Perforation Errors - Expert Help Needed

Okay. My last try. The left stamp is 10 3/4 by 11 perf. The right side stamp is 11 by 11. If one looks at the left stamp, I believe that you can see that the horizontal and vertical perforation gauges are different. But eye examination does not matter. I measured the perforations with computerized software tools to exact precision.

While I have no explanation for the two RW50 examples that you offer, I did take the time to measure the perfs on my RW50s (both of them) & RW57. I agree completely with RW57 being 11.25x11.25 ... and my measurement of the RW50s is statistically in agreement (10.9x10.9) with the official 11 shown in Scott. There is no question whatsoever that Scott rounds off perf & die-cut measurements whenever those measurements are not truly germane to identifying the stamp in question. This should explain most of your perf anomalies. But your strange RW50 begs for an explanation.
Question ... how often do your perf measurement tools show different values between the top & bottom perfs of a stamp ... or the left & right perfs of a stamp? Except for a very few exceptions, the pairs of top/bottom & left/right perfs are of the same gauge. The minute gauge differences that you give among the 4 sides seem to me small enough to possibly be explained by statistical or mechanical (human or other) error in the measurements.
That said, I still have no explanation for your weird RW50.

Here is another perforation error duck stamp...1985 US #RW52.



Perforated 10 3/4 by 11. The correct perforations are 11 by 11. This has me really confused too and I think the these stamps simply have not been well examined by the stamp community. I think that people just take them as cool specialty stamps and leave it at that. But these stamps are just as prone to errors and color omissions, etc, as any other stamps. There is something going on here with these duck stamps. What is amazing is that these duck stamps were only produced for a single year. And they were probably printed in one batch just before duck hunting season. All the stamps should be the same. But they are not. Crazy. Because the stamps are expensive, collectors have probably not performed a thorough examination of a large number of these stamps for each year. How many of these same stamps are you going to own? At prices of $10 - $60 per stamp, collectors are going to buy one stamp and call it a day.

I pay close attention to perforations whenever I buy or sell stamps and I am going through my large collection looking for perforation and other anomalies. I will locate the stamps where I had really weird perforations that did not match on each side. I will find them and post pictures.

Thanks for confirming my analysis. A lot of people in this stamp community think that I am a crack pot as I am always coming up with crazy stuff. It is nice to have someone who understands what I am seeing.

perforation analysis on the1985 #RW52.

10 3/4 by 11. No question about it. Should be 11 by 11.

Have you tried a normal perforation gauge on the stamps that appear to be perf 10.75x11? If not, do so.
I would begin to question the software used to analyze the perfs. Or rather, the mixture of analytic software with scanning software. Is the scanned image of the stamp accurate enough for the analyzer to correctly judge the perf gauge? Note that this might possibly explain the different results that you see for the two RW50s. I believe, from the images that you provide, that the two RW50s were scanned separately. If one of the stamp images were accidentally resized during the scan (or following the scan by image processing software), the subsequent perf measurements could be affected. I agree with John Becker ... try including both stamps in a single scan. Then use your software to measure the perf gauge for each stamp. If the software requires a single stamp for its analysis, carefully separate the images following the scan. Remember, no resizing allowed.
Note that since some of your perf measurements are accurate, I actually suspect the scanning/image-production process as the potential culprit as opposed to the analyzer software. An occasional error, human or otherwise, could easily alter the perf measurements for a specific stamp.
Incidentally, I checked my copy of RW52 ... as expected, perf 11 (well, actually 10.9, but the 0.1 difference could be my aging eyes).

JLLebbert
Sewall is scanning the stamps and then transferring the images back to his iPhone (as per the software instructions) so he can then use the application to gauge the perforations.**

Sewall
Please simply image both stamps in the same single scan.
Don

**This added step not only is time consuming (I could manually hand gauge 6 stamps in the time it takes to scan one, save the file, copy the file and then run it through the app) but also introduces additional potential user error.
I question the design and marketing of this particular app. It is a joke to think that a mobile device and user can take a picture with their hand held mobile device and get an accurate perforation identification. If the camera is not perfectly orientated 90 degrees directly above the stamp when the photograph is taken, the perforation gauge calculations will be incorrect.
So if you require users to go scan a stamp, save the file, and transfer the file back to the phone; what the hell is the point of making this a mobile device app? Doh.

SewallH


After reading this thread I decided to take a look at my Duck Stamps. I found the horizontal perfs on all four of the below plate blocks difficult to gauge. The closest match was perf 11 but not a perfect match. I'm wondering if this is a common deviation as it seems to appear on all of the plate blocks.

Don, thanks for the explanation/clarification as to what the software requires. This seems to me to be a "Rube Goldberg" process, fraught with the potential for inaccuracy. While I realize that duck stamps are unlikely to meet perforation gauges with any degree of frequency, I find it difficult to believe that there are multiple perforation varieties that haven't yet been discovered. When two explanations for something are available, the simplest one is almost always the correct one. If the OP wants to test his theory that there are such perf differences, he needs to apply the time-tested practice of using an actual perf gauge.
gettinold: A number of the duck stamps that are described in Scott as perf 11 are in fact slightly less than that ... my perf gauge reads to the tenths ... when I use it, I usually get 10.9.
John

Kinda like using a drone to take a photograph of your front door. Overly complicated.

The Stamp Smarter Listing Review requires users to enter three things;
1. Seller name
2. Link to Listing
3. Opinion of the listing

It has a few more additional optional fields (i.e. catalog number, country, etc.) if folks would like add more. But the three things above are the minimum for obvious reasons.
Don

Don - Wrong thread?



***LOL, Yes! My bad, sorry***

great set of duck stamp blocks! fantastic.

It was mentioned that I am transferring my stamp images to my iPhone. This is totally incorrect.

My process is as follows. I scan the stamps pressed flat on a flatbed scanner with a black background at 600 dpi and the image is scanned directly into my Mac laptop. The image is then cropped, but not resized, by the perforation software. The perforation measurements are then calculated. I do this with two separate perforation software programs to confirm the perforation calculations to see if I get consistent results. I stand by my analyses. I do not use iPhone images and never indicated such. I have degrees in electrical engineering and physics so I am not some scientific neophyte.

Part of this chat exchange is to alert board members as to my findings and to elicit some additional information and perspectives on all of this. I have a duck stamp collection of 27 stamps, but I have no duplicate stamps, so I can not do a statistical sample of many of the same stamps. Therefore, I need assistance from others to look at their stamps. I believe that I have presented enough evidence of perforation anomalies on the two stamps I have presented here to cause duck stamp owners to look at their stamps. One astute poster indicated that he thought that the specified stamp perforation gauge of 11 on a number of duck stamps is just below 11, at like 10.90. I agree with this. It may be that the standard perforation is 10.88 (rounded then to 11). But if the perforations vary slightly, one could see 10.95, 10.88, or 10.80.

However, in my stamp the precise perforations are top 10.73, bottom 10.69, left 10.91, right 10.88. The left and right sides are rounded to 11 and the top and bottom are rounded to 10 3/4. All that said, the difference between the horizontal and vertical perforate gauges is significant. The distance between the perforation teeth on the bottom is 1.869 mm versus 1.831 mm for the teeth on the left. This is not insignificant. Both the top are bottom perforation gauges are below 10.75 (10.73 and 10.69). Wouldn't those be rounded down to 10.5 to 10.75 in the stamp specification literature?

Something seems fishy here and I would love to figure it out. Just because no one has ever commented about this does not mean that my analysis is not correct. Stranger things have happened.

Can I ask members to measure their duck stamp perfs. Please measure them precisely if possible (ie to at least 0.1 tolerances - 10.6, 10.7, 10.8 etc - and if you can do it to two digits, eg 10.85, even better.

i will measure all my duck stamps precisely and post the results here. I will start by posting the results for the first 5-6 stamps later today.

From what I understand, SewallH is asserting peroration variations in the production of one or more duck stamps in the mid 1980s. I must admit that I doubt this, so may I suggest:

1. Close your catalogs and set them aside. The catalogs do NOT contain "specifications". They are not the gospel. They are secondary sources. The catalogs contain rounded-off measurements according to their editorial decisions.

2. Set aside your perforation gauges both electronic and physical. Each gauge is different and everyone uses them a bit differently.

3. Use only primary material - the stamps themselves! And nothing else. This means scanning stamps adjacent to other stamps in the same scan. No manipulation, no photo-shopping. This will visually demonstrate any relative differences. Exact measurements matter less at this point, although scans including a perforation gauge are also useful for scale.

I only own one copy of RW50, on license, scanned here with two ducks issued in the same era. The top and bottom perfs are identical across the years.



Rotating a stamp, I agree the top/bottom perfs and the side perfs differ very slightly. The top/bottom perfs are a slightly smaller gauge, maybe by 0.1 or so. Smaller than the Scott catalog editorial rounding rules.



Production-wise, it makes sense to me that the sheets were feed through one line perforator and then turned 90 degrees and run through another nearby machine set for a different wheel-width. These machines could be of different makes, models, ages. Thus slight variations between two machines is not unexpected. The point being, this is a very short-run production. It would have all been done on a single machine for each operation. All stamps would be alike.

SewallH, if your stamps are truly errors (and NOT just deviations from what the catalog states), then it should be very simple and inexpensive for you to obtain examples of the "normal" variety. So far, you have not demonstrated anything definitive other than to discovery that the catalog editors round-off their data. Compare stamps against stamps, not against the catalog. Show us two of the same duck stamp with different perforations along the same edge. If you are not convinced, then send your stamps off for certification.