2400 dpi and higher is only necessary for outputting medium-to-high quality dot rasters. Theoretically, one could switch back a RIP to 1200 dpi (or even slightly lower) for plain text only, and have the RIP use less memory and a faster output rate.

I'm not sure, though, if this is done in practice (perhaps there are text-only companies that use low res as default). I doubt my trusted magnifying glass can tell the difference on text offset-printed at, say, 800 dpi and 2400 dpi.

And when are 2400 dpi required, if not for plain text? I thought, that such a high resolution is required for letters mainly. I mean, what is more sensible than letters?

With regard to the direct comparison of printed text and ebook devices – what would you say, at which resolution could an ebook device beat a well made book, that is printed in a normal quality – I miss the right term, let’s call it a normal paperback, nothing exclusive and expensive. What resolution would an ebook device require, when it shall display body text in 8.5 points?

And can I buy printing examples with different resolutions up to 3600 dpi? I would like to check, if I am able to see the difference. More to the point, I would like to know the real resolution, that is required without castrating body text fonts like Sabon Next.

Just an idea: you can contact Taschen and ask the resolution their books are made. Art books with large, quality reproductions are the highest demanding print materials I know. And Taschen has an outstanding overall print quality –including texts at small sizes.

@ Freiberger

Yes. And I can ask in German. I better can express in German, what I would like to know. But it seems to be, that I made a mistake, when I had claimed, that an ebook device has round about 8% of the resolution as a "normal" print. There was a person on a non-typophile forum, who said, we should be glad, when the problem of turning the pages of printed books will be solved (by replacing printed books with ebooks on ebook devices.)

The original German wording was: "Was soll am Seitenumblätterns bitte schön sein? Das hat doch eigentlich schon immer nur genervt.
Das sind doch nur romantische Klischees, die Nachteile gedruckter
Bücher werden völlig ausgeblendet: Man braucht beide Hände dafür,
wenn man das Buch loslässt, blättern die Seiten zurück oder es klappt
zu, es kriegt Eselsohren und der Rücken wird verknittert, wenn man es
zu weit umknickt. Wir sollten froh sein, wenn wir diese Probleme los sind."

(I just remember, that you are from Brazil.)

And that was too much ignorance for me. But I am amateur. It would be better if professional typographers and printers would comment those things. My argument was, that the most important and most comfortable thing is the legibility. If an ebook device is legible and comfortable as printed book with regard to the legibility, then we should be glad, if the printed book dies. What I want to say is, I am not against ebook devices because of pure ideological reasons.

Another argument from my side was, that we should not compare books, that are printed in the lowest quality, with ebook devices. So I need a standard for the comparison between ebook devices and printed books.

But I don’t want to start another thread about the future of the book.

2400 dpi is a great resolution for an inkjet printer but, realistically, very few--if any--commercial offset printing presses print at a higher resolution that 200 dpi.

Arno, sorry but I can read just a few words in Deutsch.

Anyway, I don't think you're wrong. 96dpi is 8% from 1200dpi –and there are many prints in this resolution. Maybe you can better say that e-devices had near 16% of average print resolution (comparing to 600dpi output). And even this is a long distance.

In the case of smallish text in high-quality printing, one
can certainly see the difference between 1200 and 2400.

----

Today, learn about the Armenian Genocide.

hhp

2400 dpi is a great resolution for an inkjet printer but, realistically, very few--if any--commercial offset printing presses print at a higher resolution that 200 dpi.

You are surely confusing dpi and lpi. 2400 dpi is quite normal for imagesetter rasterisation of text (even for straight-to-plate these days); 150 lpi (lines per inch) is quite common for halftone images, but stochastic screening is now pretty widespread.

Anyone who thinks that

1200 dpi

as output by

a linotron

a laser printer

an inkjet printer

is the same

is mad.

The quality of the dots is drastically different. Imagesetter comes first, obviously, then, way down, laser, and way way down, inkjet.

There is nothing absolute about 1200, 2400 3600, etc. The quality of the dot controls all.

If we're talking 'normal paperbacks' I'd say ebook readers can easily give them a run for the money. Many paperbacks are printed on paper that I'd say is only a step above newsprint which, in turn, produces rather nasty ink bleed and inconsistencies.

The other thing is that thanks to both anti-aliasing and sub-pixel rendering, at the same dpi a screen is far superior to print. Obviously, the resolutions are not the same, but I'm thinking that if I had 40-50% the resolution on screen as in print, text would be about equally good.

Of course, at 8% we would have a way to go yet.

Cheers,

T

The quality of the dots is drastically different. Imagesetter comes first, obviously, then, way down, laser, and way way down, inkjet.

There is nothing absolute about 1200, 2400 3600, etc. The quality of the dot controls all.

Okay, a dot is not a dot. In case of inkjets I know, that the quality of the dots is very dependent from the paper. But that’s not all; it’s also the way, in which the dots are set, which seems to be partly dependent from the printer driver.

With regard to electronic devices Thomas already said, that the antialiasing improves the legibility. But the extra dots (pixel in this case) are still pixel.

So I would like to know more generally, when a dot is a high quality dot? Is it the shape, the size, the regularity of the grid, in which the dots are set, or the reflection characteristics of the dots? Ten lined dots on a centimeter can be ten squares, that are all much smaller than a millimeter. The I have a dotted line. If they have a size of exactly one millimeter I have a line, that is not broken between the dots.

2400 dpi is a great resolution for an inkjet printer but, realistically, very few--if any--commercial offset printing presses print at a higher resolution that 200 dpi.

200 dpi? Did you forget a zero? I can hardly imagine, that this is true.

In the case of smallish text in high-quality printing, one
can certainly see the difference between 1200 and 2400.

One argument of my counterparts was, that the text in books is often set to small for old people. And my answer was, that it is not the point size only, that makes the text harder to read, but partly the downscaled font, that was designed for a bigger point size, and partly the low quality, in which the text is printed. I did say, that we need more font families with optical sizes, and more maintenance of our cultural achievements with regard to the apprenticeship of craftsmen. It is absurd to say, that a new technology is better than the old one, if the potential of the old technology is not tapped. If a person only knows wine, that costs one Dollar per bottle, he may come to the conclusion, that alcopops taste better than wine. And that is really a pity, because it is so unfair.

Someone did tell me with regard of my pro book arguments, that we also should not use modern cars anymore, but carriages. He did not understand, that the speed, in which you can read a book, is dependent from the legibility. So the ebook device of today is a carriage in the outfit of a modern car – in the comparison with a well printed book.

So the ebook device of today is a carriage in the outfit of a modern car – in the comparison with a well printed book.

As the car has evolved over time to be more than a horseless carriage, you should expect ebooks to evolve in a similar way. For now, an ebook is simply "a book on screen" -- but take for example

.. the text in books is often set to small for old people.

which is (or should be -- I don't own any ebook reading device) a point in favor of e-devices: drag over, or pinch, or shake your reader, and the type gets bigger.

Notes in the text? Tap them and read it in a pop up window. A difficult statement? Draw a question mark gesture on the screen, and the phrase rewrites itself using shorter words ;-)

which is (or should be -- I don't own any ebook reading device) a point in favor of e-devices: drag over, or pinch, or shake your reader, and the type gets bigger.

Sure. But the argument against the printed book was, that small point sizes are not legible. And I did say, that the legibility is not dependent from the point size only.

>What resolution would an ebook device require, when it shall display body text in 8.5 points?

Assuming some hints & just average anti-aliasing (aa), and to be safe, one would want a font where all the main stems have become 2-pixels-worth of the em. For most text faces that falls somewhere around 22-24 pixels per em. For 23 pixels to be 8.5 point you'd need 195 dpi.

Assuming no hints but scary-good aa, one can obviously get it to work at 135 or so, with user distance/size of screen — key.

Cheers!

Several gotchas (missed my nap yesterday): when I said 200 dpi for halftones, I did indeed mean lpi...