sun@me.utoronto.ca (Andy Sun Anu-guest) (05/30/90)
I recently came across this problem and wonder if anybody can offer a
helping hand on this:
I tried to print a perfect square on an Apple LaserWriter II NT using
%!
/mm {72 mul 25.4 div} def
0 setlinewidth
newpath
40 mm 40 mm moveto
40 mm 90 mm lineto
90 mm 90 mm lineto
90 mm 40 mm lineto
closepath
stroke
showpage
But the square that I got was distorted. It is approximately 0.5 mm longer
in the y direction (11" direction) and 0.5 mm shorter in the x direction
(8.5" direction). It is essential for us to get an undistorted printout.
Does anybody know what is the reason for this? Is it because of the conversion
factor that I use in defining mm (doesn't make sense to me since I should
get the same error in both direction, not one longer and one shorter)? Is
there something that I should define in the above that I missed? Is it
because of the aspect ratio (does a laser printer have aspect ratio?) of the
laser printer? If it was in fact due to the laser printer itself, is there
a way to compsensate for the distortion other than specifying something
like 'x y scale' (it is not easy to measure the exact lengths to find
the scaling factors) in the code?
Thanks in advance.
Andyhenry@utzoo.uucp (Henry Spencer) (05/31/90)
In article <90May30.095014edt.18974@me.utoronto.ca> sun@hammer.me.UUCP (Andy Sun Anu-guest) writes: >I tried to print a perfect square on an Apple LaserWriter II NT ... >But the square that I got was distorted. It is approximately 0.5 mm longer >in the y direction (11" direction) and 0.5 mm shorter in the x direction >(8.5" direction). It is essential for us to get an undistorted printout. > >Does anybody know what is the reason for this? ... Is it >because of the aspect ratio (does a laser printer have aspect ratio?) of the >laser printer? ... Almost certainly that's the cause. Most "300dpi" laser printers are in fact something like 298dpi on one axis and 303dpi on the other. Usually the manufacturers don't even bother telling you about this, much less provide a way to compensate for it. (This isn't unique to laser printers, either: most photocopiers likewise have slightly different X and Y axes.) -- As a user I'll take speed over| Henry Spencer at U of Toronto Zoology features any day. -A.Tanenbaum| uunet!attcan!utzoo!henry henry@zoo.toronto.edu
woody@chinacat.Unicom.COM (Woody Baker @ Eagle Signal) (06/01/90)
In article <1990May31.001233.9322@utzoo.uucp>, henry@utzoo.uucp (Henry Spencer) writes: > In article <90May30.095014edt.18974@me.utoronto.ca> sun@hammer.me.UUCP (Andy Sun Anu-guest) writes: > >because of the aspect ratio (does a laser printer have aspect ratio?) of the > >laser printer? ... > > Almost certainly that's the cause. Most "300dpi" laser printers are in > > (This isn't unique to laser printers, either: most photocopiers likewise > have slightly different X and Y axes.) I used to work for a place that sold Copiers. I sold Lasers, and Computers. I was told by people whom I trust, that NONE of the photocopiers would produce an exact 1 for 1 copy. It was always about 3% off. This (or so I was told) is because the treasury department didnot want anyone copying anything exactly, for fear of counterfitting. Especialy color copiers. They would not release color analog copiers period. You will note that all of the current color copiers use a glossy toner. Also mandated by the treasury department. Cheers
rcd@ico.isc.com (Dick Dunn) (06/01/90)
woody@chinacat.Unicom.COM (Woody Baker @ Eagle Signal) writes: > I used to work for a place that sold Copiers. I sold Lasers, and Computers. > I was told by people whom I trust, that NONE of the photocopiers would > produce an exact 1 for 1 copy. It was always about 3% off. This (or so > I was told) is because the treasury department didnot want anyone copying > anything exactly, for fear of counterfitting... Foo. I just ran a copy on a relatively high-end copier to check. It is 1:1 on both axes to within better than 0.5%. That's greater accuracy than the dimensional stability of paper. The real reason most copiers tend to be off a little is very simple: They enlarge ever-so-slightly to compensate for positioning error and variation in paper size, so that you don't see a black line around the extreme edge of the copy where the edge of the original was. Laser printers do (as Henry noted) commonly have a pixel aspect ratio not quite 1:1. But if you're really concerned about that, you can take some precise measurements and compensate with a transformation matrix, or if it's an image, compensate in the scanning or conversion. >...Especialy color copiers. > They would not release color analog copiers period... What's a color analog copier? I'm assuming you mean a copier that is capable of continuous-tone repro (as opposed to high-contrast "line" copy). What ever would that have to do with counterfeiting? >...You will note that > all of the current color copiers use a glossy toner. Also mandated by > the treasury department. No, mandated by the process. What about standard one-color copiers with a selection of toner colors? Conspiracy theories are fun, but sheeesh! -- Dick Dunn rcd@ico.isc.com uucp: {ncar,nbires}!ico!rcd (303)449-2870 ...Simpler is better.
jaap@mtxinu.COM (Jaap Akkerhuis) (06/02/90)
In article <1289@chinacat.Unicom.COM> woody@chinacat.Unicom.COM (Woody Baker @ Eagle Signal) writes: > In article <1990May31.001233.9322@utzoo.uucp>, henry@utzoo.uucp (Henry Spencer) writes: > > In article <90May30.095014edt.18974@me.utoronto.ca> sun@hammer.me.UUCP (Andy Sun Anu-guest) writes: > > >because of the aspect ratio (does a laser printer have aspect ratio?) of the > > >laser printer? ... > > > > Almost certainly that's the cause. Most "300dpi" laser printers are in > > > > (This isn't unique to laser printers, either: most photocopiers likewise > > have slightly different X and Y axes.) > > I used to work for a place that sold Copiers. I sold Lasers, and Computers. > I was told by people whom I trust, that NONE of the photocopiers would > produce an exact 1 for 1 copy. It was always about 3% off. This (or so > I was told) is because the treasury department didnot want anyone copying > anything exactly, for fear of counterfitting. Especialy color copiers. > They would not release color analog copiers period. You will note that > all of the current color copiers use a glossy toner. Also mandated by > the treasury department. I've got the feeling about yet another urban legend being born. Copiers usually enlarge 2 to 3% for the simple reason that it makes the life of the designer/manufacturer coping with various tolerances of papersizes etc. much easier. Note that you deal with paper moving around. If you want make an exact copy, you have to make sure that the original and the paper you copy to is on exactcly the right place, else it is very easy to have light from the edges op the original show up at the copy. Furthermore the original might not be exactly up to spec in size, so again one would see this edge anomalies. For an experiment, try to make an exact 1 to 1 print of a negative using an enlarger. You'll be suprised how hard it is compared to a direct contact print. I don't know about the glossy color toner, but in general glossy surfaces look brighter then non-glossy ones. That might make producing color copies easier. It would be very silly of any treasure department trying to control the copier industry out of fear of forgeries. That would immediately start an interesting market of illegal machines. Back to the original subject. I too have experienced that the aspect ratio of various printers (and typesetters for that matter) is not constant. But why are 300dpi printers not 300 in general? At the introduction of the Agfa P400 they told me the next story: Their printer is marketed as a 400 dpi model, but if you really look in the specs, it is actually something like a 406 dpi. The 406 dpi make a nice figure in a metric system. H'm, if you get your calculator out you can see that that doesn't make a lot of sense, so it might be yet another urban legend. jaap
woody@chinacat.Unicom.COM (Woody Baker @ Eagle Signal) (06/04/90)
In article <1990Jun1.155110.3330@ico.isc.com>, rcd@ico.isc.com (Dick Dunn) writes: > Laser printers do (as Henry noted) commonly have a pixel aspect ratio not > quite 1:1. But if you're really concerned about that, you can take some > precise measurements and compensate with a transformation matrix, or if > it's an image, compensate in the scanning or conversion. > > > What's a color analog copier? I'm assuming you mean a copier that is > capable of continuous-tone repro (as opposed to high-contrast "line" copy). > What ever would that have to do with counterfeiting? Color analog. I.E. direct imaging on the drum like 99% of all black and white copiers. Current color copiers, with 1 exception that I can think of digitize and print the colors as spots like a laser printer does. The original color copiers were basically analog, with non-glossy toner. There were a few made. VERY few, and they were literaly kept under lock and key, with a strict accounting made of EVERY copy, failed or otherwise, and submitted to the treasury dept. In addition they were routinly inspected by TD investigators. They were capable of accuratly copying money, or virtually anything else. The resultant copies were quite good enough to fool automatic change machines, and enough people to get the TD rather interested. Proto types existed way back in the early 80's. I have worked at 2 places (separated by 5 years and many miles) that dealt with copiers. This info was known in both places, and it is NOT "conspiracy theory". The TD was plain flat worried about them. > > No, mandated by the process. > Dick, there are glossy and non-glossy toners. For example, the MITA 9000 (a very old copier) had one of the glossiest toners ever produced. Nearly ALL current black or single color copiers produce a flat non-glossy print very similar to laser printers. Even those that have 2 or more colors. The process does NOT mandate the glossy colors. Flat colors would (do) work just as well. > Conspiracy theories are fun, but sheeesh! Conspiracy theories???? Where did this come from? Cheers Woody.
henry@utzoo.uucp (Henry Spencer) (06/04/90)
In article <1237@mtxinu.UUCP> jaap@mtxinu.UUCP (Jaap Akkerhuis) writes: >But why are 300dpi printers not 300 in general? At the introduction >of the Agfa P400 they told me [406 dpi is a nice number in metric]. I can't answer for all the zillion 300dpi models on the market, but I do know why the HP LaserJet (and LJ+) wasn't 300dpi on the short axis, and I'm told this applied to many of its brethren. Resolution on the long axis was set by the gearing of the mechanical subsystem, but on the short axis it was set by the video rate from the controller. The video rate was ultimately set by a crystal oscillator. Exactly matching the long-axis rate required an oddball frequency that would require a custom part. However, there is a standard crystal-oscillator frequency that was *almost* right... so guess what most manufacturers used. Actually, my measurements indicated that the long-axis resolution was not exactly 300dpi either, but that might have been due to complications like the paper stretching slightly on going through the fusing rollers. -- As a user I'll take speed over| Henry Spencer at U of Toronto Zoology features any day. -A.Tanenbaum| uunet!attcan!utzoo!henry henry@zoo.toronto.edu
gillies@p.cs.uiuc.edu (06/05/90)
> (This isn't unique to laser printers, either: most photocopiers likewise > have slightly different X and Y axes.) I think that all photocopiers have different X and Y axes, so that counterfeit copies from a photocopier can be detected by the distortion.
wiml@blake.acs.washington.edu (William Lewis) (06/05/90)
In article <1296@chinacat.Unicom.COM> woody@chinacat.Unicom.COM (Woody Baker @ Eagle Signal) writes: > The resultant copies were quite good enough to fool automatic >change machines, and enough people to get the TD rather interested. Well, for *that* matter, the black-and-white copies from the corner drugstore will fool automatic change machines. A friend of mine once tried doing this, and I think his worst problem was noticing that the different sides of a bill are different sizes. -- JESUS SAVES | wiml@blake.acs.washington.edu Seattle, Washington but Clones 'R' Us makes backups! | 47 41' 15" N 122 42' 58" W
woody@chinacat.Unicom.COM (Woody Baker @ Eagle Signal) (06/06/90)
In article <1990Jun4.154728.11348@utzoo.uucp>, henry@utzoo.uucp (Henry Spencer) writes: > video rate was ultimately set by a crystal oscillator. Exactly matching > the long-axis rate required an oddball frequency that would require a > custom part. However, there is a standard crystal-oscillator frequency > that was *almost* right... so guess what most manufacturers used. Does this imply that by perhaps changing the crystal (ignoring for now the software and hardware needed) that one can increase the resolution of the printer? I'v seen several canon sx engines that have a resolution, say 800 or 600 or even 1000 dpi. Has anyone done any experimenting? While we are on the subject of Canon, have y'all heard the latest? It seems that Canon just keeps stooping lower and lower. First they created an abrasive toner to wear the SX drums out faster, and now the latest attempt to squash the reload market involves a pseudo concern for the enviornment. While I have not been able to verify this, I've been told that the new cartridges are coming with a special mailer (prepaid no less), and instructions that when the toner runs out, you are to put the cartridge in the bag and send it to Canon for "enviornmentaly safe" disposal. Has anyone seen one of these? If this is true, my suggestion is that you fill the bag with anyold thing about the size and shape and weight of the cartridge and mail it to them. By law they will have to pay the postage on the weight. Perhaps if enough people do this, and include nasty letters, they will get the idea. If this turns out to be a rumor, I appologize to Canon, but if not, they just climbed to the top of my arrogant company list. Cheers Woody
jeffe@eniac.seas.upenn.edu (George J. Jefferson) (06/08/90)
In article <99500024@p.cs.uiuc.edu> gillies@p.cs.uiuc.edu writes: > >> (This isn't unique to laser printers, either: most photocopiers likewise >> have slightly different X and Y axes.) > >I think that all photocopiers have different X and Y axes, so that >counterfeit copies from a photocopier can be detected by the distortion. Really? is this a law? an advertised feature? a defect that sombody later realized could be called a feature? George Jefferson jeffe@eniac.seas.upenn.edu george@sol1.lrsm.upenn.edu
rcd@ico.isc.com (Dick Dunn) (06/09/90)
In article <99500024@p.cs.uiuc.edu>, gillies@p.cs.uiuc.edu writes: > I think that all photocopiers have different X and Y axes, so that > counterfeit copies from a photocopier can be detected by the distortion. We seem to be cycling the urban legends faster these days. As I posted quite recently, there are good copiers whose accuracy is better than the dimensional stability of paper. -- Dick Dunn rcd@ico.isc.com uucp: {ncar,nbires}!ico!rcd (303)449-2870 ...Simpler is better.