|
|
Post by Ditch Fahrenheit on Aug 11, 2018 3:50:37 GMT -8
Recently, there has been a lot of debate about the authenticity of this Facebook claim. Here are the Facebook photos. I'll debunk this find in the next post.     |
|
|
|
Post by Ditch Fahrenheit on Aug 11, 2018 4:24:07 GMT -8
Without getting into the particulars of 'how' these books were found, or 'why' they were posted on Facebook, there are some big issues with the books in the photos. The paper stock looks incorrect, and the wear and soiling looks manufactured compared to 'real' Action Comics #1 comic books. But the easiest way to debunk these books is to compare them directly with real books and look for tell-tale signs of being inauthentic. So, let's start by looking at four of the highest graded copies of Action Comics #1.     Now let's take a close look at the font on the top edge of the front cover for each book and compare them to the two books in the Facebook photo (labeled LEFT and RIGHT).  You can easily see that the font on the two Facebook books is NOT the same as the real books. The biggest 'tell' is the width of the font in the areas marked in red in the comparison chart. Pay particular attention to the area inside the 'U' and 'J' of 'JUNE" and how much wider the fake books are compared to the real ones. DEBUNKED. I said good day!!! |
|
|
|
Post by Stu on Aug 11, 2018 6:37:00 GMT -8
XTCI strikes again!  |
|
|
|
Post by vintagecomics on Aug 11, 2018 13:58:08 GMT -8
I think you should add Stock Rotation's evidence from the CGC chat forum thread as well.
The staple placement is pretty damning as well.
|
|
|
|
Post by Ditch Fahrenheit on Aug 11, 2018 14:35:54 GMT -8
I think you should add Stock Rotation's evidence from the CGC chat forum thread as well. The staple placement is pretty damning as well.
I'm not sure I fully understand what he is saying since the staples line up with other books in the lineup too. Book #2 lines up within his margin of error, and Book #3 lines up perfectly. What am I missing?   |
|
mosconi
I made my first post!
Joined: August 2018
Posts: 1
|
Post by mosconi on Aug 12, 2018 11:37:35 GMT -8
The final nail on these two fakes are the hatching lines evident on the Superman figure in the legit Action #1's, but missing on the fakes (pic from the CGC forum) 
|
|
|
|
Post by Ditch Fahrenheit on Aug 12, 2018 12:51:27 GMT -8
The final nail on these two fakes are the hatching lines evident on the Superman figure in the legit Action #1's, but missing on the fakes (pic from the CGC forum)
It certainly looks that way doesn't it.  Actually, I think what's happening there is just an optical resolution effect where there either isn't enough distance between the hatch lines, or there isn't enough contrast between the hatch lines and the background for the camera resolution to show them. The image below shows an enlargement of that area. The fake RIGHT pic is on top, and the CGC 5.5 pic is on the bottom.  The white arrows show residual evidence of the hatching on Superman. The red arrows show hatched areas where the resolution is low enough to look like they aren't hatched (even though they are). So, I think this is just a resolution issue. Cameras can play tricks like that. |
|
|
|
Post by stockrotation on Aug 12, 2018 20:06:37 GMT -8
I'm not sure I fully understand what he is saying since the staples line up with other books in the lineup too. Book #2 lines up within his margin of error, and Book #3 lines up perfectly. What am I missing? Don't look at the staple placement in relation to the other books, look at it in relation to the sweat bead in the yellow area above the horizon line. |
|
|
|
Post by Ditch Fahrenheit on Aug 13, 2018 2:34:38 GMT -8
I'm not sure I fully understand what he is saying since the staples line up with other books in the lineup too. Book #2 lines up within his margin of error, and Book #3 lines up perfectly. What am I missing? Don't look at the staple placement in relation to the other books, look at it in relation to the sweat bead in the yellow area above the horizon line. Thanks for the clarification. I see what you're attempting to show now.  Your work on this intrigued me. My only concern was the size of the sample population, so I put a fair amount of work into this yesterday trying to increase it. I'll show my results in the next couple of posts. |
|
|
|
Post by Ditch Fahrenheit on Aug 13, 2018 3:16:40 GMT -8
Postulate: Can staple location in relation to a comic book image be used to prove that a specific book was scanned and used to produce a counterfeit Action Comics #1?
Design of Experiment
1) Obtained as many high resolution book images as possible.
2) Retained only those images which appeared to show similar staple location to the counterfeit books.
3) Culled any images which, upon enlargement, didn't have the necessary resolution to determine ...
I have to run...I'll finish this later today when I have more time.
|
|
|
|
Post by stockrotation on Aug 13, 2018 4:43:09 GMT -8
Your work on this intrigued me. My only concern was the size of the sample population, so I put a fair amount of work into this yesterday trying to increase it. How are you concerned about the size of the sample population of my evidence when your own proof is based on a sample size that's half of mine?
At any rate, you're never going to prove definitively that a crappy Facebook photo of a real-world object is sourced from a specific scan, but by showing the staple placement and the matching color, parsimony can do the heavy lifting here.
Let's also keep in mind you and I have probably spent more time talking about this than the money guys (i.e. the people who could actually buy two copies of Action 1) spent debating whether to pursue the books.
|
|
|
|
Post by Ditch Fahrenheit on Aug 13, 2018 8:14:17 GMT -8
Your work on this intrigued me. My only concern was the size of the sample population, so I put a fair amount of work into this yesterday trying to increase it. How are you concerned about the size of the sample population of my evidence when your own proof is based on a sample size that's half of mine? At any rate, you're never going to prove definitively that a crappy Facebook photo of a real-world object is sourced from a specific scan, but by showing the staple placement and the matching color, parsimony can do the heavy lifting here. Let's also keep in mind you and I have probably spent more time talking about this than the money guys (i.e. the people who could actually buy two copies of Action 1) spent debating whether to pursue the books.
Because parsimony is more easily applied to divergence from the norm, rather than convergence to a specific data point. |
|
|
|
Post by Ditch Fahrenheit on Aug 13, 2018 10:56:10 GMT -8
Postulate: Staple location in relation to a comic book image can be used to prove that a specific book was scanned and used to produce two counterfeit Action Comics #1 comic books. Design of Experiment1) Obtained as many high resolution Action Comics #1 comic book images as possible. 2) Retained only those images which appeared to show similar staple location to the counterfeit books. 3) Culled any images which, upon enlargement, didn't have the necessary resolution to locate the bottom edge of the bottom staple. 4) In order to retain as much proportionality as possible, all images were cropped to the top and bottom of the spine edge. The size of the cropped images were then adjusted to exactly 650 pixels in height. 5) For proper comparison, all images were aligned to the bottom edge of the bottom staple to show the resultant image offset. Note: Image color was not considered in this experiment due to well-known scanner/camera issues, and the fact that it is one of the easiest post processing parameters to modify. For the purposes of this experiment, no color adjustments were made to the original image sources.
The following images were used in this experiment.  CGC 1.8  CGC 1.8 (2)  CGC 1.8 (3)  CGC 2.0 R  CGC 3.5  CGC 4.0  CGC 4.5 R  CGC 5.0  CGC 7.5 R  CGC 8.0 R  CGC 8.0  CGC 9.0 WP, $3.2M, #1134755001 link
Here is the comparison image as described above. The order from left to right is the same as the list above (top to bottom). The comparison LEFT and RIGHT (counterfeit) images were added as the final images on the right respectively.  As you can see, multiple books exist which show similar image offset from the bottom staple edge as compared to the counterfeits on the far right. The books with the nearest offset to the counterfeits are marked with a white star. You will also notice that the book referenced in the prior experiment ($3.2M CGC 9.0 WP #1134755001) did not receive a white star. Conclusion: The postulate above is false.
|
|
|
|
Post by stockrotation on Aug 13, 2018 14:06:54 GMT -8
Conclusion: The postulate above is false. I'll make this brief. Your scientific method is so rigorous that you didn't notice your #1 image and #3 image are the same book. I'd bet money #5 and #6 are the same book after 5 went through a CPR (look at the staple rust migration pattern). The 9.0 scan used by the facebook guy was the pre-slab brightened version available everywhere-- you can tell from the hue of the yellow. He did not use the encapsulated scan that you used in your comparison.
You are attempting to make these comparisons at 650 pixels, which is a blob. Your posted image is 663 pixels high, so that's the actual size you were working with. I was working with images that cropped to 2500+ pixels. For comparison, this is a crop of your image at 100% dropped onto mine at 100%.
The rest of your logic and explanation follows that same thorough and careful testing.
|
|
|
|
Post by Ditch Fahrenheit on Aug 13, 2018 15:35:19 GMT -8
Conclusion: The postulate above is false. I'll make this brief. Your scientific method is so rigorous that you didn't notice your #1 image and #3 image are the same book. I'd bet money #5 and #6 are the same book after 5 went through a CPR (look at the staple rust migration pattern). The 9.0 scan used by the facebook guy was the pre-slab brightened version available everywhere-- you can tell from the hue of the yellow. He did not use the encapsulated scan that you used in your comparison.
You are attempting to make these comparisons at 650 pixels, which is a blob. Your posted image is 663 pixels high, so that's the actual size you were working with. I was working with images that cropped to 2500+ pixels. For comparison, this is a crop of your image at 100% dropped onto mine at 100%.
The rest of your logic and explanation follows that same thorough and careful testing.
You are correct about the first and third image, and I agree that the 5th and 6th are no doubt the same book pre and post dry cleaning. Good catch! I'm obviously trying to do too much multitasking right now.  So that removes two books from the comparison, but doesn't change the rather obvious conclusion. The comparison image is 663 pixels high because of the shift when aligning the 650 pixel height books to the reference line. The pixel size of 650 was deliberately chosen because of the overall sizes of the images involved, and was a good compromise which obviously provided more than enough resolution for the comparison. Unfortunately, your method of enlarging the smallest images to the size of the largest image will not increase the accuracy of a comparison since the resolution of the smallest image doesn't change. In this case, the smallest images are the most important, since they are the reference images of the counterfeit books. Magnifying the reference image by a factor of 4 or 5 is an exercise in futility unless you're writing a script for CSI and is borne out by the fuzzy picture you just posted. Thanks again for your input!  |
|
