Grid Test of Letters

When I was reading about spacing of various kinds of fonts. I got to know about grid test. It was very little information.

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z 

B C D E F G H I J K L M N O P Q R S T U V W X Y Z A 

AND SO ON...........

This is sample grid test. I wanted to know more about this kind of tests for collections. (proper credits will be given)
1. Could you help me to get more information about grid test? 
2. What other tests do you perform to proof your typeface? What is the purpose of those tests?
3. Can you suggest more reading material on spacing the typefaces? (Books/articles/videos)

Your input would be incredibly valuable to me, Thank you!


  • Grid test (never heard of it) actually doesn't seem very useful to me...
  • John HudsonJohn Hudson Posts: 2,832
    edited February 24
    For reviewing primary (non-kerning) spacing, I find the best test to be between key glyphs that do not usually require kerning. For Latin uppercase, this is usually


    Latin lowercase is a little more complicated because there are no purely symmetrical letter shapes like H, so any glyph might require kerning on one side in combination with at least some others, depending on the design. Further, one ideally wants the key glyph to be something with a counter and not, say, i or l, so that spacing can be considered relatively to width and proportion of the counter. So I typically do two tests:




    with the caveat that when you get to the diagonal letters at the end of the alphabet some of these might kern to the right side of i or n.

    For other scripts, I try to find similar key glyphs. Usually this is possible, but occasionally I do spacing tests between simple uprights as a way to check for optical centering of each glyph:



  • Chris LozosChris Lozos Posts: 1,457
    These are called test strings or as Arnold Bank used to call "necklaces".  My favorite lowercase necklace is "nu" between each lowercase glyph. As John Hudson says, for sidebearing setting.
  • John HudsonJohn Hudson Posts: 2,832
    Heh. It never crossed my mind to use a pair of letters as spacing keys, but I can immediately see the benefit:

  • Now I understand that it is a game of forms and shapes.
  • Chris LozosChris Lozos Posts: 1,457
    the left side of n and right side of u are constants.
  • John HudsonJohn Hudson Posts: 2,832
    The right side of u is often a constant, but because it isn’t symetrical in a serif type, with serif at both top and bottom, there are designs in which uv uw uy might kern.
  • John SavardJohn Savard Posts: 1,068
    edited February 27
    Grid test (never heard of it)

    Possibly he had heard of what he calls "Grid Test" in some foreign language, and his translation of the name in that language into English does not match the English-language technical term in typography for the thing he is talking about.
    It seems to me to be likely that the HAHBHCHDHEHF... test being discussed later on in the thread is a correct guess.
  • I'm not sure if this helps, but here is a site that has various ways of test proofing fonts that I have found useful.

Sign In or Register to comment.