Using TurboCad, I created two separate drawings. The 1st drawing is a tracing of the date “1909”. The 2nd drawing is a tracing of San Francisco MintMark MMS-001 that was the “S” mintmark used in 1909. I imported images of the 6 major 1909-S dies into my CAD work plane and overlaid the tracings over each image and created what I call emulated CAD drawings of the 6 major dies. The emulated drawings are attached to this thread. How accurate is the process and the drawings? Well I don’t know. Post some images of these different dies showing the whole date with the mintmark and I’ll test them.
CAD is Computer Aided Design / Drafting. It is a tool used in most all design trades. It is one of the reason technology has been able to advance so rapidly. It is used to build just about everything these days.
Not sure if this will help, but these are the 4 S VDB dies. BTW, be sure to read the note for this picture's die #1.
I assume (for the VDBs) you compared or perhaps even used the well-known position photos that have been around for many years (at least since published by ANACS as part of a counterfeit detection pamphlet in the early 80's)? Going from memory they look about right. Using photos to compare positions, even with lines added, can be a little confusing for some, so I think this an excellent idea and commend you for your efforts.
My CAD is not with me at this time - when I get home tonight, I'll overlay my VDB drawings onto the image provided by RLM. Using low resolution jpegs - importing into TurboCad then rendering back into jpeg - produces alot of pixilation - but we'll see if.
Why not just find some Heritage Archive images in excruciating resolution - finding all 4 date positions shouldn't be difficult - and do a raster-to-vector conversion into your CAD software? Let reality draw the lines.
Here are the results of the 4 S-VDB overlay with my emulated die drawings. There appears to be a problem with die #2. Either I did not emulate the die properly or I grabbed the wrong 1909-S image to emulate by mistake. Because the 1909-S VDB-002 I emulated in TurboCad cannot be the same die in the picture provided by RLM. I'll look into this. Anyway here are the results.
Where are you acquiring your true X and Y axes from which to illustrate spacing deviations? The horizontal white dots in rlm's illustration should be "true" horizontal, and the semi-vertical axis is meant to be parallel to the serifs for the purpose of showing the tilt of the mint mark. None of the horizontal lines in your CAD seem to actually be horizontal.
My axes are vertical and horizontal in relation to the date in my CAD renderings. If the date is not square and straight in the image then my axes will not be vertical and horizontal when my CAD drawing is overlaid onto that image. But mt axes will be vertical and horizontal to the date in the overlaid image. Anyway using the below images of Die 1, Die 2, Die 3 and Die 4 produces a different result.
I do not think it will affect your drawings, but looking at you MM, the two ends are supposed to be parallel. Here is a close up of the MM including the die chip.
....which is the point of the slanted orientation of the "vertical" dotted lines in rlm's illustration - those serifs having parallel faces are a primary counterfeit detection point. Don't get me wrong; I very much like the idea of reducing this graphic to accurate vector format. It'll make things a ton easier to visualize.
Nice going. NOT! The OP posts a very interesting group of images of 1909-S coins and all of a sudden the thread gets hijacked with VERY OLD INFO on 09-S VDB's and that's all that gets posted about now. Duh, 1909-S and 1909-S VDB coin are different. Posting images of an entirely different coin is of no help - not opinion - fact!(So what if the mint marks are the same). @alurid THANKS for some great work. Makes me want to see if I can confirm your die study and possibly add to it.
This is the tracing of the "S" mintmark I used to emulate the mintmark in my drawings. Not an absolute perfect tracing - I know.
That is why I titled this thread ALPHA TESTING - new in-progress rendering of MMS-001 attached. But it still doesn't change that the emulated dies VDB dies I created in CAD overlay on all four examples from PCGS but only on three of the four examples provided by RLM.