WH LFBC – YCbCr screen capture

Hermitian wanted a screen capture of the LFBC in my hex editor.

WH LFBC YCbCr

7 thoughts on “WH LFBC – YCbCr screen capture

  1. Think Hermie will actually find this?

    Time shall tell. So far he has managed to move not a single inch forward in his rebuttals, however he has shown some interesting insight into his methods.

  2. NBC

    These two strings of yours are different from mine. Before, I remember that you claimed this is simply a difference between the MAC OS and Windows OS that’s the reason.

    So answer this straight question.

    Preview writes the wrong string to the MAC OS PDF and then Windows opens this same PDF containing the wrong string in Illustrator.

    So why would each Windows graphics program necessarily handle this string mismatch the same way ?

    And what Google hits will we get if we search using your weird strings on a Windows OS computer ?

    And don’t say that all that matters is the HEX !

    The LeadTools literature states that JPEG compressed files that do not exactly conform to the JFIF standard are rendered upside-down.

    We know that the Xerox scan to PDF files do not conform to the JFIF standard because the JFIF standard requires the YCbCr color space. And we know that the Xerox WCs substitute the RGB for the YCbCr color space.

    So why would Xerox WCs insert the label “YCbCr” (as a comment) into a JPEG compressed bitmap and then substitute RGB for YCbCr as the color space used ? This doesn’t make sense because it is an intentional violation of the JFIF specification. Did Xerox insert the comment “YCbCr” just to fool their customers?

    You need to post the PDFs so rational people can get to the bottom of these inconsistencies.

  3. So why would Xerox WCs insert the label “YCbCr” (as a comment) into a JPEG compressed bitmap and then substitute RGB for YCbCr as the color space used ? This doesn’t make sense because it is an intentional violation of the JFIF specification. Did Xerox insert the comment “YCbCr” just to fool their customers?

    Ivan Zatkovich already identified that as a vestigial leftover accidentally left behind by some long forgotten programmer. It is analogous to the vestigial structures left behind by evolution that serve no current purpose, yet are invaluable to comparative anatomists trying to decipher the phylogeny of organisms. Much of human DNA (for example) contains “broken genes” (almost 600 for the sense of smell alone) that once were useful and now simply hang on as so much unnecessary “junk.” While biologically useless, they are of profound value to the investigator.

  4. Correction: It was Professor de Queiroz, not Ivan Zatkovitch that made that observation. From RC’s comment:

    I sent a short email with some links to Professor de Queiroz. He doesn’t really have the time to get involved. However, he agrees that the “YCbCr” comment is non-functional in the normal JPEG standard and could have just been something a programmer forgot to clean up.

  5. For reference, this is the Microsoft character set:

    $4á%ñ&’()*56789:CDEFGHIJSTUVWXYZcdefghijstuvwxyz‚ƒ„…†‡ˆ‰Š’“”•–—˜™š¢£¤¥¦§¨©ª²³´µ¶·¸¹ºÂÃÄÅÆÇÈÉÊÒÓÔÕÖ×ØÙÚâãäåæçèéêòóôõö÷øùú

    and this is the (Microsoft Word’s interpretation of) the Macintosh character set.

    $4•%Ò &'()*56789:CDEFGHIJSTUVWXYZcdefghijstuvwxyzÇÉÑÖÜáàâäíìîïñóòôö¢£§•¶ß®©™≤≥¥µ∂∑∏π∫¬√ƒ≈∆«»… “”‘’÷◊ÿŸ⁄‚„‰ÂÊÁËÈÍÚÛÙıˆ˜¯˘˙ˇ˛

    Note that these are the same HEX string, just different mappings of the HEX coding to character set.

    These two strings of yours are different from mine. Before, I remember that you claimed this is simply a difference between the MAC OS and Windows OS that’s the reason.

    Yes, Microsoft has one character set, Macintosh has another

    So answer this straight question.

    I’ll be sure to give a straight answer.

    Preview writes the wrong string to the MAC OS PDF and then Windows opens this same PDF containing the wrong string in Illustrator.

    So for example, instead of writing the HEX string “F8 F9 FA” Preview for some reason (not sure why Hermie thinks will happen, but whatever) writes the HEX string “E8 E9 EA”

    So why would each Windows graphics program necessarily handle this string mismatch the same way ?

    Presumably, these Windows graphics programs are using the Microsoft character set. So instead of displaying the expected “øùú” they will probably display “çèé”. Similarly, Macintosh graphics programs will likely display “ÈÍÚ” instead of the expected “˙ˇ˛”. It’s possible that a given program might use a different character set despite being on a specific platform, so it shouldn’t be assumed that it will be the character set of the platform. This is particularly true of browsers, which try to minimize cross-platform differences, and often will default to UTF-8 for their character set.

    And what Google hits will we get if we search using your weird strings on a Windows OS computer ?

    As every example I’ve seen in Google searches has been from a web image not properly displaying, I would assume that most of these would be UTF-8 encoded, rather than platform specific. Since the strings you’ve isolated have the same UTF-8 encoding as the Microsoft character set for those values, I wouldn’t expect to find a whole lot of Macintosh character set entries. So I would hazard that there would be very few examples in Google – someone would have to deliberately assign the Macintosh character set to their site in order for this string to show up.

    And don’t say that all that matters is the HEX !

    Well, only the HEX encoding gets saved to the file. The character set doesn’t get transmitted. Aside from a text object, the character map used by the graphics program is irrelevant to the PDF. (I’m not sure, and I am not going to look it up, but for the case of the text object, you may be able to include the character set, just like you can include the typeface)

  6. Excellent attempt to bring character set differences down to Herm’s level, WKV! Perhaps using an analogy to human languages that use different alphabets (English, Japanese, Russian, etc), or better yet, human languages that can be rendered into writing using multiple alphabets (Russian into Cyrillic or Roman, German into Fraktur or Roman, various Chinese alphabets …). You know, if the original is in English, a Japanese computer may render the characters into Kana (or one of several other written alphabets in use in Japan), the Russian computer may go with Cyrillic, but the original file remains English.

    Heck, you can fire up a browser, open a page in any given language, and cycle character sets.

    If Herm’s is really incapable of getting this, I suspect it’s yet another instance of a xenophobe birther who simply lacking in world experience … no experience with any culture or language beyond ‘Merican.

Comments are closed.