Options
100% accurate perfectly quantifyable grading system
Hyperion
Posts: 7,472 ✭✭✭
instead of having Michigan kill me for hijacking his thread, I created a new one...
his post got me to thinking...
(no, not how insanely overrated brett favre is)...
but numismatics could completely evolve someday (maybe soon)
we now have the ability to 'scan in'; as it were, detail, analyze, take hyper accurate "topographical" information of almost anything you could want..
and store it digitally... every detail of every texture...even luster and eye appeal could be quantified to a certain extent... we could argue any methodology,
but if we can store the data, we can analyze it as we go on...
im sure it wouldn't be out of the realm of possibility to create a completely OBJECTIVE grading system based on the factual attributes of a coin,
even balancing between techical and aesthetic components/attributes...
you could have a 100 point rating scale in any number of scientifically quantifyable categories (technical/asthetic/rare-attribute factors ) and average/combine them any way you'd like
into some overall score !
as a professional programmer, I know there are some technical things to work out, like getting accurate "baselines" for each coin type and subtype you could have..but that's just detail.
whaddayathink?! any interest in this kind of process?
(shoot. i should have patented this before posting it ! hey PCGS, wanna hire me as a programmer in your software development arm !?!?!? ill work for pocket change
)0
Comments
the die ! duh. techincal 100% would be a 100% identical copy of the source die.
might be impossible to ever get a 100%.. but that's better than a subjective PR70/MS70
his post got me to thinking...
(no, not how insanely overrated brett favre is)...
This past season is not entirely my fault, the pack has suffered
major injuries and I don't have the quality players surrounding
me that I used to have.
Hey, I do realize that I am getting older and have probably lost
a few steps and all those interceptions........ I feel really bad about
that.
At least I do have that super bowl ring, nobody can say about me
(as they do about Dan Marino...... yeah he was good but.....)
You raise a good point. As a programmer you need a base to start and continue from there.
Everything is related and graduated at each level.There is either a plus or a minus.
Before computers were considered the.. in thing...I corresponded with an electronic university
for computers with my Commodore 64.......Going way back.......
One of the questions was how you would use the computer.???
My answer was ......To use it to scan coins and record their images becase I enjoy coins as a hobby.
I also wrote that even though a computer can record and store,it cannot grade.
A grade is IMO subjective to the human eye and should be.
Very interesting process...............................................
Happy New Year
And as long as they put that eye appeal in the rating it will still be just a guess. Maybe a knowledgable guess but still a guess.
Now if we could put the computer/scanner to work to come up with the techinical grade part. And the graders eye to come up with somthing like the star over at NGC, or some other designation for the eye appeal of the coin.
Then maybe we will come up with a grading system that is more accurate but I don't see anything being 100% accurate.
You can enter all the infinite numbers all you want even decimals.
Tha fact is when I or you or some one else looks at a coin we all will find different characteristics
that we like or dislike.
As far as a grade there will always be room for a bit more up or down.
A number is still a number.
A look can tell a story.........
I've said in the past that the current TPGs have outlived their usefulness to collectors. What originally began as a noble idea has fallen victum to the mix of profit motive, and a finite supply of coins. Monsterman has made some excellent posts recently on how the game is played by the pros and how the slight differences in the top two TPGs standards create a symbiotic relationship between the two. Add to the mix that both have moved towards the Market Grading concept over the past five years what we have today is only a little better than the pre TPG days.
In my mind the only way this can be done would to be to have it spearheaded by a non profit collector organization dedicated to consumer protection. Unfortunately the destruction of the group would be a strong goal of the TPGs and dealers so it would take some deep, deep, pockets to survive. And it could take many years to have an impact. Dealers would crack and submit to thier TPGs as soon as they got their hands on any of these slabs. Only when the collector community banded together to only accept the new organizations slab's would the success be guaranteed.
which either to purchase all his coins from or grade his raw ones also.The purpose of a good TPG is to proctect the consumer
from coin doctors and fradulant coins.
Your right when you say there is a profit motive.There is somewhat of a standard.True, improvements can be made.
I do not agree on a 100 point system..........
What we have seems to be working somewhat.
I am not a professional ,just a regular collector.
PCG,NGC,ICG has given me some insight on what to expect on what a grade should look like,along with the
ANA and PCGS grading giudes along with other grading information.
Larry......................
I say collect the coins you like and quit worrying about the variances in grade so much.
Larry
My post was directed specifically to PCGS and NGC. The others are near irrelevent to the grading business or are just out right scammers. If you honestly believe PCGS and NGC have you, the collector, in mind when they make their business decisions then their marketing efforts are paying off.
<< <i>It would likely never work because we all value different attributes differently - so it's not likely to be possible for a grading company to force their program onto the market. I don't value strike very much at all - to others it's the most important thing. I do value luster, some don't care about it as long as there's few marks. Write a program for that!
I say collect the coins you like and quit worrying about the variances in grade so much. >>
much of programming is setting proper customer expectations...
we could probably provide the most detailed analysis of strike, and luster... you can balance your weights for yourself...
10/10 strike 8/10 luster wouldnt be for you, but 8/10 strike , 10/10 luster would be..
define luster for me, your eye is merely a mechanical device used by your body for detecting certain light reflection properties, no ?
if you can define it, I can quantify it.
again, this might not be a replacement for someone's subjective grading, it would just be another measuring factor or supporting.
it'd be like the BCS for coins... I think it'd be a interesting excercise..
It has 25 identifiable scratches that should not be there. XF category.
It has rub in 3 spots, AU category.
blah blah.
a lot of this technology must have already been designed in face
recognition systems.
edited to add: MIT computer science grad students would enjoy this
challenge i imagine.
Just checking out the net for microscopes... anyone using trinochular (spelling) microscopes with integrated digital recorders (cameras, etc)?
seem like cool toys. using only a loupe is killing my eyesight !
In honor of the memory of Cpl. Michael E. Thompson
but a grading system should not count eye appeal or to judge
how toning should affect the grade..
i think the hardest part will be determining original surfaces.
edited to add: i read tomb's reply below and he is right.. but if one
uses a computer to grade coins, maybe eye appeal should be left
to the seller and buyer to hammer out into a real world price. let
the computer grade what it can accurately.
In honor of the memory of Cpl. Michael E. Thompson
Excellent points.....
<< <i>MIT or an advanced study establishment would have more resources for optics, servers
Just checking out the net for microscopes... anyone using trinochular (spelling) microscopes with integrated digital recorders (cameras, etc)?
seem like cool toys. using only a loupe is killing my eyesight ! >>
yes! ......as a matter of fact.....tomorrow i'll be talking to my nikon man about a 12.6 mega pixel AND a 5.0 mega pixel ccd live camera that is used to shoot thru a microscope (trinocular) in the medical profession....(VERY HI-RESOLUTION)
can capture in up to 10 sizes......640x480...up to 4116x 3072
resolution=approx 1800 t.v. lines!.hows that for hi-def!
24 bit color
iso 320 up to 1250
software that will save to BMP,TIFF,JPEG2000 & JPEG (HI/MED/LOW) type files...(tif is best)
think i'm gonna go with the 5 mega pixel tho.....
the 12.6 was $8900.00
The best use of 3D laser scanning at present for this purpose would be to objectively "count" imperfections and base the grade on that. Purely technical grading and that's all. But if the system is consistent 100% it would fit the need (assuming there is one).
A human grader could assign an eye appeal subjective grade, or better yet leave it to buyers and sellers as others have mentioned in this thread.
I guess the current top TPGs don't really have any incentive to go in this direction. They don't seem to be hurting for submissions. The current system works from a business standpoint regardless of how frustrating it can be to collectors.
Maybe we'll see a new TPG start-up someday offering this type of grading? Would we use them?
What would also be great is if someone could come up with an accurate camera/monitor calibration system standard so when we viewed coins on the web (for purchasing mostly) the pics always matched how the coin actually looks.
collections: Maryland related coins & exonumia, 7070 Type set, and Video Arcade Tokens.
The Low Budget Y2K Registry Set
never gonna happen