snoopy@sopwith.UUCP (Snoopy) (01/16/89)
In article <10322@gryphon.COM> richard@gryphon.COM (Richard Sexton) writes: |Get it set on a 2650 dpi Linotronic ? | |/* mild personal nit to pick */ | |I'm getting real tired of seeing books, forms, and ad's in magazines |set with a laser printer instead of a phototypesetter. | |That 3 volume O'reilly series about X-Windows, for exmaple is laser typeset, |and it doesnt look like a real book somehow. More like a thick company |newsletter. | |I can't help but feel that we're making negative progress here. What is happening is that the range of reproduction quality is getting compressed towards the middle. I am happy to see less memeograph (sp?) and draft quality dot matrix output. I don't think the laserprinters themselves are necessarily to blame for the cases of lowered quality. Sometimes the software driving them doesn't take advantage of the resolution available. Sometimes an advertiser will *want* an obviously low resolution output to make it look more "computery" to an audience used to associating very low resolution with computers. (No one ever accused advertisers of having taste.) And I have to wonder if the reproduction stage(s) after the laserprinter retain what quality the laserprinter does generate. A friend of mine has a 300dpi laserprinter that generates beautiful output. Run this through a photocopy machine, (even a big Kodak one, which has been the best I've found) and it looks pretty sad. _____ /_____\ Snoopy /_______\ |___| tektronix!tekecs!sopwith!snoopy |___| sun!nosun!illian!sopwith!snoopy
chuq%plaid@Sun.COM (Chuq Von Rospach) (01/17/89)
>What is happening is that the range of reproduction quality is getting >compressed towards the middle. I am happy to see less memeograph (sp?) >and draft quality dot matrix output. And less typewriter with a bad ribbon, for that matter. >I don't think the laserprinters themselves are necessarily to blame for >the cases of lowered quality. Sometimes the software driving them doesn't >take advantage of the resolution available. I think an even more important problem is that the person using the printer and software doesn't pay any attention to the technological limits of what they're working with. I will posit that if you pay any attention at all, you can generate output on a laserprinter that is perfect acceptable in a commercial printing environment. I do it with OtherRealms. To do it, you have to remember your limitations. If you do things like 6pt type or fountains on a laserwriter, you're going to look stupid, and stupid is what I claim you'll be. But you can generate just as good a publication without fountains and 6pt. type by designing to the media. It's not an inherent problem with the software, the printers, the technology or anything. It's a problem with the people who are trying to do things they should know better than. (analogy time: if someone laid out a newsletter as though it was going to be printed out on a laserprinter and printed it on a dot-matrix, people would laugh at them. If he instead designed something that required the Lino to print and printed it out on a laserprinter, they'd bitch about the technology. So it goes....) If you understand what you're doing and design to your limitations, you can put together a reasonable document under almost any conditions. But if you ignore the limitations and design super-whizzy gosh-wows (in many cases, of marginal advantage at that in the case of things like fountains. Fountains are in most cases nothing more than a faddish "gee, look what I can do" show-off) oblivious to reality, you're going to be disappointed. Is that the fault of the system? Or the user? >And I have to wonder >if the reproduction stage(s) after the laserprinter retain what quality >the laserprinter does generate. A friend of mine has a 300dpi laserprinter >that generates beautiful output. Run this through a photocopy machine, >(even a big Kodak one, which has been the best I've found) and it looks >pretty sad. It depends. My stuff comes out pretty good. Not as good as offset (which is why OtherRealms is offset) but if you avoid some pitfalls, you're photocopying can work out quite nicely. One thing to avoid is lots of black. Replace it with something like a 60% or 70% grey instead. The effect is essentially the same, and it reproduces much cleaner.
richard@gryphon.COM (Richard Sexton) (01/17/89)
In article <88@sopwith.UUCP> snoopy@sopwith.UUCP (Snoopy) writes: I wrote: >| >| >|I'm getting real tired of seeing books, forms, and ad's in magazines >|set with a laser printer instead of a phototypesetter. >| >| >|I can't help but feel that we're making negative progress here. > >What is happening is that the range of reproduction quality is getting >compressed towards the middle. I am happy to see less memeograph (sp?) >and draft quality dot matrix output. I was talking about books, costing non-trivial amounts of money purchased in bok stores. I've never seen bot-matrix stuff there except for mauals accompanying software packages. >I don't think the laserprinters themselves are necessarily to blame for >the cases of lowered quality. Sometimes the software driving them doesn't >take advantage of the resolution available. Huh ? 300 dpi is 300 dpi. > Sometimes an advertiser will >*want* an obviously low resolution output to make it look more "computery" >to an audience used to associating very low resolution with computers. >(No one ever accused advertisers of having taste.) Sure. I have no problems with this. I'm quite used to seeing examples set in Courier. But I'd expect it to be crisp and sharp (ie, 1200 dpi) not laserish, of it's in a halfway expensize book. I've even seen a few ads in magazines that were *quite obviously* laserset. Impresive. Not in a positive sense. > And I have to wonder >if the reproduction stage(s) after the laserprinter retain what quality >the laserprinter does generate. A friend of mine has a 300dpi laserprinter >that generates beautiful output. Run this through a photocopy machine, >(even a big Kodak one, which has been the best I've found) and it looks >pretty sad. Gawd does my milage vary. I think my laserprinter generates good, not beautiful output. Phototypesetter output on resin coated paper, is beautiful. And I've seen photocopy machines (Canon) that yielded better copied than the original; they *appeared* to perform some sort of low pass filter. To summarize: laserset output, in place of typesetter output, looks cheap. -- ``In a few years, the only thing that will be made in America is a deal'' richard@gryphon.COM {...}!gryphon!richard gryphon!richard@elroy.jpl.nasa.gov
kaufman@maxzilla.Encore.COM (Lar Kaufman) (01/18/89)
In article <10844@gryphon.COM> richard@gryphon.COM (Richard Sexton) writes: >In article <88@sopwith.UUCP> snoopy@sopwith.UUCP (Snoopy) writes: >I wrote: >>|I'm getting real tired of seeing books, forms, and ad's in magazines >>|set with a laser printer instead of a phototypesetter. >>| >>|I can't help but feel that we're making negative progress here. >> >>What is happening is that the range of reproduction quality is getting >>compressed towards the middle. I am happy to see less memeograph (sp?) >>and draft quality dot matrix output. ... >To summarize: laserset output, in place of typesetter output, looks >cheap. Well, yes - and it _is_ cheap, which is the point. The O'Reilly Nutshell guides, to use an example that you mentioned earlier, have information that I want, presented in a manner that is readable and legible, in a binding that is useful if you are sitting at a keyboard. That is to say, the book opens up and stays open when laid flat. This is a characteristic that is missing from most computer manuals. No, in contrast, let me offer a beautiful paperback I recently bought: _SGML:_An_Author's_Guide_. This book looks great, with nice figures, great highlighting and layout, and so on. On the other hand, it won't stay open without physical restraint, and it cost me over $35 at my local book store (Harvard Coop). Cheap laserprinted books are a direct and suitable response to the rapid rise in cost of books. I think this is OK, myself. What I don't understand is why book costs still are rising so fast... -lar Lar Kaufman <= my opinions Fidonet: 1:322/470@508-534-1842 kaufman@multimax.arpa {bu-cs,decvax,necntc,talcott}!encore!kaufman
lee@uhccux.uhcc.hawaii.edu (Greg Lee) (01/18/89)
From article <10844@gryphon.COM>, by richard@gryphon.COM (Richard Sexton):
" In article <88@sopwith.UUCP> snoopy@sopwith.UUCP (Snoopy) writes:
" ...
" >I don't think the laserprinters themselves are necessarily to blame for
" >the cases of lowered quality. Sometimes the software driving them doesn't
" >take advantage of the resolution available.
"
" Huh ? 300 dpi is 300 dpi.
It's possible for the software to be at fault. I remade the Computer
Modern Roman fonts that came ith the TeX distribution following
Knuth's recommendations in the Metafont Book for parameter settings
for high resolution printers. The new versions look very noticeably
better on my Apple LW than the distribution versions did.
Greg, lee@uhccux.uhcc.hawaii.edufyl@ssc.UUCP (Phil Hughes) (01/19/89)
In article <88@sopwith.UUCP>, snoopy@sopwith.UUCP (Snoopy) writes: > In article <10322@gryphon.COM> richard@gryphon.COM (Richard Sexton) writes: > |I'm getting real tired of seeing books, forms, and ad's in magazines > |set with a laser printer instead of a phototypesetter. > I don't think the laserprinters themselves are necessarily to blame for > the cases of lowered quality. Sometimes the software driving them doesn't > take advantage of the resolution available. Sometimes an advertiser will > *want* an obviously low resolution output to make it look more "computery" > to an audience used to associating very low resolution with computers. Just having the choice or 300 dpi vs. 2650 dpi means you have to make another decision. We have been proofing our products (UNIX pocket references) on a laser printer for about 3 years. This is a real improvement over our previous methods. We still typeset all real products but use other methods for advertising material, software manuals and class notebooks. For example, we just completed a new catalog. It is printed on fairly inexpensive paper (50# book) and the laser print looks fine for most of it. The only problem was with screens (gee, the original subject that started this discussion). Rather than typeset the whole book on a Linotronic we just pasted in the screens (only 3) and the overall quality is very good. Some of our advertising material is set on the laser printer and then reduced to make it look decent. It is always a compromise between cost, appearance and time invested (we send out the Linotronic work). -- Phil Hughes, SSC, Inc. P.O. Box 55549, Seattle, WA 98155 (206)FOR-UNIX uw-beaver!tikal!ssc!fyl or uunet!pilchuck!ssc!fyl or attmail!ssc!fyl
rcd@ico.ISC.COM (Dick Dunn) (01/19/89)
Chuq wrote: > I think an even more important problem is that the person using the printer > and software doesn't pay any attention to the technological limits of what > they're working with. I will posit that if you pay any attention at all, you > can generate output on a laserprinter that is perfect acceptable in a > commercial printing environment... Even in low volumes, where you're reproducing with a photocopier, there are things you can do to help or hurt. Chuq mentioned not using tiny type, like 6 pt. I'd say that if you even try to use 10 pt the results are going to be marginal. I've generally encouraged people to use 11/13 for the body text of reports and such; that lets them use 9/11 for footnotes. (You can live with the 9 pt for small amounts of text.) Going up from 10 to 11 may jump at you at first, but if you're generating the typical 6.5" line (8.5" paper with 1" margin), the lines are too long at 10 pt anyway. > >...A friend of mine has a 300dpi laserprinter > >that generates beautiful output. Run this through a photocopy machine, > >(even a big Kodak one, which has been the best I've found) and it looks > >pretty sad. Then get the copier fixed! In many cases, one step of photocopying will actually *improve* the quality. Reason: it tends to smooth out some of the jaggies just a little bit and make them less noticeable, particularly on italics (where you've got lots of sloping strokes). Another thing you can do with a fancy copier is a modest reduction. For example, generate the output on regular 8.5x11 with 11 or 12 pt text, then photo-reduce. Reducing the 11" dimension to 8.5" (rotating as you copy) is a 77% factor; it increases the effective dot spacing to almost 400 dpi; the line length and type size fit better with one another; the result is easier to handle. -- Dick Dunn UUCP: {ncar,nbires}!ico!rcd (303)449-2870 ...A friend of the devil is a friend of mine.
rcd@ico.ISC.COM (Dick Dunn) (01/19/89)
In article <10844@gryphon.COM>, richard@gryphon.COM (Richard Sexton) replied to a comment about using the printer's full resolution: > Huh ? 300 dpi is 300 dpi. However, there are programs that generate output for lower-resolution printers, then just scale the bits for a laser printer--the result, of course, is that you get the quality of a dot-matrix printer for the price of a laser printer. But more than that, the "300 dpi" business is very deceptive. All that 300 dpi means is that the spacing between the "dots" the printer can put on the paper is 1/300". It doesn't say anything about the size or shape of the dots, nor about how small a dot will actually show up reliably. There's stuff like the quality of the paper surface, the texture of the toner, and all the nasty things that get out of focus, get dirty or scratched, wear, etc...all affecting the quality. The dots don't come out square--that is, you don't just divide the page up into a grid of squares 1/300" on a side and color in the squares you want. The dots are much closer to being round (which is probably actually good, since it reduces the jaggie effect). Then, consider that the dots have to be larger than 1/300" diameter, or they wouldn't join over a large area to make uniform black--there would be holes in the region inside each 4-dot square. If the dots were perfectly round, they'd need to be sqrt(2) larger than the "resolution" to join completely. -- Dick Dunn UUCP: {ncar,nbires}!ico!rcd (303)449-2870 ...A friend of the devil is a friend of mine.
chuq%plaid@Sun.COM (Chuq Von Rospach) (01/19/89)
>I'd say that if you even >try to use 10 pt the results are going to be marginal. I've generally >encouraged people to use 11/13 for the body text of reports and such; that >lets them use 9/11 for footnotes. 11/13 would be good. 12/14 would be better, actually. That is, *if* you have the space to carry it. In something like OtherRealms, where space was always as a premium, I published at 10/12 or 10/11 Palatino. Tight, but if you're careful, it can be perfectly legible. Among things to avoid: italics (use bold face instead. One nice thing about Palatino is that it has a heavy bold face, so it shows up nicely even at 10pt Xerographed. Try that with Bookman sometime....) and any non-postscript variations of a face like underline, outline, or all the other ugly algorithms Apple lets you use. Another thing would be lines: anything wider than about 2 points should be 60% grey instead of black, because a widde, black line won't reproduce without blotching where a 60% grey will. With 10pt faces, don't plan on using footnotes. If you must have them, use same-size endnotes. Learn to use whitespace with 10pt, too. Widen the column gaps, don't justify the columns. Use ragged right and don't hyphenate unless you've got a good hypenator. It generally isn't worth it and it'll help break up the text more. At this size, you need all the white you can get. The current issue of OtherRealms is set 9/10 ITC Garamond. That's *too* tight for comfortable reading, in retrospect, but I had some nasty restrictions to deal with. Next issue will probably be 9/11. Or 9.25/11.5 or something, if I can make it work cleanly. The trend in publication design seems to be towards 12pt text, either 12/40 or 12/15. Lots and lots of whitespace, too. Of course, that means they don't have to pay as much to the writers each issue, too. Some people have taken it too far into the "open and active" realm: Time, generally considered the leading edge of design, drives me crazy. And then there's "Publish!" magazine, wonderfully showing off all the state-of-the-art design technology and ideas, and why you shouldn't blindly go off and use all that crap... >Then get the copier fixed! In many cases, one step of photocopying will >actually *improve* the quality. damn straight. If your copier is botching you up, it's broke. You don't believe me, I've got back-issues of OtherRealms printed on that self-same Kodak (good boy. get biscuit. May I *never* change your toner again!) that, while you can tell it isn't offset, isn't something I feel like hiding, either. Chuq Von Rospach Editor/Publisher, OtherRealms chuq@sun.COM And now a message for the eyes of only those people with Commander Chuqui Secret Decoder Rings: 7-3-6-27-24-4-10-6-27-3-2-23-27-23-10-7-27-3-24-24-4-20-11-7-24
cplai@daisy.UUCP (Chung-Pang Lai) (01/24/89)
In article <10844@gryphon.COM>, richard@gryphon.COM (Richard Sexton) replied to a comment about using the printer's full resolution: > Huh ? 300 dpi is 300 dpi. There was a very good article by David R. Spencer in the February 8, 1988 issue of "The Seybold Report on Desktop Publishing" Vol. 2, No. 6 ISSN: 0889-9762. The title of the article was "BEYOND RESOLUTION Other Output Quality Factors." dpi is only one of the many factors affecting quality. For example, mechanical noise of the marking engine itself can ruin a high resolution output. It is not surprising that a 400 dpi printer from one manufacturor can be worse than a 300 dpi printer by the other. -- .signature under construction ... {cbosgd,fortune,hplabs,seismo!ihnp4,ucbvax!hpda}!nsc!daisy!cplai C.P. Lai Daisy Systems Corp, 700B Middlefield Road, Mtn View CA 94039. (415)960-6961
mark@lakesys.UUCP (Mark Storin) (01/24/89)
In article <2535@daisy.UUCP> cplai@daisy.UUCP (Chung-Pang Lai) writes: >In article <10844@gryphon.COM>, richard@gryphon.COM (Richard Sexton) >replied to a comment about using the printer's full resolution: > >> Huh ? 300 dpi is 300 dpi. > I missed the original posting so don't know if this is relevant, but... It is sometimes possible to increase the apparent resolution of a laser printer, beyond 300dpi, using a technique called half-bitting or dentation. From "Digital Typography" by Richard Rubinstein; "Because of the way in which toner goes onto paper, usually an isolated pixel will not produce a spot on the paper. That location will take on the color of the pixels that surround it. By intentionally creating certain kinds of ragged edges and including voids in letterforms, the toner can be coerced into occupying positions intermediate to the resolution increments. ... Thus a 300 dpi printer may produce letterforms with greater positioning accuracy than 1/300-inch. The size of the effect depends on the characteristics of the particular printer. The result for a particular printer could be the equivalent of 1/450-inch feature dimensions for the affected parts of the letterforms." -- Mark A. Storin Lake Systems, Milw., WI mark@lakesys.lakesys.COM -- Mark A. Storin Lake Systems, Milw., WI mark@lakesys.lakesys.COM
cplai@daisy.UUCP (Chung-Pang Lai) (01/26/89)
In article <2535@daisy.UUCP> cplai@daisy.UUCP I write:
]
]There was a very good article by David R. Spencer in the February 8, 1988
]issue of "The Seybold Report on Desktop Publishing" Vol. 2, No. 6
]ISSN: 0889-9762. The title of the article was "BEYOND RESOLUTION Other
]Output Quality Factors."
]
Mr. Spencer also wrote another article on output quality back in the Feb, 1987
issue of "The Seybold Report on Publishing Systems" Vol. 16, No. 11.
ISSN: 0736-7260
He presented a talk at "The Desktop Publishing Conference" in Santa Clara
last September. The material of his talk is reproduced in the current
issue of "The Seybold Report on Desktop Publishing" Vol. 3, No. 5.
Title of the article is "Output Device Image Processing."
--
.signature under construction ...
pyramid!daisy!cplai C.P. Lai
Daisy Systems Corp, 700B Middlefield Road, Mtn View CA 94039. (415)960-6961