Hi Robert,

I am no expert at all in regards to DPSpool, owever it does sound like the 
image is being resized from its original. 

Lossless bitmaps images do not scale in size well. Bitmap images are also 
created at specific resolutions, and this resolution has got to be rendered on 
different resolution devices. On monitors the image is rendered at about 100 
dpi (pixels per inch) whereas as printer will render it at perhaps 300 dpi. At 
higher resolutions more data from the image is used to display the image. 
Monitors have a built in fuzziness which means that I slightly scaled image may 
not show the true magnitude of the problem, but on a higher resolution printer 
the problem becomes even more apparent.

The issue with scaling is that the pixel position are totally digital. The 
pixel can only be placed in discrete positions, and not half way. So when an 
image is scaled the software needs to make a decision whether a pixel is 
rescaled to be in a specific position. It can't reposition a pixel to say half 
way between the pixel "quantum" positions which eventually means that data is 
lost from the image and jaggedness occurs.  Sometimes scalling an image to say 
make it exactly half its size, you actually get less jagged edges because you 
have effectively "resampled the image" to twice its original resolution.  

Some software overcomes the problem by resampling in software the image to a 
higher resolution so that resizing can work with a higher resolution image 
which improves the results of rescaling. 

A similar problems occurs with JPEGs which are a lossy compression. If you look 
very closely at say a straight black line on a white background in a JPEG the 
line actually slightly fades into the backgrount rather than being sharp pixel 
by pixel rendition of the line. When it is resized the compression algorithm 
determines how much fading, what colour and where to make it, so the image can 
lose data without necessarily becoming jagged. The issue with colour shift 
however is called interpolation by adding in dots of a specific colour. This 
process is affected by the Colour Depth of the image so the colour of a single 
pixel is determined generally by either 8 bits of data or perhaps at high 
resolutions by say 32 bits of data. The number of discrete colours that can be 
displayed is far higher with higher colours depths. High resolutions and high 
colour depths result in images taking up lots of bytes, but which are 
ultimately more manageable during resizing.

I do not know about DPSpool but perhaps you should be very accurately checking 
if the size of the image you are printing is exactly the size of the original 
image. If you have to rescale it, instead of letting DP do it, bring the image 
into a good graphics editing package and resize the image and perhaps try 
resampling it at a higher colour depth and resolution.

Regards
Brian


----- Original Message ----- 
  From: Robert Kendall 
  To: Dataperfect Users Discussion Group 
  Sent: Monday, May 19, 2008 1:59 AM
  Subject: [Dataperf] Printing images with DPSpool


  I am using DPSpool to print my letterhead on reports. The letterhead is a 
bitmap image of lines and letters in different colors and sizes composed in 
Microsoft Publisher. When I print the image directly from Publisher the lines 
and letters are sharp and crisp. The same image printed using DPSpool has 
fuzzy, jagged edges and diagonals change from lines to dots, as if the image 
was printing with a lower resolution. I tried using a different laser printer 
but I got the same result. I also tried using a .jpg format but the color 
shifted to a greenish hue and the white background developed ghost images of 
the lines and letters. Is there something I can do to get a better result?


------------------------------------------------------------------------------


  _______________________________________________
  Dataperf mailing list
  [email protected]
  http://lists.dataperfect.nl/mailman/listinfo/dataperf
_______________________________________________
Dataperf mailing list
[email protected]
http://lists.dataperfect.nl/mailman/listinfo/dataperf

Reply via email to