9

In the geometry package, what's the practical use of setting heightrounded=true? The documentation says sth about avoiding cases of "underful vbox", but I don't see why I need to worry about it. When you use the geometry package, do you enable heightrounded, and if so, what's your reason(ing) to do so?

  • 1
    The underfull vboxes that can result mean that the pages have different heights. – egreg Jul 9 '13 at 16:54
  • @egreg so do you generally enable it? or would you generally recommend enabling it? any reason not to ? – nutty about natty Jul 9 '13 at 16:56
  • 1
    Yes, I always use it and recommend using it. If there weren't other reasons, at least for avoiding the underfull vbox messages. – egreg Jul 9 '13 at 16:57
12

Whether or not you use geometry package it's always a good idea to do this. You see equivalent code in size10.clo for the standard classes (unless compatibility with LaTeX2.09 is enabled).

For any computer system it's not usually a good idea to set it to solve a problem subject to unachievable constraints. There is usually no stretchability in the baseline spacing, so if you can not fill a page completely with lines of text, but specify (via \flushbottom) that TeX should aim to achieve that, then its algorithms can not hope to find a good solution. In practice it will just leave the the page a line short and complain about the underfull box, but if you do not mind the page being short you should set \raggedbottom anyway then there is not an issue.

Reasons not to do that: If you absolutely need the baseline of the top and bottom line to be a certain distance apart and you know there will always be additional white space to allow that to happen, then you should set lengths as you need and not allow the system to "fix" them, but this is a pretty rare requirement.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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