Hermitian: NBC totally overlooks the fact that his Xerox scans to PDF consistently do a poor job of separating the monochrome text from the color background. Much more of the text characters “sticks” to the background in the Xerox scans than the human forger left on the background layer of the WH LFCOLB. Unfortunately NBC is too dumb to realize that any text that does not get lifted is downsampled from 300 PPI X 300 PPI to 150 PPI x 150 PPI. Moreover the text color is changed from binary to Grayscale and the aliased image is converted to anti-aliased. Now NBC evidently believes that this mangling of the image is all accomplished without any loss of information. Talk about missing the forest for the trees !
So that is his excuse, that the Workcentre does not do a perfect enough separation… Wow… As to the downsampling and what our friend erroneously calls aliasing were all described by me way before he had ever considered this. Of course, any time a document is ‘compressed the level of the WH LFBC, it is self evident that information is lost. But preview is not the cause of loss of this information. Other than perhaps the fact that it overwrote some metadata and rewrote the internals. If that’s the argument then yes, it is a miracle, is it not, that we still have been able to track it back to a Xerox workflow…
Another great strawman… I have mentioned the separation issues, I have addressed the downsampling, I have addressed the cause of ‘aliasing’, even before I had identified the Xerox WorkCentre. Anyone understanding MRC could have explained why Zebest’s claims ignored the effects of MRC, which includes downsampling, JBIG2 encoding and JPEG encoding of the background.
Hermitian has served his purpose but he has failed to raise any real objections to my well document work flow.
Oh and Hermitian is also rewriting history about me reporting that my Xerox WorkCentre copies did not match up with the quantization tables. Such is the burden of a scientist but to report on all the data. What Hermitian sees as a weakness, is in fact a strength to those interested in applying the scientific method to establish a theory. Of course, our poor friend forgot that I, soon thereafter, was informed that the 7535 tables matched and, that I, on a hunch, analyzed the WH 7655 Tax forms which also showed a matching Quantization Matrix.
Which led me to hypothesize that the differences can be explained by differences in firmware. But that hypothesis adds no degrees of freedom, as I base my conclusion on the actual Xerox WorkCentre found owned by the White House and used to scan in President Obama’s Tax Returns.
Again our poor friend overlooks all the evidence to propose his ill supported ‘forger’ who just happens to mimic a work centre…
For that I thank him, we could not have achieved this success without him and he should be remembered for it.