Hello, all! I have a rather broad question. Let me start by saying that I am not a type designer by trade, nor do I have any formal training in type design. I'm a graphic designer who fell in love with type years ago and has been drawing it ever since. I bought FontLab shortly after my infatuation began, but, due to the combination of my perfectionist approach to everything and FontLab's rather steep learning curve, I've yet to produce any fonts other than those for my own personal use on various projects.
I have a handful of typefaces fully drawn that I'd love to turn into full-fledged, high-quality fonts that could then be sold either on my own website or through a distributor. They were drawn in Illustrator, and then I imported the glyphs into FontLab. Now I'm stuck on hinting and kerning, and no matter how carefully I read the FontLab manual, I still end up feeling out of my depth and with no clue where to start. Does anyone have any tips or resources for a novice like myself?
Related question: As most of these are display typefaces meant for use in graphic design applications, how in-depth do I realistically need to get with all of this to make the fonts market-ready?
Thank you in advance for indulging me!
0
Comments
If you want to start out without feeling overwhelmed, start kerning the large basic faults: (Cap-lowercase comps, Y-A, etc) After the basic combinations run some type tests/rivers and see what jars. Don't try to fix it all in one go.
If most are display types, I would drop the hinting currently since it might not even add any value (or perhaps will even make the overriding-hinting worse than current auto-hinting does)
Perhaps even retina will be in full spread by the time you release it.
But do remember you will spend 75% of the design time on perfectionism though, but that's the life of designing type!
;-) :-) :-D
Laughed from iPad
Kerning is a correction on the fitting, which is sometimes required because the harmonic systems of the Latin script are inconsistent. I tend to disagree a bit with James when it comes to ‘A poorly drawn, well-spaced design is much more useful than a beautifully drawn, poorly spaced one.’, because I believe that harmony and rhythm in type are inextricably connected with each other. One never gets the spacing completely right if there are inconsistencies in the design pattern.
We didn’t see anything of your design so far on TypeDrawers. If you are interested we could do a test here with artificially spacing your type. One way would be to calculate the space between the letters, which is based on the type-foundry practice since roughly the seventeenth century. We use the IKARUS-based Kernus program for this. Another option –if applicable for your design– would be to distill the underlying rhythmical pattern from your type and use this for the fitting. This is done with Kernagic (with a newer version then is currently available at http://www.lettermodel.org). This approach is based on what I distilled from Renaissance type.
In both cases the actual calculation is done in a split second, the conversion of the formats will probably take a couple of minutes. Interested?
Best, Frank
To illustrate my point, I just ruined the spacing of DTL Fell (under development since 1997) by setting the side bearings to zero.
Subsequently I applied Kernus and Kernagic. Kernagic wins here because it understands the pattern and recognizes the stems and curves. Also the individual widths are very easy to tweak on the fly (not done here though).
Perhaps you will see things that can be improved. We see them too, but this makes a great starting point for editing the fitting, and in case of Kernagic also the complete design pattern.
We can generate the kerning in a jiffy too.
Sorry, just couldn’t resist.
However, if a design clearly deviates from the archetypal models, like Comic Sans, then overall Kernus seems to do a better job. Trying to achieve an equilibrium of white space by (basically) purely calculating this, seems to make more sense in this case than trying to distill (and subsequently apply) the rhythmical pattern.
Similarly, Georgia performs incredibly well without any kerning pairs whatsoever.
James: ‘Frank, please stay on-topic. If you want to start a thread about the merits of DTL’s auto-spacing software start a new thread in the software category.’
¿Qué? What is exactly off-topic about my posts addressing the backgrounds of spacing translated into artificial stuff if someone asks for advice concerning spacing? And Kernagic isn’t DTL software, it’s Open Source and it’s for free too! And it supports UFO! WOW!
Further, the "optical kerning" in Adobe apps starts with the initial spacing of some key letters such as O and H. If they are "wrong" then you make optical kerning fail dramatically.
Plus, kerning every possible combination would be vastly more work! Kerning is a correction for unusual shape interactions, not an all-the-time thing.
So you really need to get the spacing right.
I will second the motion that you could work with all your fonts and get them perfectly spaced and pretty much ready to release before you start kerning any of them. That is way more sensible than anything else.
The reason is, if you change the spacing after you've started kerning, you will have to throw out some or all of your kerning. That's potentially many hours of work lost. Sometimes repeatedly!
I have never heard any type designer ever in my life say "I started kerning that font too late." But I have heard (and myself said) the opposite more times than I can count. I have frequently called it out as the #1 most common beginner mistake in type design.
(There is useful information in the whole "book")