1851-1861 3c Washington Compression Charts

Jaxom - It looks like the compression image for position 95R2L is incorrect. Can you confirm?
And thanks for putting together the right side of Plate 6!!

Harper1249

Hi Jaxom --

Here are reference copies of the two positions on R8 that you asked for -- Both at 1200 dpi -- and thanks again to Don for hosting these images on StampSmarter as they both exceed the SCF file size limit.

Regards // ioagoa


Here is 44R8:



And here is 59R8:

Thank you Ioagoa for the images. It is a big help.

It looks like I will be doing plate 8 next, before plate 7.
I have images for plate 8 ready to start.

Don, thanks for compiling the images for plate 6 right.

Harper1249,
The image appears to be correct for 95R2L. There is a slight variation on the right side where the lines come together. But I do see some concerning variations. Here is an image of 95R2L with cancel and with cancel removed and with the Chase image. The compression images are shown at 20% and 10%. I also showed the top right of 95R2L with the bottom right of 85R2L to show the match on the extension of the right frame line. The stamp used for the image is the only one in the database. Maybe Ioagoa would be willing to confirm the plating on the image shown for 95R2L in the database.

The SSD copy does seem to be correct. It looks like removal of the cancellation tweaked it enough for me to think it was different. It definitely changed the spacing between the left frameline and the upper left triangle and smoothed the curves on the right framelines. Compressed images have their limits I guess. Thanks for looking into it though.

Harper1249

Removing the cancel does tweak it a small amount sometimes but gives a much better appearance on the charts. I compare the before and after images and try my best not to alter the lines.

Jaxom --

Further to Harper's post on the 95R2L compressed images -- and in regard to "making improvements" to the compressed images -- although discussed previously in this thread and others -- a few things bear repeating when it comes to the limitations of using compressed images as a plating tool.

First -- compressed images can be distorted by any bends, wrinkles, tears, creases, etc., etc. in the paper. In other words, unless the stamp paper is completely flat, artificial curves, bends, spacings, etc. can be introduced.

Second -- the stamps should be perfectly straight up and down in the source image -- as stamps tilted at an angle -- one way or the other -- can likewise distort comparability across compressed images.

Third -- my view -- and this is only my view -- and is in no way meant take away anything from all the effort you have put in to creating the charts (which are a major contribution to the plating community) -- is that no image enhancement of any kind should be made in the process of creating the compressed images -- other than of course rotating the image to get it perpendicular to the X/Y axis -- and then resizing the image to compress it horizontally or vertically as the case may be.

I am curious -- in those cases where a cancel covers up a particular plating feature, once the cancel is digitally removed -- do you then interpolate to "fill in" the missing portion of the applicable line? If so, and again, this is just my opinion and others may disagree -- the filled in portion(s) of the applicable line may, or may not, be representative of the actual position? On stamps with recut inner lines, this is less of an issue -- but on the no-inner line plates -- especially plates 4, 6, and 7 where some of the differences among the various positions are ever so subtle -- the image enhancement, despite, as you say, "giving a much better appearance to the charts", can impact the usefulness of the images as a plating tool.

For the reasons stated above -- I agree with Harper that compression analysis is a very useful tool -- but it does have limitations - and anybody using compression analysis needs to be aware of those limitations.

Again -- all of this is just my "3-cents" -- and meant to be constructive -- and in no way meant to take anything away from the contributions that have been made.

Regards // ioagoa

Ioagoa,

Thank you for posting your thoughts. I am always open to hear what others are thinking regarding my charts.

Firstly, I understand what wrinkles and bends can do to alter the images. For this reason, I try to avoid any "on paper" images. And I take that into account when I search my images to plate a stamp. I hope that others are also aware of this factor. Many of the frame lines are unique enough to only find one match. More so on the ones with IFLs. My guide dot charts also help on the B relief stamps. If the match is not dead perfect, I copy the compression position and keep looking for a better match. One thing that my images can tell you is if it is not that position.

Secondly, I always rotate every image that I compress according to the right side frame line. Including the Chase images that am putting in the database.

Thirdly, My OCD has caused me to start removing the cancels. That is also what is slowing down my progress. I am spending an hour with each plate position. And you are correct that it may be causing errors no matter how hard I try to avoid it. Maybe I should stop trying to remove the cancellations. It would save a lot of time and effort. There are times when part of a frame line is missing and it is unavoidable. I have also use left side and right side from different stamps. I am making my best effort to match the Chase image when finished, but I am human and make mistakes. That is why I ask people to let me know of any errors they may find and I try to address the issue. I also have limited resources for images and it is going to get harder to find images for #10s.

However, I am willing to forego the cancel removal effort to speed up the process like when I started off. I am curious what others think.

Everyone one that uses my charts, give me your opinion about the cancels. Should I remove them for eye appeal or leave them for better accuracy (I do my best to be accurate when I remove cancels)?

I'd say leave the cancels.

Harper1249

I agree...
less time spent = better
and
less image modification = better
Don

I agree that the less modification of the source, the better.

I will not remove any cancel that affects the lines. This is making the process much faster.

Ioagoa, can you find a copy of 28L8 showing the guide dot?

Hi Jaxom --

Here you go -- position 28L8 with GD showing -- (sitting mostly on top of the BFL and pushed up against the RFL).

Thanks again to Don for facilitating the posting of this 1200 dpi scan!

Regards // ioagoa

That you Ioagoa for your time and effort to get me what I need.
I posted the plate 8 left guide dot chart on StampSmarter with the first two rows.
I could use any left guide dots from the first column that you have.

Hi Jaxom --

I checked all my plate 8 stamps -- and the only one that is large enough to capture a left GD in the first vertical column from either pane is a copy of 51R8 -- which is an A relief from the 6th row and does not have any GD's -- but this particular copy is large enough to capture a small portion of the adjoining 41R8 above -- and despite showing only the BFL and a trace of the LLB of 41R8 -- it does show the positioning of the 41R8 LL GD.

Not sure if you would want a scan of this one or not? LMK and I will post a scan if you want -- otherwise I am not going to be able to be of much help with your request.

Regards // ioagoa

I recommend leaving the cancels. Removing them can distort the lines.