Body text point sizes: whole increments or in between?

[lucadelcarlo]

This topic was imported from the Typophile platform

A question I've never directly posed, waiting for the answer to become apparent over time (still a relative newcomer), is whether there are conventions regarding body text sizing in whole or perhaps half point size increments? In book layouts, I often find myself using the work of others as a template, and the point size of the body sometimes appears to be in half increments or even something odder than quarter point size increments. Do my eyes deceive, and do most designers/typsetters adhere to standardized sizings for body text (10, 11, 12 pt)? At present, I'm tempted to set something in 11.25 Dante MT.

[Té Rowan]

There is nothing sacred about whole-number points. If 11¼ points are the best choice to keep readability up and paper usage down, go for it.

[Joe Pemberton]

Type sizes in modern font menus are just a leftover from a time when type was made of lead and sorted in wooden trays.

[eliason]

If delta hints are hints that trigger adjustments at certain sizes, are there ramifications for delta-hinted fonts of using fractional point sizes?

[aluminum]

The size of individual blocks of text is somewhat arbitrary. 10pt is as valid as 11.23482pt.

What matters more is balance between other typefaces used in the composition. Which, again, can be arbitrary, but needs to fit your aesthetic best judgement.

[quadibloc]

It is true that computers can ask a laser printer to produce type in odd point sizes. However, if type in a document does not seem to have an even point size, another possible cause is that it was set to an exact size... of Didot points.

[brianskywalker]

I like to measure my type in inches.

[rs_donsata]

If there is any value in the point size system is the idea of proportional increments: having a scale relating all type styles in a given design.

Fibonacci, piramidal numer increments (4, 9, 16, 25), modular...

[jacobsievers]

In short: no.

[JamesM]

I agree with the others that it doesn't matter.

The only argument I can think of against it is that odd sizes may make it harder to notice a sizing mistake when you glance at the character palette. If you're setting 10/12, a quick glance at the palette will confirm if you're using the right size, but if you're setting 10.25/11.75, those extra digits may make it easier to overlook a sizing mistake when you're tired or in a hurry. But I don't think that's anything to really worry about, especially since you'll probably be using style sheets anyway.

[Frode Bo Helland]

If delta hints are hints that trigger adjustments at certain sizes, are there ramifications for delta-hinted fonts of using fractional point sizes?

Yep.

[quadibloc]

@rs_donsata:
If there is any value in the point size system is the idea of proportional increments: having a scale relating all type styles in a given design.

Fibonacci, piramidal numer increments (4, 9, 16, 25), modular...

It is true that before the point size system we now use was adopted, other schemes were tried based on different principles.

In 1822, George Bruce proposed a system in which every size of type would be related to the next one smaller by a constant ratio, the sixth root of two:

Pica: 12 points
Small Pica: 10.69 points
Long Primer: 9.52 points
Bourgeois: 8.49 points
Brevier: 7.56 points
Minion: 6.73 points

In 1824, James Fergusson proposed a system in which 14 lines of Nonpareil type was the fundamental unit, and all the other sizes of type were aliquot parts of that unit:

Pica: 12 points
Small Pica: 10 1/2 points
Long Primer: 9 1/3 points
Bourgeois: 8 2/5 points
Brevier: 7 7/11 points
Minion: 7 points

So it is indeed true that other mathematical relationships between the different sizes of type were considered before the simplest system, multiples of a basic unit, was accepted. Of course, that system was the most convenient for allowing a limited number of thicknesses of leading to be used to align different type sizes.

John Savard

[washishu]

When I served my time as a comp. in the 1960s, my boss was an old guy aged 63. He knew the old point sizes and would often refer to primer (he pronounced it primma), long primer and minion.

Large woodletter type, above 72 pt was referred to as, for example, 12 line. This would be 144 pt or 12 picas.

Our standard range of point sizes for most faces went 6—8—10—12—14—18—24—30—36—48—60—72. For a few faces we also had intermediate sizes, usually 9 and 11 pt. Mostly, the size range was dictated by the foundry, for founders' type and, of course, the needs of the work in hand.

Odd but somehow I still feel very slightly uncomfortable using a size outside of this range. I think it's because I became so familiar with these sizes that I know exactly what I'm getting so to speak and going outside this feels slightly anarchic. Weird.

[PabloImpallari]

10, 11, 12 pt does not means nothing really, it all depend on the Vertical metrics of each font and how they where set up by the Font Designer.

[hrant]

True, but if somebody (like probably "washishu") is used to
a "feeling" given by a specific font at specific sizes, that's an
instinctive dimension worth leveraging.

hhp

[Bert Vanderveen]

Anything that works for you and the project!

I have used type defined to a thousandth after the full value — because it was what the design warranted.

I need to add that using ‘full sizes' makes your life much easier, because who can remember a number with three digits after the comma?

[washishu]

Something of an aside but, as I sit here wearing my reading glasses I also recall how that same boss would sometimes hand me a piece of type he was working with and say "What's this?" I would identify the character as, maybe, "A 6pt Bodoni Bold lowercase i", meanwhile thinking "You mean you can't see that? Isn't it obvious?" Now that I am the age he was then, I know how he felt.

[Renaissance Man]

I use WordPerfect. Fonts can be sized in tenths of a point.

Last time I used Word, it sized fonts only in half-points.

If you want to fit type to to fill a page, you need better than full or half-point sizes.

The "best point size" is akin to "best religion." Or do I repeat myself?

[oldnick]

I used to use WordPerfect exclusively, until Corel bought them and royally screwed the program up. As a former dedicated-system typesetter, I liked WP's direct access to the underlying code. Corel's latter-day approach appears to be making their programs idiot-proof, which in turn seems to guarantee that only idiots will use them.

[hrant]

Which of course ties in to idiocy being held up as an ideal these days.

hhp

[Renaissance Man]

@ idiocy being held up as an ideal
but the real ideal is.... whole or fractional/decimal point sizes?

[dberlowgone]

The ideal is what looks right. With x hts and cap hts not uniformly sized to the em, people who size text for newspapers and magazines looking for just the right size almost always end up in tenths of points. That's what I've seen.

Also:
"If delta hints are hints that trigger adjustments at certain sizes, are there ramifications for delta-hinted fonts of using fractional point sizes?"

Delta hints trigger adjustments at certain pixel per em sizes, not at point sizes, so no, fractional point sizes don't ramify, as such a size is rounded to a whole pixel value by the time deltas strike.

And, when you're dealing with screen fonts, unless the per inch resolution of the device is a multiple of 72, almost none of the sizes one sees or selects actually are whole numbers if considered in points. E.g. a 10 point font on a 72 dpi device is 10 pixels. One point is one pixel. But a 96 dpi device has 1.333 points per pixel. So at 10 point, 10 x 1.333 = 13.33. Since this rounds to 13 pixels, what you see is really 9.7 point, or whatever.