2

If I have two virtually similar files:

  1. 1200DPI Bitmap TIFF, placed in InDesign with 100%K
  2. 1200DPI CMYK TIFF, same art as #1, lineart thresholded and now at 100%K

And both are on white backgrounds (so no need for the benefits of Bitmap having transparency), would there actually be a difference in submitting to the printer the bitmap file, rather than the CMYK file? Why does every printer suggest using bitmap files? Do bitmap files print differently than CMYK files, even though both have the same amount of pixels and are set up the exact same way?

The only reason I can see using a bitmap is that it's a smaller file, but also (I guess this is my second question), why are bitmap files smaller than the CMYK file?

My coworker says that something about Bitmap files make them print better (as in the lineart is crisper), and that's why everyone uses it, but I can't see there being any difference.

5
  • No one in the printing industry cares about files sizes (kb). The size of a file (kb) in printing is not a factor to be concerned about. Small or large, files are as big (kb) as they need to be to meet quality standards.
    – Scott
    Mar 26, 2022 at 2:02
  • Do you realize, that a CMYK tiff file is also a bitmap? I think you are referring to a 1-bit image on your first case.
    – Rafael
    Mar 26, 2022 at 3:07
  • @Rafael, I agree that Bitmap is an ambiguous term, but it is nonetheless what Adobe calls 1-bit images.
    – Wolff
    Mar 26, 2022 at 9:59
  • Facepalm... Those guys of adobe... They should come to graphicdesign.stackexchange.com from time to time...
    – Rafael
    Mar 26, 2022 at 16:41
  • @Rafael It’s not just Adobe. It’s perfectly commonplace to call 1-bit mappings ‘bitmaps’ and multi-bit ones ‘pixmaps’. Mar 26, 2022 at 22:34

1 Answer 1

7

Normally you would compare a 1200 PPI 1-bit bitmap image with a 300 PPI 8-bit (or more) grayscale image (or a CMYK image with 100%K which is basically the same except the file size is larger).

The 1-bit image only has black and white pixels so it has to have a large resolution to be able to print without the individual pixels being visible. The 8-bit image has 256 tones which can be used to make anti-aliasing, so it doesn't need as high resolution to give the illusion of smooth lines without visible pixels.

A printing device can only print 1-bit. Either you apply ink or you don't. So all tints of the pure ink must be made using some form of pattern. In offset you commonly use halftone screening and in digital print you often see stochastic screening.

Let's compare how a 1200 PPI bitmap image and a 300 PPI grayscale image with anti-aliasing look on offset print.

If the images look like this (each image is around 5×5 mm):

They could look something like this on offset print:

Click images to see full size.

The grayscale version gets halftone dots along all the anti-aliased edges and seem a little blurred. The bitmap version gives a superior sharp result comparable to vector graphics.

You probably know this already, so let's get to the point about using a thresholded 8-bit image instead of a 1-bit image. In theory, you should be able to get the same result. But in practice, there are many things that could go wrong (meaning that unwanted screening would be introduced).

  • Scaling or rotating the image in Adobe InDesign or Illustrator shouldn't introduce anti-aliasing because they just scale and rotate the pixels as they are, but I can't say if this goes for all layout applications. If a layout application actually interpolates the image, anti-aliasing might be introduced.

  • When exporting a PDF, an 8-bit image ends up in the Color Images or Grayscale Images category and might be downsampled to 300 PPI (or whatever resolution you choose). Downsampling would introduce anti-aliasing, so none of your color or grayscale images can be downsampled on export.

  • Likewise, your color or grayscale images can't have JPEG compression (or any other lossy compression), because it will introduce artifacts that will end up being screened.

  • The RIP (raster image processor) used by the print house might by default scale color and grayscale images down to 300 PPI to save resources. Bitmap images would be unaffected, but the RIP doesn't recognize your "fake" bitmap images.

  • The RIP makes automatic trapping which behaves a little differently on 1-bit and 8-bit images. Some RIPs might even not do trapping on 8-bit images, but only on 1-bit images. In your case with purely black and white line art, it wouldn't make any difference, but if you use multiple inks, it might.

  • There might also be an issue with rounding caused by the page being placed at some coordinates on the printing plate and your image being placed at some coordinates on the page that doesn't fit the grid of dots on the plate. A bitmap image will just be placed at the nearest gridline, but some RIPs might introduce anti-aliasing. I'm a little unsure about this, but I've seen a halftone dot here and there on images where there shouldn't be any and I can't explain it any other way.

Finally a short explanation to why bitmap images takes up less disk space than a similar looking CMYK image. Bitmap images only use 1-bit per pixel meaning it can only have black or white pixels. The CMYK image on the other hand use at least 8-bits per CMYK channel (256 steps for each channel) which totals in 32 bits per pixel (4 million different possibilities). So in a totally naïve and uncompressed implementation a black pixel would be saved as 1 in a bitmap image and 00000000 00000000 00000000 11111111 in a CMYK image. Besides that, the CMYK image might also have a color profile embedded.

1
  • A compression if 1:32
    – joojaa
    Mar 26, 2022 at 9:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.