Didn't you mean *increasing* the x-height?

I think Beorcana has a couple of different optical sizes.
And, you might want to look at Tim Ahrens's book if you haven't already.

Josh Darden’s Freight family comes in optical versions. I’m pretty sure that Matthew Carter’s proprietary version of Franklin Gothic for the Museum of Modern Art has optical sizes.

And I’m pretty sure that you should be raising the x-height for smaller sizes!

I was thinking my grotesk would benefit from longer asc/descenders, as they are quite low. Why would you increase the x-height?

I knew about Freight, but Freight Sans doesn't – does it?

"Why would you increase the x-height?"
Larger counters. You need more light inside and around the black* in smaller sizes. This should be the case in pretty much any old metal face (and/or do get that book).
* Which typically also gets heavier.

This might be interesting, or maybe not:
http://typophile.com/node/56000

I'll read! Thanks Nina. Asking silly questions benefits the learning process :) I'm considering the need for more space, though I'm also considering the need for more distinct wordshapes.

I knew about Freight, but Freight Sans doesn’t – does it?

Sorry, you are correct.

FF Absara has text and display versions that vary in this way. While FF Meta Headline differs from the original by its tighter spacing and retracting of tails for a more solid fit. Cronos does optical sizes in yet a different way, adding more detail and contrast as it gets larger.

Following your link, Stephen, the "n" illustration for Rumba does a lot of what I'm hoping to do:

Many metal typefaces had an increased x-height in smaller sizes. In many of the examples I have studied the variation of the descender length is literally the only change that was made to the letter shapes.
The interesting thing is that in some cases the descenders are longer in large text sizes than they are in very small fonts but they get shorter again in the display sizes, having a maximum at around 14 pt or so. This makes sense, of course: making the x-height as big as possible in small sizes on one hand, and making the vertical proportions more "compact" in the display sizes to allow for tighter headline setting.

Given that adjusting the descender height is not much effort, especially in the digital world, I think including different options would make sense and give the typographer more flexibility – you may want super-tight newspaper-style headlines, or an airy layout with elegant long extenders and more generous leading. I think using an OpenType feature such as titling alternates would make things easier, you wouldn't need separate fonts.

What good does just lowering the asc/descenders? That doesn't automatically give larger counters than you get by not lowering the asc/desc. It's relative to point size, which again is a rather vauge term. The only change will be the option to set lines tighter.

The difference in descender lengths may have more to do with conforming to Standard Alignment than it does to optical sizing.

Kent: I thought that at first, but then later it occurred to me that Standard Alignment might have been created with optical size considerations in mind, in the first place....

T

What good does just lowering the asc/descenders?

Perhaps the shorter ascenders and descenders serve, like the heavier strokes, to darken the color of the smaller sizes on the page?

Thomas -- If I'm not mistaken, Standard Alignment is basically Point Line (Updike relates them thus), which was established to allow types of different sizes to be aligned using 1-point leads or their multiples. I don't think optimizing proportions for readability at differing sizes played any part.

Let me amend that.

I believe what Tim describes as the broader trends in descender proportions from very small sizes to text sizes to display sizes probably are indeed related to size-relevant considerations. And I agree with the probable reasons that Tim gives.

I cite Standard Line as an explanation for any micro trends of descender variation between proximate sizes.

Okay, I can buy the second post. :)

To clarify my own previous comment: the descender/ascender proportions established in Standard Alignment were likely based on the proportions in common use at the time for the various sizes in question, and those in turn would have generally reflected optical size variation. Folks could have invented the standard alignment numbers based on measurements of actual type, and hence reflected pre-existing optical size considerations, and promoted future optical size proportions, yet that need not have been a deliberate objective as such.

T

*Maybe*. On the other hand, there were those who decried Standard Alignment when it was first adopted.

Updike states the case thus: "There are mechanical advantages inherent in this scheme, but they are attained at the expense of the correct proportion of certain letters. . . . This disregard for proportionate length of descenders of types of different sizes pervades the whole scheme; and consequently no size of type (excepting possibly 7-point and 11-point) which is cast on Standard Line can have descenders of adequate length." [D. B. Updike, Printing Types Vol 1, p.35. Cambridge: Belknap Press, 1922]

When Updike starts ranting, you have to take it with a grain of salt; but this makes me question just how much Standard Line conformed to traditionally accepted optimum proportions for sizes.

To provide context for others, here is a diagram of the Standard Alignment/Point Line scheme:

 
And a link to a previous discussion.

Frode -- Sorry, Thomas and I have taken this far afield of your original query.

Don't worry Kent! I just wish you'd explain with simpler words, as I'm having a hard time following you.

Sorry, Frode, I know I can be rather verbose and often use complex sentence structures. I'll try to get simpler.

I apologize to non-native speakers. If it's any consolation, I know even native English speakers can find me hard to follow. ;-)

Please feel free to ask for clarifications, here and elsewhere. (But I'm going out of town and won't be online for a few days.)