice. All three play important roles and I wouldn't argue that one is more important than the other. Just that there are several paths to reach unmasking Utopia. :’)

Again, many people will find one of these three factors (separation, recovery speed, or discrimination level) more important to them than others, so I don't argue that one is more important than the other. Largely that's based on the strengths and weaknesses of the machine being used and thus what you need to address or improve on to enhance unmasking ability. The only thing I would conclusively argue is that the most important thing is the ability of the coil to separate by way of a sharper or smaller field. Everything else is built on top of that foundation, and that's were there are then several paths to take to reach the end results. Again, those paths are largely dependent on what your machine can and can't be changed about, and so you may be forced to improve one or the other to reach the final goal. Any one of these three factors can be improved when others can't, to enhance the end result.

A bit controversial in terms of even how I always thought about the factors involved in unmasking for years. Certain in field tests on undug targets has opened my eyes to other ways unmasking can be improved when certain things involved can't. Just saying there are several things involved, and each play a different but equally important role (separation, recovery speed, and discrimination level). One can compensate for the other when it can't be changed. For example, if your machine is unable to lower discrimination to accept all iron, then by using a sharper or smaller detection field from what I've seen it can seemingly do just as well as a low or no iron rejection but not using as coil with as sharp of a detection field, or a smaller one of course. Also, since you can't change recovery speed on most machines, you can change your sweep speed as well as how you use the coil to sniff between targets to make it pretty much a non-issue. Know the strengths or weaknesses of your machine and then try to address one of the three things to increase unmasking in ways that you can alter. Just remember that the foundation that it all is built on is how good the coil can "see" between targets.



Edited 5 time(s). Last edit at 06/24/2012 07:17PM by critterhunter.

Good write up Critter..

And yes there is confusion in the area of unmasking and terminology used to discribe it..

You can take a slow response machine and unmask as long as the targets are allowed to blend response...

An AT-PRO is not the fastest or does not have the narrowest detection field in coil design but it unmsk very well in a blending way...

Where the greatest unmasking occurs is when machine's are using ...Fast recovery/razor edge iron disc and tight coil design..

we are just now getting into this realm...

Believe me I know where you are coming from with the BBS technology and I have seen it first hand for years...The ability to unmask as well as other machines known for there unmasking ability...but The European line of detector's like the XP's have turned the curve on unmasking...

No matter how hard you think you have hunted a site just short if using entrenching and sifter's the XP//maybe some other's in europe? will unlock target's with the correct setting's...and it happens through razor edge disc....fast recovery. and fast coil seperation design...and there's a certain amount of BLENDING going on also..

I have also seen first hand machine's like the Blender's become shutdown yet the user is not aware of it...

Keith

Funny you should post this now, after I have spent hours running some iron nail - coin tests with the CTX (Norfolk Wolf style - 2" square nail in line and as close to a dime as I could get). Basically, my results above ground and below (but with freshly buried targets) showed me that the CTX without a doubt did better with discrimination (I run it from 28 on down like I did on the E-Trac). I didn't want to do that - I had been running an open combine screen with the 5 tones. It worked fine and it was all the rage and the recommended pattern over at Finds. But, I didn't like the super smooth tones and sort of liked the 50 CO sounds and their warble - pretty much E-Trac like. edit - I should add 50 CO with the same disc and settings outperformed Combine mode - but with fresh targets.

Here is a 8 minute test video of the Norfolk Wolf like test where I played with the settings a lot: [www.youtube.com]
Here is a short 3 minute version with the best settings I came up with and I compared it to an open combine screen. Point was clear, at least in a test. [www.youtube.com]

So, I compared recovey fast on/off and ferous coin with high trash. Interestingly, the CTX in my opinion doesn't clip targets like the E-Trac did with recovery fast on (need to double check this though). It has a noticeably better effect than on the E-Trac as well (meaing it seems to make it noticeably faster). Ferrous coin seems to up average ferrous signals. They will false like crazy (in testing) with an open screen, BUT NOT WITH DISCRIMINATION. Seems to do better in iron than high trash does due to the up averaging (which again, doesn't seem to false when running disc).

I was digging bottle caps in combined. They sounded ok with the open screen and looked mostly ok. Then when I put on some discrimination the sound was mostly gone except for a typical broken tone. The crosshairs now no longer showed in the non ferrous range. See here: [www.youtube.com]

I should get a hunt in, in the next two days. I'm gonna go back to hunting in 50 Conductive mode with 28 down disc'd out with recovery fast on and ferrous coin enabled. I'll compare that to an open Combine mode with setting changes just as in my tests but initially everything but the 2 Modes (50 CO vs Combine) will be the same.

Albert



Edited 2 time(s). Last edit at 06/24/2012 08:38PM by earthmansurfer.

Nice write-up, critter.

Keith -- if you have a minute, can you tell me what you mean by "blending?"

Steve

Explorers like to blend and by this I mean the tones instead of having a definite start and stop between them they sort of morph into the next tone like a flute going from note to note without pausing between ..

It allows say a colocated target to instead of being say a high tone or a low tone it will be a BLEND of tones even becoming one tone on top of the other like a calliope musical instrument does..

Some machine's can only report one tone at a time and have definate start's and stops between the tones even in multiple tone's..

Some machine's will report tones in a BLENDING sound will sound like one tone is intermingled on top of another tone..

Say you have a nickle and a nail real close...Well if the machine is not fast enough to shutdown between them and signal say high tone and low tone...then a machine will have to do a composite and that would give say a low tone high tone bounce..sort of like a iron fasle will do..Now have a machine that is slow with no quick shutdown between target's and but has a tonal blend when you go across same target instead of a bounce high low it will give a sound like it's one target signal that changes tones as you go across it without a break in the tone's...Tune your ear to the subtleties and you can pick item from iron or trash on a very productive scale sometimes equal to other machiens known for unmasking...


Keith

Gotcha, Keith, thanks. I know just what you mean, now. Explorers do indeed "blend" tones, as you say...and I am just beginning to learn to utilize that...

Steve

I wonder if the XP's Deus seperation ability comes from the design of putting the electronics in the coil. As there may be an advantage of processing a signal that is so close to the coil rather than a signal that has to travel the length of the detector cable which may degrade the characteristics before the electronics and software can process it in other detectors.

I am sure the XP Deus has really fast response electronics and processor also.

I speculate 'no' due to the XP GMP having the exact same separation abilities.