Is there a book or a website somewhere that lists the average number of
characters per inch for different typefaces? I need to choose a typeface
that takes up approximately the same amount of space as Palatino, yet
doesn't look like Palatino.

Ken Benson

Ken Benson wrote:
> Is there a book or a website somewhere that lists the average number of
> characters per inch for different typefaces? I need to choose a typeface
> that takes up approximately the same amount of space as Palatino, yet
> doesn't look like Palatino.
> 
> Ken Benson
> 
> 


Hi Ken,

Try here:

http://gtsgraphics.com/cppd/cpp.html

Regards,

Dick

Hi Dick

That's pretty cool, but I was hoping for something that listed numbers at a
consistent size, so I can fairly quickly get four or five choices.

For instance, I've got an old Agfa Compugraphic Type Book, that lists all
the typefaces you could buy for your Compugraphic system in 1986. Each
typeface has a CPP number. The CPP number keys to a chart at the end that
lets you look up the CPP at different sizes. But if I wanted to find a
typeface that would fit about the same amount of copy as CG Palacio
(Compugraphic's version of Palatino), I would look up the CPP number for CG
Palacio (314) and then look for typefaces that had a CPP number around 314.

There must be type reference like this. I'd be happy to buy a book, but my
ideal would be some kind of database of typefaces I could just arrange in
ascending order of CPP.

Ken Benson

Ken Benson wrote:
> Hi Dick
> 
> That's pretty cool, but I was hoping for something that listed numbers at a
> consistent size, so I can fairly quickly get four or five choices.

Ken,

I share your frustration. I found that after you posted, by Googling for 
it. And it was the only such tool I found. However, it is usable. You 
just have to pick several fonts that you have and pick a consistent 
point size across them.

The Adobe site talks about characters per pica as a measure, but it 
doesn't seem to offer the number for the fonts it sells, which seems a 
little odd until you think about it.

> 
> For instance, I've got an old Agfa Compugraphic Type Book, that lists all
> the typefaces you could buy for your Compugraphic system in 1986. Each
> typeface has a CPP number. The CPP number keys to a chart at the end that
> lets you look up the CPP at different sizes. But if I wanted to find a
> typeface that would fit about the same amount of copy as CG Palacio
> (Compugraphic's version of Palatino), I would look up the CPP number for CG
> Palacio (314) and then look for typefaces that had a CPP number around 314.

That was a holdover from what was a standard practice in both foundry 
catalogs (for handset type) and hot metal catalogs, because you were 
dealing with a metric that was not subject to puttering by the 
compositor. Lots of old specimen books provide the information you want.

What happened with the introduction of phototypesetting, is that machine 
makers gave compositors control over tracking and, eventually, optical 
compression and expansion of glyphs. So while there might theoretically 
be a nominal figure for characters per em, in practice one compositor's 
usual settings might be significantly different from another. Print 
buyers discovered that they had to have long, hard conversations with 
comps--at least certain comps--about how many pages their books were 
going to make up.

So I think the font suppliers dropped the practice, to avoid getting 
caught in the middle of those, er, discussions.

How about, in your present circumstance, you pick a couple of fonts that 
are, in terms of period and connotation, in the area you want, check the 
numbers on them, and then see what they look like if you fudge the 
tracking and condense/expand percentage a smidge? Obviously, if you have 
to track something at minus 75 and set it at 85% width, it's going to 
look like crap. But if it's just a matter of nudging the tracking by a 
couple percent and narrowing or expanding the width by one or two 
percent, who's gonna know? I don't mean changing from line to line, 
which can be quite noticeable; I mean changing it for the whole book. 
Give it a shot.


> 
> There must be type reference like this. I'd be happy to buy a book, but my
> ideal would be some kind of database of typefaces I could just arrange in
> ascending order of CPP.
> 

Well, periodically we have a discussion over on alt.binaries.fonts about 
how great it would be if someone would design a public relational 
database that captures all kinds of font attributes (Panose 
characteristics and more), but it's a massive task. A couple of people 
are actually sketching something out along those lines right now. But I 
think we'd all be more comfortable if ATypI would take up the challenge 
and put some money behind it. The foundries might even cooperate and 
upload data to such a database rather than our having to rely on 
volunteer font collectors.

But I don't know of any extant resource. I don't think old hotmetal 
specimen books would be useful, as digital fonts have been redrawn 
pretty radically since those days.

Dick

> What happened with the introduction of phototypesetting, is that machine
> makers gave compositors control over tracking and, eventually, optical
> compression and expansion of glyphs. So while there might theoretically
> be a nominal figure for characters per em, in practice one compositor's
> usual settings might be significantly different from another. Print
> buyers discovered that they had to have long, hard conversations with
> comps--at least certain comps--about how many pages their books were
> going to make up.
>
> So I think the font suppliers dropped the practice, to avoid getting
> caught in the middle of those, er, discussions.

Not necessarily. I think they quit using CPP number because it is not used that much any more. Also CPP number has it weaknesses. It is just a rough estimate and the error may be considerable depending on the number of characters on the line or even on language.

When I started my career as graphic designer at magazine publiher CPP number was a very important tool. The lenght of text had to be calculated and the layout was made with that estimate. After that the text was actually set. The estimate usually did not go wrong very much, just a few lines, but the printing house we used had 10 ten list of miscalculations. The best score was made by my boss, AD of the company, and it was about 1,2 meters of text that did not fit into the layout.

Today deciding what typeface to use on the basis of CPP number is exactly the wrong way to do it and inevitably leads to bad typography, I think. Now it is fairly easy to select any typeface and with a good layout program like InDesign create just as many nice looking and readable pages as necessary.

Jukka

Armadillo wrote:
> 
>> What happened with the introduction of phototypesetting, is that machine
>> makers gave compositors control over tracking and, eventually, optical
>> compression and expansion of glyphs. So while there might theoretically
>> be a nominal figure for characters per em, in practice one compositor's
>> usual settings might be significantly different from another. Print
>> buyers discovered that they had to have long, hard conversations with
>> comps--at least certain comps--about how many pages their books were
>> going to make up.
>>
>> So I think the font suppliers dropped the practice, to avoid getting
>> caught in the middle of those, er, discussions.
> 
> Not necessarily. I think they quit using CPP number because it is not 
> used that much any more. Also CPP number has it weaknesses. It is just a 
> rough estimate and the error may be considerable depending on the number 
> of characters on the line or even on language.
> 
> When I started my career as graphic designer at magazine publiher CPP 
> number was a very important tool. The lenght of text had to be 
> calculated and the layout was made with that estimate. After that the 
> text was actually set. The estimate usually did not go wrong very much, 
> just a few lines, but the printing house we used had 10 ten list of 
> miscalculations. The best score was made by my boss, AD of the company, 
> and it was about 1,2 meters of text that did not fit into the layout.
> 
> Today deciding what typeface to use on the basis of CPP number is 
> exactly the wrong way to do it and inevitably leads to bad typography, I 
> think. Now it is fairly easy to select any typeface and with a good 
> layout program like InDesign create just as many nice looking and 
> readable pages as necessary.
> 
> Jukka


Yes, I think Ken and I both understand that. Nonetheless, I can see how 
a situation can arise, in interacting with a customer, where a handy 
reference guide would be helpful for making a rough cut at font 
selection. It is not always the case, as I'm sure you know, that the 
designer starts with a manuscript and a blank screen. Often there are 
arbitrary constraints imposed by the customer.

> selection. It is not always the case, as I'm sure you know, that the
> designer starts with a manuscript and a blank screen. Often there are
> arbitrary constraints imposed by the customer.

I know, yes, but I still think that with current layout technology relying on CPP number in selecting typeface or font is not the way to do it and I can't imagine what the constraints could be.

Jukka

Armadillo wrote:
> 
>> selection. It is not always the case, as I'm sure you know, that the
>> designer starts with a manuscript and a blank screen. Often there are
>> arbitrary constraints imposed by the customer.
> 
> I know, yes, but I still think that with current layout technology 
> relying on CPP number in selecting typeface or font is not the way to do 
> it and I can't imagine what the constraints could be.
> 
> Jukka


Well, take Ken's original problem. (I actually don't know what Ken's 
problem is in this situation; I'm just creating a hypothetical on the 
basis of what he told us.)

A customer gives you a Word file of a novel that he wants to 
self-publish. He used Palatino, 10/12 and had enough sense to turn on 
hyphenation; he went to the trouble of defining a custom page size and 
measuring the margins in some paperback novels on his bookshelf (too 
narrow, but what does he know?); and the book came out, with allowance 
for frontmatter and a couple of blank leaves, to exactly 320 pp. He's 
already contacted printers and figured out his retail price based on 
that page count and is unwilling to vary from it. He just wants you to 
apply your good typographic knowledge and judgment and use a layout 
program to turn his manuscript into real typeset pages.

Well, if the customer is a reasonable sort of person, you can have a 
discussion about the changes you think are necessary. But some customers 
aren't reasonable, as you know. All you can do in this hypothetical 
situation is talk the customer into using a more traditional book font. 
Otherwise, your hands are tied. Yes, you could just pick a more 
appropriate font at random and play with text attributes until you got 
the right fit. But if you had a list of fonts sorted in characters per 
em order, it would be a matter of seconds, rather than an hour or more, 
to find something appropriate in the right numeric range.

Does this situation come up often? I doubt it. This is the first time 
Ken has brought it up and he's been setting type for quite a while. It 
hasn't come up for me, either. But might it be useful under some 
circumstances? I think the answer has to be yes.

>>> selection. It is not always the case, as I'm sure you know, that the
>>> designer starts with a manuscript and a blank screen. Often there are
>>> arbitrary constraints imposed by the customer.
>>
>> I know, yes, but I still think that with current layout technology
>> relying on CPP number in selecting typeface or font is not the way to do
>> it and I can't imagine what the constraints could be.

> A customer gives you a Word file of a novel that he wants to
> self-publish. He used Palatino, 10/12 and had enough sense to turn on
> hyphenation; he went to the trouble of defining a custom page size and
> measuring the margins in some paperback novels on his bookshelf (too
> narrow, but what does he know?); and the book came out, with allowance
> for frontmatter and a couple of blank leaves, to exactly 320 pp. He's
> already contacted printers and figured out his retail price based on
> that page count and is unwilling to vary from it. He just wants you to
> apply your good typographic knowledge and judgment and use a layout
> program to turn his manuscript into real typeset pages.
>
> Well, if the customer is a reasonable sort of person, you can have a
> discussion about the changes you think are necessary. But some customers
> aren't reasonable, as you know. All you can do in this hypothetical
> situation is talk the customer into using a more traditional book font.
> Otherwise, your hands are tied. Yes, you could just pick a more
> appropriate font at random and play with text attributes until you got
> the right fit. But if you had a list of fonts sorted in characters per
> em order, it would be a matter of seconds, rather than an hour or more,
> to find something appropriate in the right numeric range.
>
> Does this situation come up often? I doubt it. This is the first time

Yes, this is very hypothetical. If I were offered a job like that I would just say "You do not need me". On the other hand this customer probably would not notice any of the typograchic improvements made.

Also using CPP values in this case would fail because same value would not mean same character widths. Therefore for example hyphenation would change inevitably.

Jukka

Armadillo wrote:
> 
>>>> selection. It is not always the case, as I'm sure you know, that the
>>>> designer starts with a manuscript and a blank screen. Often there are
>>>> arbitrary constraints imposed by the customer.
>>>
>>> I know, yes, but I still think that with current layout technology
>>> relying on CPP number in selecting typeface or font is not the way to do
>>> it and I can't imagine what the constraints could be.
> 
>> A customer gives you a Word file of a novel that he wants to
>> self-publish. He used Palatino, 10/12 and had enough sense to turn on
>> hyphenation; he went to the trouble of defining a custom page size and
>> measuring the margins in some paperback novels on his bookshelf (too
>> narrow, but what does he know?); and the book came out, with allowance
>> for frontmatter and a couple of blank leaves, to exactly 320 pp. He's
>> already contacted printers and figured out his retail price based on
>> that page count and is unwilling to vary from it. He just wants you to
>> apply your good typographic knowledge and judgment and use a layout
>> program to turn his manuscript into real typeset pages.
>>
>> Well, if the customer is a reasonable sort of person, you can have a
>> discussion about the changes you think are necessary. But some customers
>> aren't reasonable, as you know. All you can do in this hypothetical
>> situation is talk the customer into using a more traditional book font.
>> Otherwise, your hands are tied. Yes, you could just pick a more
>> appropriate font at random and play with text attributes until you got
>> the right fit. But if you had a list of fonts sorted in characters per
>> em order, it would be a matter of seconds, rather than an hour or more,
>> to find something appropriate in the right numeric range.
>>
>> Does this situation come up often? I doubt it. This is the first time
> 
> Yes, this is very hypothetical. If I were offered a job like that I 
> would just say "You do not need me". On the other hand this customer 
> probably would not notice any of the typograchic improvements made.
> 
> Also using CPP values in this case would fail because same value would 
> not mean same character widths. Therefore for example hyphenation would 
> change inevitably.
> 
> Jukka

I think you are being intentionally obtuse today, Jukka. I said nothing 
about preserving the hyphenation. I said preserving the quantity of text 
in the page count allotted. In any case, there are times when I have the 
luxury of turning down work and there are times when I don't. Perhaps 
you are fortunate enough to have that luxury all the time. In that case, 
more power to you; but you might consider that not everyone shares your 
good fortune.

> I think you are being intentionally obtuse today, Jukka. I said nothing

No, you got it all wrong. My point has been all the time that there is no sense whatsoever using CPP values any more. Hyphenation was just another example.

> luxury of turning down work and there are times when I don't. Perhaps
> you are fortunate enough to have that luxury all the time. In that case,

I can't afford to turn down work, nobody does, but if it were like you described there would not be any money in it. Compared to the amount of work at least. I've learned the hard way that customers that do not understand what designer does are never willing to pay for their work. They pay plumber, they pay grease monkey, they pay personal trainer, whatever, but they do not pay designer. So sometimes it is best just to tighten the belt and say no.

Jukka

Armadillo wrote:
> 
>> I think you are being intentionally obtuse today, Jukka. I said nothing
> 
> No, you got it all wrong. My point has been all the time that there is 
> no sense whatsoever using CPP values any more. Hyphenation was just 
> another example.
> 
>> luxury of turning down work and there are times when I don't. Perhaps
>> you are fortunate enough to have that luxury all the time. In that case,
> 
> I can't afford to turn down work, nobody does, but if it were like you 
> described there would not be any money in it. Compared to the amount of 
> work at least. I've learned the hard way that customers that do not 
> understand what designer does are never willing to pay for their work. 
> They pay plumber, they pay grease monkey, they pay personal trainer, 
> whatever, but they do not pay designer. So sometimes it is best just to 
> tighten the belt and say no.
> 
> Jukka


Ah, I see where the misunderstanding is. You're a designer. I'm just a 
lowly typographer. People don't pay me to be a designer. They pay me to 
create readable books. It's different. I design books for people to 
read, based on the constraints my clients present to me. I don't express 
my personal artistic vision at the client's expense and to the reader's 
rue. So it is natural that you and I would have different ways of seeing 
the situation.

On Wed, 26 Apr 2006 21:20:18 +0300, Dick Margulis <margulisd@comcast.net> wrote:

> Armadillo wrote:
>>
>>> I think you are being intentionally obtuse today, Jukka. I said nothing
>>
>> No, you got it all wrong. My point has been all the time that there is
>> no sense whatsoever using CPP values any more. Hyphenation was just
>> another example.
>>
>>> luxury of turning down work and there are times when I don't. Perhaps
>>> you are fortunate enough to have that luxury all the time. In that case,
>>
>> I can't afford to turn down work, nobody does, but if it were like you
>> described there would not be any money in it. Compared to the amount of
>> work at least. I've learned the hard way that customers that do not
>> understand what designer does are never willing to pay for their work.
>> They pay plumber, they pay grease monkey, they pay personal trainer,
>> whatever, but they do not pay designer. So sometimes it is best just to
>> tighten the belt and say no.
>>
>> Jukka
>
>
> Ah, I see where the misunderstanding is. You're a designer. I'm just a
> lowly typographer. People don't pay me to be a designer. They pay me to
> create readable books. It's different. I design books for people to
> read, based on the constraints my clients present to me. I don't express
> my personal artistic vision at the client's expense and to the reader's
> rue. So it is natural that you and I would have different ways of seeing
> the situation.

Lowly, don't say that. Many professional designers probably can't do what you can. I know because my 'hobby' is to create as good body text as I possibly can. Many of my colleagues often ask me about text typefaces hyphenation settings etc. which they are not that familiar with. Also if a clients do not undrstand what typographer does they are not willing to pay typographer.

Anyway, isn't current programs like InDesign good tools for typographers too?

Jukka

Armadillo wrote:
> On Wed, 26 Apr 2006 21:20:18 +0300, Dick Margulis 
> <margulisd@comcast.net> wrote:
> 
>> Armadillo wrote:
>>>
>>>> I think you are being intentionally obtuse today, Jukka. I said nothing
>>>
>>> No, you got it all wrong. My point has been all the time that there is
>>> no sense whatsoever using CPP values any more. Hyphenation was just
>>> another example.
>>>
>>>> luxury of turning down work and there are times when I don't. Perhaps
>>>> you are fortunate enough to have that luxury all the time. In that 
>>>> case,
>>>
>>> I can't afford to turn down work, nobody does, but if it were like you
>>> described there would not be any money in it. Compared to the amount of
>>> work at least. I've learned the hard way that customers that do not
>>> understand what designer does are never willing to pay for their work.
>>> They pay plumber, they pay grease monkey, they pay personal trainer,
>>> whatever, but they do not pay designer. So sometimes it is best just to
>>> tighten the belt and say no.
>>>
>>> Jukka
>>
>>
>> Ah, I see where the misunderstanding is. You're a designer. I'm just a
>> lowly typographer. People don't pay me to be a designer. They pay me to
>> create readable books. It's different. I design books for people to
>> read, based on the constraints my clients present to me. I don't express
>> my personal artistic vision at the client's expense and to the reader's
>> rue. So it is natural that you and I would have different ways of seeing
>> the situation.
> 
> Lowly, don't say that. Many professional designers probably can't do 
> what you can. I know because my 'hobby' is to create as good body text 
> as I possibly can. Many of my colleagues often ask me about text 
> typefaces hyphenation settings etc. which they are not that familiar 
> with. Also if a clients do not undrstand what typographer does they are 
> not willing to pay typographer.
> 
> Anyway, isn't current programs like InDesign good tools for typographers 
> too?
> 
> Jukka
> 

Yes, I've only recently switched from PageMaker to InDesign, and I'm 
generally pleased with it. So far I've only found a couple of bits of 
nastiness I have to watch out for. In general, though, it's a powerful 
tool. Like a car, though, it's dangerous in the hands of someone who 
doesn't know how to drive it ;-)

"Dick Margulis" <margulisd@comcast.net> wrote in message
news:124uqrmqf5hno07@news.supernews.com...

> A customer gives you a Word file of a novel that he wants to
> self-publish. He used Palatino, 10/12 and had enough sense to turn on
> hyphenation; he went to the trouble of defining a custom page size and
> measuring the margins in some paperback novels on his bookshelf (too
> narrow, but what does he know?); and the book came out, with allowance
> for frontmatter and a couple of blank leaves, to exactly 320 pp. He's
> already contacted printers and figured out his retail price based on
> that page count and is unwilling to vary from it. He just wants you to
> apply your good typographic knowledge and judgment and use a layout
> program to turn his manuscript into real typeset pages.


This is surprisingly close. The client used 11/13 Palatino, typeset his book
in Word, and then decided he wanted it professionally designed. Most of the
typefaces I want to use take up considerably less space than Palatino, and I
don't like stretching type. I can't go bigger than 11 pt, although I may
change the margins.

I finally used my old Compugraphic Type Book to come up with some choices.
ITC Garamond Book is really close, or at least it was in 1980.

Ken Benson