8 bit alignment pictures

Process Select the objects, add an outline, red color and 1 pixel. Save as png and select 300 ppi. Open in photoshop. Set grid to every 8 pixels. Snap screenshots focusing on the top right corners.. I hope that this finally helps our dear friend Hermitian understand… Doing the simple calculations is more accurate and much easier… So Hermitian, did you forget to set the resolution to 300 ppi? I think you did as changing resolution in Illustrator can be quite painful. Apologies accepted🙂

None signature Mostly_textUpdate:

Even Illustrator can be encouraged to reveal the alignment

illusnoneillusdate

53 thoughts on “8 bit alignment pictures

  1. NBC

    “8 bit alignment pictures

    “Posted on August 18, 2013 by NBC

    “Process Select the objects, add an outline, red color and 1 pixel. Save as png and select 300 ppi. Open in photoshop. Set grid to every 8 pixels. Snap screenshots focusing on the top right corners.. I hope that this finally helps our dear friend Hermitian understand… Doing the simple calculations is more accurate and much easier… So Hermitian, did you forget to set the resolution to 300 ppi? I think you did as changing resolution in Illustrator can be quite painful. Apologies accepted:”

    Let me see if I am getting this right. You have posted several screen shots of the WH LFCOLB object boundaries that were taken in Illustrator at a low zoom factor. Based on your purported success you are claiming that I must have forgot to properly set the pixel resolution when I did my analysis of the WH LFCOLB object boundaries which I posted on Scribd.com over a month ago.

    NBC: Nope: I am saying that you obviously forgot to do so in your analysis of the 7535 pdfs]

    http://www.scribd.com/doc/151738307/Analysis-of-Rectangular-Object-Boundaries

    Now my analysis proved that the mostly text layer satisfied the 8 MOD 0 condition for resolutions of 150 PPI x 150 PPI and 300 PPI x 300 PPI and the 16 Mod 0 condition for resolution of 300 PPI x 300 PPI. So maybe you can explain how I did that without ever setting the pixel resolution. By the way changing the pixel resolution in all Adobe Products is a simple operation that takes maybe 30 seconds (or less).

    Looks like you have finally caught up to my work on the WH LFCOLB for the 8 x 8 blocks. I suppose that’s progress. Now we can look forward to your posting your results for the 16 x 16 blocks.

    [NBC: No need to do so. The 16×16 bit alignment is one of your own imagination]

    By the way, any screen shot taken with Illustrator at low zoom factor is unreliable when the pixel resolution is 300 PPI x 300 PPI because the grid does not display accurately at low zoom ratios. The zoom factor needs to be either 3200% or 6400% before the 300 lpi x 300 lpi grid displays accurately.

    [NBC: Have you done the experiment yet? Because all my data aligns with 8×8 bt, while you made some ill informed claims about it based on a 72 PPI grid. Am I right?]

  2. Based on your purported success you are claiming that I must have forgot to properly set the pixel resolution when I did my analysis of the WH LFCOLB object boundaries which I posted on Scribd.com over a month ago.

    No, based on our success we are claiming that you must have forgotten to properly set the pixel resolution when you did your analysis of the 7535 object boundaries which you have never posted. Where are your results for the 7535 that you kept bragging about? The 7535 is not the same file as the WH LFBC. Why do you pull up your results for the WH LFBC whenever we ask about your results for the 7535?

    Now we can look forward to your posting your results for the 16 x 16 blocks.

    Divide by 2. The WH LFBC does not show alignment on 16×16 blocks at 300 ppi, there is no particular reason to expect that it would, and therefore there is no reason to expect that any other Xerox WorkCentre file would show that alignment.

  3. No, based on our success we are claiming that you must have forgotten to properly set the pixel resolution when you did your analysis of the 7535 object boundaries which you have never posted.

    Hermitian may have forgotten how he did it the first time and he mentioned himself that he looked for visual alignments…

    He still does not understand the simple fundamentals of the alignments…

    See how he continues to insist on 16×16 alignment, not realizing that this only exists at the 600 PPI resolution at which the document was scanned.

    Is he just stalling or his he just not understanding this? Others seem to be having no problems understanding..

  4. Now my analysis proved that the mostly text layer satisfied the 8 MOD 0 condition for resolutions of 150 PPI x 150 PPI and 300 PPI x 300 PPI and the 16 Mod 0 condition for resolution of 300 PPI x 300 PPI. So maybe you can explain how I did that without ever setting the pixel resolution. By the way changing the pixel resolution in all Adobe Products is a simple operation that takes maybe 30 seconds (or less).

    Still arguing a strawman. I accept your WH LFBC analysis, which matches the one I had posted before. You claim that since a few layers match 16×16, this needs to be a rule as well, but fail to understand why it fails several times.

    If changing the resolution in Illustrator takes 30 seconds (and I agree), did you fail to properly do so? Because when I did it for the 7535 and other documents, the 8 bit alignment was quite obvious. Given that your instructions failed to include that important step and given that you claim that they 8 bit alignment does not exist, it seems reasonable to conclude that you forgot to do so…

    So now that you have been told how to do it correctly, can you report your findings? Numbers, screen shots etc would be quite helpful in showing some support for your claims.

    You do understand how the scientific method works?

  5. WKV

    “”Now we can look forward to your posting your results for the 16 x 16 blocks.””

    “Divide by 2. The WH LFBC does not show alignment on 16×16 blocks at 300 ppi, there is no particular reason to expect that it would, and therefore there is no reason to expect that any other Xerox WorkCentre file would show that alignment.”

    That’s funny because I found that the mostly text layer satisfies all three modulo conditions. And the results have been posted here for over a month:

    NBC: But not all of them… Sigh

    http://www.scribd.com/doc/151738307/Analysis-of-Rectangular-Object-Boundaries

    Of course that’s true only for an x,y coordinate system with origin point at the upper-left corner of the background page. The x axis is positive to the right and the y axis is positive from top to bottom of the screen.

  6. NBC

    “Still arguing a strawman. I accept your WH LFBC analysis, which matches the one I had posted before. You claim that since a few layers match 16×16, this needs to be a rule as well, but fail to understand why it fails several times.”

    The critical image layer is the mostly text layer. All three modulo conditions are satisfied for this image layer and several others. Most of the layers which do not satisfy the 16 x 16 block alignment are not critical as to their positions. Examples of these are the State Registrar’s signature and date stamps and the two white spot object layers.

  7. That’s funny because I found that the mostly text layer satisfies all three modulo conditions. And the results have been posted here for over a month:

    Given that it meets the first modulo condition, there is a 25% chance that it will meet the other two (and if it meets one of the other two, it’s gauranteed to meet all three).

    Since there are 9 objects, we expect 2-3 to meet all three conditions. There is no significance to the 16 mod @ 300 condition or the 8 mod @ 150 condition, beyond the 8 mod @ 300 condition.

  8. Since there are 9 objects, we expect 2-3 to meet all three conditions. There is no significance to the 16 mod @ 300 condition or the 8 mod @ 150 condition, beyond the 8 mod @ 300 condition.

    Yeah but this goes way over Hermitian’s head.

  9. So Hermitian has found one layer that meets all three but no evidence that ALL meet the mod 16 condition.

    Such a fool

  10. W. Kevin Vicklund says:

    August 19, 2013 at 04:19

    “That’s funny because I found that the mostly text layer satisfies all three modulo conditions. And the results have been posted here for over a month:

    “Given that it meets the first modulo condition, there is a 25% chance that it will meet the other two (and if it meets one of the other two, it’s gauranteed to meet all three).

    “Since there are 9 objects, we expect 2-3 to meet all three conditions. There is no significance to the 16 mod @ 300 condition or the 8 mod @ 150 condition, beyond the 8 mod @ 300 condition.

    HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH
    How soon you conveniently forget !

    I have already posted this twice before. And it’s from a link that YOU posted.

    http://en.wikipedia.org/wiki/JPEG

    “Lossless editing[edit source]

    “See also: jpegtran and Commons:User:Cropbot

    “A number of alterations to a JPEG image can be performed losslessly (that is, without recompression and the associated quality loss) as long as the image size is a multiple of 1 MCU block (Minimum Coded Unit) (usually 16 pixels in both directions, for 4:2:0 chroma subsampling). Utilities that implement this include jpegtran, with user interface Jpegcrop, and the JPG_TRANSFORM plugin to IrfanView.

    “Blocks can be rotated in 90 degree increments, flipped in the horizontal, vertical and diagonal axes and moved about in the image. Not all blocks from the original image need to be used in the modified one.

    “The top and left edge of a JPEG image must lie on a 8 × 8 pixel block boundary, but the bottom and right edge need not do so. This limits the possible lossless crop operations, and also prevents flips and rotations of an image whose bottom or right edge does not lie on a block boundary for all channels (because the edge would end up on top or left, where – as aforementioned – a block boundary is obligatory).

    “When using lossless cropping, if the bottom or right side of the crop region is not on a block boundary then the rest of the data from the partially used blocks will still be present in the cropped file and can be recovered.

    “It is also possible to transform between baseline and progressive formats without any loss of quality, since the only difference is the order in which the coefficients are placed in the file.

    “Furthermore, several JPEG images can be losslessly joined together, as long as the edges coincide with block boundaries. jpeg supports 12-bit and 32-bit color as RGB.”

    This is why the forger applied his compression after he rotated each image by 90 degrees counterclockwise and before he placed each image within a new PDF document in Adobe Illustrator. So then when he applied the 90 degree clockwise rotation in Illustrator the image was not affected.

    It’s really not complicated.

  11. This is why the forger applied his compression after he rotated each image by 90 degrees counterclockwise and before he placed each image within a new PDF document in Adobe Illustrator. So then when he applied the 90 degree clockwise rotation in Illustrator the image was not affected.

    Let us count the ways in which this fails:
    1) Only three of the objects actually meet the 2nd and 3rd modulo criteria – expected if only the first was required, an abject failure on the part of the ‘forger’ if all three were required.
    2) This only applies to jpegs. Only the background image is a jpeg – and it doesn’t meet the 2nd or 3rd modulo criteria.
    3) This only applies to the internal layout of a jpeg, not the external alignments of jpegs with respect to each other. None of the foreground images have dimensions that meet any of the three criteria, and the back ground does not meet mod 8 or mod 16 criteria for its native resolution of 150 ppi. If this were necessary to rotate the images without affecting them, it was an abject failure on the part of the ‘forger’.
    4) This only applies to editing a jpeg. If you open a jpeg, edit it, and then save the changes in a jpeg format, then the 8 mod and 16 mod criteria are required. But the rotation of the jpeg is not being saved to the jpeg. The image extracted from the jpeg encoding is being shifted and rotated. but the changes are not being saved back to that jpeg bitstream.
    5) The Preview versions of the trial scans do not meet the 2nd and 3rd modulo criteria, yet the image is not affected.

    I’m sure there are others, but the point remains: lossless editing fails as an explanation for what was done, unless the ‘forger’ was incompetent.

  12. W. Kevin Vicklund says:

    August 19, 2013 at 13:55

    “This is why the forger applied his compression after he rotated each image by 90 degrees counterclockwise and before he placed each image within a new PDF document in Adobe Illustrator. So then when he applied the 90 degree clockwise rotation in Illustrator the image was not affected.

    “Let us count the ways in which this fails:
    1) Only three of the objects actually meet the 2nd and 3rd modulo criteria – expected if only the first was required, an abject failure on the part of the ‘forger’ if all three were required.
    2) This only applies to jpegs. Only the background image is a jpeg – and it doesn’t meet the 2nd or 3rd modulo criteria.
    3) This only applies to the internal layout of a jpeg, not the external alignments of jpegs with respect to each other. None of the foreground images have dimensions that meet any of the three criteria, and the back ground does not meet mod 8 or mod 16 criteria for its native resolution of 150 ppi. If this were necessary to rotate the images without affecting them, it was an abject failure on the part of the ‘forger’.
    4) This only applies to editing a jpeg. If you open a jpeg, edit it, and then save the changes in a jpeg format, then the 8 mod and 16 mod criteria are required. But the rotation of the jpeg is not being saved to the jpeg. The image extracted from the jpeg encoding is being shifted and rotated. but the changes are not being saved back to that jpeg bitstream.
    5) The Preview versions of the trial scans do not meet the 2nd and 3rd modulo criteria, yet the image is not affected.

    “I’m sure there are others, but the point remains: lossless editing fails as an explanation for what was done, unless the ‘forger’ was incompetent.”

    HHHHHHHHHHHHHHHHHHHHHHHHHHHHH
    Boy ! ow wrong can one person be ? You see the reason that your entire comment is worthless is that we don’t know the file format of the nine individual image files as they were created one-at-a-time by the human forger. The files may have all been JFIF files. JFIF files can be compressed with several different filters. I lean toward this possibility because of the easy availability of the JPG_TRANSFORM plugin to IrfanView. If the forger created his images in Irfanview, then he would most likely have also had access to the JPG_TRANFORM plugin. Otherwise he would be a dummy for not downloading the Irfanview plugin package.

    With this plugin he could then apply his counterclockwise rotation losslessly and then compress each JFIF image with whichever filters that he had on hand.
    Then given the 16 x 16 block alignment he could then apply the 90 clockwise rotation in Illustrator after he placed each JFIF into the same new PDF file. He then scaled each image and embedded each to sever all links to the external JFIF files.

    So your assumption that each individual image was created as a PDF or some other (bitmap format) is yet to be verified.

  13. NBC

    “and the back ground does not meet mod 8 or mod 16 criteria for its native resolution of 150 ppi.”

    Wrong again Vicklund ! The background also satisfies the 8 x 8 blocks for the upper-left origin point. Just because you and NBC insist on using the wrong coordinates doesn’t alter the facts.

    It’s about time that you and NBC actually read my carefully written report which is posted on Scribd. It also includes the background image which you and NBC conveniently leave out of all your tables.

  14. Wrong again Vicklund ! The background also satisfies the 8 x 8 blocks for the upper-left origin point. Just because you and NBC insist on using the wrong coordinates doesn’t alter the facts.

    ROTFL… Are you still struggling with your analysis here? And yes, the background trivially satisfies the 8×8 block because that’s the one for which the image was padded…

    But there is no consistent 16×16 alignment and if you understood why there is an 8×8 bit alignment, you should also understand why this is the case.

    Your ‘carefully written report’ shows full support that there is no consistent 16×16 bit alignment.

    How are you doing on your 7535 alignment analysis btw?
    Figured out what you did wrong?

  15. Hermetian: It’s really not complicated.

    You forget a simple observation: The WH LFBC does not abide by your ‘rule’ of 16×16 bit alignment. And if you understood the complete workflow, you too would have known why.

    But it does not matter, your own work shows that there is no consistent 16×16 bit alignment at 300 PPI.

  16. The critical image layer is the mostly text layer. All three modulo conditions are satisfied for this image layer and several others. Most of the layers which do not satisfy the 16 x 16 block alignment are not critical as to their positions. Examples of these are the State Registrar’s signature and date stamps and the two white spot object layers.

    ROTFL, Hermitian admits that the 16×16 alignment is a figment of his imagination and that it applies only sometimes… I doubt that he understand why… Vicklund explained…

  17. 3) This only applies to the internal layout of a jpeg, not the external alignments of jpegs with respect to each other. None of the foreground images have dimensions that meet any of the three criteria, and the back ground does not meet mod 8 or mod 16 criteria for its native resolution of 150 ppi. If this were necessary to rotate the images without affecting them, it was an abject failure on the part of the ‘forger’.

    Wrong again Vicklund ! The background also satisfies the 8 x 8 blocks for the upper-left origin point. Just because you and NBC insist on using the wrong coordinates doesn’t alter the facts.

    Coordinates are irrelevant to the dimensions of an object. Remember this?

    The top and left edge of a JPEG image must lie on a 8 × 8 pixel block boundary, but the bottom and right edge need not do so. This limits the possible lossless crop operations, and also prevents flips and rotations of an image whose bottom or right edge does not lie on a block boundary for all channels (because the edge would end up on top or left, where – as aforementioned – a block boundary is obligatory).

    Which of the 9 objects have all four edges on an 8×8 pixel block boundary? Those whose dimensions (in pixels) satisfy the mod 8 condition. I have listed each object, its dimensions (in pixels), and the dimensions divided by 8. Only whole numbers satisfy the mod 8 condition (bolded).

    Img0 1652×1276 206.5×159.5
    Img1 1454×1819 181.75×227.375
    Img2 199×778 24.875×97.25
    Img3 42×274 5.25×34.25
    Img4 123×228 15.375×28.5
    Img5 47×216 5.875x27
    Img6 34×70 4.25×8.75
    Img7 243×217 30.375×27.125
    Img8 132×142 16.5×17.75

    Not only do none of these satisfy mod 8 for both dimensions, only 1 manages to satisfy it for one dimension!

  18. Perhaps Hermie is unaware that the 8×8 block boundaries are internal to each jpeg and can’t be imposed by external coordinate systems? It’s like he thinks there’s a global ruler that controls the block boundaries of each jpeg in a pdf.

  19. It also includes the background image which you and NBC conveniently leave out of all your tables.

    Oh no! The background image aligns with itself! Surely my argument is destroyed! Whatever shall I do!

    That was sarcasm, btw.

  20. Then given the 16 x 16 block alignment he could then apply the 90 clockwise rotation in Illustrator after he placed each JFIF into the same new PDF file. He then scaled each image and embedded each to sever all links to the external JFIF files.

    He could do this even without the alignment. The external alignment has no relation to internal rotation.

  21. He could do this even without the alignment. The external alignment has no relation to internal rotation.

    It will take Hermitian some time to understand this.

  22. He hasn’t even figured out the clipping path (Illustrator name is clipping mask), despite being told at least once a week for the past two months where to find it.. You expect him to understand this?

  23. You expect him to understand this?

    Well, who knows. I am doing my best helping him understand things that you and I have learned in the last few months. It is not that hard to look up the PDF and JPEG format and understand what’s inside the file.

    But applying knowledge to new situations can be tricky and Hermitian needs his time. Eventually, he too will get it. I am certain of that.

  24. W. Kevin Vicklund says:

    “August 19, 2013 at 13:55

    “”This is why the forger applied his compression after he rotated each image by 90 degrees counterclockwise and before he placed each image within a new PDF document in Adobe Illustrator. So then when he applied the 90 degree clockwise rotation in Illustrator the image was not affected.””

    “Let us count the ways in which this fails:
    1) Only three of the objects actually meet the 2nd and 3rd modulo criteria – expected if only the first was required, an abject failure on the part of the ‘forger’ if all three were required.

    HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH
    That’s three more than you or NBC have ever produced with your Xerox forger. And that for sure includes the files “wh-lfbc-scanned-xerox-7535-wc.pdf” and “wh-lfbc-scanned-xerox-7535-wcpreview -Shifted Origin.pdf”.

    You are dead in the water unless you can duplicate the precision that the forger achieved in placing his images.

    So you have no answer for a human forger having created all nine images in a computer graphics program.

    First time out of the box the human forger created a PDF image with greater precision in the placement of his images than you and NBC have ever produced using your Xerox / Preview forger.

    Here are the three Boundary Alignment Tables for the WH LFCOLB.

    Resolution = 300 PPI X 300 PPI / 8 X 8 Blocks
    Top and Right Sides Obey 8 MOD 0
    Left and Bottom Sides Align with Grid Lines
    (and also Touch Pixels)
    Layer
    N (x,y)(w,h)(x+w,y)((x+w)/8,y/8)
    1 (0,0) (2552,3304) (2552,0) (319,0) Background
    2 (373,880) (1819,1454) (2192,880) (274,110) Mostly Text
    3 (1270,2848) (778,199) (2048,2848) (256,356) Onaka Signature
    4 (710,2928) (274,42)(984,2928) (123,366) Onaka Date
    5 (1836,2160) (228,123) (2064,2160) (258,270) Reg. Gen. Date
    6 (432,2240) (216,47) (648,2240) (81,280) Loc. Reg. Date
    7 (1458,1960) (70,34) (1528,1960) (191,245) Non
    8 (735,2528) (217,243) (952,2528) (119,316) Bottom Speckle
    9 (1050,32) (142,132) (1192,32) (149,4) Top Speckle

    Resolution = 300 PPI X 300 PPI / 16 X 16 Blocks Layer
    N (x,y)(w,h)(x+w,y)((x+w)/16,y/16)
    1 (0,0) (2552,3304) (2552,0) (159.5,0) Background
    2 (373,880) (1819,1454) (2192,880) (137,55) Mostly Text
    3 (1270,2848) (778,199) (2048,2848) (128,178) Onaka Signature
    4 (710,2928) (274,42)(984,2928) (61.5,183) Onaka Date
    5 (1836,2160) (228,123) (2064,2160) (129,135) Reg. Gen. Date
    6 (432,2240) (216,47) (648,2240) (40.5,140) Loc. Reg. Date
    7 (1458,1960) (70,34) (1528,1960) (95.5,122.5) Non
    8 (735,2528) (217,243) (952,2528) (59.5,158) Bottom Speckle
    9 (1050,32) (142,132) (1192,32) (74.5,2) Top Speckle

    Resolution = 150 PPI X 150 PPI / 8 X 8 Blocks
    Layer
    N (x,y)(w,h)(x+w,y)((x+w)/8,y/8)
    1 (0,0) (1276,1652) (1276,0) (159.5,0) Background
    2 (186.5,440) (909.5,727) (1096,440) (137,55) Mostly Text
    3 (635,1424) (389,99.5) (1024,1424) (128,178) Onaka Signature
    4 (355,1464) (137,21)(492,1464) (61.5,183) Onaka Date
    5 (918,1080) (114,61.5) (1032,1080) (129,135) Reg. Gen. Date
    6 (216,1120) (108,23.5) (324,1120) (40.5,140) Loc. Reg. Date
    7 (729,980) (35,17) (764,980) (95.5,122.5) Non
    8 (367.5,1264) (108.5,121.5) (476,1264) (59.5,158) Bottom Speckle
    9 (525,16) (71,66) (596,16) (74.5,2) Top Speckle

    HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH

    2) This only applies to jpegs. Only the background image is a jpeg – and it doesn’t meet the 2nd or 3rd modulo criteria.

    HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH

    You don’t know the format of each image file as it was initially created by the forger. For example if you save a bitmap in JFIF format from Photoshop CC it places the file type label Exif (rather that JFIF) in line one. I have personally extracted the background layer as JPEG, PSD, BMP and PNG format. It also depends on the setting of the MCU block (Minimum Coded Unit). The MCU can either be 8 or 16 depending on the chroma subsampling. There is no reason that the MCU could be 8 x 8 blocks for the background image and 16 x 16 for the non-background images.

    Just remember that 16 x 16 boundary alignment is a precision which is out of reach for your Xerox forger.

  25. Just remember that 16 x 16 boundary alignment is a precision which is out of reach for your Xerox forger.

    ROTFL… Our friend is still so clueless.

    Remember that you will expect certain 16×16 bit alignment randomly and the fact that your forger could not even manage to get it right, shows that 16×16 bit is not a requirement.

    You are so clueless my friend.

    And the MCU is only relevant to JPEG… The others are clearly bitmaps and show no evidence of any jpeg artifacts. JPEG is a poor image solution for text as anyone should have known.

    If only our friend had the ability to look at the raw PDF and he would know how foolish his claims are.

    And yet the MCU was likely 16×16 at 600 DPI, the resolution with which the WH LFBC was scanned… This translates to 8×8 at 300 DPI.

    You’re so funny but have forgotten to look carefully at the specs for the Xerox scanner.

    The question you have to ask yourself is simple: Why is there no 16×16 bit alignment in most of the instances at 300 DPI. The answer is trivial and you can even calculate how many coincidental alignments you would expect.

    Come on Hermitian, the WH LFBC shows that 16×16 bit is not a requirement @300 ppi , and we see the same in the Xerox workflow.

    You were so close to figuring it out but your need to have a ‘forger’ for which no evidence has ever been presented, has made it impossible for you to explore more rational alternatives.

    I have personally extracted the background layer as JPEG, PSD, BMP and PNG format. I

    Sure, you can convert it to whatever you want but the PDF stores it as JPEG encoding and pure bitmaps for the foregrounds.

    Now if you only had access to better tools.

  26. Rats this comment editor doesn’t even do tabs.

    Operator error. If you had only taken time to familiarize yourself with how to properly enter text in the WordPress comment interface… Stop blaming others my friend, it’s not that hard to do it correctly.

  27. That’s three more than you or NBC have ever produced with your Xerox forger. And that for sure includes the files “wh-lfbc-scanned-xerox-7535-wc.pdf” and “wh-lfbc-scanned-xerox-7535-wcpreview -Shifted Origin.pdf”.

    You are dead in the water unless you can duplicate the precision that the forger achieved in placing his images.

    So you have no answer for a human forger having created all nine images in a computer graphics program.

    First time out of the box the human forger created a PDF image with greater precision in the placement of his images than you and NBC have ever produced using your Xerox / Preview forger.

    Yawn. 7535 alignment for the mod 16@ 300 ppi condition. The mod 8@ 150 ppi condition is mathematically equivalent. Mod 16 criteria met indicated by bold.

                              top x  top y  x_px y_px  ofx  ofy ofx/16 ofy/16
    Mostly text              236.64  96.72   986  403  400  408   25    25.5
    signature                 71.52 317.52   298 1323  312 1328   19.5  83
    Date Apr 25               88.8  175.44   370  731  384  736   24    46
    Block right Oahu/African 369.12 375.12  1538 1563 1552 1568   97    98
    Aug 8 1961 Right         248.16 434.64  1034 1811 1048 1817   65.5 113.5
    Aug 8 1961 left          246.24 110.16  1026  459 1040  464   65    29 
    Above left date          280.8   83.28  1170  347 1184  352   74    22
    Lower right blocks       246.24 430.8   1026 1795 1040 1800   65   112.5
    None                     315.36 346.32  1314 1443 1328 1448   83    90.5
    Maternity                444    179.28  1850  747 1864  752  116.5  47
    Kapiolani                449.76 112.08  1874  467 1888  472  118    29.5
    Right Top                737.76 484.56  3074 2019 3088 2024  193   126.5
    Below right top          687.84 490.32  2866 2043 2880 2048  180   128
    Triplet                  488.16 200.4   2034  835 2048  840  128    52.5
    Below Center Top         735.84 298.32  3066 1243 3080 1248  162.5  78

    5 objects meet the condition, not including the background image (which by definition meets all modulo conditions). #Expected if only mod 8@ 300 ppi condition is required: 16*.25 = 4.

    Another Hermie fail.

  28. Excellent calculations Kevin, I was going to do the same. Our poor friend has no clue… And Hermitian, note how Kevin manages to properly align and bold his information…

    Again, he seems to be way ahead of you

  29. “And yet the MCU was likely 16×16 at 600 DPI, the resolution with which the WH LFBC was scanned… This translates to 8×8 at 300 DPI.”

    HHHHHHHHHHHHHHHHHHHHHHHHHH
    Nice try but you still fail to deliver the 16 x 16 blocks (achieving 16 MOD 0) in even one PDF.

    Looks like another show stopper for NBC’s folly.

    That 600 DPI scanning resolution is an assumption on your part. Even if the WH LFCOLB was created by scanning a paper document (which it wasn’t) you wouldn’t know the DPI unless you are (or at least know) the Xerox operator.

    Other wordpress sites don’t have the problems that you allow to go unattended. I suspect that you have tailored your site to facilitate your posting of meaningless long printouts as images and other graphics such as screenshots and photographs.

    You still haven’t produced a single PDF image with multiple layers which satisfy all three modulo conditions for any single layer.

    If you could produce just one example of 16 MOD 0… But I guess your dumb machine just doesn’t understand your instructions. You have completed dozens of trial scans using every conceivable placement orientation of the original and have failed to produce even one PDF that duplicates the placement precision of my human forger. And he accomplished the task on the first attempt.

  30. Nice try but you still fail to deliver the 16 x 16 blocks (achieving 16 MOD 0) in even one PDF.

    Time for Hermie’s theme song! Bum-bum-badUmm, BWWWaaaa…..

  31. (Sorry, it was the Price is Right failure sound that I had meant to propose as Hermie’s theme song. Can I get a sad trombone?)

  32. You still haven’t produced a single PDF image with multiple layers which satisfy all three modulo conditions for any single layer.

    Because there is no requirement for this as the WH LFBC does not show all three conditions being satisfied. What I and Kevin have done is shown how mere chance creates the right number of alignments expected.

    As to the 600 PPI that is the standard scanning resolution for Xerox WorkCentre scanners. And my explanation, once again, matches all the observations.

    Such fail

  33. NBC says:

    August 20, 2013 at 03:53

    “”You still haven’t produced a single PDF image with multiple layers which satisfy all three modulo conditions for any single layer.””

    “Because there is no requirement for this as the WH LFBC does not show all three conditions being satisfied. What I and Kevin have done is shown how mere chance creates the right number of alignments expected.”

    “As to the 600 PPI that is the standard scanning resolution for Xerox WorkCentre scanners. And my explanation, once again, matches all the observations.”

    Matches all the observations except for the only one that matters which is the WH LFCOLB.

    So you are actually claiming that the Xerox created the WH LFCOLB by scanning a paper original but it can never produce the same precision image from a copy ?

    That’s really lame NBC.

    And I also remember when you swore that a 300 DPI print produces a 300 PPI image.

    Now you are singing a different tune. And that is a 600 DPI print produces a 300 PPI image.

    Were you lying then or now NBC ?

  34. So you are actually claiming that the Xerox created the WH LFCOLB by scanning a paper original but it can never produce the same precision image from a copy ?

    That’s really lame NBC.

    And I also remember when you swore that a 300 DPI print produces a 300 PPI image.

    Now you are singing a different tune. And that is a 600 DPI print produces a 300 PPI image.

    Were you lying then or now NBC

    That’s it my friend, next time you call me a liar, you are no longer welcome. Perhaps I may have said something like that, perhaps you misunderstood what I said, but you are under the flawed impression that one cannot change one’s position based on the available evidence.

    And the fact that

    1. The default scanning resolution of the Xerox is 600 ppi
    2. There is consistent MCU alignment (16 bits) at 600 ppi, which causes the 8×8 bit alignments at 300 ppi

    All support my workflow, and are consistent with what the LFBC released by the Whitehouse shows.

    We can of course, never recreate the PDF with the same precisions, but we can show that all features are present or can be explained by the Xerox work flow.

    Consistent alignment at 8 bit boundaries at 300 ppi
    Inconsistent alignment at 16 bit boundaries at 300 ppi

    Both are seen in the WH LFBC and the Xerox WorkCentre scans.

    So before you accuse someone of lying, you may want to ask yourself some questions. You apparently failed to do so and now have to defend the indefensible: the accusation of a lie.

    Shame on you my friend for having sunk so deep.

  35. So you are actually claiming that the Xerox created the WH LFCOLB by scanning a paper original but it can never produce the same precision image from a copy ?

    That’s really lame NBC.

    And I also remember when you swore that a 300 DPI print produces a 300 PPI image.

    Now you are singing a different tune. And that is a 600 DPI print produces a 300 PPI image.

    Were you lying then or now NBC

    That’s it my friend, next time you call me a liar, you are no longer welcome. Perhaps I may have said something like that, perhaps you misunderstood what I said, but you are under the flawed impression that one cannot change one’s position based on the available evidence.

    And the fact that

    1. The default scanning resolution of the Xerox is 600 ppi
    2. There is consistent MCU alignment (16 bits) at 600 ppi, which causes the 8×8 bit alignments at 300 ppi

    All support my workflow, and are consistent with what the LFBC released by the Whitehouse shows.

    We can of course, never recreate the PDF with the same precisions, but we can show that all features are present or can be explained by the Xerox work flow.

    Consistent alignment at 8 bit boundaries at 300 ppi
    Inconsistent alignment at 16 bit boundaries at 300 ppi

    Both are seen in the WH LFBC and the Xerox WorkCentre scans.

    So before you accuse someone of lying, you may want to ask yourself some questions. You apparently failed to do so and now have to defend the indefensible: the accusation of a lie.

    Shame on you my friend for having sunk so deep.

  36. I have finished checking the Object boundaries for two of the objects contained within the Xerox 7535 PDF “wh-lfbc-scanned-xerox-7535-wc.pdf”. This PDF file is one of the two files that NBC has released. The other PDF of these two is the Preview PDF file obtained by opening this first PDF file in Preview, rotating the image and then printing the resulting portrait image to a second PDF image file.

    NBC has not released any scan to PDF files produced by the Xerox 7655 Workcenter.

    I have checked only the mostly text layer and the Onaka signature stamp for the PDF file “wh-lfbc-scanned-xerox-7535-wc.pdf”. The results from the analysis of these two objects are representative of the other objects.

    The results are for the case of the global origin point (of the screen x,y coordinates) set to the upper-left corner of the background image page in Adobe Illustrator CC. This page is not the clipped page that is reported by Adobe Acrobat XI Preflight. The certificate image is in the landscape orientation such that a 90 degree counterclockwise rotation is required to obtain the same orientation as the WH LFCOLB. Equivalent results for an upper-right origin position are also given in the second table. The screenshots from Illustrator which graphically verify the reported results are contained in one five-page PDF document which can be downloaded from scribd.com here:

    The Top and left sides of each rectangle satisfy the 8 MOD 0 condition.

    The results are summarized in the following table for the upper-left origin position.

    Layer x y W H x/8 y/8

    Mostly 1000 408 1403 1749 125 51
    Text

    Onaka 312 1328 196 694 39 166
    Signature

    The results for the upper-right origin position are given in the table below.

    Layer x y W H x/8 y/8

    Mostly 2328 408 1403 1749 291 51
    Text

    Onaka 3016 1328 196 694 377 166
    Signature

    Hence we find that only the residual 166 is divisible by two. Thus, only this one side of Onaka’s signature object rectangle meets both the 8 x 8 and the 16 x 16 modulo conditions.

    The first page of the five-page PDF posted on Scribd is the Xerox 7535 scan to PDF image.

    Page two of this same report shows the graphical analysis (with the origin point set at the upper-left corner of the background page) for the mostly text layer. The left-hand window shows the upper-left corner of the mostly text rectangle highlighted in Blue. The grid consists of 8 x 8 blocks (major grid lines) and 300 PPI x 300 PPI resolution (minor grid lines). As shown the top and left sides of the rectangle are congruent with two intersecting major grid lines. This is graphical proof that the 8 x 8 block alignment (i.e. 8 MOD 0) is satisfied. The right hand window on this same page is the lower right hand corner of the same rectangle also highlighted in Blue. The lower-right corner is not congruent with the major grid lines.

    Page three is the same two figures for the Onaka Signature Rectangle. Again the top and left sides are congruent with major grid lines but the bottom and right sides are not.

    Pages four and five are the same as pages two and three except the grid blocks are 16 x 16 (major grid lines) and 300 PPI x 300 PPI resolution (minor grid lines).

    In this case, we find that only the top edge of Onaka’s signature rectangle is congruent with the 16 x 16 major grid lines. Thus only this top side satisfies both the 8 MOD 0 and the 16 MOD 0 conditions. However the left side only satisfies the 8 MOD 0 condition.

    The setting of the origin point involved two steps because of the overhang of the background page to the artboard. The artboard was re-sized to match the background page dimensions and then the entire composite image was moved to place the origin point at the point of intersection of x and y major grid lines.

  37. So Hermitian once again confirms that the 8×8 pixel alignment is universal and that the 16×16 alignment is guided by chance (1 in 4 chance that both x and y are on an 16×16 bit boundary)

    Congratulations…

    Thanks for verifying my results my dear friend. Again, my workflow stands a little bit stronger because of your analysis.

  38. NBC

    “”And I also remember when you swore that a 300 DPI print produces a 300 PPI image.””

    “”Now you are singing a different tune. And that is a 600 DPI print produces a 300 PPI image.””

    HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH

    So which is it NBC ? As usual you never answered my question.

    The default orientation for single-page originals is right-side-up. You seem to celebrate an up-side-down goof by the Xerox operator but the same operator could not have possibly set the DPI to other than the default 600 DPI.

    Not exactly consistent logic NBC.

    This is simply you making up the facts to fit your storyline.

    Now how many rotations of the original and the created images does your Xerox friend require today?

  39. The default orientation for single-page originals is right-side-up. You seem to celebrate an up-side-down goof by the Xerox operator but the same operator could not have possibly set the DPI to other than the default 600 DPI.

    I let the data guide me, not my wishes. And I do not believe that your ignorance or failures to understand have any relevance to my work flow.

    Can you tell me how you set the DPI on a Xerox WorkCentre?

    The maximum resolution is 600×600 dpi and the workcentre will downsample if necessary, but it makes sense that MRC happens at the highest resolution.

  40. And there is a big difference between upside down scanning and setting the resolution to a different setting.

  41. NBC says:

    August 21, 2013 at 00:13

    “And there is a big difference between upside down scanning and setting the resolution to a different setting.”

    HHHHHHHHHHHHHHHHHHHHHHHHHHH
    I agree ! To get your upside-down orientation requires a drunk paralegal who has never used a Xerox Workcenter. And we are talking about a one-page original here.

    To set a scan resolution to other than the default 600 DPI requires an intentional act by the operator rather than a dumb goof.

  42. And I also remember when you swore that a 300 DPI print produces a 300 PPI image.

    No, he said that you can’t print a 300 PPI image at a higher DPI and get any better resolution than 300 PPI.

    Now you are singing a different tune. And that is a 600 DPI print produces a 300 PPI image.

    Scan, not print, and look up the term “downsample.”

  43. I agree ! To get your upside-down orientation requires a drunk paralegal who has never used a Xerox Workcenter. And we are talking about a one-page original here.

    Yes, and it is not that uncommon. It’s again what the data tell us my friend.

    Simple logic.

  44. To set a scan resolution to other than the default 600 DPI

    Where does it say that 600 DPI is the default setting?

  45. No, he said that you can’t print a 300 PPI image at a higher DPI and get any better resolution than 300 PPI.

    Ah that makes sense..

    Now you are singing a different tune. And that is a 600 DPI print produces a 300 PPI image.

    You are confused the scan was sampled at 600 ppi and down sampled. This is beginners stuff my friend.

  46. No, he said that you can’t print a 300 PPI image at a higher DPI and get any better resolution than 300 PPI.

    Ah that makes sense..

    Hermitian: Now you are singing a different tune. And that is a 600 DPI print produces a 300 PPI image.

    You are confused the scan was sampled at 600 ppi and down sampled. This is beginners stuff my friend.

  47. Where does it say that 600 DPI is the default setting?

    It’s not the default per se but rather the max resolution for scanning. When doing MRC you would do the separation at the highest resolution before downsampling.

  48. The setting of the origin point involved two steps because of the overhang of the background page to the artboard. The artboard was re-sized to match the background page dimensions and then the entire composite image was moved to place the origin point at the point of intersection of x and y major grid lines.

    Of course, if we moved things around, Hermie would go batshit insane over it. Just look at how upset he was that NBC hovered over anchor points to get their exact coordinates.

  49. So you are actually claiming that the Xerox created the WH LFCOLB by scanning a paper original but it can never produce the same precision image from a copy ?

    Yes. Yes. A thousand times, YES!

    See also, down-sampling, lossy compression, scanner artifacts, printer color calibration.

  50. Of course, if we moved things around, Hermie would go batshit insane over it. Just look at how upset he was that NBC hovered over anchor points to get their exact coordinates.

    Oh he surely wished he had thought of that🙂

  51. See also, down-sampling, lossy compression, scanner artifacts, printer color calibration.

    Does Hermitian really expect a perfect scan every time? Seems he has not been doing many experiments in that area.

Comments are closed.