Yours is a tell me everything I need to know in twenty five words or less question - the simple answer is get a copy of Logo Font and Lettering Bible by Leslie Cabarga - study it and then come back with any questions. FontLab or Fontographer is probably what you are going to need, and also ScanFont - all from FontLab.

Sorry, didn't mean to come across that way, just full of questions, trying to provide as much context as possible. I'll take a look at that book.

Note that if you stick to tiny straight line segments only, you can go much further in terms of texture detail.

hhp

Is there any rule of them for determining if a typeface is too detailed to vectorize?

I don't think that's so much of an issue these days.
However, you might experiment by acquiring some fonts that have complex glyph outlines and testing them on various devices.
Or do your own tests, by seeing how much performance difference is made by variations with different numbers of BCPs.

Thanks hrant - upon vectorization is there a option for converting or simplifying curves to smaller straight line segments?

Nick - That's good to know. How would one go about testing performance? Also, what are BCPs?

Bezier Control Points, the thingies by which font outlines are defined...

ah, got it thanks.

I would recommend using ScanFont in conjunction with FontLab.

I would recommend using ScanFont in conjunction with FontLab.

I second that motion. It takes a lot less time to clean after Scanfont/Fontlab than it does to clean up after Illustrator. And of course the automatic letter separation trick saves oodles of time.

This might get you started:

http://ilovetypography.com/2007/10/22/so-you-want-to-create-a-font-part-1/

Thanks for all the recommendations and feedback everyone. I'll keep everyone posted on my progress.

I have a question regarding TypeTool vs FontLab.

For what I am doing, it TypeTool robust enough, or am I going to hit roadblocks getting the font tuned to a high enough quality? Where is FontLab going to do for me that Typetool won't?

I guess I'm just hesitating on laying down $650 for the full program if I don't need it.

It depends. You wrote I work with a very talented calligrapher. If your calligrapher writes alternate versions for various characters which later should be accessed through OpenType features (Stylistic Sets, Contextual Alternates, Ligatures ...), it can't be done in TypeTool alone, because TypeTool can not add OpenType features to a font.

Up top you wrote:
> I'd like to create an online site that would sell his fonts.

This is clearly not a one-off project, so I'd say the full shebang is called for.

hhp

Thanks everyone for the advice.

Is fontforge a valid option, or am I just going to be hitting bugs all the time? I've read through the tutorial and it seems fairly straight forward.

I am on a mac - so I know that makes things easier as far as installation goes.

More iHype.

See creepy video here: http://www.bbc.co.uk/news/business-13416598

hhp

Is fontforge a valid option…

Only if your time is worth nothing. FontForge is like Fontlab, but the interface is twice as confusing and twice as ugly. And the program itself seems to be twice as unstable.

I'm getting the feeling I might just have put the money out. I'm thinking I might still give fontforge a try - at least until it drives me crazy.

If you're curious - here is a look at the different calligraphic options I'm assessing - http://typophile.com/node/82327

There is no clear way to evaluate the amount of time it will take to digitize these fonts if you're doing it yourself from the start.

I agree that FontLab is your best bet. Like with anything though, it really depends on how much you're willing to invest in time and energy vs. what you want in outcome. Calligraphers typically write and connect letters in different ways depending on the context. To capture all the nuance and subtlety accurately would require going the longer route of learning the program well and doing the fonts in OpenType. Getting feedback from your calligrapher along the way would also be helpful.

Maybe it would be worth your while to find someone in the Chicago area that does font development to work with on the project.

Those are good thoughts Stephen. I'll definitely be working closely with the calligrapher. I might need to get some help as well.

I know opentype provides some amazing possibilities, depending on the specific font, I might need to look into that. It seems like with alternates combined with kerning pairs, it could easily get crazy in the workload. I do want to make sure it's a quality product though.

James: FontForge is like Fontlab, [...] the program itself seems to be twice as unstable.

Although I don’t really see it as my task to defend FontForge, I actually think that this statement is a bit unfair. Programs may have overlapping functionality, but this does not mean by definition that the applied (quality of) technology is always identical. For instance the auto-hinter for TrueType in FontForge is one of the best I have seen so far. As you know, the program is also used for making the conversion to the Cufón format (locally or on the web).

Perhaps it is not too simple to get FontForge running under X11 on the Mac or under Cygwin on the PC, but it installs very easy on, and runs solidly under Linux-distros, like for instance Ubuntu (I am actually impressed by version 11.x). Furthermore it is conveniently scriptable and it is for free (as are the Linux-distros).

So far the advertisement for a competing product.

Josh: Ideally, I'd like to vectorize each letter form [...]

This requires a program that converts the scanned letters into single glyph databases in batch, I reckon. In DTL’s TraceMaster one can select a (theoretically unlimited) number of tiff’s and these will be converted into one font file, as shown in this small YouTube movie.

So far the advertisement for one of our own products.

FEB