Talk:Halftone

From Wikipedia, the free encyclopedia

This article is supported by the Color WikiProject, a project that provides a central approach to Color-related subjects on Wikipedia. Help us improve articles to good and 1.0 standards; visit the wikiproject page for more details.
B This article has been rated as B-Class on the quality scale.
Mid This article has been rated as mid-importance on the importance scale.

Contents

[edit] dots get too small?

The effect has its limits, when the dots get too small or spaced too far apart, the eye starts seeing dots again.

How so? Obviously, when dots are too far apart, they will look like individual dots, but when they are too small? I'd think the smaller they get (that is, the greater the lines-per-inch of the halftone screen), the more smooth they would appear to the human eye. Can someone clarify this? -- Wapcaplet 01:28, 10 May 2004 (UTC)

When the halftone frequency is too high for the reproducing medium's resolution (be this the Laser Printer or Image Setter dpi resolution or the repro film's grain size) an effect of posterization occurs. Maybe this is what the comment is getting at. Another possibility is due to the flaw that this article refers to 'Screened halftoning' only and not other forms of halftone such as dithering. Under Floyd-Steinberg or Error-Diffusion dithered halftoning (halftoning processes used in inkjet printers) areas of low intensity (ie where the dots are spaced wide apart) have the dots much clearer to the eye (due to high contrast). This is the reason that reverse grey component replacement is used. It is also the reason for the new generation of 6 or 7 colour inkjet printers (often called photo printers). One day I will explain this clearly in the article itself, but it's a big ask as I'm new to wiki... --Bb3cxv 7 July 2005 14:15 (UTC)

[edit] halftone before digital processes

I've always been fascinated to learn how this process actually took place before digital. How did a newspaper in the 70s take a photograph and turn it into this dots pattern? Can someone add this in - a brief descrption of the actual process? TeamCoachingNetwork 06:43, 30 November 2006 (UTC)

It was originally done in about the 1880s by using closely-spaced parallel lines etched onto glass - hence the 'lines-per-inch' measurement and the 'screen ruling' term. 2 sheets of ruled glass were glued face to face at 90 degrees at first. This was called the 'screen.' Dropping an expensive glass screen was a disaster. Aligning the rules diagonally rather than at right angles would give diamond-shaped cells that generated elliptical dots, which were useful for printing certain types of image. By the 1970s when I was doing this at print college, the 'contact screens' were single cross-ruled sheets of thin clear plastic.
The screen was placed in a very large 'process' camera. It was positioned ruled-side down in close contact with a sheet of unexposed film, and the whole sandwich held flat by a vacuum plate. A photograph was then taken of the continuous tone photographic original. A high contrast 'lith' film was used, which gave essentially pure black or pure white images, ie close to binary. As the original greyscale image was projected by the lens, it fell onto the screen where it was split up by the cells into small discrete areas which formed the basis of individual dots.
The dots were round despite the square screens because of diffraction effects around the cell edges, coupled with the high contrast film/paper which ignored the grey tones due to light fall-off around the edges of the cells. The dots were therefore sharp-edged, with different sizes depending on the light intensity of the particular area of the original that they represented.
The haltoned film image was a negative, which was then placed face-down in contact with a photo-sensitive surface on the printing plate (which might be letterpress, lithography, flexography or gravure). Placing the film face-down was vital for maximum image transfer quality, but meant that the image was mirror-image reversed. To allow for this, a reversing mirror set was placed in front of the lens of the process camera, so the initial screened negative image was reversed to start with. This meant that the final plate was correct.
So actually the halftone dots were originally formed by a relatively simple optical effect and the challenge of early electronic systems was to recreate this effect digitally for scanners and laser film recorders. Simoneccles 03:48, 3 January 2007 (UTC)
Newspapers created "Screen Prints"
Many newspapers could not afford a process camera, a PMT processor, and a huge darkroom which many process cameras required. Therefore, a different halftoning process was created for newspapers. (I believe that this process was created by Kodak since most of the original supplies were available from Kodak and sold exclusively to newspapers. However, by the time I began to use this process, most people bought their screens from a Japanese company.)
Rather than take a postive image and turn it into a negative halftone, the low-budget newspaper process used the same darkroom and the same photograph enlarger to create a positive image which could be placed on a page of typesetting to be photographed by a process camera onto a printing plate (without any additional halftoning.) This positive image was called a "screen print" or "screened halftone print" or erroneously, "paper halftone".
A black and white roll of film (usually Tri-X) was developed normally. Then a negative was placed in the photographic enlarger but rather than printing onto photographic gray tone paper, the image was printed onto a high contrast paper. To create the halftone, a specially designed plastic screen was placed over the high contrast paper from Kodak and other companies. The paper and the contact screen was held in place on a tiny frame by a tiny vacuum pump or for really inexpensive operation, the screen and the high contrast paper were held down by a piece of glass with extra pressure applied by the darkroom operator. The light from the enlarger created a positive image on the high contrast paper. The image was developed with a very high contrast developer to get pure back and white dots. This was called a "Screen Print". As long as the resolution was 85 dpi or less (some low quality newspapers used 65 dpi which is the same as the shading used in newspaper cartoons), a printing plate could be photographed directly from the screen print. Some very high quality newspaper printers could capture higher resolution than 85 dpi but it was better to use a negative halftone for higher quality work. Most instant presses (using paper plates or, better yet, plastic plates) could easily use page layouts which contained screen prints without doing any halftoning of the printing plate as long as the screen print was not more than 85 dpi.
Care had to be taken in creating the screened print because of dot gain of the printing plate running on the printing press. A screen print ranged in dot sizes of roughly 15% to 85%. Any dots smaller than 15% would fade out entirely when photographed onto a printing plate leaving big white splotches in the final image from the newspaper printing press. And dots larger than 85% turned to solid black which not only looked bad but also were hard on the newspaper printing press. The dot gain was different for each newspaper printing press (and for each instant printer printing press). Calibration had to be done for each press. Robert Elliott 16:56, 23 July 2007 (UTC)
2,000 dpi resolution for positive halftones (Screened Prints made on an photographic enlarger)
A very special note: The resolution of a screen print or "positive halftone" (unlike the resolution of a negative halftone) is over 2,000 dpi. Computer programmers do not know this.
While the dots on a screen print are spaced 85 to the inch, the resolution of the center of the dot (where the half tone screen for a screened print is almost completely transparent) is the resolution of the very high contrast photographic paper (not the line resolution of the screen used to make the screened print.)
  • For big, fat dots, the resolution is only 85 dpi since the dots are big and round (the same as the halftone screen.)
  • In the center of dots (which is the highlights on a screen print and contains the most important details of the image), there is a posterized image at the center every dot smaller than about 25%. This posterized image includes a very accurate image of that portion of the dot. The resolution of that posterized image was limited only the the limits of the photographic emulsion on the halftone paper which is at least 2,000 dpi. If there was an image with a sharp line, that line will clearly be seen in the center of the dots at full 2,000 dpi resolution.
This effect is not important with negative halftones since the tiny dots of a negative halftone are the shadow areas so you only see detail in black areas where there is little detail, not sharp lines or detailed images that you see in the center of a screened print dot (the highlight.)
When computer software companies began creating software for halftoning on a computer, they programmers did not know about the screen print effect. They just assumed that a screen print was simply the reverse of a negative halftone. Not so! Programmers did not know that a postive halftone can be made which combines a half tone with a posterized image of the highlights to give an huge increase in halftone quality (2,000 dpi).
Therefore, even today, all computer software for creating positive halftone dots is wrong. Hopefully, someone will read this and correct their software to allow 2,000 dpi digital positive halftones just like the screen prints made long ago on photgraphic enlargers. Robert Elliott 16:49, 23 July 2007 (UTC)

[edit] talk page purpose

While I very much enjoy and benefit from the technical discussion, below, in my confusion I'd thought that this talk page's purpose was for writers to discuss the merits of the current wiki page. This writer, for example, longs for a definition or three of "halftone," a definition that will precede the descriptions of "halftoning" that dominate the wiki page. Walter Dufresne 20:21, 3 January 2007 (UTC) walterdufresne Brooklyn, New York, 3 January 2007

Talk pages can also be used for relevant general questions about the subject. Such questions often help improve articles as much as pointed critiques, as they show the gaps in the article's current explanations. --jacobolus (t) 10:58, 9 May 2007 (UTC)

[edit] seeking a working definition of "halftone"

The page does a wonderful job of defining, at length and in detail, such compelling processes as "halftoning." This seems good. I miss reading any kind of definition of a slightly different thing, the "halftone" of the page's title.

What is a halftone? How might we define such a thing? Is it possible to write a working definition of "halftone"?

Walter Dufresne 17:21, 6 February 2007 (UTC)Walter Dufresne, Brooklyn, NY, USA

Yes - something is missed in this article. An actual definition of "halftone", the names derivation, the invention and history of the process. This article seems to be written backwards from the modern process and doesn't seem to dig very deep. Needs a little work. Fountains of Bryn Mawr 04:37, 3 May 2007 (UTC)

[edit] Missing title in references

I have just completed a missing title in the references, but I'm not sure it is correct:

Twyman, Michael. Printing 1770-1970: an illustrated history of its development and uses in England. Eyre & Spottiswoode, London 1970.

I noticed the title was missing. Library of Congress has only two titles by this author in the year of 1970; this one and Lithography, 1800-1850: the techniques of drawing on stone in England and France and their application in works of topography.

I made an educated guess (that section deals with printing, not litography), but I'm not sure this is correct. Could someone confirm? —Preceding unsigned comment added by Miguel Vieira (talk • contribs) 21:33, 9 December 2007 (UTC)