Can we trust current slab coins?.

that's applied to u.s.postal service. where everything is automated. it can easily send 1/2 of its employees sit down and doing nothing but with pay. total employee stood at 700,000 strong.

 

Log in or Sign up to hide this ad.

Such technology exist.... I have seen it at coin shows....

I'm not sure if it is the popular thing to do tho or how accurate the technology is, but it definitely does exist.

 

I think the mint should stick to what they were designed to do... make circulating coinage... The minute the mint starts buying back it's own junk it opens up a whole new can of worms... not that it'll ever happen.

Who do you think orders more US mint product new.... the average Joe on the street???

 

The two most recent posts on computer grading is this one by Doug
http://www.cointalk.com/t42332/#post440257
but his url of the dissertation by Rick A. Basset on computer grading is dead now and should be this
http://www.rickbassett.com/pace/dissertation/Manuscript.pdf
which I have saved and read a couple of times.



and also this post by myself in the post following Doug's
http://www.cointalk.com/t42332/#post440412

which has this url:
http://www.freepatentsonline.com/5220614.html

Which is a patent for an Automated coin grading system ,1993 (and it is assigned to Guess who! Right !!! :secret: PCGS.

This URL is very interesting in that it has references to other patents of similar claims, back to 1967 "Silhouette coin grader", including 1989 "Numismatic detector" and 1990 "Method and system for objectively grading and identifying coins".

Since PCGS has held the patent since 1993 it would be interesting to know if they ever tried implementing it in real tests or if it was eventually discarded. Even holding this patent would discourage many TPGs from trying to implement it themselves, IMO.


Jim

 

The way it was put to me is that they know down to the number how many pieces will come back in certian grades... Very well could be a fib, or a stretch of the truth, or they just have someone that good at grading modern stuff... who knows. They didn't elaborate.

 

I really think being "against" slabbed coins is crazy. Personally, I won't buy anything over $50 online that isn't slabbed by either NGC or PCGS. At the same time, I've never bought a slabbed coin in-person. In an age when a great deal of coin buying/selling is happening online, slabbing is only going to grow.

I don't know of any people out there can accurately grade a coin based on ebay pictures, especially without knowing the lighting, etc., used, whether the photo was doctored, whether its clear, etc.

 

Slabs were created with the intent of verifying the authenticity and grade of CLASSIC collector coins. This new fad of grading modern pieces has IMHO been a detriment to the reputation of "the graded coin". You have several camps when it comes to this stuff... you have the everything (including my moderns) should be graded people, the nothing should be graded people, the certian coins should be graded people, and the who gives a hoot people... did I miss anyone?

If the grading companies got back to the basics and realized what they were created for and quit seeing the dollar signs... we wouldn't be having this discussion or the hundreds like it across all the boards. It's not gonna happen...

 

1993 was the stone age.:computer: Many watches have more computing power than the computer I was using in 1993. I have no doubt that with all of the advances in optical scanning and computer power over the past 16 years that a machine could be built that would grade well enough and consistently enough to match the TPGs. However, the cost would probably be prohibitive and not economic for the hobby. Anything you see at a coin show is probably only 1% as functional as what is possible. But if the military ever decided that they needed one for some unknown reason, they probably could reach into the budget and buy a very nice system.

 

The system I saw was at a Baltimore coin show this year. The reason I remember it was this year was because I was looking for a Lincoln dollar and he had one in his display. I asked him how much it was and he told me it wasn't for sale, he was using it to show off his computerized coin grading system. He then demonstrated it for me. He put 2 of the same coins in the machine and then showed me on the screen how they were different. His machine honestly had the shape of one of the small George Foreman grills. I was in a hurry so I didn't get into details with him... but this technology is still being tinkered with these days.

 

The Submit and resubmit until it gets the grade I want people.

Matt, I could see where the device you describe could work at differentiating with a "sample coin". With scanners, and modern photo comparison software, a point to point comparison could be presented as a resulting photo, but I think it would be limited unless it had 3-D scanning and recognition software, and some kind of algorithm to produce a standard grade. I agree the modern computer devices are exponential to what was available then, but the software to "Grade" a coin would probably merge into an artificial intelligence type of decision.

And sometimes I think we already have that around

Jim

 

I imagine this is a combination of the dealer screening the coins as GD said and a law of averages. With ASE and similar moderns, they can expect a certain percent to come back a certain grade due to the care of the production. With individual screening on top of it...they can probably nail down pretty close to exact numbers of 69s and 70s they will get back. But, as far as a corrupt system where the dealers buy the grades...I just don't believe it. If that ever came out true publicly...it would ruin the TPGs. They have too much to lose and not enough to gain.

 

But a computerized system that could distinguish wear only and be able to give even an Unc/Circ decision would be very useful for the average collector. But I agree there are currently too many factors for an accurate grade and people watch CSI type of programs and think software is far more advanced than it is.

kinda like the old Buster Brown shoe fluoroscope machine I remember as a kid, Put your foot in and see if it fits.....and if you develop 6 toes afterwards.....

 

I remember those fluoroscope shoe machines. And when I retired some years back was working with a tech known an NIR. (But you had to keep the camera soaking in liquid nitrogen, kinnda a hazard unto itself.)

NMR & MIR (nuclear magnetic resonance/mircopower impulse radar) have come a way since then, I bet. Actually, MIR is a precursor of Bluetooth and can be had for a pentance these days. I bought my last chip in the form of a Stud Finder from Zircon for $8. Anybody here like pico pulse examination?

NIR (near infrared) was a different animal and required a super computer to place the location of how many protons of what flavor were in proximity.

...course the Sparc Sun Station I used back then was less capable than my duel core laptop I use today.

It's not all fantasy. AU & MS may some day be determined by protein analysis to determine whether or not a human has been close to the coin.

 

Dang! That's why I have 11 toes?

I guess if the shoe fits, bare it.:crying:

 

Of course it's the hardware, the word "computer" being too restrictive. 20 years ago the photoelectric digital sensors and optics probably didn't exist or were not advanced enough to provide the input from the target coin that the software needs, and the hardware wasn't fast enough to process the inputs. You are correct that the software and hardware haven't been designed specifically for this application -- but the technology is advanced enough to do so. There just isn't enough money in coin grading to justify the R&D.

 

Who brought this thread back to life?

 

I'd actually tend to believe it. He said they send in a LOT of eagles. Well it's just a matter of statistics. Send in 5K or more per quarter, do that for several years and yo can work up a pretty good idea what percentage will come back 68, 69, and 70. So when you send in the next batch you will be able to make a fairly accurate prediction of how many of them will come back in each grade.

And the PCGS computer grading system was called The Expert. Compugrade was a different company usuing their own system.