Attrition rate for re-slabs

[fantastic_four]

Based on the coding, if you resubmit the book, CGC has no choice but to give the book a 9.2 (or worse if they believe the book has been damaged in th case).

 

They've got plenty of choice, for the exact reasons you've already stated--they might have overgraded before so they can just grade it independently again. If what you're saying is true, then whenever anybody resubmits a book right now WITH the label, then CGC would simply give it the exact same grade again. This isn't what they do.

 

The serials would be to prevent significant grade raises due to cleaning/pressing, not to reinforce potential undergrades. I only see the serials being used by a certification company AFTER the re-grade to make sure the book wasn't altered, not before it.

 

However, Chrisco's argument makes sense...they're making some nice green on clean/press resubmits and probably wouldn't want to screw it up AND tick some people off with the serials.

[FFB]

Not only that, but if a book is scanned, then pressed and cleaned, the second scan won't show up as a match because the dirt that was there before would be gone. It'll show up as a different book and the computer won't know it's the same. The software thing won't work.

 

Pardon my ignorance....but isn't a template defined once a particular scan has been entered into the computer? Is it really that difficult to search for a match ?

 

It would be quite complex.

If you scan a book 100 times, you will get 100 different pixel variations.

No. It's not as easy as looking for a simple match.

You could get some that are close. But, depending on how much you spent fine tuning it, you would probably get every other specific issue of whatever book you were scanning.

If you scan an ASM #1, it would return ALL ASM #1's.

 

No matter how tight you may be able to get the tolerance, you'd still have to pay an employee to sit there all day and scan books.

Then another 5 employees to compare scans of books.

 

If you had the software fine tuned enough to nail it down to 10 or 20 like issues, chances are the original book wouldn't be one of them if the book *had been PRESSED* in the first place.

 

My guess the chances of this technology actually working is between slim and none.

[FFB]

The purpose is not to prevent a book on the cusp from regrading .2 higher. They could grade it blind as to the prior grade for chrissakes. The important thing is if the scan looks markedly different from the re-subbed book, they could tell that there is otherwise undetected restoration.

 

While a good idea, IMO, the "invisible stamp"/scan bar/etc... does not make sense for CGC. They would wipe one of their services all-togther. I would guess that a large majority of their re-submissions come from those with books that are on that "bubble": graded 9.2, but a good shot at 9.4 on re-sub, etc... From a business stand-point, it would be foolish of CGC to implement this kind of change.

 

Chris

[FFB]

The fact that the code is in the book does not necessarily have to mean that the prior grade is followed. The book could be in WORSE condition, so it has to be regraded on its merits anyway. My point is to use this as a restoration detection device, not as a way to hold CGC to a prior grade.

 

While a good idea, IMO, the "invisible stamp"/scan bar/etc... does not make sense for CGC. They would wipe one of their services all-togther. I would guess that a large majority of their re-submissions come from those with books that are on that "bubble": graded 9.2, but a good shot at 9.4 on re-sub, etc... From a business stand-point, it would be foolish of CGC to implement this kind of change.

 

Finally a little bit of reason.

 

STOP THIS TALK OF IDENTIFYING BOOKS THAT HAVE BEEN GRADED.

 

Why?

 

Because your logical conclusion to that is:

 

Once a book has been given a grade, IT COULD NEVER GO UP IN GRADE.

 

Yet, as I have stated many times, CGC's opinion (which I value a lot) is only the OPINION of THREE GRADERS at an EXACT MOMENT IN TIME.

 

Do you BELIEVE that some graders might be more strict than others? I DO!

 

Therefore you submit a book and it gets a 9.2 and a code is put on the book. It just so happens that the two STRICTEST GRADERS at CGC are reviewing books that day. They both give 9.2 and one other person gives a 9.4.

 

Based on the coding, if you resubmit the book, CGC has no choice but to give the book a 9.2 (or worse if they believe the book has been damaged in th case). But if you submitted the book the day before, when maybe only one of the STRICTEST GRADERS is working, the book gets a 9.4.

 

CGC has always stated that if you disagree with a grade given to a book, ask for the graders notes and if you still disagree then resubmit. While obviously they will be glad to take your money again, I look at like an appeals process in court. They will give the book another chance.

 

[sterlingcomics]

Sorry, coming in late on this interesting conversation. Here's my 'Final Thoughts' on what I've digested:

 

1. The cost of doing business will probably prevent CGC or other professional grading company from implementing software to detect previous submitals. I've worked in IT for about 15 years, and "right now" I think the software is out there, but it needs to be customized. Problem is, what kind of market exists for such specialized software like this? If you can't resell it, it's going to cost the one customer a bucket load of money.

 

2. Even if such software exists, costs of doing the extra work/searches would delay service longer, reducing submissions. Doubled with that would be a most certain price increase to pay for the latency and to pay for the technology. That's a sensitive area for all of us who already have books in their warehouse right now (or in the past).

 

3. The invisible serial number is a cool idea (and actually something I was thinking about when discussing the resubmits) but, as demonstrated already, this will be a passionate discussion with folks that truly want their collectibles untouched.

 

4. I've personally been concern with the re-submits only because no one is probably sending in their old labels. This of course artifically inflates the number of books in the census, thereby potentially reducing the value of books at that level. Granted, it is more than likely at the lower grades, but I've heard it's not uncommon for folks to press and resubmit a 9.0 or 9.2 book. If it's a pricey book, the monetary rewards are obvious.

 

My humble submission,

Sterling

[comicwiz]

What made you bring up vector graphics? Not sure how they'd help with comic covers...if you're just generally geeking out, then that's fine too, welcome to the club.

 

 

Actually, in case you guys didn't know, there are bitmap tracing utilities that can take a pixelated image, and convert it into a vector-based images -- within seconds!

I've actually seen scanner software that comes with this feature, but I mostly use the utility that ships with Flash. I can show a few examples later for comparison.

 

I wanted to point out the advantage to vector-based imaging, the mathematical formulas, and how fingerprinting software uses the same premise to determine variability in prints, and its ease in finding print matches. I also wanted to draw on this mainly to point out that any discussion on comparing comic covers, or other imaging related endeavours should not exclude the possitibility, and power of vector-based graphical imaging.

[sborock]

Cool thread, lots to think about.

 

BUT...because some people are saying "cleaning and pressing" and "restoration" in the same breath, I have one comment:

 

CGC does not consider DRY cleaning (earsure, not water or solvent) and pressing (when not taken apart and done correctly/safely) restoration.

[drbanner]

CGC does not consider DRY cleaning (earsure...)

 

But does CGC downgrade for ear-wax residue?

[FFB]

Does dry cleaning include this "Wonder bread" thingy that I've been reading about? If that's not listed as restoration, I hope that you guys have at least conducted experiments on the effects of bread mold on paper inside of the CGC case. How much does a comic get downgraded if it has a yeast infection?

 

Cool thread, lots to think about.

 

BUT...because some people are saying "cleaning and pressing" and "restoration" in the same breath, I have one comment:

 

CGC does not consider DRY cleaning (earsure, not water or solvent) and pressing (when not taken apart and done correctly/safely) restoration.

[fantastic_four]

CGC does not consider DRY cleaning (earsure, not water or solvent) and pressing (when not taken apart and done correctly/safely) restoration.

 

Are you sure you guys are able to tell consistently when books have been taken apart? We think we've found examples of some Mile High resubmits from before-and-after Heritage scans where there's a high chance the cover had been removed to do the work.

[comicwiz]

Below are two scans. One is a rasterized image (bitmap, pixel-based graphic), and the other is a vector-based graphic. Can you tell which one is which?

 

Example A	Example B
Click Image to Zoom-In, Click again to Zoom-Out	Click Image to Zoom-In, Click again to Zoom-Out

[fantastic_four]

The poor text rendering leads me to guess that the one on the right is the vector-based graphic. If comic covers started out as vectors instead of having to be interpreted into them, I would expect the text to look better than a scan, but I'm guessing that having software try to turn a 2D image into a set of points leads to that kind of fuzziness.

 

I don't really know because I hadn't thought much about creating vectors from 2D images before.

[comicwiz]

The poor text rendering leads me to guess that the one on the right is the vector-based graphic. If comic covers started out as vectors instead of having to be interpreted into them, I would expect the text to look better than a scan, but I'm guessing that having software try to turn a 2D image into a set of points leads to that kind of fuzziness.

 

 

You are correct in saying that covers would be sharper/crisper if they started out as vectors, but it also depends largely on the resolution with which the bitmap is being traced. I have had examples of vector images which are sharper than a rasterized version. The Plop 5 was traced at a 72 dpi, so the text looks rather crude in its interpretation -- but not much better on the rasterized version to the left either.