1851-1861 3c Washington Compression Charts

Here are the compression charts for the left pane of Plate 6. The A relief, B relief and C relief charts are not compressed and the full pane chart is compressed at 75% as needed to post here. 1100 images complete, only 1500 more to go. Work is going slow but still progressing. These charts have been uploaded to the database at full size for your use. I have been working on these charts for 14 months now. I never figured that it would take so long. I will keep pushing on and eventually get them completed. Thank you all for your patience. Comments appreciated.

Don, you can recompile the left pane of plate 6 now. All images are there.

Stan Shepp, thank you for your plate reconstruction images. Without them I could not have completed this pane.

A lot of the images in the Stamp Smarter database are of the actual Chase images for this plate. It was a pleasure finding these images. Those images are in the database with a dark green border instead of black.

Hi Jaxom --

Nice work!

I was looking at the "Chase" images on L6 that you highlighted with the green border -- and noted that the ex-Chase 99L6 is doubled up as the image for 98L6 also. A minor "typo", but worth fixing when you have the time.

Again -- nicely done !!!

Regards // ioagoa

Done.
Don

Thanks for catching that Ioagoa. It is fixed. I got in a hurry trying to finish up last night.

Ioagoa, you use the Chase images often (I assume), so do you think it would be helpful if I sharpened the images or leave them as they were originally presented? I can do this with only 1 command. The sharpened image is not as good as the 1200 dpi original images but better than the original Chase image (I think). Anyone else can feel free to give their opinion.

Hi Jaxom --

I do not know enough about image management technology to give any thoughts on "sharpening" up the Chase images -- as I do not know what actually happens to the image during that process?

I will say that when it comes to plating the 1851 3c stamp -- the differences between some of these positions are so subtle -- (especially on plates 1E versus 1i -- and on plates 4, 6, and 7) -- that even a seemingly trivial manipulation runs the risk of distorting the plating characteristics. For that reason alone, I would be hesitant to do anything -- but again -- I say this without understanding what "sharpening" would actually do to any given image?

In any case -- I have found that plating from actual confirmed copies of a given position (or from high-resolution scans, meaning at least 1200 dpi and with jpg compression no greater than 16%) is easier than plating from the Chase photo's -- and between StampSmarter and StampPlating there is a growing data set of high-resolution images available to the collecting community. I will also say that -- even with multiple confirmed copies of the same position in hand -- differences in inking, impression, and plate wear can sometimes distort comparability to varying degrees. Which is why I like the fact that the StampSmarter data set has room for up to 3 reference copies of each position (i.e., in addition to the Chase copy).

In addition to potential differences in inking, impression, and plate wear -- the Chase photographic prints bring another element into the equation -- which is the relative photographic quality -- and variations in the exposure of the 1:1 contact negative prints that the Smithsonian made up over the years -- which is then further compounded by the final set of "digitized" prints they made when they decided to stop the production of traditional contact negative prints.

As an aside -- I know that there are errors of plating in both the StampSmarter and StampPlating data sets that can be annoying at times -- but these are getting discovered and fixed as more and more platers use the data sets. Even the Chase platings had errors of plating -- (all of which are documented on the USPCS website). Conversely, every time the data sets are used to successfully confirm a plating, their validity is further confirmed.

Wish I had more to offer you on the "sharpening" question -- but I just don't have the image management skill set to even begin to provide input on that subject.

Regards // ioagoa

In my opinion, sharpening is a 'no - no'. In fact, I would discourage ANY image manipulation and if any image manipulation is done it ought to be documented and published with the image. Any digital image, no matter what resolution or quality, is only representation of the actual stamp in hand. It is bad enough that we assume it is an accurate representation but adding additional manipulation only pushes the representation even further away from the actual stamp.
Don

I think the sharpened image is100% better than the original. It sharpens the details that are blurred in the photo image. It should be much easier to use and compare with actual stamps than the horrible photo image that is never sharp and sometimes too dark. Sharpening does not add or remove detail. I think it would attract more people to use our Chase images. I was amazed at the results of sharpening the old b/w photos. I never tried it on them before because it does not do very well with color images. I was very much hoping that you guys would approve of the results so we could have better Chase images than anyone else. I am considering making a full set of sharpened Chase images for myself. I hate looking at those blurry images and I do not trust the accuracy of the Lund images because I have found way too many wrong Lund images as I go through these plates. I am keeping track of the wrong images for my use. There were 8 wrong Lund images on plate 6 left.

PS: Don, without image manipulation none of the charts that I am making would be possible. Many of the Chase images have been manipulated so much that they are barely usable now. This technique clears up the low quality photos that Chase made and have been handed down to us. This is why we are making these pages, so we can have real images to compare to instead of the low quality Chase images. It does not alter the image in any way.

For what it's worth (prob not much) I have "sharpened" the chase images using the High Pass filter in Photoshop. I find it improves these images considerably and creates less eye strain for me. The camera was not in focus in a number of areas on these plates when the photograph was taken, so they are altered to begin with. Having said that I am not plating very many 1E and 1i stamps and I could see how further altering these images could impact being able to plate them from the Chase images alone.

In the end I try and compare my stamp with Chase, SSD, Lund and other high res images provided by you all. In most cases SSD has the best images, but other times it's Chase, and other it's Lund. In my most recent it was Ioagoa's example.

Just an opinion from and apprentice plater.
Harper

I am looking for a copy of 30R6 and 48R6 for the database and compression chart.
If anyone has one, I would appreciate it if they posted it here at 1200 dpi.
The 48R6 I need for the guide dot chart.

Hi Jaxom --

Attached are scans of both 30R6 and 48R6 -- both at 1200dpi.

Regarding the GD on 48R6 -- this is one of the positions from the 8th vertical column on R6 that has an extra line outside of the RFL running from the top of the plate to the bottom (i.e., Recut Variety #10) -- and the GD on 48R6 sits on top of this "extra line" -- at a level slightly below its BFL. This shows best on the Lund copy on StampPlating.com -- and once you know where to look -- the GD is visible on the Chase copy as well (as at first glance the GD on the Chase copy could easily be mistaken for a bit of cancellation ink). Unfortunately -- my copy is cut off in this area -- so, I suspect that this scan may not be suitable for your purposes -- but I cropped the scan to capture the GD which shows at the top of the adjoining stamp below.

Regards // ioagoa

Scan of 30R6 is here:


Scan of 48R6 is here:

Thank you Ioagoa.