Postscript Printing.

http://www.cs.wisc.edu/~ghost/
Any use?

Keeping it simple, think of it as a printer/document layout language that can be used to get data between applications on totally different platforms, a bit like PDF is platform independent.

Paul

OK, so postscript only becomes a concern if you need to insure output on say a windows machine when it was created on an apple.

Why not just use PDF then?

PostScript is a device-independent language for depicting the visual contents of a page or screen. Adobe's Portable Document Format (PDF) utilizes this page-description language as an underlying part of its structure.

PDFs can be generated in OS X from the "File:Print..." menu item–you'll see a "Save to PDF..." button in the dialog box. This will generate a PDF of whatever size page has been selected under "File:Page Setup..."

Raw Postscript files (.ps) do exist, though they're not used as much these days. They must be converted to a bitmap image by a raster image processor (RIP) in order to print.

There is an open-source Postscript RIP called Ghostscript which, when combined with suitable printer drivers, makes a great page proofing solution:

http://gimp-print.sourceforge.net/MacOSX.php3

This is built into Mac OS X 10.2.x and 10.3.x, though you may need to confirm that your printer has a supported driver. The gimp-print drivers access the device at a more sophisticated level than many of the original printer drivers provided.

Yea I downloaded Gimp-Print, thanks for the link, unfortunatly it would appear every printer on earth is supported except mine!

Still never mind it was just a question.

Gary -- Dave was pretty much on the button.

As he says, PostScript is a device-independent page description language. In combination with one-piece Macs (some of them containing up to a staggering 4MB of RAM), Aldus PageMaker and an Apple LaserWriter (which contained the PostScript interpreter hardware) you could produce printed matter which could bring down governments (and sometimes did). Of course, no-one would deny the modern improvements wrought by Apple concerning themselves with much more important things such as diseminating Britney Spears singles.

The key thing about PostScript is that it describes the page. PostScript uses points as a unit of measure, which makes sense as this is how type is measured. Traditionally there were 72.3-ish points to the inch, but for convenience, PostScript always used 72. This was why the original Macs had a screen resolution of 72dpi -- when working with 12-point type, the period character (.) could be accurately depicted on screen using a single pixel. PostScript describes the position of every element (every character in every word, every dot in every halftone, etc.) to a precision of 1/256th of a point if required, but the whole point is that it deals with descriptions of the individual elements, without breaking them down into dots.

Creating a PostScript file is a "Print" process. You can therefore create a PostScript (.ps) file from any document which can be printed; just use "Save as file" within the print dialogue. It is not a file as such, but actually a PostScript program -- a set of instructions which can be "run" on any computer programmed to interpret PostScript, such as that within a PostScript output device such as an Apple LaserWriter, a Linotype imagesetter, or a Heidelberg digital printing press. When the PostScript file is run in this way, the interpreter within (or sometimes without) the output device will "interpet" the PostScript code into a set of device-specific instructions for printing the document on that particular device. Thus the same PostScript file can be fed to each of the three devices above, and the output will be the best of which each device is capable, even though the source file is indentical in each case.

If you had a Postcript printer connected to your Mac, it would by default be a network device. Printer Setup Utility would see it using AppleTalk or LPR. The basic driver is included with the MacOS, and the only thing needed is a small text file called a PPD (PostScript Printer Description) which describes the individual characteristics of the printer (such as paper handling, colour capability, resolution, RAM, etc) to the system PostScript driver, and is responsible for the printer-specific options that appear in the Print dialogue. What happens when you print is that a PostScript file is created on-the-fly and "run" on the chosen printer's interpreter (RIP) without any further internvention by you.

Many applications only officially support printing to PostScript printers, as they use PostScript internally in their documents -- QuarkXpress and Illustrator are two big examples from which printed output can be unpredictable using "dumb" printers. The System's "Save as PDF" option will result in a limited-resolution PDF which is quite useful for most things except high-resolution output. Installing Acrobat (full version) will also install a "printer" called "Adobe PDF". Printing to this results in a PDF with whatever specification is chosen within Acrobat Distiller, and can incorporate "off the page" elements such as bleed, crop marks, density strips, etc. Mostly these can still be opened by Preview.

ALL PDFs must be created from a PostScript file, whether this is obvious when you do it, or hidden from the user. Thus, a PDF is not truly device-independent as the device (Adobe PDF) has already been chosen, and they are usually optimised during the creation process for whatever step is next in the workflow. A PostScript file, however, is truly portable (and hugely bigger). I actually put one here at the end of this post for you to try, but the whole thread page went wobbly, so I've deleted it -- sorry!

Best;

Mark

Thanks very much for the description Mark.

If for instance I had created a photoshop document A4 Size but a resolution of 300dpi how would postscript interpret that?

I think basically your description kind of hints at computer Voodoo, which the average user need not worry about.

There was a reaosn behind my question the output on screen rarely matches that of the printer.

I think there are a number of reasons.

The printer is far from professional (Canon i560) and the screen brightness is significantly more than what could be outputed from the printer.

I have played in colour sync and there as been an improvement, but in somecase I have had to increase the brightness of the print settings to get what I want.

With the pictures in the 'hifi 2005' thread, one was taken to Kodak for printing (the wife works there) and the out put was frankly abismal. This was on a 'proper' four pass photo printing machine, the out put was a lot lot darker particualry on the reds which came out maroon. I wonder if this had been a postscript file, if we would have had better results, baring in mind it was increased to a file and image size of around 34 Cms tall, then increased again at the print stage to around 60 cms tall (i know confusing innit) It was in native PS format.

It basically comes down to the blind leading the blind in accurate prints in my household, LOL.

Gary - what happened to your question about colour calibration etc of your screen?

Derek

<< >>

I didn't think I had any replies on that??

I have followed the recommended paths on this Derek, but I think perhaps the canon just prints a bit dark, there dosn't appear to be a lot I can do short of reducing the brightness of the screen when working, but I like it bright!

quote:

---

Originally posted by garyi:
If for instance I had created a photoshop document A4 Size but a resolution of 300dpi how would postscript interpret that?

---

Actually that's a really bad example (not your fault!) as Photoshop only uses PostScript at all when you add text to images, and then only if you're using a PostScript font. If you're printing an image straight from Photoshop, there would be no point in having a PostScript printer. It works, but is generally slower and more hassle. Photoshop just addresses the printer dot-for-dot, as it were.

If OTOH you were to save your image from Photoshop as an EPS, then it could easily be dropped into a publication assembled using Quark or InDesign (or even the unjustly much-malligned PageMaker) and output to high-resolution colour-separated film or plates for commercial printing, with no further work.

quote:

---

I think basically your description kind of hints at computer Voodoo, which the average user need not worry about.

---

PostScript is as geeky or as transparent as you want it to be. An A4 PostScript laser is still the most hassle-free way for a office of (say) 5 users to get predictable output of letters, etc. from standard apps, without using fiddly "print servers", etc. Up to about 10 years ago, it was just about the only reasonable way for computers to share printers over a network at all. Commercially, PostScript is also not only the cornerstone of the entire publishing industry today, but also the reason that Xerox, Inc., and by extension, both Adobe and Apple exist in their current forms. It is not in any way a niche product, but being "middleware" and therefore not particularly visible, it is easy not to notice it -- particuarly when well implemented.

Your colour problem is classic. Firstly, source-first works! You can never add any information to a photograph which isn't already in it, and any editing you perform is always, always subtractive -- it removes information. It may selectively remove stuff which makes other information more noticeable (as in sharpening), but it is still removal rather than addition.

Secondly, your image should have sufficient resolution (pixels per whatever) in it to make it reproducible on your chosen output device at the size you want it. In practice this means at least 300 pixels per linear inch of final size, although you may sometimes get away with less.

Thirdly, any colour image should nowadays have embedded in it the colour profile of the device which generated it (often either a digicam or a scanner). The colour profile describes the characteristics of the originating device to anything reading the image which can pick up this information. It might say something like "I do blues really well, but I go over the top on reds". Reading this information, your computer should know to keep the blues as in the image,, but ease off on the reds a bit. In showing this image to you, it should take into account the colour profile of your monitor, and modify accordingly so that if you edit the image at all, you are not effectively working blind. If you then print the image, your computer will need to know the profile of your printer, so that it work out how best to print what you have seen on your monitor. (Please note that this gross over-simplification!)

Of course, this is potentially fraught with errors, as your monitor is effectively a large back-lit transparency using only red, green and blue "added" to black so that if all three are full on, the result is white. Printed output OTOH is a reflective medium using cyan, magenta, yellow and black ink in dots of variable sizes which selectively subtracts (or filters) the light reflected from the white of the paper to produce the result you see.

This is why, to stand the faintest hope of getting anywhere near predictable results from an inkjet printer, you need to use the recommended ink and paper (usually lots of both) as the printer driver then knows what it's dealing with, and many loops are closed. Suddenly you find your inkjet ain't that cheap any more...

quote:

---

I wonder if this had been a postscript file, if we would have had better results, baring in mind it was increased to a file and image size of around 34 Cms tall, then increased again at the print stage to around 60 cms tall (i know confusing innit) It was in native PS format.

---

PostScript would have made no difference at all, I suspect. If you started out with a JPG, then you are already hamstrung for large-format output. If starting with a RAW, or TIFF file you would have had more info in there to begin with. From the sound of it, most of the problem is with relative colour gamuts, and/or resolution.

If you can get your camera, monitor and printer working in a common colour space you will achieve a lot, but you will probably find that your monitor won't be very nice to use for other work when set up properly for this.

Best;

Mark

Gary;

I've just had a look at the pics you mention -- didn't realise they weren't "photo-real". They should have come out really well using the process you describe, BUT:

a) you should have been working on a calibrated monitor, and

b) you should have chosen to "embed colour profile", using the monitor profile when you saved the file. Personally I would have saved a TIFF as well as .psd file.

c) Find someone who has an Epson 7600 Pro printer. This will give you much more brightness and impact than a chemical process (since you are really starting with a digital image), and will probably fade less!

Wicked pics!

Thanks Mark for the detailed answer.

I have been practising alot with various software to achieve the results I am interested in. I do a lot of art work, usualy from original photos, unfortunatly the raw capabilities of the E10 is such that its a bit impractical (specifically the fact it takes around 30 seconds to write the file) so I am using JPEG at the highest resulotion.

As to how this effects the final project I believe is negligable in terms of the pictures I produce, as typically I am seeking to get only the line art or basic idea of the original.

Sheila has managed to obtain the very best photo quality inkjet paper from Kodak, which although a bonus is not the original Canon product as recommended by canon (No suprise) The kodak does have specific 'drivers' for PC to allow other printers to use the kodak to its best, but inevetably this is not avilable for OSX.

So I am kind of in a quandry, I do not wish to use masses of ink and paper to find the best setting, but may well need to. It would be great if there were more detailed software to allow one to get the best from the printer, the canon/OSX software is a bit agricultural, i.e., fast/medium/best sort of thing.

What I am learning from this is the art of the graphics designer, i.e. its not just down to talent with the software but how this is interpreted to paper, publications and indeed web application, each one requires specialist knowledge of the print and display process.

So big up to the GD basically!

I wil revisit coloursync in terms of /camera/screen/printer but feel confident they are all working to the same song sheet, its just that damn screen brigntess, I need to turn it down I think! Can you blieve though that the controls for screen calibration won't adjust the brightness of the screen so you seem to have to do it every time, unless I am missing something fundemental.

quote:

---

Originally posted by garyi:
I am using JPEG at the highest resulotion.

---

No problem for this application. As you say, you are basically only using the outlines from the original photo.

quote:

---

Sheila has managed to obtain the very best photo quality inkjet paper from Kodak, which although a bonus is not the original Canon product as recommended by canon (No suprise) The kodak does have specific 'drivers' for PC to allow other printers to use the kodak to its best, but inevetably this is not avilable for OSX.

---

Typical Kodak! Advise you to try Canon paper. It doesn't matter how "good" it is, the important thing is being able to select it in the print settings. Most inkjets drop more ink when using their "own" paper, but they need to know things like absorbency and reflectivity to get it right.

Also, make sure you're using the latest drivers for your printer (v2.32, from memory, but they might have done a new one).

quote:

---

What I am learning from this is the art of the graphics designer, i.e. its not just down to talent with the software but how this is interpreted to paper, publications and indeed web application, each one requires specialist knowledge of the print and display process.

---

Of course! That's why non-specialist, box-shifting dealers do nobody any good. People just say "it doesn't work", throw their hands in the air and walk off disgruntled. Anything workthwhile requires input, so that you understand what you're trying to achieve. Only then can you understand if you've achieved it --instant gratification is invariably temporary (grasshopper... )

quote:

---

...the controls for screen calibration won't adjust the brightness of the screen so you seem to have to do it every time...

---

The colour calibration routine does not include brightness and contrast -- it has to start from a given set of brightness and contrast characteristics, rather than settings. This is because these vary according to things like the age of your monitor, the proximity of radiators, the ambient light, etc. The first step in the calibration routine makes you set your brightness and contrast so that you are starting from a known point, and how they are set will radically affect the results of calibration.

All the best with it!

Gary

I tried the two software calibration techniques and decided that I could not adjust the screen to the reccomended brightness - because I could not compare the brightness of a sheet of paper compared to the brightness of the screen so I bought the SpyderPro with the OptiCal sofware to set up the screen. You can also get a device that also will calibrate the printer by measuring the printed image.

I get my pictures printed at Photobox.co.UK and so far the first set of prints under the new regime are better. I have also obtained calibration data for the Photobox printers and will next attempt to use them in my process.


Also check on the Olympus SLR Talk Forum, The Retouching Forum and the Mac Tools Forum on the www.dpreview.com site, there are q and as on this subject in these forums.

(I know that Count.d hold no truck with what is posted on the internet - but at least it gives you extra information to accept or reject).

Derek

<< >>